E acm (2) - Pondicherry University DSpace Portal

Jawaharlal Institute of Postgraduate Medical Education and ResearchMinistry of Health and Family Welfare (Government of India)Pondicherry, 605 006, INDIACERTIFICATEThis is to certify that the thesis entitled uClinical Prevalence, Identification andMolecular Characterization of Aminoglycoside Resistant Enterococci" submitted byP. Vittal Prakash is a bonafide record of original research work done by thecandidate during the study period under my supervision. The work was planned,organized and executed by the candidate in the Department of Microbiology,Jawaharlal Institute of Postgraduate Medical Education and Research [JIPMER].Pondicherry. This study has not previously formed the basis for award of anydcgree, diploma, associate ship, fellowship or other similar title.I further certify that entire thesis represents the independent work ofP. Vittal Prakash, and all the Microbiological and Molecular biologicaltechniques employed in this work were actually undertaken by the candidatehimself under my supervision and guidance.Prof. R. SAMBASIVA RAO, MD.,Former Director, JIPMERPresently Vice-ChancellorNTR University of Health SciencesVijayawada, Andhra Pradesh, IndiaVICE-CHANCELLORd, T, R. Uniwity Of Health ScimVIJAYAWADA-62000&*P.b

Jawaharlal Institute of Postgraduate Medical Education and ResearchMinistry of Health and Family Welfare (Government of India)Pondicherry, 605 006, INDlADECLARATION BY THE CANDIDATEI hereby declare that the thesis entitled "Clinical Prevalence, Identification andMolecular Characterization of Aminoglycoside Resistant Enterococcin submitted tothe Pondicheny University for the fulfillment of the requirement for the award ofdegree of Doctor of Philosophy in Medical Microbiology, is a bonafide record ofthe original research work carried out by me at Jawaharlal Institute ofPostgraduate Medical Education and Research [JIPMER], under the supervisionand guidance of Prof. R. SAMBASIVA RAO, MD., Former Director, JIPMER,Pondicheny.This study represents the independent work conducted by me and has notformed the basis for award of any degree, diploma, associateship, fellowship orother similar title.P ddzz 'Tab&P. VITTAL PRAKASHResearch ScholarDepartment of MicrobiologyJIPMERPondicheny, lndia

ACKNOWLEDGEMENTS1 express my deep sense of gratitude to my mentor and supervisorProf. R. Sambasiva Rao M.D., Former Director, JIPMER for his assiduous guidanceand incessant support, and encouragement over the years of my research work. Being afatherly figure he would always remain a source of inspiration for me in every aspect andwalk of life.I owe my profound thanks and appreciation to Dr. Subash Chandra Parija,M.D., Ph.D., Director-Professor and Head of the Department of Microbiology, JIPMERfor his unconditional support, simplistic approach and the belief he had in me throughoutthis study, for which I am indebted.I am grateful to the Present Director, Medical Superintendent and Dean,JIPMER, for providing institutional facilities and permitting me to carryout this work.I owe my sincere thanks to Dr. Jayaehandran, Professor and Head, Departmentof Biotechnology, Pondicheny University for his valuable suggestions and critical reviewof my work proceedings being a doctoral committee member, all through this study.I express my heartfelt thanks to Dr. B.N. Harish, Professor of Microbiology forhis valuable suggestions, and Dr. S. Sujatha. Associate Professor of Microbiology,JIPMER for the academic gestures and warmth she showed me all through this study.I am highly thankful to all the former faculty members of the Department ofMicrobiology, JIPMER. I am deeply indebted to Dr. Reba Kanungo, for her pricelesscompassion and persistence in guiding this work to fruition. My sincere thanks toDr. Badrinath, Former Medical Superintendent, Dr. Vijayalakshmi, & Mr. Natarajanfor their solicitous and invaluable help during the initial stages of my work.I wish to express my deep sense of gratitude to Dr. Richard.R.Facklam, Centerfor Disease Control and prevention [CDC], Atlanta, Georgia, U.S.A for his generosity toprovide me with the CDC reference strains of Enterococcus spp., and several reprints ofhis works. Many academic gestures rendered by him readily in the form of much neededdiscussions and suggestions at various stages of this study were highly motivating, whichkindled my passion for Enterococcus research since the very beginning of this study.1 owe my sincere thanks to Dr. Lalji Singh, Director, Center for Cellular andMolecular and Biology [CCMB], Hyderabad for permitting me to cany a part ofMolecular biological studies included in this study. '

1 am greatly indebted to Dr. Malay Ray for his authentic guidance,encouragement and hospitality he showed during my stay in CCMB that is imperishable.It was truly a learning and rewarding experience in his laboratory with his lab members,which would be nostalgic forever.I am grateful to Dr. Yasuyoshi Ike and Dr. Koichi Tanimoto, Gunma UniversityGraduate School of Medicine, Japan for providing me the standard genetic strains ofEnterococcus for gene transfer assays, and valuable suggestions during the initial stagesof my work to start with.My sincere thanks to Dr. Nathan Shankar, Dr. Arto Baghdayan andDr. Michael Gilmore, University of Oklahoma Health Sciences Center, U.S.A. for theirgenerosity to provide me with the control strains for "esp" genotypic assays, as wellpractical suggestions and reprints of their work.I express my deep sense of gratitude to Dr. Neil Woodford, Central Public healthlaboratories [CPHL], London for lending his expertise in the form of discussions andsuggestions to interpret DNA hybridization results during the final stages of this study.It was my fortune and pleasure to be amidst the company of Dr. Malay's labmembers in CCMB: Ajit, Raj, Reddy, Pavan, Bhuban, Suman, Shaheen, Manjula, andother mates Vivek, Kasturi, Asish, Kiran, Prabhakar, Meenal, RamPrasad and manymore. The assistance by Ajit and Raj to perform PFGE and DNA-hybridizations and theirhospitality made me feel ecstatic and indebted, which I'd cherish forever.I was fortunate to have a reliable and unsolicited companionship in every aspectduring my Ph.D journey in my colleague Sowmya Ranjan Sahu for which 1 am indebted.1 owe my special thanks to Swama and Madhulika for all their priceless gestures duringthis study. My junior colleagues Krishna, Vaidya, Chandrashekar and Smitha deserve myappreciation and acknowledgement for being kind in their own way.I would like to express special words of thanks to the past and present residents ofour department Dr. Prabhakar, Rakhi, Annie, Sanjay, Sheela, Judy, Kokila, Venkatesh,Surya, Murali, Siva Sangeetha and Uma for their solicitous assistance and friendship.My owe my earnest thanks to my Seniors: Dr. Prashanth for his invaluablesuggestions, help and encouragement all through, and Dr. Geethalatha, Dr. Rajalakshmiand Dr. Raghava for their incessant support in their own way.

I take this opportunity to express my sincere thanks to Dr. Adithan, Professor ofPharmacology, Dr. Neigi, Associate Professor of Medicine, JIPMER and Dr. S.L. Hoti,Deputy Director, Vector Control Research Ccnter, Pondicherry for their assistance andsupport during different stages of this study.I would like to thank Dr. Ajay, Dept. of P & SM, for his solicitous assistance instatistical analysis of the data, and Selvaraj and Vinayagamoorthi, Ph.D Scholars, Dept.of Biochemistry, JIPMER for their unreserved assistance all through this study.My sincere acknowledgments to Mr. Palani, Storekeeper, for his immensecooperation and Mr. Natarajan, Stenographer and all the past and present technical staffof the Department of Microbiology for their invaluable help and support all through.1 am indebted to my friends Thangadurai and Narasimha Rao, PhD Scholars, andother friends and colleagues for their love, suppon and inspiration they have showed meall through that made my stay in JIPMER a memorable and valuable experience.1 am enthralled to express my heartfelt thanks to Dr. J. Muralishwar Rao, Dept. ofRadio diagnosis for his unconditional love, passion and support which was impulsive anddrove me through these years in JIPMER, for which 1 am eternally indebted.I take this opportunity to express my deep gratitude and reverence for all myteachers in school and college over these decades. I render my special thanks to Mr.Muniaraj whose teachings provoked interest in Microbiology during my undergraduation.I express my earnest thanks to my very special friends in school and college VijayBhaskar, Karthi, Shiva, Sabu, Rajesh, Namachi, Ramya. Vino, Geetha, Saravanan, Sushi,Pavani, Ammu and many more who had adored, motivated and inspired me in one way orthe other, for which 1 am deeply indebted forever. My heartfelt homage to MukeshKumar Singh [Late] and Deepak [Late] who are no-more to see their wish coming true.1 profoundly express my deepest gratitude & appreciation to my family membersSisters Praneetha, Bhargavi, uncle, aunt and grandparents for all their all love, affection,and innumerable sacrifices over these decades, which makes me privileged and indebted.Lastly I owe and dedicate everything to my PARENTS for their unfathomablelove, sacrifice, motivation & patience over these decades to make me what I am today. Itwould be my privilege to emulate them in every aspect of my life for which I'm indebted.P. Vittal Prakash

~edicatedTo NyBeloved Parents

LIST OF TABLESTable 1 .Table 2 .Table 3 .Table 4 .Table 5 .Table 6 .Table 7 .Table 8 .Table 9 .Table 10 a .Table 10 b .Table 10 c .Table 10 d .Table 10 e .Table I1 a .Table 11 b .Table 11 c .Table 12 .Table 13 a .Table 13 b .Table 14 .Table15 .Phenotypic tests used for identification of Enterococcus species .......... 27Susceptibility profiles of genes that mediate resistance toaminoglycoside synergism in enterococci ..................................... 62Characteristics of the types of resistance to glycopeptide antibioticsfound in enterococci ............... . ...................................... 64NCCLS guidelines for disk diffusion and dilutionsusceptibility testing for Enterococcus species ............................... 67Screening methods for detecting vancomycin and high-levelaminoglycoside resistance in enterococci ..................................... 68Prevalence of different species of enterococci ................................ 94Prevalence of Enterococci among various clinical specimens .............. 95Distribution of Enterococnrs species amongdifferent clinical specimens ...................................................... 95Conventional biochemical Phenotyping results of enterococci ............. 97Antibiotic susceptibility pattern of enterococciby Kirby-Bauer disc-diffusion method ........................................ 114Antibiotic susceptibility pattern of urinary enterococci ..................... 114Antibiotic susceptibility pattern of E .. faecalis and E . faecium ............. 115Antibiotic susceptibility pattern of unusual Enterococcus species ......... I16Antibiotic susceptibility pattern of unusual En!erococcus species ........ 116MIC testing results of enterococci byagar dilution agarlscreening method .......................................... 117MIC testing results of E..faecalis and E.faecium ........................... 119MIC testing results of unusual species of enterococci ....................... 120Multiplex-PCR results for HLAR enterococci .............................. 121Plasmid REA and DNA hybridization results ................................ 151Plasmid REA and DNA hybridization results ............................... 152Prevalence of Virulence factors among Enterococci ....................... 164Distribution of Virulence factors among various clinical specimens ..... 168

Table 16 . Clumping response of E . faecalis ............................................. 186Table 17 . Filter-Mating experiments of HLGR-PCR positive E.fuecalis. ........... 187Table 18 a . Molecular typing results by PFGE and Plasmid REA ...................... 216Table 18 b . Molecular typing results by PFGE and Plasmid REA ...................... 217Table 19 . Statistical analysis-Simpson's index of diversityand Confidence interval ......................................................... 218

LIST OF FIGURESFigure 1 .Figure 2 .Figure 3 .Figure 4 .Figure 5 .Figure 6 .Figure 7 .Figure 8 .Figure 9 .Figure 10 .Figure 11 .Figure 12.Figure 13 .Figure 14 .Figure 15 .Graph 1 .Phylogenetic relationship of Enterococcus species based oncomparative analysis of 16s rDNA sequences ................................ I0Pheromone-responsive conjugative system of E . fuc~colis ................... 39Antimicrobial-drug regimens for the treatment of drug resistantenterococcal infections ............................................................ 49Schematic diagram of the mechanism of resistance to vancomycin ........ 64Computational Cluster analysis of WCP gels of atypicalenterococcal strains .............................................................. 100Multiplex-PCR results of HLAR strains of Enterococcus .................. 122Whole Plasmid DNA analysis of HLAR enterococci ....................... 147Plasmid-restriction endonuclease analysis andSouthern hybridization of HLAR enterococci ............................... 148Screening representative strains of Enterococcus forBacteriocin production ........................................................... 166"Esp" PCR results of representative test strains of Enterococcus ........ 166Molecular Characterization of E . fhecalis donors, recipientsand transconjugants .............................................................. 189Fluorescent and Phase contrast Microscopy images of Clumpingand Mating Assays of Enterococcus .......................................... 195SmuI macrorestriction analysis of HLAR Enterococci by PFGE ......... 209Computational Cluster analysis of PFGE gels of HLAR enterococci .... 211Computational Cluster analysis of Plasmid RE gels ofHLAR enterococci ............................................................... 214Representation of Simpson's index of diversity and 95% confidenceintervals for Plasmid REA and PFGE analysis .............................. 218

TABLE OF CONTENTSI . INTRODUCTION .............................................................................. 1AIMS AND OBJECTIVES ............................................................... 6REVIEW OF LITERATURE2 . DISCOVERY OF ENTEROCOCCI . A ROADMAP ....................................... 73 . MOLECULAR TAXONOMY AND PHYLOGENY OF ENTEROCOCCUS ........... 8................................3.1. Reclassification of some Enterococcus species 114 . HABITAT AND ENVIRONMENTAL SIGNIFICANCE OF ENTEROCOCCI ....... I15 . EPIDEMIOLOGY OF ENTEROCOCCI ..................................................... 125.1. Nosocomial epidemiology of enterococci ..................................... 135.2. Molecular epidemiology of nosocomial enterococci ......................... 155.3. Epidemiology of multidrug resistant enterococci ............................. 175.4. Epidemiology of Enterococcal colonization ................................... 185.5. Epidemioloby of Enterococcal Superinfections ............................... 195.6. Epidemiology of Nosocomial environmental reservoirs of enterococci ... 195.7. Epidemiology of Community Reservoirs of enterococci ..................... 206 . CLASSIFICATION AND IDENTIFICATION OF ENTEROCOCCUS6.1. Genus definition and metabolic Characteristics ............................... 216.2. Cultural characteristics and Morphology ...................................... 226.3. Laboratory media for isolation and enumeration of enterococci ........... 236.4. Conventional species identification methods ................................. 256.5. Commercial rapid species identification methods ............................ 296.6. Molecular Phenotyping methods for species identification ................. 306.7. Other Biotyping methods ........................................................ 326.8. Molecular Genotyping methods for species identification .................. 337 . GENETICS OF ENTEROCOCCI ............................................................. 357.1. Gene transfer mechanisms in Enterococcus ................................... 357.2. Role of Mobile genetic elements ................................................ 367.3. Conjugative gene transfer mechanisms in Enterococcus .................... 388 . ENTEROCOCCAL INFECTIONS8.1. Types and their Clinical significance .......................................... 418.2. Management of enterococcal Infections ....................................... 47

9 . PATHOGENESIS AND VIRULENCE OF ENTEROCOCCI9.1. Pathogenesis of enterococci ...................................................... 499.2. Virulence factors of enterococci ................................................. 5010 . ANTIMICROBIAL RESISTANCE IN ENTEROCOCCI ................................ 5710.1. Intrinsic resistance ................................................................ 5710.2. Acquired resistance ............................................................... 5910.3. Laboratory detection of Antimicrobial resistance in enterococci ........... 65I 1 . EPIDEMIOLOGICAL TYPING METHODS FOR ENTEROCOCCI .................. 70I I . 1 . Conventional Phenotyping methods ............................................ 711 1.2. Molecular typing methods ....................................................... 7212 . ENTEROCOCCUS- THE INDIAN PERSPECTIVE ..................................... 7612.1. Non-human reservoirs of enterococci .......................................... 7712.2. Significance of nosocomial enterococcal infections .......................... 7712.3. Role of enterococci in polymicrobial infections ............................... 8012.4. Hospital reservoirs of enterococci .............................................. 8112.5. Antimicrobial resistance among enterococci .................................. 8112.6. Need of the Hour for Extensive Multifaceted studies on Enterococci .... 8313 . CHAPTER I . ENTEROCOCCI IN CLINICAL INFECTIONS .................... 84OBJECTIVE ................................................................................ 85MATERIALS AND METHODS ........................................................ 85RESULTS ................................................................................... 94DISCUSSION ......... .......................................................... I01SUMMARY ............................................................................... 10814 . CHAPTER I1 . ANTIMlCROBlAL RESISTANCE IN ENTEROCOCCI ...... 109OBJECTIVES ............................................................................. 109MATERIALS AND METHODS ....................................................... 110RESULTS ................................................................................. I13DISCUSSION ............................................................................. 123SUMMARY ............................................................................... 132

15 . CHAPTER I11 . MOLECULAR CHARACTERIZATION OF HIGH-LEVELAMlNOGLYCOSlDE RESISTANT ENTEROCOCCI ...... 134.............................................................................OBJECTIVES 134MATERIALS AND METHODS ....................................................... 135RESULTS ................................................................................. 146DISCUSSION ............................................................................. 153SUMMARY ............................................................................... 15816 . CHAPTER IV . VIRULENCE FACTORS IN ENTEROCOCCI ................. 160OBJECTIVES ........ ..... ..................................................... ..I61MATERIALS AND METHODS ...................................................... 161RESULTS ............... ....................... ................................ 164DISCUSSION ............................................................................ I69SUMMARY ............ ............................................................... 18017 . CHAPTER V . CELL-CELL COMMUNICATION AND GENE TRANSFERAMONG AMINOGLYCOSIDE RESISTANT ENTEROCOCCI ... 182OBJECTIVES ............................................................................ 182MATERIALS AND METHODS ....................................................... 183RESULTS .................................................................................. 186DISCUSSION ............................................................................. 190SUMMARY ............................................................................... 20018 . CHAPTER VI . MOLECULAR TYPING OF HIGH-LEVELAMINOGLYCOSIDE RESISTANT ENTEROCOCCI ........................ 201OBJECTIVES ............................................................................. 202MATERIALS AND METHODS ....................................................... 202RESULTS ................................................................................. 208DISCUSSION ............................................................................. 219SUMMARY ............................................................................... 22819 . SUMMARY .................................................................................... 23020 . CONCLUSIONS ............................................................................. 235REFERENCES .................................................................................. I-XLADDENDUM

ABBREVATIONSaac(6 ')AFLPAMEanr(6)-laaph(2 '7APSASBacBEABHIBSICDCCFUCHEFDAPldATPddldNTPEB SEDTAESPFAMEGelHLARHLGRHLRHLSRIEKbKD6'- aminoglycoside acetyltransferaseAmplified fragment length polymorphismAminoglycoside modifying enzyme6'- aminoglycoside adenylyltransferase2"- aminoglycoside phosphotransferaseAmmonium per sulfateAggregation substanceBacteriocinBile-esculin azideBrain heart infusionBlood stream infectionsCenter for Disease Control and preventionColony forming unitsContour-Clamped homogenous electric field4'-6-Diamidino-2-phenylindoledeoxy adeonsine 5'- triphosphateD-alanine D-alanine ligasesdeoxy nucleoside 5'- triphosphateEnterococcal binding substanceEthylene diamine tetra acetic acidEnterococcal surface proteinFatty acid methyl estersGelatinaseHigh-level aminoglycoside resistantHigh-level gentamicin resistantHigh-level resistanceHigh-level streptomycin resistantInfective endocarditiskilo base pairskilo Daltons

MDGMDRMHAMICPg1.11MLEEMLSMRSANCCLSNICUNNISPBPPCRPFGEPYRRAPDREASDS-PAGESSTlTBETETEMEDTMPlSMXTnUPGMAUTIVanVREVRS AWCPMethyl-a-D-glucopyranosideMulti drug resistanceMueller-Hinton agarMinimum inhibitory concentrationmicrogrammicroliterMultilocus enzyme electrophoresisMacrolides, Lincosamides, and Streptogramin BMethicillin resistant S!aphvlococcus aureusNational Committee for Clinical Laboratory StandardsNeonatal intensive care unitNational Nosocomial Infections SurveillancePenicillin-binding proteinsPolymerase chain reactionPulsed-field gel electrophoresisL-pyrrolidonyl- aaphthylamideRandomly amplified polymorphic DNARestriction endonuclease analysisSodium dodecyl sulfate - Polyacrylamide gel electrophoresisSkin and soft tissue infectionsTris borate EDTA bufferTris EDTA bufferN,N,N',N'-tetramethylene-ethylenediamineTrimethoprim-SulfamethoxazoleTransposonUnweighted pair group method with arithmetic averagesUrinary tract infectionsvancomycinVancomycin resistant enterococciVancomycin resistant S1aphy1ococcu.u aureusWhole cell protein

INTRODUCTIONOne of modem medicine's greatest success stories is the triumph of antibiotics overdisease-causing bacteria. Since their introduction during the era of 11 World War, theyhave saved innumerable lives and blunted serious complicat~ons of many feared diseasesand infections. AAer more than six decades of widespread use, however, many antibioticshave lost their punch [I]. At the beginning of a new millennium humanity is faced withanother crisis, a phenomenon called "ant~micmbial reslstance". Joshua Lederberg, NobelPrize has quoted "Anrrhlorlc resisruncc as u phenomenon IS, in ifsol/, nor.~urprrsing, nor rs if nrm. 11 I.\. homvlar, nw/v wortying kcuute il rs uccumulunng undrrccc,lrmrrng, u,hrlir the norldi rtx11.c for comhutrng rr decri,u.sc, rn power und number" [2].whlch unfortunately emphaslm that the development of new antimicrobial drugs maynot be In pace with the abtl~t); of bacterial pathogens to develop resistance. Since the firstrepon of pcn~clllin reslstanl ~luph).k~~~cc. jwl a few years after penicillin wascommercially ~ ntndud. the problem of antimlcmbial reslstance has snowballed into aserious publ~c health concern WII~ a-onomic. soc~al and polltical implicat~ons that areglobal In scope As quoted anonymously In the history of medicme."21HH) 8.C. t/t*rc. c,ur rho rtwrIWO A. I) Thur rcn~l u hc.arh~*n. llcrt .rhrs pruverI I . 7hur pruy1.r I.\ ~upc.rsrrrton. hen^. Jnnk rhrs porronIY?O .4.D.I'huf porron r.r .wke 011. Here, mullo om this prllIY45 .4.D. lhurprll 1-5 rncfl~~crr~r~ Ilc*n., rulic. rhr~ penrc.rllrn1955 A.D. Ot p.r.... hug.\ mururc*d. Hcsrc.. rake rhi.r rcrruo.clrtrt.1960- IWV .ZV morc "tn~pr " IA-n*. rukcf 1hr.r tnorc.po\rerful unrrhiorrcJIM ,I./). The hug.\ hu~r mnw.' ll~~n*. tvur rkrs rcnu "treating ~nfcctious, diseases may reach the juncture where its primitive journey oncebegan. due to h e phenomena of "antimicrobial res~stance" ~f uncondnd. [2].h is m unprccedentcd level of concern about increasing drug resistance bothin community and in hcwpids involving a wide variety of human pathogens. whichnflccls he gmwing frustration wirh the intractability of this dtifaccted problun. The

INTRODUCTIONNational Foundation of Infectious Diseases had ranked antimicrobial resistance andemerging infections as the first among the top ten problems in infectious diseases onwhich it will concentrate its efforts [3]. No population is more vulnerable to multi drugresistancethan those admitted to hospital wards. Of the resistant organisms nowproliferating around the world, none cany more potential for destruction and threatenexisting medical interventions than the emergence of hospital-acquired super-infections[2]. Someone who never heard of antibiotics ironically offered the clearest eluc~dation ofwhy antibiotic resistance increases so much in hospitals; Charles Darwin observed thatcompetition occurs In every environment and that "nature selects the strain or speciesmost suited to survive wlthtn a particular environment"The lncreaslng rate of reslstance among nosocom~al pathogens IS particularlydl~oncenlng over the vean Of the -. 2 mlll~on nosocornla1 lnfectlons occurring eachyear In the Lntted States 50 to Woo are caused by antimrcmb~al-res~stant stralns ofhacterta, and thew nowomlal lnfectlom are thought to contnbute to or cause >77.000deaths per year and cost appmnlmatelv $5 to 510 blllton annuall, [4] Surveillance datareponed bj the CDC National Nosocomtal Infecttons Sunelllance (NNIS) System for1990- 1990 showed a Lnntlnutng increase In ant~m~crobtal-resls~ parhogens associatedw ~th nosocomtal lnfcctions (in ICl patlenu) from U S hosprtals The Increase waspan~cularl) mdcd for \ancomvc~n-mlstant cntemcoccl (\'RE) (55%). methlclllln-reststant Slrrphtl(x tn t u\ urrrrur (MRSA) (3 1 %), thtrd-generatton cephalosponn-res~stantF3thrrrthru to11 (2900). tmlpenem-reststant Pinrdomonur ucrugfnoto (32%). andquinolone-mlstant P uc.rugrno\u (890/0) (5)Research and studies about the genetics of rrslstance to antimlcroblal drugs. hadshown clearly that bacteria have a remarkable array of tools at their disposal to overcameantibiotics. They can undergo chmmsomal mutattons. or acquire new genetic resistancematerial through dlmt exchange of DNA by conjugation or transformation, or throua abacteriophage-mediated tmwhuXlon. Many of the genes that mediate antimicrobialresistance arc found msfrnble plasmids. or on msposons [6. 7). The resistancemay be exhibited h g h three major mechanisms: production'of y enme that will

INTRODUCTIONinactivate or destroy the antibiotic; alteration of the antibiotic target site to evade actionof the antibiotic; or prevention of antibiotic access to the target site [S]. Variousorganisms possess different types of resistance mechanisms, although it is not unusual fora single bacterial strain found in a hospital to possess several of these mechanismssimultaneously. It is of major importance to establish whether it is the resistant bacteriumthat spreads in a clonal fashion, or whether the resistance determinant (genetic element) isproficiently transferred. The worst-case scenario is provided by the possibility toencounter a microorganism that is capable of both efficient spread and efficient resistanttransfer. The best example for the latter phenomenon is enterococci, which has emagedas a prominent nosocom~al pathogen since last decade (91.Though longstanding monitoring of nosocomial infections has defined severalbacterial spates as the mosl prevalent pathogens. a clear shift in prevalence from gramnegativeto gram-posit~ve spares as the maln cause of In\,aslve ~nfections has beenohsewed over the past dcrade (10). Enterococci are currently ascendant nosocomialpathogen. and were rcpned as the xcond most common organism recovered fromnosocomial unnar). tract and wound infect~ons. and the th~rd most common cause ofnosocom~al hactmia in the Unlted Stata, and overall they were the third leading causeof nosocom~al infections In Llnlted Stata dunng last dccade ( I1 ]Entcmcocc~ are gram-p~s~tive cocci that occur In a remarkable array ofenvtronments. like soil. food, water. and a wide variety of I~ving animals because of theirability to grow and suwivc unda harsh conditions. although the11 major habitat appearsto be the gastn)intestinal (GI) tract of humans and of other aninials where they make up asign~ficant portion of the normal gut flora (I?. 131 Since the inception of separate genusEnrcrocwcu~, chat are 28 spla of en~emcocci with clinical significance to date [I41 ofwhich Enrcrm~/&,~~/i.~ Enrcmcoc.c.us ,facciurn account for upto 90% of clinicali!i~lstes belonging to [his genus [13]. Ncvenheless. the incidence of other species ofenterncocci from clinical sourca shows an alarming increase with the pmperties ofintrinsic resistance to many antibiotics including beta-lactams and glppeptida [ 15. 161.

INTRODUCTIONOne of the major reasons enterococci have thrived in the hospital environment istheir intrinsic resistance to several commonly used antibiotics and, perhaps moreimportant, their ability to acquire resistance to all currently available antibiotics, either bymutation or by receipt of foreign genetic material through the transfer of plasmids andtransposons [6, 7, 171. Because most enterococci are tolerant to the bactericidal activityof 0-lactam and glycopeptide antibiotics, bactericidal synergy between one of theseantibiotics and an aminoglycoside is needed to treat the most serious enterococcalinfections, such as endocarditis and meningitis. This effect is lost if there is resistance toeither class of drug [a, 131. Enterococci generally cause wide range of nosocomialinfections such as urinary tract infections, endocarditis, bacteremia, and intra-abdominaland pelvic infections. They may also cause neonatal sepsis. meningitis and wound andsoh tissue infections, while the common risk factors associated in acquiring enterococcalinfections are a prolonged hospital stay, prior antibiotic therapy with cephalosporins.aminoglycosides or vancomycin. [I 3. 18-24].Enterococci, being lactic acid bacteria (LAB) plays a vital role in environmentaland food microbiology too. These bacteria play an important beneficial role as a starterculture in the production of various traditional fermented food products in Europe, andhave been successfully used as probiotics, while the detrimental activities of enterococciare assoc~ated with spoilage of foods, especially meats. This 'dualistic' nature ofenterococci, as well the demonstration of transfer of virulence determinants from clinicalenterococci to starter enterococci by conjugation gives rise to concern about their use asprobiot~cs or as starter cultures in the food industry 1251.The multi-drug resistant enterococci have become a serious clinical problem,since the antib~otic-resistant strains gets transmitted easily fmm one individual to anotherin a nosocomial setup. Antibiotic-resistant strains have the advantage to flourish andlikely to cause the infection among compmmised patients. since they easily proliferatewhen antibiotic therapy eradicates patient's normally sensitive microflora leaving an eco-vacuum. So the emergence of multidrug resistance in any health care setting, underscoresthe importance of proper and rapid identilication of the isolates to species level and their

INTRODUCTIONantimicrobial susceptibility panern, for initiating appropriate infection control measuresat the earliest to reduce the health care catastrophes.Antimicrobial resistance although irrefutably contributes to the prominence ofenterococci in nosocomial infections, many other factors augment the prospects of theirpathogenicity. The pervasiveness of enterococcal infections is attributable to enhancedvirulence, a prospect for which there are evidences 1261. The factors encoding virulencetraits have been shown to play a major role in transformation of this innocuouscommensal into an opportunistic nosocomial pathogen, thereby increasing their statisticalprobability ofcausing disease [I 3, 25, 261.Studies worldwide show the emergence of enterococci as a potent nosocomialpathogen with a gradual Increase in morbidity and mortality. With the dearth ofsignificant information on this nosocomial pathogen from Indian subcontinent it hasbecome the "need of the hour" to undertake an in depth study about the prevalence ofenterococci and their clinlcal significance with special emphasis on their antimicrobialresistance and virulence Further. molecular analysis of multiple isolates of enterococcican contribute to developing new insights into both the epidemiology and thepathogenesis of enterococcal infections from Indian perspective.

AIMS AND OBJECTIVESThe aims and objectives of the present study are:To determine the prevalence of enterococci from clinical specimens in atertiary care hospital in South India.To detect the antimicrobial resistance among enterococci.To Molecular characterize the determinants encoding antimicrobial resistance.To investigate the presence and clinical significance of various virulencefactors in enterococciTo study the transferability and characterization of the genetic determinants inenterococci.To document the epidemiological pattern of enterococci by Molecular typingof the isolates.

Review of Literature

RE VIEW OF LITERATURE2. DISCOVERY OF ENTEROCOCCI - A ROADMAPThe history of enterococci dates back to a century when Thiercelin in 1899 used the term"enterocoque" in a French publication to describe bacteria seen in pairs and short chainsin human feces. Later MacCallum and Hastings in the same year described a fatal case ofendocarditis from the John Hopkins hospital caused by a bacterium that was "very hardand tenacious of life" which they termed as "Micrococcus zymogenes". They confumedthe pathogenicity of the new organism by satisfying Koch's postulates to reproduceendocarditis using the organism in a canine model [13, 261. However. Andrews andHorder first coined the name Srrep[ococcus ,faccalrs in 1906, for an isolate recoveredfrom the blood of a patient with endocarditis, and considered that this streptowccus was"so characteristic of the human intestine that the term Srreptococcus,faetalis may justlybe applied to it". Later in 1918. Orla-Jensen described a second organism of this groupStreprococcus faecium that difTered from the fermentation patterns of Streprococcusfaecalis but was not formally recognized as a separate species for several decades Athird species Streptococcus durans was proposed In 1935 by Sherman and Wing. whichwas similar to Strcptocomus,fac~c~urn but had less fermentation activity. In 1937 Shermanemphasized that the term E~ll~~rococclts had been used to mean different things rangingfrom the broad definition of any fecal streptococcus to a restricted definition oforganisms that appeared to be identical to S, faecalis. He proposed a classificationscheme, which separated streptococci into four divisions: pyogenic, viridans, lactic, andEnterococcus. The term Enterococcus was used for organisms that grew at 10 and 45'~,in 6.5% NaCI, and at pH 9.6 and which survived 60'~ for 30 min. Many of thesecharacteristics became widely used to distinguish between enterococci andnonenterococcal streptococci, such as S. Iw~is, and some are still used today to helpidentify enterococci. Sherman's classification scheme also correlated with the serologicalscheme originated by Lancefield in the early 1930s. In that system, the enterococcireacted with group D antisera, while the pyogenic streptococci reacted with group A, B,C, E, F, or G and the viridans streptococci were nongroupable; S. bovis, classified by

REVIEW OF LITERATURESherman as a viridans streptococcus, was later shown to react with pup D antiserum. In1967, Nowlan and Deibel added a new species S. avium to the enterococcal group, whichwas found to react not only with Lancefield's group D antiserum but often with group Qas well [12, 131. Through the following years a number of other types (species) ofenterococci were recognized and added to the enterococcal group. To date, a proposed 28species of enterococci belong to the growing list of this genus (141.3. MOLECULAR TAXONOMY AND PHYLOGENY OF ENTEROCOCCUSMicrobial differentiation and identification IS a developing area, which is based on'class~cal' and 'advanced' methods Phenotypic differences among enterococci maymisrepresent the phylogenetic relationships, complicating the classification ofenterococcal species. With the advenr of novel DNA-based identification techniques,several methods focus on the unlque nucleic acid composit~on of the microorganismrather than on phenotypic expression of products that IS encoded by the respective gene.Polyphasic taxonomy comb~nes phenotypic and genotypic information and forms thebas~s for actual systematic bacteriology [27, 281 Over most of the last century barring thelast three decades, enterococci were classified as group D streptococci on the basis oftheir colony morphology and reaction with group specific antiserum. In 1970, Kalinaproposed a separate genus "Enterococcus" for the enterococcal streptococci based oncellular arrangement and phenotypic characteristics. But. the use of genus nameStreptococcus continued since no action on this proposal was taken [12]. Later Schleiferand Kilpper-Balz in 1984 provided genetic evidence using DNA-DNA and DNA-rRNAhybridization to prove that Streptococcus faccalis and Streprococcus faecium weresufficiently different from other members of the genus Streptococci including S. bovis,and suggested to merit a separate genus. The DNA G + C content ranged from 37 to 45mol% as revealed by them. It was similarly proposed that the group N lactic streptococcishould be transfemed to a new genus Lactococcus [29]. Shortly after the proposal ofSchleifer and Kilpper-Balz, Collins and his colleagues used similar methodology to showthat strains priorly denominated as S. avium. S. carseliflovus. S. durans. S. fiecalissubspecies malodoratus, and S. gallina~m were sufficiently closely related to other

REVIEW OF LITERArUREmembers of the genus Enierococcus to be transferred to this genus but sufficientlydistinct to be considered separate species. They proposed the names Enierococcus avium.E. casseliflavus. E. durans, E. malodoratus and E. gallinarum for those species [30].Following these two pioneering studies most other novel species of enterococci reportedthereafter, involved the application of DNA-DNA reassociation, 16s rRNA genesequencing, Whole cell protein (WCP) analysis, Fatty acid methyl esters (FAME)analysis and long-chain fatty acid analysis as genetic evidence for establishing any newerspecies of enterococci, in conjunction with conventional biochemical and physiologicaltests. Recently, authoritative journals have required that the sequence of the 16s rRNAgene be deposited in the Genbank when reporting a new species [29,31,32].To date, as of August 2004 [14] there are 28 species of enterococci proposed withappropriate genetic evidences. They are E. faecalis, E. faecium [29], E, avium, E.cu.\seliflavu.s, E. dlrrans. E. gallinorum, E. malodorarus (301 which are the commonlyreported species. The other specles subsequently proposed with appropriate geneticevidences as summarized recently by Caravalho et al. [14] are E. canis, E. hirae, E.mundrii, E. rufinosu.~. E. psc~udourrum, E. solirariur, E. cecorum, E. rolumhoe, E.saccharo!viicus, E. dispar. E. slrlfitre~ts, E. asini. E. ~illorum. E. hoemoperoxidus, E.mora\mic3nsis, E. rarri, E. porcinus, E. gihw. E, palletls, E. seriolicida, E. flovescens, E.phoc,~licrrlicolo, En~erococcus sp. Nov. CDC PNS-El, Enierococcus sp. Nov. CDC PNS-E2, and Enicrococcus sp. Nov. CDC PNS-E3. The Phylogenetic relationship among 28species of enterococci reported to date, and other species of related genera is depicted inFigure 1, which is based on comparative analysis of 16s rDNA sequences.

REVIEW OF LITERA TUREFigure I. Phylogenetic relationship of Enterococcus species and other species of relatedgenera based on comparative analysis of 16s rDNA sequencesDcndrogram rtpmduced with kind permission of Dr.Richard Facklam.CDC. Atlanta. U.S.A [Id].

REVIEW OF LITERA rURE3.1. Recla~slfication of some Enterococcus SpeciesSome of the previously proposed enterococcal species are not validated as newEnrerococcus species based on the results of recent genetic studies conducted by thesame, or other investigators as depicted in a recent review [12]. E. seriolicida showedhomology to Locrococcus gan~ieae by WCP analysis, rRNA gene sequence analysis andDNA-DNA reassociation experiments, and henceforth, the former has been denominatedas L. ganieae currently. Similarly. E. solitarius was later proved to be more closelyrelated to Tcrra~enococcus halophilus than 10any other species of the genusEnrerococc~rs by 16s rRNA sequence analysis and DNA-DNA reassociationexperiments. Likewise E flave~cms was shown to be closely related to E. casseli/lo~~usbased on DNA-DNA reassociation experiments and hencefonh denominated as E.c~us,scli/7a1~u.\, while same was the case of E porcinus designated later as E. ~~illorum asboth showed homology by genetic studies [I214. HABITAT AND ENVIRONMENTAL SIGNIFICANCE OF ENTEROCOCCIEnterococci occur in a remarkable array of environments, since they are able to grow andsurvive under harsh conditions. They can be recovered from water. soil, food, and avariety of animals, birds and insects 112. 131. In humans. the major habitat of enterococciappears to be the gastrointestinal tract although iso!ated less frequently from other bodysites. Entemcocci are numerous in the large intestine where concentrations of 10' to 10'bacteria per gram are typical. Several studies carried worldwide indicate that enterococciare found in feces of most healthy adults as well infants, with E. faecalis as thepredominating species followed by E,foccium [33, 341. Enterococci though prevalent, areless commonly isolated from other sites such as vagina, anterior urethra, skin, and otherportions of gastrointestinal tract including oropharynx and bile ducts. Several studieshave indicated that enterococci can be isolated from hospital environmental surfaces. buttheir role in strain transmission remains debatable [I 2. 131.

RE VIEW OF LITERA TUREEnterococci are considered to be good indicators of fecal contamination of foodand water since they are present in the feces of humans and warm-blooded animals. Theterm "fecal enterococci" include four species: E. furcalis. E. ,faecium, E. durans andE. hirae [IZ, 131. With their emergence as a nosocomial pathogen, environmental orother nonhuman sources could contribute to the dissemination for antibiotic resistantenterococci, which pose a therapeutic challenge. Apart in-vitro transfer, in-vivo transferof antibiotic resistance genes under natural conditions between E. ,faecalis in sewagewater treatment plants has been described [35] creatlng awareness about the role ofenvironmental strains in the dissemination of antibiotic resistance. While the significanceof enterococci In health care settings have been studied extensively, focus onenvironmental enterococcal strains could help to complete epidemiological studies and todetermine the role of these strains in antibiotic resistance dissemination in community.5. EPIDEMIOLOGY OF ENTEROCOCCINosocomial infect~ons have become onc of the major health care problems globally,attributing to the morbidity and mortality of hospitalized patients. "A nosocomialrnfcction IS one for which there is no evidence that the infection was present or~ncubat~ngthe trme of hospital admission To he classified as an infection, thecondition musl be manifested as a clinical disease and not merely colonization. However,an asymptomatic patlent may he considered infected :f pathogenic microorganisms arefound in a body fluid, or at a body site that is normally sterile. such as the cerebrospinalfluid or blood I I I]. Furthermore, the emergence of drug resistance among many bacteriahas raised an unpraxdented level of concern, since hospitals serve to be the idealbreeding ground for the development and spread of several multi-drug resistant bacteria.Data from several studies shows. over the past decade there is a clear shift in prevalencefrom gram-negative to gram-positive species as the predominant cause of nosocomialinfections, among which entemcocci has become one of the top three pathogens causingvarious infections [I 1, 131.

RE VIEW OF LITERATURE5.1. Nosocornill Epidemiology of EnterococeiThe clinical microbiology laboratory in collaboration with several teams of the respectivehospital generally carries out the surveillance of nosocomial infections. Federal bodieslike the Center for Disease Control and prevention (CDC)- National NosocomialInfections Surveillance NNIS), or the bodies approved by them carry out nationwide andinternational nosocomial surveillance progams through a network of laboratoriesworldwide. While many regional bodies in different countries, replicate the functioningof CDC's NNIS, according to their setup following the standard guidelines fornosocomial infections surveillance [20. 36. 371. Data from most of these programs showsthat enterococci continues to be one of the top three pathogens causing different kinds ofnosocomial lnfect~ons like urinary tract infections (UTI), blood stream infections (BSI),sk~n and soft tissue infections (SSTI) and other miscellaneous infections since lastdecade, and this has been supported by several other independent studies carried out invarious laboratories worldwide.Nosocomial surveillance studies conducted in different pans of the world byvarious groups showed that enterococci ranked among the top three pathogens causingnosocomial UTI [20, 38-40] In 1984 the NNIS listed Enterococcus as the third mostcommon cause of nosocomial infection. enterococci caused approximately 10% of allsuch infections, including I4 7% of UTls and 7% of bacteremias [13], the fact which wassupported by Morrison and Wenzel [19], who conducted a comprehensive 10 years studyof nosocomial urinary tract infections (LITI) due to enterococci from 1975 through 1984in a U.S hospital. and concluded that enlerococcus was the second most frequent cause ofnosocomial UTI with a crude mortality rate of IS%, and the proportion of nosocomialUTls due to enterococci increased from 6% to 16%. Davies et el. [41] in theirepidemiological study of nosocomial UTls in ~ediatric population showed E. coli (26%)and Entcrococcus species (1 5%) as the predominant pathogens causing nosocomial UTI,and showed catheter related infections accounted for 48% of all the nosocomial UTIs,while secondary bactcremia occurred m ly, with an incidence of 2.9%. While in anotherstudy Moulin et al. 1421 showed that E. coli (39%) and Enterococcus species (12.1%)

REVIEW OF LITER4TUREwere the etiogens of nosocomial UTI among pediatric age group in their retrospectiveone-year study in a pediatric hospital. Bouza et al. [38, 391 conducted a study on behalfof Co-operative group of the European Study Group on Nosocomial Infections toestimate the incidence of nosocomial UTI in Europe. The results from 141 hospitalsspread over 25 European countries, which participated in the study showed Enrerococcusspecies as the second common pathogen isolated next only to E. coli from nosocomialUTI. The SENTRY study of nosocomial UTI showed the pathogen occurrence andsusceptibility profiles in different geographic locations. Seven pathogens accounted for90% of all isolates, and Enterococcu.~ species ranked second accounting to 13% next onlyto E, coli (47%) [40]. Most of these studies showed that bladder catheterization, priorantibiotic therapy as the common risk factors for acquiring nosocomial UTI.Enterococc~ were the third most common cause of nosocomial BSI in the SCOPEProgram of IWI, conducted in 41 U.S. hospitals. They accounted for 11.7% of all~solates reported, with E. fueculr.\ as the most common species (60%). followed by E.fueci~trn (20%). Vancomyc~n resistance was observed in 36% of all participating medicalcenlcrs, with van-A and van-B genotypes equally prevalent [20]. The study conducted byCDC's-NNIS to describe the epidemiology of nosocomial infections in combinedmedical-surgical (MS) intensive care units (ICUs) between 1992 and 1998, revealed thatenterococci contributed to 11% of primary BSI's in the study group, and concluded thesenosocomial infections (BSI) in MS ICUs were almost associated with use of an invasivedevice (431. The SENTRY surveillance program conducted in 1997 revealed, thatenterococci ranked second, next only to Staphylococci as a cause of nosocomial bloodstream infection (BSI) during a one-year study conducted in different laboratoriesworldwide during 1997. They also demonstrated 22-76% of HLAR among enterococci[37]. A study by Diekema et al. [44] to find the age-related trends in pathogen frequencyand antimicrobial susceptibility of BSI in North American medical centers, showedenterococci as one of the common pathogen, while vancomycin resistance amongEnferococcus species predominated among nosocomial BSI in patients over SO yeam ofage. Another prospective surveillance study of nosocomial BSl's in pe&aaic patients at49 hospitals during a 6-year period conducted by SCOPE, showed enterococci as the

REVIEW OF LITERATUREsecond common cause of BSl's occurring predominantly in very young and/or criticallyill children, with vancomycin-resistance in 1% and 11% of E. faecalis and E. faeciumisolates respectively [45].In six years study of surgical wound infection surveillance at a tertiary carecenter, Weiss et al. [46] showed wagulase-negative Staphylococcus and Enterococcus asthe two most frequent isolates before and after antibiotic restriction policies wereimplemented in their hospital. Vanwmycin resistance was exhibited by 2.4% ofenterococci identified between 1996 and 1998. While in another study conducted to findthe epidemiology of nosocomial infections in combined medical-surgical (MS) intensivecare units (ICUs), Enterococcu.\ was found to be the single most frequently reportedpathogen accounting to 17% in patients with surgical-site infections [43].5.2. Molecular Epidemiology of Nosocomial EnterococciInitially, enterococcal infections were suggested to have evolved from patient's own floraconsidering the natural habitat of enterococci in humans, and due to lack of concreteepidemiological markers. But with the advent of molecular epidemiological toolsresearchers were able to present genetic evidences for exogenous acquisition ofenterococci mustly fmm nosocomial setup. In 1986 Zervos et al. [I81 reported thenosocomial transm~ssion and exogenous acquisrtion of S. foecolic. for the first time. Theyconducted a case-control study by comparing patients with gentamicin-susceptible andresistant S. ,/aecalis infections between 1981 - 1984 and concluded that all highly resistantstrains appeared to be nosocomial since I2 cases were clustered on a surgical floor and ina bum unit, using plasmid DNA content as an epidemiological marker. Subsequently inanother pmspective study of 100 patients hospitalized on the surgical and thoracicintensive care units (ICU) and a general medical floor, the same laboratory showednosocomial acquisition and inter-hospital spread of gentamicin-resistant enterococci.using plasmid DNA analysis as an epidemiologic marker. Their results showed that tenpatient's cultures grew colonies of gcntamicin-resistant enterococci-six after admission tothe ICU and four after hospitalization on the medical ward. Resistant entemocci were

REVIEW OF LITERA 7UREisolated from the hands of hospital personnel and were frequently isolated fromenvironmental surfaces 1471.Later Weems el al. [48] showed genetic homogeneity among few HLGR isolatesusing a gentamicin-resistance gene probe, suggesting nosocomial transmission ofenterococci. Wells et al. [49] investigated an outbreak of infections due to betalactamase-producing,high-level gentamicin-resistant (HLGR) E. ,faecalis. Restrictionenzyme digests of total chromosomal DNA showed nearly identical panems for selectedisolates of beta-lactamase-producing HLGR E. /ueculis, suggesting disseminationthrough the hospital of a single strain of E. .faacalis Thereafter, several nosocomialepidemiologic studies utilized molecular epidemiological tools for determining thegenetic relatedness of the enterococci. and showed that the most likely way these resistantenterococci spread among hospital patients were, via transient carriage on the hands ofhospital personnel, patient-to-patient and inter-hospital transmission of strains.Molecular epidemiological tools like PFGE were highly instrumental in studyingthe clonality of nosocomial infections caused by HLGR enterococci and VRE in manyhospitals. as shown by several independent studies [SO-551. Nourse et al. [56] usingPFGE and van-PCR suggested that environmental contamination played an importantrole in patient-to-patient transmission of VRE in a pediatric oncology unit, andinterventions including implementation of infection control measures were associatedw~th a decreased incidence of gastm-intestinal colonization. While another study showedthat the molecular epidemiological tools like PFGE was highly useful for detecting theendemic strains responsible for colonization, or infection over a prolonged time period1531. Bopp et al. [52] analyzed 116 VRE isolates obtained from patients in six New YorkState hospitals by PFGE, Plasmid profiling and PCR in a Molecular epidemiologicalstudy, which revealed genetic heterogeneity among isolates from within each of the sixhospitals. They concluded that PFGE typing could show that nosocomial VREtransmission had occurred in some hospitals. In another study, arbitrarily primedpolymerase chain taction (AP-PCR) was successfully used to evaluate cross infectionand a possible outbreak of E. faecalis UTI in a urology ward in the period 1982-1996.

REVIEW OF LITERATUREThey revealed that five of the nine isolates had the same pattern, which had caused theoutbreak of E. faecalis UTI and suggested that cross infection had occurred via urinarycatheters or by hand contact in that ward [54]. Thus the application of molecular toolswere shown to be highly useful for studying the nosocomial epidemiology of enterococci,which in many instances unraveled the outbreak sources in hospital as well communitysettings, that might have gone unmasked if failed to apply.5.3. Epidemiology of Multidrug Resistant EnterococciEnterococci show a remarkable ability to acquire and disseminate antibiotic resistancegenes by a variety of routes especially in a hospital setup, the property that has propelledthem to become a predominant nosocomial pathogen [I 3, 17. 571. Of prime importanceare the emergence of resistance to high-level gentamicin andlor streptomycin, ampicillinand glycopeptides among nosocom~al enterococci, since exhibition of resistance to anycombination of the above-mentioned antimicrobials poses a great therapeutic challenge.Various studies worldwide had shown that enterococci resistant to these antimicrobialswere responsible for several nosocomial outbreaks and clonal dissemination. Thusepidemiological investigations of these drug resistant strains and appropriate infectioncontrol procedures, helps in reducing the morbidity and mortality due to these multi-drugresistant enterncocci among hospitalized patients.Vancomycin resistance among enterococci has emerged to be one of the majornosocomial problems to deal with. Since it's first description in enterococci in 1986.there is a steady increase in vancomycin resistance among nosocomial enterococciworldwide. From 1989 through 1993, the proportion of enterococcal isolates resistant tovancomycin (VRE) reported to CDC's National Nosocomial Infections Surveillance(NNIS) system increased from 0.3% to 7.9% [58]. Subsequent studies showed a 47%increase of VRE from 1994 to 1998, with 26% of nosocomial enterococci exhibitingvancomycin resistance [59]. While a surveillance data reported by the NNIS srjtem for1993-1997 compared with January-November 1998, showed a marked increase (55%) inVRE associated with nosocomial infections in ICU patients from U.S. hospitals [S].

REVIEW OF LITERATUREBoyce et al. (601, in their study showed the incidence of ampicillin-resistantenterococci (ARE) increased sevenfold at a university-affiliated hospital between 1986and 1988. The plasmid and chromosomal DNA analysis of the ARE isolates revealed thatthe increase was due to an epidemic of 19 nosocomial infections that yielded closelyrelated strains of E, jaecium, whereas the non-epidemic strains were identified as E./irocium. E, raflnosus. E. duran~, and E. gallinarum, and concluded that the increase inthe incidence of ARE was due to the selection of various strains of resistant enterococciby the use of imipenem. Hanhug et al. [61] described the first nosocomial outbreak ofampicillin-resistant and vancomycin-resistant E../irecium in Norway during 1995 to 1996.Their results revealed that 149 patients were infected with ARE, and isolates from 1 15cases were genomically related to the outbreak strain, while four infections were causedby a van-B type VRE that was genomically related to the ARE outbreak strain. Theyconcluded that one year after the onset of outbreak. VRE occurred in wards which had arelatively high consumption of vancomycin, and the first nosocomial outbreak caused byARE observed in 1995 was still ongolng.Another prospective study revealed that 9% of enterococcal clinical isolates wereampicillin resistant, while ARE were not isolated from hospital personnel orenvironmental surfaces, they were common in the rectal flora and could spread to theurinary system to become an emerging clinical problem 1621. Torell et al. [63. 641showed the incidence of ARE among enterococcal isolates at a University Hospital inSweden increased from 0.5% to 8.1 % beween 1991 and 1995. and suggested multifocalorigin of the majority of the infecting ARE strains. They concluded that. non-recognizedfecal colonization and silent spread of ARE among many patients over a prolonged timeperiod as a major reason for the increase of ARE infections in their hospital.5.4. Epidemiology of Enterococcnl ColonizationBonten et al. [65], investigated colonization of patients and environmental contaminationwith VRE in an endemic setting to assess the importance of different sources ofcolonization. They revealed that persistent VRE colonization in the gastrointestinal tract

REVIEW OF LITERATUREand on the skin. the presence of multiple-strain types of VRE, and environmentalcontamination might contribute to the spread of VRE. However, they concluded that oncecolonization pressure was high, it became the major variable affecting acquisition ofVRE. Oon el al. 1661 conducted a study from January to March 1997 at a 900-bedteaching hospital in Singapore to determine the prevalence of intestinal colonization ofVRE in the patient population. Out of the total of 299 consecutive stool specimensscreened. VRE were detected in the stool of 35 patients (12.3%). This group consisted offour isolntes with Van-B (one E. .fueculis and three E.,faecium) and 31 isolates with Van-C (30 E, casselr/lu~v.s and one E, gallinurum). Except for isolates from the same patients,PFGE patterns were diverse, suggesting that the VRE isolates were genotypicallyd~fferent and possibly introduced from many sources. Thus epidemiological studies ofentemoccal colonizers exhibiting drug resistance in GIT, helps in initiating appropriateeradication approaches to decrease the risk of VRE infections in humans. especially thosehospitalized [67].5.5. Epidemiologv of Enterococcal SuperinfectionsI'nor use of ant~m~crobial agents lacking enterococcal activity has been quoted as an~mportant factor in the development of enterococcal superinfections in hospitalizedpatlents [13]. Several authors have shown that enterococcal superinfection andcolonization developed in 2-59.0 hospitalized patients, after therapy with broad-spectrumantibiotic like moxalactam. aztreonam and ciprofloxacin. However. other disposingfactors like urinary cathetenzation and wound infections have also been shown tocontribute to this fac~ [68.69].5.6. Epidemiology of Nosocomirl Environmental Reservoirs of EnterococciHospital environment plays a major role in transmission of entemocci, mostly resistantto lhe commonly used antimicrobials in hospitals. The versatile nature of enterococcienables them thrive in adverse conditions like those of hospitals. Noskin et al. [70]showed that VRE are capable of prolonged swvival on hands, gloves, and environmental

REVIEW OF LITERATUREsurfaces. Their study revealed that E. faecium isolates survived seven days oncountenops. 24 hours on bedrails, 60 minutes on telephones and 30 minutes onstethoscopes. Freeman et al. [71] showed that nosocomial enterococci were resistance toheat (upto 80'~for one minute) which enables them to survive the temperatures andholding times specified by the health authorities for the disinfection of used or infectedlinen. Also they could withstand 150ppm available chlorine for five minutes, thetreatment suggested by the health authorities for the disinfection of heat labile materials,which underscores the significance of enterococci to survive and disseminate in thehospital environment.Several studies have shown that health-care workers and environmental surfacesin hospital settings were responsible for transmission and spread of gentamicin andvancomycin resistant enterococci Freeman et al. in thelr study using Pyrolysis massspectrometry authenticated that nosocomial spread of enterococci had occurred viafluidized microsphere since the decontamination process was inadequate [72]. Whilesome studies have shown thermometers. hlood pressure handcuffs, IV fluid pumps,bedrails and linen as different sources of contamination with VRE contributing to theirspread in hospital scttings [65. 73, 741. U'hile many studies have shown concordancebetween hospital environmen~al strains and the patient isolates often resistant lovancomycinigentamicin using molecular epidemiological tools [75-801. Most of thesestudies have funher suggested and proved that appropriate infection control measures,minimizes transmission of these multi-drug resistant enterococci in hospital settings.5.7. Epidemiology of Community Reservoirs of EntermciSeveral epidemiological studies conducted in human subjmts from community haveyielded entcrococci resistant to various antimicrobials like ampicillin, gentamicin andvancomycin. Most of the human subjects from community who carried drug resistantmterococci were previously hospitalized and had used antibiotics. Vancomycinresistance among community subjects had been the focus of study by several authorssince laat decade, and most of the studies have concluded that previous hospitalization

RE VIEW OF LITERATUREandl or prior use of vancomycin as the common factors for community dissemination ofvRE [&I-831. While other studies have quoted that transmission of resistant enterococcior resistance genes takes place between humans and animals in the community. Onepossible explanation they have given for the animal to human transmission was the use ofgly~opeptide growth promoters like Avoparcin in feed animals [84-871. This fact wasfurther authenticated by another study, which showed a decreased incidence of VREisolated from poultry meat and from fecal samples of humans in the community afterdiscontinuation of avoparcin usage in animal husbandry [88]. But most of theseepidemiological studies have been conducted in Europe during last decade, where therewere no restrictions for using Avoparcin as growth promoters in farm animals, unlikeother countries. But the impact of recent ban on avoparcin as growth promoter inEuropean countries has not been studied much to give a clear picture of this less studiedproblem6. CLASSIFICATION AND IDENTIFICATION OF ENTEROCOCCUS6.1. Genus Definition and Metabolic CharacteristicsThe genus fiinti*rococcrr.s consists of gam-positive. facultative anaerobic organismsassigned as chemo-organotrophs, that are ovoid in shape and may appear on smear inshort chams, pairs. or as single cells. They are strict fermenters since they lack a Krebscycle respiratory chain. Their metabolism is homofermentative and produce lactic acid asan end product via. Embden-Meyerhof-Pamas pathway. Their predominant end productof glucose fermentation is the L (+)- lactic acid enantiomer. Like streptococci, theseorpanisms do not have cytochrome enzymes and are thus catalase negative. althoughsome strains do produce pseudocatalase [12. 13. 27. 281. Most strains produce a cell wallassociated glycerol teichoic acid antigen-the streptococcal group D antigen, but thedetection rate varies depending on the extraction procedure and the quality of antiseraused, while some rract also with group Q antisera [89, 901 Hydrolysis of L-pyrrolidonyl-3-naphthylamide (PYR) is a characteristic feature that is seen also with group-Astreptococci but not other streptococci. Most strains in the newly defined Enterococcus

RE VIEW OF LITERATUREgenus possess the characteristics summarized by Sherman in 1937 such as the ability togrow in 6.5% NaCl and at pH 9.6, to grow at 10 and 45'~, and, for the most part, tosurvive at 60'~ for 30 minutes. Although the above screening tests appeared to besufficient in the pas1 to identify enterococci presumptively, it is now recognized that otherless commonly encountered gram-positive cocci can also give a positive reaction in someof these tests [90]. For example, some cultures of Laciococcus, Aerococcus, Pediococcus,and Leuconostoc spp. are bile-esculin positive or can grow in 6.5% salt or both. Strains ofPcdiococcus and Leuconosioc spp. can be group D positive, and some lactococci andaerococci are PYR positive; lactococci and aerococci, however, should not react withgroup D antisera. The phylogenetic analysis based on 16s rRNA gene reveals that thegenus .Enrerocor.clr.c is more closely related to Vogococcus, Teiragcnococcus andC'arnohacic~riurn than they are to the genus Sireprococcus and Lociococcu.~ [I 2,29,90].6.2. Cultural Characteristics and MorphologjEnterococci generally produce well-circumscribed. smooth. raised colonies about 1-2 mmIn d~ameler, although some variants may appear smaller on prlmary isolation media suchas blood agar. Recently mucold encapsulated strains of E. ,faeculis have been isolatedfrom unne specimens. on 5% sheep blood agar and Muller-Hinton agar plates, whichemphasizes the versatil~ty of the colony morphology of enterocwci. Hemolys~s can beobserved around enterococcal colonies if the agar contains horse. rabbi6 bovine. orhuman erythrocytes, but sheep erythrocytes are largely refractory to the effects of thecnterococcal hemolysin [26]. Some stralns of E. ,faecalis may be a-hemolytic on agarcontaining rabbit, horse, or human blood but non-hemolyt~c on agar containing sheepblood, while some strains of E. durata are p-hemolytic regardless of the type of bloodused. All other species are usually a-hemolytic or non-hemolytic. Strains that are a-hemolytic arc actually non-hemolytic strains that produce peroxide. which acts on theblood cells in the medium and results in 'greening' of the agar, and not merely due to theproduction of a toxin by the strain. E. cossel~/lo~ws, E. mundiii and E. sulJureu~ produce ayellow pigment on Trypticasc blood agar medium, which can be detected by using awhite conon swab to pick up growth and examining the swab for a yellow color [12, 131.

RE VIEW OF LITERA TURE6.3. Laboratory Media for Isolation and Enumeration of EnterococciA diversity of media has been described and proposed for the isolation and enumerationof enterococci, owing to their importance in different foods, feeds, and clinical andenvironmental samples. Because of their requirements for several vitamins and aminoacids, enterococci cannot be grown easily in synthetic media. Profuse and rapid growth isonly achieved if rich complex media such as Brain Hean Infusion (BHI) broth orTrypticase Soy (TS) broth are used. A recent review by Domig et al. details differentmedia used for the enumeration and isolation of enterococci from various sources [27].Although various selective media are used for isolating enterococci from differentsources, we focus mainly on media used for cltnical specimens. For primary isolation ofenterococci from clinical specimens any blood agar base containing 5% animal bloodsuppons the growth. The hemolysis of enterococci depends on the type of blood usedalong with the basal media. but predominant enterococcal species are usually a-hemolyticor non-hemolytic. All enlerococci grow at 35 to 37'~ and do not require an atmospherecontaining increased levels of carbon dioxide [I 21Bile esculin test was applied to differentiate between enterococci (group Dslreptococci) and non- group D streptococci (90). Bile-esculin azide (BEA) agar(available with various commercial manufacturers) is an excellent primary isolationmedia, for ennchrnent and isolation of enterococci fmm samples that might bepolynicrobial in nature containing gram-negative bacteria. The azide in the mediuminhibits the gram-negative bacteria and enterococci appear as black colonies byhydrolysis of esculin. However, Listerio monoc:~'~opcne.~ may exhibit a similar colonialmorphology on this medium after 48 h of incubation. Most other bacteria either growweakly or appear as colonies of different shape. Media like Columbia colistin-nalidixicacid agar (CNA) or phenylethyl alcohol agar (PEA) are used for successfi~l isolation ofenterococci. CNA is advantageous over PEA, since hemolytic reaction can be read fromCNA if supplemented with blood, but not from PEA [12]. The Cephalexin aztreonamarabinose (CAA) agar allows the isolation of E.furcium from heavily contaminated sites.and in comparison with CNA agar it can differentiate E. .foecium from E. .foecolis and

REVIEW OF LITERA TUREE. durans by its ability to ferment ~binose [91]. While other commercially availablemedia like Kanamycin-aesculin-azide (KAA) medium, Membrane filter enterococcus(ME) agar medium contains various antimicrobials which suppresseslinhibits the growthof other microbes, but enhanceslfacilitates the gowth of enterococci [27].With increasing incidence of antimicrobial resistance among enterococci.isolationldetection of vancomycin resistant enterococci (VRE) and aminoglycosideresistant enterccocci are of clinical significance. For the detection of VRE in differentspecimens, numerous variations of media and isolation procedures have been published.Most of them are variat~ons of selective media, which differ with regard to the antibiotic,or their concentrations used (271. Willey and colleagues demonstrated that VRE could bedetected by using Muller Hinton agar (MHA) supplemented with varylng concentrationsof vancomycin (6-12 pglml) [92]. Swenson et al, in a multilaboratory evaluationdetermined that BHI agar supplemented with 6 pgiml of vancomycin as an optimalmedium for detection of VRE [93]. Many clinical microbiology laboratories worldwideare using this medium for detection of VRESeveral techniques and media have heen developed for isolating enterococci fromInanimate surfaces in hospital environment. Premoistened swab is the most commonlyused method for surface swabbing in hospital environments that are placed in enrichmentbroth and then plated appropriately for isolating enterococci 112. 271. Recentlycommercial manufacturers have come up with agar imprint methods (Rhodac-imprintmethod), where the agar surface is directly applied to the environmental surface to becultured and examined for the growth of enterococci. Unfortunately, despite theavailability of an array of med~a to isolate and enumerate enterococci, there is no singlemedium which equally meets all requirements. since in most cases a pronouncedselectivity is only achieved if lower recovery rates are tolerated and vice versa.Mortova, the performance of each method depends largely on the matrices of thesamples, and on the associated microflora.

REVIEW OF LITERA TURE6.4 Conventional Species Identification MethodsClassic species identification of enterococci involves assays for a combination ofbiochemical and morphological characteristics of the unknown organism. After Kilpper-Balz's1291 newer classification of enterococci based on the molecular andchemotaxonomic approaches, Facklam and Collins in 1989 [90] described anidentification scheme utilizing conventional biochemical and physiological tests forspeciation of enterococci. In subsequent years, other authors described tests involving thecleavage of methyl-a-D-glucopyranoside, the susceptibility to efrotomycin, and thefermentation of D-xylose (94, 951 for difrerentiation of Enrerococcus species. Thereafter,several authors devised various algorithms with minimal number of biochemical tests forldentification and speciation ol'enterococci. However, most of the tests were based on. order~ved from the conventional identification schema devised by Facklam and Collins [12,90. 961. Today even though a broad array of novel identification methods prevails, formost clinical microbiological laboratories worldwide. the primary method of identifyingCrircro~.occ.~c.~ strains relies on phenotypic characterization based on the conventional~dentificat~on schema dev~sed by Facklam and Collins [90].6.4.1. Facklam 's Conwntional Phenotyping SchemaIn 1972, Facklam published a summary of 26 physiological tests to differentiate group Dstreptococci, and thereafter over the years enterococci has undergone dramatic taxonomicchanges 1291. However. in 1989 Facklam and Collins devised an array of biochemicaland physiological tests for proper identification of twelve Enrerococcus speciesprevailing during that period. The grouping is based on key phenotypic tests, and doesnot necessarily conform to the grouping by 16s rRNA sequencing or grouping by othermolecular techniques. The identification schema has been constantly updated by Facklamand his colleagues to fit in all entemcoccal species reported to date [12, 14. 31. 901 andthe details of the schema are presented elaborately in the forth coming pages. Most oftheir phenotyping studies wm performed on lactococci and enterococci isolated fromhumans.

REVIEW OF LITERA TUREAs mentioned in the "Genus definition", any unknown catalase-negative grampositivecoccus presumptively identified as Enterocnccus, are subjected to theconventional phenotypic tests listed in Table I to identify the species as per standardprocedures. The five species E. cecorum. E. calumbue, E. pollens, E. succharoly~icus andEnlcrococcus sp. Nov. CDC PNS-E3 are included even though they are PYRasenegative,and may grow very poorly in broth containing 6.5% NaCl (except Enterococcussp. Nov. CDC PNS-E3) [12, 14, 311. The identification schema is based by separation ofthe enterococcal species into five groups, based on acid formation in mannitol andsorbose broths and hydrolysis of arginine. Although identification of enterococcal speciesby conventional tests is not rapid most identification can be made after two days ofincubation, while some may require incubation of the tests up to ten days.Group lThis gmup consists of E, u~,i~tn~. E, malodorutus. E. raffinosus. E, pallens, E, gih*us, E.pwudrw~.ium. E. .r~c.i.h~r(~!r.li~li.\ and Enteroc,occlc.~ sp. Nov. CDC PNS-E3.

REVIEW OF LITERATUREhilrpld lam rc1om.c I4 vllh rmmukrn' h14h rmnnttot. SOP. nmvr mc,. arp,n,rr 4 ~ amh,m* 4 sn1. *,m,t,,i R ~ . raninos. I TI I. o (I.. tcliun~c. MOT. molll~r!.I'I(* pl-l S(I(. s u m PYI' pru\nr M(;F ~ ~ \ I ~ . I ~ ~ I + W -u(l%.~~~,~~arcrni~~\c~ ~ ~ R ~ ~ I ~ - C ' Ilr.slnlns anpoS118vcI ~ a h (I k I 10 195 ~k unm. uc pa,tlrcl. . WCLIII~I c,cqn~,mr ,rcur (.?a. 0rem8ns shou a h n l rmcllunr)' kmwr rhuu~ntwcx M on d.1. inm I>V .tntns''[alr rmnnl~ol p n l t ~ l&p c toItr)(uh~~jnl

RE VIEW OF LITERA 7UREGroup 11This group consists of E, faecaiis, E. haemoperoxidus. E. ,fhecium. E. casseliJavus. E.mundlii. E. gallinarum and Enterococcus sp. Nov. CDC PNS-E2.(;roltp 111This group consists of E, drrruns. E. Irrrcrc~. E, dirpur. E rutti and E. ~illorum (E. rillorurnand E, porcincts constitute a single species as proven by genetic studies).

RE VIEW OF LITERA TUREGroup-IVThis group consists of E. o.~ini, E. sulfureus, E. cecorum. E. phoenicuiicolo andEntcroc0ccU.Y Sp. NOV. CDC PNS-E2.(iroup I'According to the latest classification 1141, this group consists of E cunis. E. columhae,and E, moru~'~i~n.v~s.6.5. Commercial Rapid Species Identification MethodsThe convcntional phenotypic identification of enterococci is rather time-consuming andoRen ambiguous, hence based on the recent developments miniaturized test kits havebeen introduced and used for rapid phenotypic differentiation of enterococci. Most of thekits and strips allow an indicator-based determination of sugar utilization or show

REVIEW OF LITERATUREreactions based on their specific micrhial enzymes. There are several publicationsdescribing comparative studies using these devices, but we have tried to give an accountof only the recent updates of the commercial systems available, within the scope of ourreview. The most commonly used testing systems, or kits usedltested and evaluated areAPI 20 Strep (Bio-Merieux, France). AP1 50 CH and Rapid ID32 Strep (Bio-Merieux),API Rapid ID 32 system, Crystal Gram-positive and Crystal Rapid Gram-positive kits,Vitek Gram-Positive ldentification Card (Vitek-2 system), Microscan Gram-Positiveldentification panel and Phene Plate (PhPlate Microplate Techniques, Sweden) [14, 31 1.However, the errors encountered in identification with these commercial systems or kits,were related io misidentification of E. gallinarum and E. cas.srl~flavus which neededsupplemental tests for motility and pigmentation to improve their identification by aconsiderable perceniage 1951In summary most of the systems or kits could detect only a limited number ofenterococcal species, and fuflher rests were necessary for a higher level of identification.Even though these rap~d ~dmtification systems have revolutionizedmicrobial~dcntification In clin~cal microbiology laboratories, it is still out of reach for most of theclln~cal micmbiological laboratories In developing countnes. ldentification of novel andunusual species has become unreliable with commercial systemu'k~ts w~th the increase in\he number of enterococcal species and phenotyp~c similarities among enterococcalspecies 114. 3 1. 97). Funhennore no commercial kit includes the whole set of tests usedIn conventional testtng schema for tdent~ficatton of Dlterococcus species [96]6.6. Molecular Pbenotyping Methods for Species IdentificationPhenotyping or bioiyping, meaning the detection of carbohydrate fermentation andenzyme pattern, can be regtuded as a traditional way of microbial differentiation.According to merit devclopmen~s, alternative procedures have been sought to simplifyto speedup Uuse methods that are alternatively termed as "Molecular phenomyping"since the procad- used. helps in identifying and exploiting the phenotypiccharactniaics of the organism, unlike genotypic methods which exploits the genetic

REVIEW OF LITERATUREcontent (DNA) of the organism. An ovefiiew of various methods available for speciationof entemuxci, are reviewed briefly in the following sections.6.6.1. Whole Cell P rorein (WCP) Fingerprinting By SDS-PA GEThe comparison of whole-cell protein panems obtained by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) under well-defined conditions has beenshown to be reliable method for identification and classification of enterococci, especiallythose of unusual or atypical enterococcal species [98]. Results of some studies suggest ah~gh correlation between similarity in protein patterns and DNA-DNA homology values.Teixeira et a!. analyzed E. ,fuecium strains of atypical phenotype and showed that theprotein profiles were similar to those of typical stra~ns and easily dist~nguishable fromthose of other enteroccxcal specles [97] Preliminary studies from our laboratory haveshown that atyp~cal h~ochemlcal variant strains of E fuccium. E, fuecu1i.c and E.~~ac.t~~liflu~~us, exhibited the same whole-cell protein profiles, as do typical strains, apartfrom minor qualitative or quantitative d~fferences In the WCP profiles among strainsIVV] Studies show that E. ~ull~nuntm and E, cusscliflu~~u.\ strains that are very closelyrelated by 16s rRNA gene sequencing arc clearly separated by WCP profiling, whilemany novel enterococcal spsles have been proposed based on WCP profiles and otherDNA-DNA hybridization methods [ 14. 3 I]. With the advent of gel documentationsystems and software programs, the analys~s of proteln fingerprints have become muchcaster, and as evident fmm current literature this technique is being followed by manylaboratorin worldwide6.6.2. Olhcr Molecvlor Phenotyping methodsWhile several clinical micmbiological laboratories are using WCP profiling foridentifying or authenticating the taxonomic status of enterococcl, some researchers havetned other novel mahods for identification and speciation of entemcocci. A recentreview summarizfi, other less commonly used methods for this purpose [28]. Pyrolysismass spectrometry (pyMS). Vibrational spectroscopic methods and Proton magneticresonance spatroscopy (1" MRS) have been used fo~ rapid identification ofSlreptococnu; species and En,ercroccu.c species, and the results were compared and

REVIEW OF LITERATUREconfinned by 16 S rRNA sequencing and species-specific PCR [28]. While, Multilocusenzyme electrophoresis (MLEE), Long-chain fatty acid analysis and Fatty acid methylesters (FAME) analysis were used as an adjunct to other methods, for the description ofthe novel enterococcal species [3 I]. However, as of now their applications are restrictedto experimental studies with medical isolates, or food isolates.6.7. Other Biotyping MethodsThere are several other biotyping methods for characterization and identification of themterococcal species although less frequently used, besides the above-describedconventional biochemical phenotyping. or molecular phenotyping methods.6.7.1. SrrotypingThe histop of serotyping dates back to 1933. when Lancefield reported that p-hemolytickcal streptoci~ci are typical for pnssesslng the group D antigen. With the advent ofno\,el techn~ques. the Polyphasic taxonomy has combined chemotaxonomic and genetictechniques w~th less emphasls on serolog~cal cnteria for classification 1281. Not all thestrams produce a cell wall assoc~ated glycerol teichoic acid antigen-the streptococcalgroup D antigen 112. 141. Moreover the detection rate banes depending on the extractionprocedure and the qual~ty of antisera used, while some strains react also with group Qantisera 1891.6.7.2. Anrimicrobial Susceptibility TestingEnterc~occi are intrinsically resistant to a large number of antimicrobial agents. TheVanC phenotype (low-level resistance to vancomycin. susceptible to teicoplanin) is anInherent property of Egullinu~m and E. ~u.vsi~l~(lu~~~~v. This propeny is not transferableand is related to the p-nceof species-specific genes vane-l and vanC-2 respectively.which maks this propaty an excellent criterion for their species identification [12, 171.The susceptibility or rcgistance 10 Efmtomycin-an elfamycin dnrg is another typicalattribute that was applied in the identification scheme of entemocci proposed byCawalho ct al, [94].

REVIEW OF LITERATURE6.7.3. Bacteriocin TypingThe production of inhibitory substances that are effective agalnst closely related speciesIS an interesting phenomenon among certain bacterial groups. In 1963, Brock and hiscolleagues were the first to describe the inhibitory substances produced by enterococci asentermins [loo, 1011. Later some studies explored the ability of bacteriocin to typeenterococci from clinical specimens. Pompei and his colleagues exploited and evaluatedthe potential taxonomic value of the bacteriolytic activity of enterococci. The detection ofenrerococcal lyo-groups was proved to be as reliable for species identification asconventional methods. and the bacteriolytic pattern groups correlated well with speciesgroups phylogenetically determined [102]. But studies on bacteriocins from enterococciin clinical setup are fbcused towards their role in bacterial pathogenicity although few innumber.6.8. Molecular Genotyping Methods For Species IdentificationSince last decade, there has ken a rem men do us advancement in the field of clinicalinlcrobiology with advent of molecular techniques. Many novel genotypic techniques~argeting the nucleic acid material- DNA or RNA of the organism were used mostly forlaxonomic purposes in special laboratones in late 80's and early 90's. However since mid90's. the application of molecular techn~ques for the identification of Enterococclrsspecies has expanded dramatically, and a brief review of vanous methods potentiallyadaptable for use in microbiology laboratones are summarized in the follow~ng sections.6.8.1. SpecicsSpcciifir PCR AssaysSeveral variants of PCR based nucleic acid amplification assays prevail for species~dentification of enterococci followed by clinical microbiology laboratories. A MultiplexPCR assay was developed by Dutka-Malen et al. (151 to amplify internal portions of the"ddl" and the vancomycin resistance (van) genes: vanA, vanB, vanCl and vanC21 3.Since the vancomyin resistant genes "vanA" and "vanB" could be found in E. faecalisRS well E. faecium, the "ddl" genes for these two species were used for speciesidentifica~ion. vagl and Vanc~3 arr specific for E, gallinarum and E. cassel~/7a1ws

REVIEW OF LITERA 7URErespectively and thus identification of these genes, were equivalent to speciesidentification. Thus a total of six primer sets were used to identify four species ofclinically relevant enterococci. This multiplex PCR showed an excellent agreement withother phenotypic identification methods for species identification. Subsequently, severalauthors worldwide used this method for species identification of enterococci fromdiflerent clinical specimens with minor modifications [61. 103. 1041. Kariyama et al.[lo51 presented primers specific for E. jueca1i.s. which were modified based on theinformation obtained from Dutka-Malen et al. (151. Enterococcus protein A (efaA) genesspecific for E. faeco1i.s and E. juecium were amplified by PCR and probed withcorresponding sequences in immunoblots without any cross-reactions with otherenterococcal species 11061. Amplification of the pEM1225 gene was repofled to be aspecies-specific PCR assay for identification of E. .fuecrum using a nucleotide sequencethat is strongly conserved in this species as a target for the PCR. (1071. Species-specificpninen for E. dttrun.\ and E, hirue targeting the "ddl" gene were also developed andvalidated by PCR assays [108].6.8.2. Mi.~cellaneous PCR Based assaysIn a Randomly amplified polymorphic DNA (RAPD)-PCR (also called as Arbitrarilyprimed (AP)-PCR) amplificat~on of random segments of genomlc DNA of enterococciwas d~rected by using a single oligonucleotide pnmer of arbitrary sequence, whichd~splayed polymorphism by the lengths of the amplified sequences obtained. However,only few ditTerent primers were proved to be reasonably successful for speciesicientificat~on of entmocci [109. IIO]. A Broad range (BR)-PCR assay was developedlo ampl~fy a ponlon of 16s rRNA gcne. which pro\,ed to be species-specific only for sixspecies of enterococci (1091.Tyrrell and colleagues successfully identified elevenenterococcel species by an Intergenic spacer (ITS)-PCR using primers to amplify rRNAspacer regions. They performed gel electmphoresis analysis of the PCR products afterrestriction digestion without sequencing the PCR product [I 1 I]. Amplification andsequencing of he Dala:D-ale ligases (ddl) genes was performed to identify all thespaies tested except for E. mundtii and E ruJ7nusu.c [I 121. Squence analysis of thesmall-subunil ribosom] rjbonucleic acid-16S rRNA gene for identification of novel

REVIEW OF LITERATURELnrerococcus species was performed alon'gwith conventional phenotyping and analysis ofwhole-cell protein profiles [14, 311. However, none of these PCR based assays weresuccessful in identifying all enterococcal species.6.8.3. Other Nucleic acid based mefhodsA commercially available non-radioactive DNA probe- Accuprobe system (Gen-Probe,San Diego, USA) has been shown to be successful to verify the identification of majorityof enterococcal species of human origin by many clinical microbiology laboratories. Thisnon-radioactive DNA probe was based on a DNA oligomer having a structurecomplementary to a segment of enterococcal rRNA 112, 14. 311. Even with the growingrange of molecular techniques available to identify, characterize and type enterococci, thecho~ce of selecting the appropriate technique depends on the ~ndividual's need and theirsetup.7. GENETICS OF ENTEROCOCCIThe fr~quency and spectrum of multidrug resistant enterococcal infections have Increaseducirldwide dunng the last two decades This Increase has been attributed to a combinationof versatile genetic characteristics of enterncoccl, the selective pressure of antimicrobialuse. and envirnnmental changes that enhance the transmission of resistant organisms. Wewill briefly review the unique and versatile genetic mechanisms present in enterococci inthe following secttons.7.1. Gene Transfer Mccbanisms in EnterococcusIn entcmcocci primarily, acquirtd resistance to antibiotics occurs by horizontal transferof resistance genes located on various typs of mobile DNA elements (plasmids andtranspsons) and, secondarily 10 a lesser extent by mutations on the bacterial genome.The horizontal transfer of genetic elements involves the phenomenon of "conjugation"driven by a unique genetic mechanism present only in enterococci. Although in-vitmstudies on eIcctrotrmsfomation or E. ,/aat.o/is were tried (1131, natural gene transfer

RE VIEW OF LITERATUREmechanisms like transformation or transduction that are common in other gram-positive"rganisms are not well established in enterococci. Hence, "conjugation" remains to be theunique genetic mechanism present in enterococci for gene transfer.7.2. Role of Mobile Genetic Elements7.2.1 PlasmidsThe extra chromosomal mobile genetic elements like plasmids play a major role intransfening the genetic determinants encoding antibiotic resistance and virulence inenterococci in a hospital setup. Although different classes of plasmids have beendescribed in enterococci. the conjugative (pheromone-responsive) plasmids. and thebroad-host range plasmids are of clinical significance. The presence of plasmid DNA inS fuaculis was first reported in 1972, subsequently Jacob and Hobbs [I 141 presentedcv~dence for direct involvement of plasmid DNA in the conjugal transfer of multiple drugresistance from S. fat~culrs. Thereafter, several conjugative or non-conjugative butmohilimble plasmids In group-D-streptococci were discovered and characterized. Mostof the plasmids reporred encode determinants for resistance to antibiotics or UV light,hemolysin and bacterioc~n. Some of the medically significant plasmids, which have beencharacterized in detail. are plasmid pAMBl encoding eqzhromycin resistance [I IS].plasmid pADl encoding hemolysin-bacteriocin, plasm~d pAD2 encoding erythromycin,streptomycin and kanamycin resistance [I 161, plasmid pPDI, plasmid pCF-I0 encodingtetracycline [6], plasmid pIP800 encoding gentamicin. chloramphenicol and kanamycinresistance [I 171 and plasmid pAM373 [I 18).The pAMBl is a conjugative plasmid that determines eqzhromycin resistance.This resistance is representative of the so-called MLS phenotype (i.e., resistance tomacrolida, lincosamides, and streptogramin B). Plasmid pAMPl was originallyidentified in S. fueculjs strain DS5 [I IS], which exhibited a broad host range with itstmsferability into nine different species of streptococci, as well to Lactobacillus casei.Staphylococ~~~~ ourcus, and Bacillus subtilis [6]. The salient feature of this pup of

REVIEW OF LITERATUREplasmids is that they exhibit transferability at relatively lower frequencies only on solidsurfaces such as filter matings, and not in broth matings.The pheromone responsive plasmids like pADI, pPDI, pCFlO and pAM373 onthe other hand, transfers eficiently at a very high frequency (10' - per donor cell) inhroth matings. These plasmids confer mating response to small sex pheromones secretedby potential plasmid-free recipient cells. This mating signal induces synthesis of a surfaceaggregation substance by the donor cell, which elicits formation of mating aggregates andplasmid transfer between the donor and recipient cells (6, 11 81. Recent studies reveal thatthese pheromone responsive plasmids give rise to aggregation response when exposed tosub-inhibitory concentrations of antibiotics used in treating enterococcal infections [I 191.Apan from these two major types of plasmids. several other conjugative plasmidshave been reported from different laboratories and named accordingly. Of these, apl~eromone Independent plasmid pMGl encoding gentamicin resistance has beendescribed to exhibit transferability at a higher frequency during hroth matings from Japanwhich ~ndicates there exists an alternate gene transfer system. which differs from thepheromone- induced plasmid transfer system [120].7.2.2. TransposonsTransposons are gene sequences that can move from one locat~on on the chromosome toanother, the genes required for this movement being contained within the Transposon(Tn). All transposon classes have heen described in enterococci. of which the conjugativetransposons are exclusively found in E. faecalis predominating the other classes [6. 12 I].Hodel-Christian and Murray identified a 4.7 kb composite transposon, designated Tn5281In E. ,fuecalis which conferred HLGR by virtue of the 2 kb bi-functional geneaac(6')+aph(2") flanked by 1.35 kb IS elements, designated IS256 [122]. Rice et al. [I231reported a 26 kb composite transposon Tn5384 in E. ,foeculis. which conferred HLGRand erylhromycin resistance. A simple transposon Tn1546 belonging to Tn3 family,Possessing inverted repeats and encoding van-A type glycopeptide resistance wasdescribed in E. ,ficcium. Later the same group described a composite transposon Tn1547,

RE VIEW OF LITERATUREflanked by IS16 and IS256-like elements conferring vancomycin resistance in E. faecalis[124].While the former two classes of transposons jump between cells by the processtransposition, the conjugative transposons described predominantly in E. fueculisdoesn't jump, rather move between genomes through intercellular contact betweenentemcocci. Some of the well-characterized conjugative transposons in enterococci arethe resistance transposons Tn916, Tn917 (61. Tn1546 like transposon [I 251 and Tn154911261. Tn916 and Tn917 were identified originally in a clinical isolate of E, faecalisDS16. 'h9 16 is chromosome borne, conferring tetracycline resistance and has beenshown to insert into several sltes in different conjugative hemolysin plasmids like pADland pAMyl (61. while Tn IS49 encodes van-B type glycopeptide resistance inE~lrc.rococc~rs [ 1261.7.3. Conjugative Gene Transfer Mechanisms in EnterococcusThe s~gnificance of Enrcrococcus as an ascendant nosocomial pathogen lies in itspropensity lo acquire resistance to multiple antibiotics, used to treat enterococcal~nfect~ons. by a varie~y of conjugative mechanisms. The following sectlon briefly reviewsthe different conjugative gene transfer mechanisms described in enterococci.7.3.1. Pheromone-Inducible Plasmid Transfer in EnterococcusThe bacterial sex phemmone mediated conjugatton was described exclusively threedecades back only in the gmup-D-streptococci [127], as depicted in Figure 2. In agenerally accepted model of this system, plasmid free strains of E, faecalis typicallysecrete into the culture medium a number of different small peptide sex pheromonesspecific for different types of plasmids. When a potential donor cell containing apheromone-responsive plasmid (generally encoding antibiotic resistance) comes intocontact with its corresponding pheromone, transcription of a gene on the plasmid isturned on, resulting in the synthesis of a sticky substance called aggregation substance(4s) on ils surface, which can pmmote attachment to recipients via a complementary

REVIEW OF LITERATUREreceptor called enterococcal binding substance (EBS). The close cell-to-cell contactresulting from AS-EBS binding facilitated by random collision of cells, allows forsubsequent formation of some son of mating channels between the cells enabling thepheromone-responsive plasmids to transfer from the donor to the recipient (mostlyplasmid free strains) bacterium. Since a cell-free filtrate of a recipient also elicits anaggregation (clumping) response when it is mixed with donors, this substance has beenreferred to as clumping-inducing agent (CIA). During in-~itro experiments in test tubes,clumps of cells actually fall to the bottom of the tube, resulting in a visible aggregate.Once the recipient cell has acquired this particular plasmid, the synthesis of thecorresponding sex pheromone is shut off to prevent self-clumping. This system ofconjugation, which occurs primarily in E. faecalis, is highly efficient and results intransfer of plasm~ds in both filter and broth matings [6. 17, 127, 1281.Figure 2. Pheromone-responsive conjugative system of E. fueculi~.Pheromone A released from the potential rec~pient cell (right) interacts with plasmid A inthe potential donor cell (len) to induce synthesis of aggregation substance. Attachment ofaggregation substance to binding substance causes the cells to clump into visibleaggregates. Once the pheromone-responsive plasmid A has transferred from donor tomip~ent cell. synthesis of pheromone A is shut off.Figure reproduced with kind permission of Dr. B.E. Murray, MD.,University of Texas Houston-Medical School, Houston. Texas. U.S.A [I 71

REVIEW OF LITERATURESeveral studies worldwide have shown that clumping response was exhibitedpredominantly by the antibiotic resistant E. faecalis from hospitalized patients possessingpheromone responsive plasmids, and proved that these plasmids encoding antibioticresistance transferred between the experimental strains in-virro 1129-1351. Further thesepheromone responsive enterococci producing hemolysin-bacteriociniaggregationsubstance have been shown to enhance the pathogenicity of enterococci [136, 1371.7.3.2. Pheromone Independent Plasmid TransferAS reviewed previously, some broad host-range plasmids can transfer between species ofenterococci and other gram-positive organisms such as streptococci and staphylococci.The coexisting pheromone responsive plasmids can greatly increase the transferfrequency of these plasmlds even though these plasmids transfer at a lesser transferfrequency on sol~d surfaces like filter matings (1381. Recently, Ike and his colleagueshave described an alternate conjugative plasmid transfer system, which was "pheromoneindependent" but efliciently transferred the antibiotic resistant plasmid pMGl duringhroth matlngs [120, 125. 1391.7.3.3. Transposon mediated Conjugation (Conjugative Transposons)Several conjugative transposons reported to be prevalent in enterococci play a major roleIn dissemination of antibiotic resistance genes to many different species. Studies confirmthat such transposons may have evolved from a Tn916 ancestor; the first reportedconjugative transposon in enterococci 1122, 1251. Several transposons encodevancomycin and gentamic~n resistance in nosocomial enterococci, thereby enhancingglobal dissemination of these resistant traits. All these studies imply that the genetics ofcntertxoccl is much complex and versatile than perceived currently, and underscorestheir significance and impact in healthcare settings.

REVIEW OF LITERATURE8. ENTEROCOCCAL INFECTIONSg , ~ Types . and their Clinical Significnnce~nterococci lead commensalism in the gastrointestinal tract of humans and severalmammals. The transfonnation of this commensal into an opportunistic pathogen in healthcare settings remains enigmatic to alarger extent. Even though enterocwcalpathogenicity has been addressed a century back still several aspects of this nosocomialpathogen remains unraveled. Their intrinsic resistance to multiple antibiotics, and uniquegenetic mechanisms allows them to survive and proliferate in patients receivingantimicrobial chemotherapy. This accounts for their ability to cause superinfections inpatients receiving a number of broad-spectrum antimicrobial agents [140]. Although E./uc~culi.r and E. fuecitrrn account for more than 95% clinical infections [13]. otherenterncoccal species are repnrted to cause clinical infections rarely [16. 1411. Recentstudies show that enterococci are the third most common pathogen isolated fromblondstream infections, the single most frequently reported type of pathogen in surgical-site infections in intensive care units. and the second most common nosocomial pathogenIn the US [20] underscoring their clinical significance. The imbalance within the hostsystem (human body) disrupts the commensalism and makes it an opportunistic pathogen.Funher, several predisposing factors along with the virulence factors, facilitatessuccessful evasion of the host defense leading to an array of clinical infections byen~erncocci [26] Although the aspects of multidrug resistance and virulence needs to beaddressed concomitantly to elaborate their role in enterocwcal infections, this sectionrrviews the array of infections caused by enterococci.8. I. I. Urinary noo infedionsUrinary tract infections (UTI) an the most common of all infections caused byentemocci panicularly in hospitalized patients. It continues to be the second mostcommon nosocomial uropathogen accounting to an average 12-15% of nosocomial UTInexl only to E. cnli since two decades, as depicted by several studies conductedworldwide [13. 19, 38-40, 1421. Although E. colr has been topplng the list since decades,

REVIEW OF LITERATUREthe frequency of it's isolation has declined over time from 35.6% in 1996, to 32.5% in1998 and thereafter to 26.6% in 2001, on the other hand enterococci were isolated at anincreasing frequency over the years from 11.8% in 1996, to 15.3% in 1998, and thereafter22.0% in 2001 which authenticates it's emergence as a significant nosocomialuropathogen [ 1431.The increased prevalence of nosocomial enterococcal UTI is probably the resultof many disposing factors like increasing use of catheterization and broad-spectrumantibiotics mostly among patients of older age group as shown by several studies [38. 39,42, 1441, but lesser prevalent (< 5%) among younger age group where catheter relatedinfections accounted for upto 48% of all the nosocomial UTI and secondaly bacteremiaand cerebral palsy occurring rarely [41, 421. The crude mortality rate due to nosocomial(;TI remains to be approximately 18% where enterococci play a major role [19]. Further,reviews show that enterococci can cause prostatitis and perinephric abscess in addition touncomplicated cystitis or pyelonephntis 113. 681, Some studies had shown that some\ ~mlcnce factors like "enterococcal surface protein-esp" increase the pathogenicity of theurinary enterococcal lsolates thereby altering the outcome of the infection [145].Apart fmm independent laboratories, several official bodies (SENTRY, SCOPE)across the globe conduct multicenter antimicrobial Surveillance Programs to assess thechange in the bacterial profile and pattern of antibiotic resistance of nosocomial urinarytract infections, which leads to more effective prescribing practices for entemcoccal UTI.119, 38-40, 1431. Results of most of these studies show that ciprofloxacin andnttrofurantoin remains to be ideal antibiotics for treating enterococcal UT1, althoughsome studies raise a concern on this option 11461. Fraimow et al. [I471 had proven thatsome E, faecalis isolates requires vancomycin for growth, a phenomenon which couldhave evolved due continuous exposure to high concentrations of vancomycin in the urine,that narrows down the treatment options in case of multidrug resistance by enterococci.Thus by and large cnterococci continues to play a a major role in the nosocomial UTI.

REVIEW OF LITERATURE8.1.2. BacterernbThe bacteremia exists in two fonns in hospitalized patients: enterococcal bacteremiawithout endocarditis, and bacteremia with endocarditis. The management and mortalityof either form of enterococcal bacteremia to a larger extent depends on the antimicrobialsusceptibility profile of the isolate, apart from other predisposingirisk [actors. Severalstudies over these years elucidates that enterococci remains one of the top three-nosocomial bloodstream pathogen till date since a decade [20, 21, 371. Although E./uc~culis remains the predominant species causing enterococcal bacteremia followed by E./accium, other species like E. orium. E. cusseliflu~~us, E. durons, E. gullinorum and E.ruffinoslls have been reponed to cause bacteremia although very rarely [90, 141, 148-IS I ] Only 2% cases of enterococcal bacteremia are associated with endocarditis, whichare mostly community acquired [68]. However. few studies depicts that 8-32% ofenterococcal bacteremia associated with endocarditis were predominantly communityacqu~red, but their case-selection criteria were totally different [23, 241.Although high-level gentamicin resistant bacteremia has been described since twodecades, several studies depict the emergence of vancomycin resistant enterococcalbacteremia among imrnunocompromised hosp~talized patients since last decade in U.Sand several European countries [I 8.20. 1.521. But the frequency of vancomycin resistantenterococci causing systemic infections in developing countries like India is not veryhigh and alarming, as in west [153. 1541 A prolonged hospital stay by patients withsevere underlying diseases and life threatening conditions remains to be the mostcommon riskipredisposing factor associated with bacteremia as depicted by severalstudies. The diseases and conditions generally include cancer. diabetes mellitus, chronicrenal failure, major trauma, surgical procedures and urinary and vascular catheterization(22-24. 150. 155- 1571. While patients who had previously received broad-spectrumcephalosporins, gcntamicin/vanwmycin are at greater risk of acquiring enterococcalbacteremia [36. 44. 47, 155. 158, 1591. Mostly enterococcal bacteremia tends to bepolymicrobial isolated along with staphylococci and enteric gram-negative bacilli. Theinfection by enterococci mainly occurs in the form of urinary tract infection and

REVIEW OF LITERATUREsecondary bactmemia mostly related to abdominal and soft tissue infections, with E./aL.ca/is as a predominant species in all foci [22,47].Mortality associated with enterococcal bacteremia remains high, since it occursperally in debilitated patients. However, it is hard to define the exact role of bacteremiain the fatal outcome of these patients, as several other complicating factors too areaccountable in these cases. Several studies worldwide show that mortality ranges upto50% in most cases, although there are exceptions with antimicrobial resistance andvirulence determinants of enterococci playing a sipificant role in the outcome (152,1601. In a comparative study mortality rates of 38.3% and 30.5% were shown for patientsw~th enterococcal bacteremia due to strains with and without high-level resistance togentamicin respectively [149]. Another study depicted that 34% patients with seriousunderlying illness like bums, malignancy or granulocytopenia, diabetes and renal failuredied with enterococcal bacteremia [23], while Malone et al. 1241 showed a mortality rateof 44% and found that male sex and a fatal underlying disease were significantlyassociated with increased mortality. Huycke et al. (1611 showed that patients withhemolytic, high-level gentamicin-resistant E. farcu1r.s bacteremia showed a five foldIncreased (3 1%) mortality rate within 3 weeks of bacteremla compared with patients withnon-hcmoly~c, gentamicin-susceptible strains in a cohort study. Thus the mortality rateson a whole depend on the cohort group studied. apart from other influencing factors.8.1.3. EndocarditisInfective endocarditis remalns to be the most senous of all infections caused byenterococci with a higher monality rate. Wtcrococctr.~ remains third most common causeand account for approximately 5-20% of all cases in infective endocarditis, but may varyaccordingly [13. 162-1641, Several studies have documented E. faecalisas thepredominant species causing enterococcal endocarditis, although other species too havebeen reponed to cause this disease. Among isolates sent to the Saeptococcus laboratoryat the Centers for Disease Control, Atlanta, endocarditis was the diagnosis given forPatients hrn whom E, avium. E. cosseli/lo~us. E, durans. E. gallinamm, and E.ra/linmus wen isolated, apart from E. ,foecolis and E. ,faecium [go]. Although most cases

REVIEW OF LITERATUREare predominantly community acquired, recent studies show a steep increase in thepercentage of enterococcal endocarditis acquifed in a hospital setup [165-1671.While several factors have been depicted as predisposing factors for enterococcalendocarditis, the male patients of older age with prior urinary tract infection or urinaryinstrumentation and bacteremia are more prone to this disease when compared with theirfemale counterpart. as reviewed [13, 1621. Several virulence determinants andantimicrobial (gentamicin) resistance in enterococci were shown to predict the outcomeof this disease by altering the mortality rate [129, 165, 1671. Hence early diagnosis ofenterococcal endocarditis is very imponant to initiate appropriate antimicrobial therapydepending on the susceptibility profile of the isolate based on the blood culture report.8.1.4. Neonutul infectionsThe emergence of enterococci among the neonates as a significant cause of nosocomialinfections has been depicted by many studies since last decade. The results of a recentpoint prevalence survey of nosocomial lnfect~ons in 29 Pediatric Prevention NetworkNlCll conducted by CDC in U.S. has shown that enterococci ranks second only tocoagulase-negat~ve staphylococci as the leading cause of nosocomial infections ofnennates in ICU [168]. On the other hand, the emergence of vancomycin resistantentcrcxocci (VRE) In neonatal infections as shown by other recent studies poses alherapeutic challenge and emphasizes the urgency for more effective prevention~nterventions[ 169- 17 1 1.Sepsis remains to be most commonly reported from neonates, with focalinfections including meningitis. scalp abscess and pneumonia in this group [169-1781.The most common predisposing factors for neonatal enterococcal infections as reportedby recent studies are nosocomial acquisition due to prolonged hospital stay, long-termantibiotic therapy, low birth weight, pre-term birth, central intra-vascular catheters,prolonged ventilation and patients receiving total parenteral nutrition [172, 174, 1751.Few reports show that cntcrococci causes other neonatal infections like urinary tractinfection accounting to 15% [4]], and conjunctivitis and peritonitis although less

REVIEW OF LITERATUREfrequently [13]. While the mortality rates of neonatal enterococcal infections, especiallythose of sepsis ranges on an average from' 8-18% with higher rates for necrotizingenterocolitis-associated infection [172, 1731. While appropriate infection controlmeasures have shown to reduce the incidence of neonatal enterowccal colonization andinfections from 67% to 7%. which emphasizes the impact of good infection controlpctice [l71]. Most of the neonatal enterococcal infections especially those of sepsis,respond well to appropriate antimicrobial therapy according to the susceptibility patternof the isolate.8.1.5. Intra-iabdominal and pelvic infectionsThe clinical significance of enterococci in cases of bacteremia and super-infections inselected patient populations has been well established. But, being a resident flora of theGI tract in humans. their role as primary pathogens in polymicrobial intra-abdominal andpelv~c infections remains controversial. Although the clinical significance of enterococciIn intra-abdominal and pelvic infections remains murky, several reports depict their roleIn peritonitis. intra-abdominal or pelvic abscess, surgical site infection, suppurativethrombophlebitis.acute salpingit~s. and endometntis [179-1841. Although surgicaldrainage remains the cornerstone of therapy for enterococcal infections involving ad~screte focus, in the circumstances typified by the compromised surgical patient, specificantibacterial therapy directed against enterococci based on their susceptibility profileclearly reduces the mortality rate [179, 18 I].8.1.6. Skin and sop tissue infectionsEnterococc~ rank the second most common gram-positive pathogen, and third mostcommon isolate associated with skin and soft tissue infections (SSTI) in hospitalizedpatients next only to S. uureuv and E. coli [185-1871. These studies conclude thatenterococci account for approximalely 8-10% of SSTls in hospitalized patients, althoughthere are variations in different geographical areas. Like in intra-abdominal and pelvicinfections, the role of cnterococci remains controversial in SSTIs since they arefrequently isolated Fmm mixed cultures with staphylococci and gram-negative bacilli insuwcal and bum wound infections, decubitus ulcers, and diabetic foot infections. An

REVIEW OF LITERA 77JREenter~~~~al bacteremia accompanying any of these infections authenticates the clinical,ignificance of enterococci in SSTls [13. 22,' 1861. While multidmg resistance amongenteroc~~~i varies geographically, vancomycin-resistant enterococci among SSTls areuncommon outside the USA (17.5%) and Italy (7.4%) and the susceptibility profile ofenteroco~~i need to be considered when selecting therapy for treating enterococcal SSTls,since empirical therapeutic options like broad spectrum cephalosporins are ineffectiveagainst enterococci [I 851.8.1.7. Miscellaneous InfectionsThe involvement of enterococci in other infections appears although exceedingly rarely.Enterc~occal meningitis remains to be more common in neonates than adults [173, 1781.Although rarely reported it generally seems to be related to an underlying disorder whichrends to be a complication of bacteremia in patients with endocarditis, or severe~mmunodeficiency syndromes [13. 1771. The reports of respiratory tract infections likepneumonia and lung abscess shows that it appears In hospitalized patients who wereseverely debilitated and ventilated, rece~ved prolonged antimicrobial therapy coupledw~th enteric feedlng [ 173). Enterncocc~ were rarely reported to cause other infections likeotitis, osteomyelitis, and septic arthritis although their role tn these infections remains tohe murky [188, 1891. While most of the cases are treatable with appropriate antibiotictherapy. the emergence of multidmg resistant enterococci has narrowed down thetreatment options.8.2. Management of Enterococcal InfectionsEnterncocci are intrinsically resistant to many antimicmbial agents and readily acquireadditional resistances, which have propelled them to become a prominent nosocomialpathogen [I 1, 131. Although the treatment of enterococcal infections depends on severalfactors, the major factor to be considered is the antimicmbial susceptibility panem of theisolate for choosing optimal therapeutic regimens for the treatment of infections causedby muttidrug resistant strains. The standard treatment since several decades, for seriousinfections like mtcrococcal endocarditis and meningitis. tends to be a bactericidal

REVIEW OF LITERATUREantibiotic therapy, which includes a cell-wall active agent(penicillidvancomycin) and an aminoglycoside (gentamicid streptomycin) if the isolateis susceptible to both the antibiotics, while bacteriostatic agents are sufficient to treatmost other enterococcal infections. Treatment of infections caused by strains resistant tobeta-lactams, glycopeptides and aminoglycosides has become problematic due to thel~mited number of therapeutic options. No medical therapy is reliably effective forendocarditis caused by strains resistant to all cell wall-active antibiotics and allaminoglycosides [I 1. 13, 140, 1901. Although recently introduced novel antimicrobialagents such as linezolid and quinupristiddalfopristin are effective in managing multidrugresistant (including vancomycin) enterococci, their activity that is mainly bacteriostaticraises a concern at times [191].Ciprofloxacin and norfloxacin have been used successfully for enterococcal UTI,but the organism's susceptibility is marginal for treating systemic infections with thisdrug [ I I. 13. 140. 1901. Most of the serious infections, caused by VRE were successfullytreated with newer antimicrobials like linezolid and quinupristinidalfopristin, asmonotherapy, or In combination with other antimicrobials [57, 1911. While Murray 1571has proposed a schema (Figure 3). depicting the possible antimicrobial-drug regimens forthc treatment of clinically important ~nfections caused by enterococci resistant to bothampicillin and vancomycln. which was based on a review of the literature relating toVRE, personal observations and current clinical practice.The colonization of the intestinal tract with VRE may predispose patients toinfections by this organism and may contribute lo its nosocomial spread. Thus anempts toeradicate VRE from colonized patients were tried with different oral antimicrobialregimens, ~ncluding bacitracin, novobiocin and a novel antimicrobial agent ramoplaninthat have has shown partial success [192]. In addition to systemic antibacterial therapyand decolonization efforts, adequate debridement of devitalized tissues and drainage ofabscesses and fluid collections remains to be an integral pal of appropriate managementof enterococcal infections [I 791. Thus, this enigmatic pathogen presents a stifftherapeutic challenge ahead for health care professionals.

REVIEW OF WTERA TUREFigure 3. Antimicrobial-drug regimens for the treatment of drug resistantentnococcal infections [resistant to both Arnp~cill~n and Vancomycin]MC dmmprllnM lb fdlanng.Olnupnmln dmlbpnan.l*ld.r Invr*).lan bnad'duetntmcn qumupmmd.lbpnllnrnpnllnbpmwcn LYplaB. Wbn .r.W IW+npsnemnn or smparyanHgh do"amprclln nrc mlwlmr olllr Idk*n,,(.r\rnp~#l!n ~n~pmm11 .ndOC-d~ln,,I tu~hghh ,wan.ludy rrelght h sv-buothbh, k be rurrtub re!*lurrd!1t+. C-ry.nl.mnn u .lnptomvn.*a*an tmslol,otu~ .o.~L +I.Cons&, rabb lplrrmrntFlow-chart reproduced with kind permission of Dr. B.E. Murray. MD.,University of Texas Houston-Medical School. Houston. Texas, U.S.A [57]9. PATHOGENESIS AND VIRULENCE OF ENTEROCOCCI9.1. Pathogenesis of EnterococciAntimicrobial resistance has been cited as one of the major reasons for evolution ofentmocci as a prominent nosocomial pathogen, however there are evidences forentemcoccal infections a century back even before the introduction of antibiotics [13].This raises a concern about the versatility of entemocci to cause infections even withoutantimicrobial resistance. The pathogenicity of enterococci was addressed a century back-

REVIEW OF LITERATLIREin 1899, by MacCallum and Hastings, who isolated an organism from a case of acuteendocarditis, and designated it Micrococcus' zymogenes based on its fermentativeproperties. It was also found to be lethal when injected intraperitoneally in white mice,and capable of producing endocarditis in a canine model. In subsequent years, studies andreports had shown that this "fecal streptococci" was responsible for infections likeendocarditis, UTI, and wound infections [13,26].The difficulty in treating enterococcal infections has driven several researchers tobcus their efforts in studying the factors that undermine the commensalistic relationship,since these factors in one-way or other contributes to the establishment of the organism ina particular ecological niche. Infection occurs once when the host-enterococciequilibrium is disrupted aided by various factors. In order to infect, enterococci must firstbe able to colonize, primarily by adhering to the host mucosal surfaces with the help ofadhesins and surface carbohydrates. After adherenceicolonization the organism must thenevade the host clearance to cause any infection and ultimately produces pathologicchanges in the host with the help of an array of (virulence) factors [26. 1931. While mostol'the colonizers or the commensals. do not posses all the (virulence) factors needed forsuccessful evasion and subsequent infection process, those that posses emerge to becomea successful nosocomial pathogen facilitated by other predisposing factors. Our reviewelaborates the incidence and role of various abovementioned virulence factors in thepathogenesis of enterococcal ~nfections in the following section.9.2. Virulence Factors of EnterococciAlthough the pathogenicity of enterococci was described a century back, the significanceof' enterococcal virulence in several cases is still debated. Unlike other gram-positivepathogens like S. aureus and S~reprococcus species, enterococci do not posses classicvirulence fac~ors. But their predominant role in nosocomial infections with their intrinsicand aquid antimicrobial resistance had provoked researchers to study the role ofpossible virulence factors in the pathogenicity of enterococci. A brief review of the

REVIEW OF LITERATUREvirulence factors proposed and studied in enterococci pertinent to humaninfections, are elaborated in the following sections.9.2. I. Aggregation Substance - ASAdhesive properties act as virulence factors in the pathogenesis of enterococcal infectionslike UTI and endocarditis. Studies suggest that UTI strains showing the highest invasionand adhesive potential invade the kidneys, cause bacteremia. and, after having expressedthe serum-dependent surface modification, colonize the heart [194]. Aggregationsubstance (AS) remains to be the most well studied adhesin of enterococci. AS is apheromone-inducible surface protein that mediates binding of donor cells to plasmid freerecipients. and is essential for high-efficiency conjugation of sex pheromone plasmidsand also acts as a virulence factor during host infection. Several different functions havebeen attributed to AS in addition to bacterial cell aggregation as reviewed previously 161.The other major function exhibited by AS, is adherence of the E, fueculir isolates to hosttrssues. In vitm studies have shown that AS mediates adhesion to various eukaryotic cellsurfaces, such as cultured pig renal tubular cells. and promotes internalization by culturedhuman intestinal cells, suggesting that AS-expressing cells may likely form largeraggregates in vivo than cells not expressing this trait [195-1981. While Hirt et al. [I371showed that AS contributes adhesion to fibrin and increased cell surface hydrophobicity,which may induce localization of cholesterol to phagosomes and prevent or delay fusionwith lysosomal vesicles. Furthermore, AS exhibits resistance to killing bypolyrnorphonuclear leukocytes and macrophages, thereby promoting intracellularsurvival of E. ,faecalis inside neutrophils [198, 1991. AS was also studied in the rabbitmodel of E. Jaeculis endocarditis and found to be associated with greater vegetation sizecompared to vegetations caused by isogenic AS-defective strains, although theseinfections were not observed to be lethal [200]. The same study has shown that mostcytolytic strains of E. ,faecalis also express AS, by which both the virulence factors maywell work synergistically in the host system [ZOO]. Jea et al. [I361 have shown that E.fuecalis strains bearing AS were pathogenic in endophthalmitis models, while studies bylsenmann a al. 1201, 2021 have shown that interaction of fibronectin and aggregationsubstance promotes adherence of E. (aecalis to human colon.

RE VIEW OF LITERATURESeveral researchers in different geographical locations worldwide have studietlthe incidence of the virulence factor "AS" among clin~cal as well non-cli~~ical isolates toprove this factor contributes substantially to the increased pathogenicity of enterococcl[25. 129, 203. 2041 Most of these studies revealed that AS was prcscnt or~lv among 1firL,~,trlis and the prevalence varied between 20-60°/u based on the geopraplncal location aiwell the type of infections from which cnterowcci were isolated tlowcver the role of ASin adhesion and virulence of enterococci remains debatable since a recenl study h~Johnson el al [205] has shown that AS and binding substance are not inalor contributorsto urinary tract colonization in a mouse model ol'ascending uliobstr~lcted (!TI9.2.2. Enterococcal Surface Protein - EspA recently identified cell wall-associated protein of E. fuecolis called Enterococcalsurl'ace protein-Esp has been shown to act as an adhesin like aggregation substance,which revcaled a significant enrichment in infection-derived E. foecalrs isolates. Funhers~ructurnl analysis of the gene revealed that Esp exhibited global organizational similarityto the Rib and C alpha proteins of group B streptococci [206]. In their study Shankar etal 12061 did not identify the Esp gene in any of non-E. fu~colis species. But later studiesshowed that a subpopulation of epidemic vancomycin-resistant E. ,faecium isolatescontained a varlant of the Esp gene that was absent in all non-epidemic and animal~sc~lates 12071 Following their study Esp variants, were reported in E, forcium clinical~solates from different parts of the world [208, 2091. Subsequently Shankar et al. 11451using an animal model of ascending urinary tract infection showed, while Esp does notinfluence histopathological changes associated with acute urinary tract infections, butcontributes to colonization and persistence of E, fuecolis at this site.Later Esp was shown to be highly associated with the ability to form a biofilm atabiotic surfaces [210]. The Esp induced biofilms have increased antimicrobial resistanceand are important in infections involving catheters [21 I]. But the result of a recent studycontradicts the earlier hypothesis. The authors demonstrated that in vim biofilmformation occurs, not only in the absence of Esp, but also in the absence of the entirepathogenicity island that harbors the Esp coding sequence, and concluded that E. faecalis

REVIEW OF LITERATUREforms complex biofilms by a process that is sensitive to environmental conditions anddoes not require the Esp surface protein [212]: Another study showed that endocarditisisolates of E. fuecali.~ produced biofilm significantly more often than non-endocarditisisolates. Furthermore, their results showed that Esp was not required, but its presence wasassociated with higher amounts of biofilm (2131.Waar et al. [21 I] hqpothesized that Esp might be associated with colonization andspread, because it was more frequently isolated from feces of healthy volunteers andtransplant patients, a fact, which was supported since an epidemic Esp gene-positive E./ueculi.s strain was found among liver transplant patients. Shankar et al. [214] showedthat Esp along with another virulence factor cytolysin was a pan of a pathogenicity islandin E. /ueculis. Recently, Leavis et al. [215] had described a novel putative enterococcalpathogenicity island in E. faccium for the first time, which was shown to be linked to thevariant Esp virulence gene of E. faetium and associated with epidemicity. This wascompletely different from the Esp-containing pathogenicity island previously disclosed inI: fuc~c.ulrs In another study, infection-derived E /uccium strains enriched with Esp had~ncreased ability to adhere to Caco-2 cells and were less genet~cally diverse than Espnegativeisolates. But the authors indicated that additional factors are of importance foradhesion, since Esp-negative E. ,fuccium fecal isolates from healthy individuals adheredsignificantly better than Esp-negative infection isolates [216].Se\eral studies have shown that the "Esp" vilulencc Ihctor was pre\aIzii~ arnoi;i_?0-70% of enterococci and tllc prevalence rate varied accordilly to the geo@rapllic;~llocation, as well other contributing bctors, wlllle the prevalellce was pltdominant in I:,htc.c.ulrs than in 1.. ,j~tec.rrrrn isolates I 157. 2 17, 2 I X] Most of tlic L'SI) positive strains wereresistant to one or more antibiotics [2 11). 2201 Samc studies have shcjw11 that Esp gclieplays a major role in dissem~nation of vancnmyci~l-resistant I:.predominant epidemic strains harbored the Esp penc. whilc mc~st ot'f~r~-c'rrllrl. sitice11o1l-c01dcrn1cstrains were Esp negative (220-2221. A recerit study bv Oancea et al (?2.;1. demonstratetlin vitro conjugative transfer of the Esp gene among I*,: ,~wc~itmr and 1,'. ,/irc,c.iilr.c fromclinical samples reemphasizing the ~i~nificattce of Esp in nosocorninl settings

RE VIEW OF LITERATURE9.2.3. Adhesin of Collagen from E. foecalis -Ace~ce-adhesin is a 74kDa protein identified. in E. jaecalis, which is significantlyhomologo~s to the collagen binding protein- Cna, of S. aureus [224]. Subsequentcharacterization of the Ace adhesin revealed, that apart from mediating attachment to~xtracellular matrix (ECM) collagen proteins, the Ace proteins were produced duringenteroc~~~al infections and was confirmed since anti-Ace antibodies were detected infour different molecular sizes in Western blots [225]. Recently researchers from the samelaboratory had identified a new collagen-binding adhesin of E. ,faecium- Acm, thatshowed 62% similarity to the S. aureus collagen adhesin-Cna, than to the E. ,faecaliscollagen-binding adhesin-Ace. Their results demonstrate that Acm, which encodes apotential virulence factor, is functional only in certain infection-derived clinical isolatesof E, fuecium, and suggest that Acm is the primary adhesin responsible for the ability ofE, fuc~cium to bind collagen [226]. Even though Ace adhesin have proven role in adhesionof ECM, further multicenter studies are needed to unravel their exact role in enterococcal~nf'cctions.9.2.4. Cyiolysin/BacteriocinAfter the initial adherence to host tissues, enterococci invade and cause systemiclnfect~ons and modulates the host inflammatory responses with the help of potentiallytox~c secreted products causing direct tissue damage. The major factors well studied toenact this role. are the secreted factors cytolysin and the gelatinase (zincmetalloprotease), which contributes to the seventy of enterococczl infections [26, 227,2281. The cytolysin expressed by some strains of E. ./a~colis is a unique bacterial toxinthat is distantly related to ]antibiotic bacteriocins, a family of small, post-translationallymodified antimicrobial peptides. In addition to toxin activities, the cytolysin of E.lacca1i.s possesses bacteriocin activity against a broad range of gram-positive bacteria.which may provide a selective advantage for E. faeculis strains expressing this trait. Theoperon encoding the cytolysin and the bacteriocin, are either encoded on a plasmid, orintegrated into the chromosome [26, 2291.

REVIEW OF LITERATUREAlthough the significance of enterococcal cytolysin remains high in a clinicalsetup, it was the bacteriocin production by enterococci reported half century back [loo],[hat triggered researchers to probe into the molecular basis of this property. The study byrock et al. [loll depicted that 50% of the E. fueculis strains expressed a bacteriocinactivity against gram-positive bacteria, but not gram-negative bacteria. They alsodemonstrated that the cytolytic and bacteriolytic activities were simultaneously lost bysome E. ,/oeculis strains after exposure to UV irradiation, but both the properties weresimultaneously regained upon reversion and concluded that a single enterococcal productis responsible for cytolytic and bacteriolytic activities [100, 101]. Bacteriocins althoughlnlt~ally explored for their antagonistic effects in bacteriotherapy, attracted muchattention in subsequent years because of their potential use as food preservatives. Severalstudies were carried on food and other non-clinical isolates of enterococci, to study theslgn~ficance of bacteriocin production and to assess their biotechnological potential asbod and feed additives, and were found to possess the same [25.204.230].On the other hand, the studies carried out in clinical settings focused more on thesiyl~ficancc of cytolysin property, while only few studies explored the role ofhacteriocins exclusively among clinical isolates of enterococci. An initial study hadshown that there was a relation between bacteriocin production and virulence of E.luccu1i.c. which was later confirmed by many other studies [231]. Galvez et al. [232]round that among 90 enterococcl strains of human origin, 36 strains producedhacteriocins. Later Libenin et al. [233] screened the clinical isoiates for bacteriocinproduction, and concluded that hemolysinhacteriocin produced by enterococci could beconsidered as a marker of pathogenicity. Del Campo et al. [I041 had shown thathacteriocin was more among vancomycin-resistant enterococci, than vancomycinsusceptibleisolates of diflerent origins. The preliminary results of our study showed thatscreening of enterococcal isolates for bacteriocin production, may be useful to elucidatethe virulence of nosocomial enterococci, since the property confers an ecologicaladvantage for the producer strain [234].

REVIEW OF LITERA 77JREUnlike bacteriocins, the role of cytolysin in the pathogenesis of enterococcalinfections has been well established and several studies have shown that the pADIencodedcytolysin contributed to virulence in E. faecalis [26, 200, 2281. Even before theof the cytolysin the vimlence of enterococci was studied in a mice modelas hemolysin, which depicted their pathogenicity. The same authorsshowed a high incidence of hemolysin production by E. .farcalis strains associated withhuman parenteral infections [US]. Diagnostically, this toxin causes a beta-hemolyticreaction on human and horse blood agar, but does not hemolyze sheep blood agar, whichIS frequently used in clinical microbiology laboratories. Various studies on enterococciisolated from patients with different infections showed that cytolysinihemolysinlbacteriocin occurred at a frequency of 10-60% 1129, 152, 160, 161, 235, 2361. Most of~hese studies emphasized that presence of these virulence traits played a role in thepathogenicity of entemocci.9.2.5. GelarinasdProtease(ielatinases are proteases of enterococci, which were recognized a century back when theproteolytic and the hemolytic activities were used to divide the Strcprococcus furculisinto subspecies. The proteolytic isolates were designated as S. Juecul~~ var. liqu

REVIEW OF LITERATURE9.2.6, Miscellaneous FactorsFew other factors have been shown to contribute to the pathogenicity of enterococci.E ,fuecolis antigen - A (EfaA) contributes to the host tissue adhesion as well for speciesidentification of E. ,faeca/is isolates [106]. A capsular polysaccharide antigen (cps) fromclinical isolates of E. .faecolis and E. .foeciurn was shown to contribute to the pathogenesisof enteroco~ci by enabling them to survive the phagocytosis within the host system.Further, antibodies to this capsular polysaccharide (cps) were shown to have prophylacticand therapeutic efficacy 12421. Other studies show that extracellular superoxide,lipoteichoic acids, and few other cell wall components contribute to the pathogenicity ofenterococcal infections by modulating the host immunity [243, 2441. While all thesestudies emphasize the significance of these virulence factors in enterococcal infections,the possibility to explore these newer therapeutic targets would be a promising area tolook upon, since enterococci has become tough to crack due to their ruggedness to allavailable antimicrobials currently.10. ANTlMlCROBlAL RESISTANCE IN ENTEROCOCCIThe property of antimicrobial resistance has complemented the emergence of enterococcias a predominant nosocomial pathogen since last two decades. The rapidity with whichenterococci exhibit resistance even to the recently introduced novel classes of antibiotics:linezolid and quinupristinidalfopristin. along with glycopeptide resistance. has become amajor issue of serious concern. especially in nosocomial settings [17. 57. 2451. Theantimicrobial resistance in enterococci can be classified into two types:I) InherenUlntrinsic resistance and.2) Acquired resistanceThe details of both types of resistance are reviewed in the following sections.10.1. lntrln~ic ResistanceThe inherent or intrinsic antimicrobial resistance is a property that is present in all, ormost of the strains of enterococcal species. The genes for intrinsic resistance, like other

RE VIEW OF LITERA 77JREspecies characteristics, appear to reside mostly on the chromosome. These intrinsic traitsexpressed by enterncocci enable them to exhibit'resistance to p-lactams, especially semisyntheticpenicillinase-resistant penicillins, cephalosporins and aztreonam, as well to lowlevels of aminoglycosides and clindamycin [13, 57. 681. The reports of treatment failurewith penicillin in enterococcal endocarditis were reported soon after the introduction ofpenicillin in the early 1940s. since MIC of E. faeculis and E. faecium were generallybetween 2 to 8 pglml and 16 to 32 pgiml respectively, which were at least 10 to 100times greater than those for most streptococci proving that enterococci were considerablyless susceptible to penicillins than streptococci [57]. Later, studies showed with evidencethat the intrinsic resistance of enterococci to b-lactams (penicillins) is a characteristicfeature of these organisms and it appeared due to low affinity of the pen~cillin-bindingproteins [246-2481 Most of these studies have shown that the MIC of penicillins was onlybacteriostatic on enterococci. and very higher concentrations (>I00 pg/ml) of p-lactams(penicillins) were required in order to be bactericidal. While none of the cephalosporinstested show activity against enterococci at least in-vivo, making them ineffective intreatlng enterococcal Infections [ 13, 571.Entemcocci exhibit intrinsic resistance to low-levels of aminoglycosides andclindamycin, with MlCs for various aminoglycosides ranging between 8 to 250 pg/mlgenerally, although occasionalexceptions were also reported. The low-levelam~noglycoside resistance among enterococci appeared to be due to low uptake of theseagents by enterococci, which is associated with the proteins involved in electrontranspon. While some E. ,faccium strains were reported to show higher MICs for specificgroup of aminoglycosides than E. ,faeculis, and the combinations of penicillin plus theseaminoglycosides fail to show synergism against those E. ,faecium strains. [13. 57, 140,249, 2501. Although intrinsic resistance is exhibited for low-levels of aminoglycosides,high-level resistance is of more clinical significance and tends to be acquired and aregenerally encoded by more than one plasmid [251].The intrinsic resistance to the glycopeptides- vancomyciniteicoplanin is exhibitedby few species of entemcocci. The genes encoding the VanC type of vancomycin

REVIEW OF LITERATUREresistance are endogenous, species-specific components of E. gallinamm encoding vanC-I gene and E. casseliflovus/E. flavescens encoding vanC-21vanC-3 genes respectivelywhich enables them to exhibit low levels of vancomycin resistance. The MICs are muchlower when compared to other types of acquired inducible vancomycin resistances,ranging from 2 to 32 and 0.5 to 1 pg/ml for vancomycin and teicoplanin respectively [13,571. While some reviews report that wild tqpe strains of enterococci exhibit intrinsic lowlevel resistance to fluoroquinolones, clindamycin, lincomycin, and trimethoprimsulfamethoxazole(TMPISMX) due to reduced drug accumulation either by decreasingthe drug uptake or Increasing eflux of the drug [I 31.10.2. Acquired ResistanceThe most important type of antimicrobial resistance exhibited by enterococci is theacquired resistance either due to a mutation in the existing DNA (that occurs lessfrequently), or by acquisition of new DNA, such as plasmids or transposons (that occursvery commonly in enterococci). The conjugative gene-transfer system plays a major roleIn acqulsltlon of plasmids or transposons encoding resistances to different antibiotics asrev~ewed previously. Enterococci exhibit acquired resistance to an array of ant~microbialsas reviewed in the following sections [I 3, 571.10.2.1. Resistance to Chloramphenicol, MLS, Tetracycline and FIouroquinolonesThe transferable chloramphenicol resistance encoded by plasmid was reported in E.luecalis, and upto 40% of clinical isolates of enterococci exhibited chloramphenicolresistance mediated by chloramphenicol acetyltransferase [252]. Two decades back, in1974 Clewell et al. [I 151 first demonstrated that a broad host range conjugative plasmidpAMBl encoded erythromycin resistance in a S. ,fuecalis strain, which was arepresentative of the so-called MLS phenotype (i.e.. resistance to macrolides,lincosarnides. and streptogramin 8). Some studies showed that erythromycin resistanceexists on different determinants (ermB gene) canied by Tn917 that is widespread inhuman and animal isolates of enterococci [6]. The strains exhibiting MLS phenotype, in

REVIEW OF LITERATUREaddition to erythromycin resistance confers high-level resistance to clindamycin [6]leading to serious therapeutic problems.Although less significant, studies have shown that different genes like tetL, tetM,tetN tetO and tetS encoded by plasmidsiconjugative transposon (Tn916) exhibitedtetracycline resistance. These genes mediate tetracycline resistance either by active effluxof tetracycline from cells, or by a mechanism that protects the ribosomes from inhibitionby tetracycline [6, 131. While studies have depicted that some E. jaecalis and E. ,faeciumisolates exhibit flouroquinolone resistance by mutations on the genes encoding DNAgyase and topoisomerase IV leading to alteration of the target site [253]. The versatilityof enterococci has been authenticated further, since some clinical isolates of enterococciexhibit resistance to the recently discovered novel class of antibiotic calledoxa7olidinones (linezolid) by mutations of the target site [245], which has created aserlous impact in management of vancomycin resistant enterococcal infections for whichthis newer antibiotic was reserved.10.2.2. Acquired Aminoglycoside ResistanceApart from intrinsic (low-level) resistance to aminoglycosides, many clinical enterococciwere shown to exhibit acquired high-level resistance to aminoglycosides (streptomycinand kanamycin) three decades back, with MlCs generally >2.000 pg/ml, andconcomitantly resistant to synergism with cell-wall active agents (penicillins) [249, 2501.Subsequently Jacob et al. [I 141 demonstrated that high-level resistance to streptomycinand kanamycin were carried by conjugative plasmids, after which the acquiredaminoglycoside resistance was proven to be transferable by many other reports. andshown to be mediated by both streptomycin adenylyltransferase and aminoglycosidephosphotransferase [I 171. In 1979 Horodniceanu et al. reported the first high-levelplasmid-borne resistance to gentamicin in three strains of S~reptococcvs .faecalissubspecies zymogenes, which were also resistant to kanamycin, sisomicin, netilmicin, andtobramycin, to macrolide antibiotics, chloramphenicol. and tetracycline, and were able totransfer the plasmids encoding d ~ resistance g canied by them [254]. Since then several

REVIEW OF LITERATLIREstudies were conducted to reveal the genetics of the aminoglycoside resistance inenterococci.Generally, aminoglycoside resistance in gram-positive cocci like enterococci isdue to synthesis of enzymes, which modify the antibiotics [8]. Table 2 depicts variousaminoglycoside resistance genes prevalent in enterococci and their susceptibility profilesfor synergistic activity [8]. High-level resistance (HLR) to kanamycin (withoutgentamicin) is a fairly common trait and is due to the production of a 3'-phosphotransferase-APH(3')-III. This enzyme is important because it also eliminatessynergism between cell-wall active agents and amikacin through phosphorylation of the3'-hydroxyl group, although it does not necessarily confer HLR to amikacin. Ferretti et al.12.551 have shown that HLR to gentamicin results from the bifunctional protein (AAC(6')-IbAPH(2")-I), encoded by a single gene with two active sites, one with 6'-acetyltransferase activity and the other, 2"-phosphotransferase activity. The combinationof these activities results in HLR or resistance to synergism for all commerciallyavailable aminoglycosides ~ncluding gentamicin, sisom~cin, netilmicin, tobramycin,kanamyc~n, and amikacin, except streptomycin, which is not modified by this enzyme.However, HLR to streptomycin may be due to streptomycin adenylyltransferases:ANT(6')-la or ANT(3")-la, which can coexist with the gene(s) for HLR to otheraminoglycosides, or can either be due to ribosomal resistance [256]. Eliopoulos et al.12561 showed streptomycin MlCs of 4.000 to 16,000 pgiml for strains with streptomycinadenylyltransferase. while MlCs were up to 128,000 pgml for strains with ribosomalresistance. Further, the bifunctional enzyme AAC(6')-IIAPH(2")-I does not modifyspcctinomycin too. but this agent, which is not a true aminoglycoside, is not generallybactericidal against enterococci and does not appear to show synergism with cellwallactive agents [8, 171. Studies show that the bifunctional aac(6")-le-aph(2")-la is no longerthe only aminoglycoside resistance gene in enterococci, since, new aminoglycosideresistance genes aph(2")-ic [257], aph(2")-Id and aph(2")-lb [8] are known to encoderesistance to gentamicin, and a new approach for detecting resistance to aminoglycosidesynergism may be required in these cases.

REVIEW OF LITERATURETable 2. Susceptibility profiles of genes that mediate resistance to aminoglycosidesynergism in enterococci.Aminoglycoside antibioticsResistance gene -GentamicinTobrrmycin Amikacin bnamycin Netilmicin DibckacinStreptomycin Arbekacin-u~~(6~-I~-~phf?"l~l~ RR R R R R SS"uphf?l-lh RR S R R R SSuph/?l-kR R S R S S S Suph(?'i-IdRR S R R RSSaphf37-lllu SS R R S SS SUUL /c('i-i~ s R sN-I s N-IR l Ronrf3'i-luSS SSR Sun1/4 ?-loSR RSSun~~fi'i-~u s sS ~ S ~ S I R S ,sNT- not tested; R- resistant to synergism; S- susceptible to synergisma Fony percent of ~solates tested susceptible to synergismTable adapted from Chow (81.Thus acquired aminoglycoside resistance In enterococci gains more clinicals~gnificance, since several stud~es carried out over the last two decades shows that manyof the strains have the ability to transfer especially the gentamicin resistance to othersrrains, even though different types of aminoglycoside resistance genes are prevalentamong enteroccxci [ I 31.10.2.3. Beta-Lacemuse and Nan-Beta-Lacramase mediated Penicillin ResistanceBeta-lactamase mediated penicillin resistance in enterococci was first described in 1983.It was encoded by a transferable pheromone responsive plasmid, and was constitutivelyproduced and cell bound (2581. Thereafter several reports of beta-lactarnase producingEnlerococcus were published, although most of these isolates were From the UnitedStates [259, 2601. While Rice et a]. [XI] showed that some E.,faecalis strains exhibitedchromosoma~ mediated beta-lactamase production, which also encoded HLR to

REVIEW OF LITERATUREgentamicin. The beta-lactamase production goes undetected by routine laboratorytests such as MIC or disk diffusion testing due to the relatively low levelsof' beta-lactamase produced constitutively by enterococci. Hence a specific betalactamasetest such as the chromogenic cephalosporin based methods is recommended[13, 1901. While low-level penicillin resistance has been reported to be an intrinsicproperty of E. faecium due to low-affinity of penicillin binding proteins-PBP [246, 2471,some studies reported strains with much higher levels of penicillin resistance that werebeta-lactamase negative 117, 621. Recently. a novel p-lactam resistance mechanism wasproposed in E. .faecium due to a bypass of DD-transpeptidation, which did not involvePBP (DD-transpeptidases) [262]. Thus enterococci adapt newer mechanisms forovercoming the action of p-lactams, to combat them10.2.4. Acquired Glycopepiide ResistanceThe most worrisome resistance trait to emerge in enterococci is the resistance to~ancomycin. Since the first report of Vancomycin resistant enterococci (VRE) in 1988 bylittley ct al. 12631 in England, and Leclercq et al. [264] in France, which was shown to beplasmid mediated and transferable. a large body of research has been stimulatedworldwide to elucidate the genetics, epidemiology and causal factors responsible foremergence of vancomycin resistance in enterococci. Five types of vancomycin resistancehave been reported in enterococci: VanA, VanB, VanC, VanD, and VanE as shown inTable 3 [57. 2651. The mechanism of resistance has been best characterized for the vanAcluster of seven genes found on the transposable (mobile) genetic element Tn1546 asshown in Figure 4. In the presence of an inducer like vancomycin, transcription of thegenes necessary for resistance to vancomycin is activated as a result of the interactions ofa sensory kinase and a response regulator. The transcribed genes are translated intoenzymes, some of which make cell-wall precursors ending in D-alanyl-D-lactate (D-Ala-D-Lac), to which vancomycin binds with very low affinity. Others prevent synthesis of,or modify endogenous cell-wall precursors ending in D-alanyl-D-alanine (D-ala-D-Ala),to which vancomycin binds with high affinity. All but one of the genes in the vanAclusters have homologues in van9 gene clusters, that in turn have a unique gene notfound in the vanA clusters.

REVIEW OF LITERATUREhyrrds,. h~wn .~.*=d q r ejr. I , b-, ,k,rrd -ddl en.*'l.hruum,r o,'h * cm,t. .UI n h L! N, D la. ,? ,U. D t r. L I .,,*,rri*ma.. I".+.& ,,"*.,l,".dd~rr.~) rmwburrr,nlu,m I**\,...,n.' 1 .,a ...lh ,064 ID21114 0.5< L -.I, ulmake cell w.1 pnch~ah aIf#nllv lor vanelnhibnaon O( wlC",al smthu'Mns-n -tan .nl.~. sn t hVancornyc~n-hubccpt~hle crrtertroccl r)nthcz,c ccll-uall prccursorr ending in 1)-Ala-1)-Ala.which. aner t~iln,locution from the cytapla\rn to the cell surlkcc. bind \unconrycin nith highatlinil). ollue hound. thew prccurwrs cunnol pnntcrputc In cell-wall s~nthcsis.Vunconrycin-rc*istunt enlerwoccl. In the prcrcncc of'sn inducer lihc vmcomycin. pencrateprecursors wilh dilkrenl tertnlnl (I>-Alu-1)-l.ilc, 1)-Alu. or IJ-Alu-1)-Scr). which have lowaninit) lor vunconiycin and thus C .U~ c(>nt~nue. in large pun. lo hr used lo synthrsi~e cell wallAlu dcnotcs alnnyl or slunine. and X luctulc fi,r VanA. Vunt), und Vann types ofresi5rancc andrcrinr Ibr VUII~' und Vunl typch?c#rSdr~4~**hklrdpa\hd#ld&.S.&.Muc$yr~.~ ~ ~ . t T u a r r H u r ( n - ~ ~ , U J A ~ Y l u r o y ~ .64

RE VIEW OF LITERATURELess is known about VanD24 or VanE23 types of resistance, but the genes fortypes A, 8, D, and E all appears to be acquired. Jn contrast, the genes encoding the VanCtype of vancomycin resistance are endogenouslintrinsic, species-specific components of6. gallinarum (vanC-I ) and E. cassel~~uvuslE. flovescens (vanC-2ivanC-3) respectivelyas reviewed previously [57, 2511. Recently McKessar et al. [266] has described anothermechanism of vancomycin resistance in some E. .faecalis isolates encoded by VanG gene,with MIC ranges from 12-16 pdml and theteicoplanin MIC < 0.5 pg/ml.In-vitro studies have shown the transferability of vancomycin resistance gene(vanA) from enterococci to S. aureus, S. epidermidis, and other gram-positive organismsvia plasmid-mediated conjugation [131, 2671, which created panic among health careprofessionals. As anticipated, the first documented case of infection caused byvancomycln resistant S. oureus-VRSA with MIC >I28 pgiml was reported in a patientfrom Michigan in the United States in June 2002 [268]. Subsequent molecular geneticstudies have authenticated that the vancomycin resistance was transferred fromenterococci to S. alrreus. Thereafter several reports of VRSA were published from US,Japan and other countries, although not at the pace of VRE. Thus the ability ofenterococci to acquire and transfer newer traits has raised serious concern for~mplementing stringent infection control measures to prevent the spread of VRE in healthcare settings, span from appropriate use of antimicrobials.10.3. Laboratory Detection of Antimicrobial Resistance in EnterococciThe diversity of the emerging antimicrobial resistance traits among enterococcal isolates,and occasionally the intrinsic resistance exhibited by certain species of enterncocci tosome antimicrobials, creates an additional need for accurate identification at the specieslevel and continuous surveillance of the resistance characteristics [17, 57, 2511. On theother hand, several of the specific drug resistance characteristics acquired by enterococcalstrains are not adequately detected by the routine susceptibility tests most commonly usedin the clinical microbiology laboratories, so they require modifications of the usualprocedures, or to use the most accurate methods recommended for the recognition of

RE VIEW OF LITERATUREresistance in enterococci. Hence, early detection of these strains would be of great valueto the selection of the more appropriate antimicr~bial therapy for the treatment of seriousentero~~~~al infections leading to better patient management, and to minimize treatmentfailure and spread of antimicrobial resistance. Although the National Committee forClinical Laboratory Standards (NCCLS) has published the standard susceptibility testingfor enterococci [190], different laboratories have adapted several modificationsof this standard method. In summary laboratory detection of antimicrobial resistance Inenterococci can be performed by two methods I) Phenotypic identification as per NCCLSguidelines, and 2) Genotypic identification of antimicrobial resistance. A brief review ofboth the methods would be presented in the following sections.10.3.1. Phenotypic Methods for Detecting Antimicrobial Resistance-NCCLS GuidelinesSmcr several years, controversy and confusion exist regarding antimicrobialsusceptibility testing of enterococcal isolates, particularly about the reliability ofphenotypic methods for detection of HLR to aminoglycosides and resistance tovancomycin [269]. Updated guidelines for the selection of antimicrobial agents arefollowed for routine testing and repofling, as well as performance, and interpretativecriteria of susceptibility testing for enterococci have been published by the NCCLS [I901and are summarized in Table 4 and 5. Although NCCLS guidelines is regarded as thestandard for selection of the agents to be tested, it has been emphasized by several studiesthat the selection can vary depending on the antibiotic usage pattern by the clinicians inany particular hospital setting [I 31. The choice of testing depends on the need, althoughdlsk diffusion testing is performed more widely than dilution testing. While other testingmethods like E-test (PDM epsilometer. AB Biodisk, Sweden) have been used in somestudies for quantitative antimicrobial susceptibility testing of enterococci, where apreformed antimicrobial gradient from a plastic coated strip diffises into an agar mediuminoculated with the test organism. The MIC is read directly from a scale on the strip at thepoint where the ellipse of growth inhibition intercepts the strip. Furthermore. phenotypicantimicrobial testing is ansidered very helpful for routine detection of resistance inenterococci since they do not discriminate the resistance markers rather detect themqualitatively irrespective of their mechanism of action.

RE VIEW OF LITERATURETable 4. Interpretive criteriaa for disk diffusion and d~lution susceptibility testingb for Enreroroccusspecies, according to NCC1.S recommendations [I901Disk diffusion teatsdZone dinmeter Dilution tests'(mm)MIC (~dml)Report Antimicrobial Disk --____ - -group' agent content S I R S I RAHCI1Penic~llinsPcnicill~n lounits >IS . 514 9 . ~ 1 6Ampicillin lop9 117 - 516 58 - >I6StreptograminsQuinoprist~nldalfopr~stinr 15pg 219 16-18 C15 '1 2 14GlycopcptidesVancomyc~n 30pg r17 15-16 514 9 8-16 t32Te~coplan~n 30pg :I4 11-13 510 9 16 Z32An~inoglycoside\-H1.RGentamicin 120pg '10 7.9 56 5500Streptomycin 300pg r10 7-9 -4 -- >SO0Macrol~desEtyhromycin 15pg 123 14-22 :I3 50.5 1-4 >8TetracyclinesIetracycl~ne 30pg .I9 15.18 514 :il 8 216Doxycycl~ne 30pg 1 6 13-15 r I? 14 R 116Minocycl~ne 0 219 15-I8 514 54 8 816Anuaniyc~n\Kifampin 5pg 220 17-19 ,-I6 51 2 14Phcnicol\Chloramphenicol 30pg 218 13-17 112 4 16 232Fluoroqu~nolonesC~pmfloxacin Spg :21 16-20 515 51 ? r4Levofloracin 5pg 117 14-16 513 52 4 28Norfloxacin IOpg :I7 13-16 112 14 8 216Nitrofuranto~nsN~trofurantoin 300pg 517 15-16 114 532 64 ?I28FosfomycinsFosfomyc~n ?OOpg ?I6 13-15 512 34 I28 256.- .-.-.-.-.-.-.-p'Interpretative cntena for susceptibility testing : S, susceptible. I. ~ntermediate; R, resistant.h~encral comment. Ccphalosponns, aminoglycos~des (except for high level resistance screening).clindamycin, and Tnmethopnm~sulfamethoxazole may appcar active In vitro but are not effect~veclinically, and ~solate should not be reported as susceptible.'Report group (ant~microbials): group A - recommended for inclusion in routine primary testing panel:group B - cl~nically important agents particularly for nosocomial infections & may warrant pnmar). testing:group C . alternative agents that may . requlrc . testing if situation demands. viz., for endem~c!epidemicmulidrug resistant st&ins; group 11 - primarily used for treating unnary tract infections;d~onditions for disk diffus~on testing: med~um. MHA; inoculum. 0.5 McFarland turbidity standard (growthmethod or d~rect colony suspens~on); incubation, 3S°C. amb~ent air. 16 to 18 h (24 h for vancomycin);quality control: S, aureus ATCC 25923 (E faecal~s ATCC 29212 & E,faecalis ATCC 51299 for HLAR)"onditions for dilution testing: medium. MHA for agar dilution or cation ad~ustcd Muelier Hinton brothfor broth dilution; incubation, 35'~. ambient air, 16 to 20 h (24 h for vancomycin): quality control:E faecalis ATCC 29212 (E faecalis ATCC 51299 for HLAR and vancomycin resistance screening).'For reporting against vancomycin-resistant E .forcium

REVIEW OF Ll TERA TUREa. Detection of High-Lewl Aminoglycoside ResistanceThe agar dilutionhroth dilution screening method and the disk diffision screeningfor detection of HLR to aminoglycosides are generally used to predict theeffect between an aminoglycoside and a cell wall-active agent [190, 2691.several laboratories are routinely using these methods worldwide successfully, and thedetails of the performance and interpretation of the tests are as shown in Table 5.Gentamicin and streptomycin are the two agents that are tested by several laboratories ona routine basis, since, enterococcal isolates that are resistant to gentamicin can also beconsidered resistant to other aminoglycosides except streptomycin, although exceptionsarise very rarely where strains susceptible to gentamicin may be resistant to kanamycinand amikacin. In such instances, susceptibility testing for these alternate aminoglycosidesmay help in choosing the ideal synergistic combination for therapy 12691.b. Detection of Vancomycin ResistanceSince the first report of vancomycin-resistant enterococci (VRE) [263, 2641, it hasbecome a serious therapeutic problem. With newer details regarding the genetics andclin~cal significance of vancomycin resistance emerging constantly [13], the laboratorydiagnostic procedures and interpretive criteria of vancomycin resistance in enterococclkeeps changing. or is subjected to modifications. Many methods commonly used byclinical laboratories, including the conventional disk diffusion method and automatedsystems, have had problems in detecting enterococcal strains with low to moderate levelsof vancomycin resistance in enterococci [270]. While most of the modifications andrecommendations for disk diffusion testing and agar dilution screening for detectingvancomycin resistance suggested by various studies 193, 2711 were further evaluated andadopted by NCCLS as shown in Tables 4 and 5.c. Detection of B-Lacm ResistonceThe routine antimicrobial susceptibility tests like disk diffision or dilution methodsperform well in detecting the usual resistance due to changes in PBPs. But the resistancedue to the production of klactamase by enterococcal strains is not detected by these tests,since an inoculum 100-fold greater (10' cfuiml) than routinely recommended be

REVIEW OF LITER' TUREnecessary for demonstrating resistance mediated by p-lactamase. Hence, in the absence ofpgp-mediated resistance, p-lactamase producing enterococci should be consideredresistant to penicillin, ampicillin. and the ureidopenicillins, since several reports of p-lactamase positive enterococci causing infections have been published [259, 2601. Thedetection of $-lactamase production by enterococci can be detected by using achromogenic cephalosporinase based (nitrocefin) method as recommended by themanufacturer.10.3.2. Genotypic Methods for Detecting Antimicrobial Resistance in EnterococciThe genotypic methods can rapidly detect specific antimicrobial-drug resistance genesand substantially contribute to the understanding of the spread and genetics of acquiredenterococcal resistance. But because of their high specificity and versatility ofantimicrobial resistance in enterococci [8. 17, 571, genotypic methods have their ownllmitations, since they will not detect antimicrobial resistance due to a mechanism that isnot included in the testing, as well as emerging resistance mechanisms. The high-levelgentamicln resistance is a good example to authenticate this fact, since they have morethan one genetic mechanism to exhibit resistance [a]. But these problems too have beenpart~ally overcome by using alternate procedures like multiplex-PCR, using which morethan one gene could be amplified at the same time [105, 2691. Several laboratoriesworldwide are using the PCR based genotypic methods routinely. to identify variousaminoglycosideivancomycin resistance genes to aid in diagnostics, and to study thegenetics and epidemiology of acquired enterococcal resistance [20, 105.2691.I I. EPIDEMIOLOGICAL TYPING METHODS FOR ENTEROCOCCIMicrobial typing is the first and foremosr step in establishing nosocomial epidemiologicalsunreillance system after preliminary identification of the nosocomial pathogen Theresults of the microbial typing depicts the degree of relatedness among the isolatesstudied, which in turn helps in initiating and executing appropriate infection controlmeasures in health care settings. Microbial typing system can be categorized into twogroups

RE VIEW OF LITERATURE1) Techniques those determine the phenotypic characteristics- phenotypic techniques2) Techniques those characterize genetic determinants- genotypic techniques.As reviewed in previous sections, the emergence and dissemination of multipleantimicrobial resistance traits among enterococci, and the evidence supporting theconcept of exogenous acquisition of enterococcal infections have generated an additionalneed for typing the isolates as a means of assisting infection control and epidemiologicalstudies both within and among different healthcare settings [13, 17, 571. Thus severalstudies were initiated and being carried out, to study the epidemiology, andoutbreaWclonality analysis of drug resistant enterococci in nosocomial settings, as well incommunity using appropriate techniques. A brief review of the different epidemiologicaltyping procedures followed by several laboratories worldwide is presented in thefollowtny sections.11.1. Conventional Phenotyping MethodsThe conventional phenotyping techniques were used since 1960s for epidemiologicalinvest~gations of enterococcal infections based on phenotypic characteristics ofenterococci, even though they were not highly reproducible and sufficientlyd~scr~minatory. The conventional/classic phenotypic typing methods used to investigatethe diversity among isolates ot nosocomial enterococci include biotyp~ng, antibiotyping,serotyping, bacteriocin typing and bacteriophage typing, the details of which have beenrev~ewed previously (Section 6). Most of these typing techniques were used extensivelybefore the advent of molecular techniques for characterizing enterococci. But todaymolecular techniques are used extensively along with biochemical techniques, forepidemiological typing of enterococci [272].Biochemical typing based on physiological characteristics has been used byseveral studies for outbreak analysis as well, subtyping of enterococcal strains [272-2751.While, several studies have utilized antimicrobial resistance as a trait (antibiotyping) fore~idemiolo~ica~ typing of mterococci. along with biochemical typing [259, 276, 2771.

REVIEW OF LITERA TUREFew laboratories have also used other typing systems like serotyping, bacteriocin typingand bacteriophage typing, although less extensively [12, 131. Since biochemical typingwas showing a higher discriminatory index than other phenotyping techniques, automatedbiochemical fingerprinting systems were introduced commercially for strain typing [272].~ost of these phenotypic typing systems although yield usehl information, are generallytlme-consuming, difficult to reproduce and interpret. Some of these techniques frequentlyfail to adequately discriminate among strains because strains of several enterococcalspecies do not usually show enough physiological variation to be useful [97], and forsome others the reagents are not readily available; therefore, they have limited value inepidemiological studies. Thus today. most of the laboratories use phenotyping methods inassociation wrth more recent genotyping techniques, for epidemiological investigation ofenterococcal isolates. which enables to track dissemination and evolution of multi-drugresistant strains more efficiently in health-care settings.11.2. Molecular Typing MethodsThe applications of novel molecular genotyping tools have provided critical insights intoep~demiological aspects of nosocomial enterococcal infections, especially in relation tooutbreaks due to strains that exhiblt clinically Important antimicrobial resistance. Thegenotyping methods are comparatively more discriminative than other phenotype-basedtyplng methods, since the approach focuses on genetlc determinants rather than merelyphenotypic characteristics. Over the last two decades several molecular techniques likePlasmid and genomic DNA profiling. restriction enzyme analysis-REA of genomic andplasmid DNA, chromosomal DNA restriction endonuclease analysis by pulsed-field gelelectrophoresis (PFGE) [278-2801, Ribotyping, Multilocus enzyme electrophoresis(MLEE). and PCR-based typing methods such as, the random amplified polymorphicDNA-RAPD-PCR assay, and repetitive element sequence REP-PCR have been used toinvestigate the genetic relationship among enterococcal strains [281, 2821 . While fewstudies have used DNA sequencing based tools especially to determine differences in thenucleotides among specific resistance genes in entemocci. However results from a largenumber of investigations over the last decade have indicated that restriction digestion of

REVIEW OF LITERATUREgenomic DNA by PFGE has an edge over other genotyping tools for typing enterococcaland is being considered as a "gold standard tool for epidemiological typing ofno~~comial enterococci. Another recently introduced technique called amplified fragmentlength polymorphism-AFLP analysis shows a similar power of discrimination whencompared with PFGE for typing drug resistant nosocomial enterococci [55, 2831. A briefreview of few important genotyping methods used widely for epidemiological typing ofenterococci is presented in the forth-coming sections.11.2. I. Plasmid DNA Based TypingThe first report of plasmid DNA in 1972, encoding erythromycin and tetracyclineresistance in enterococci [6], initiated several researchers to probe into the molecularbasis of plasmid mediated dug resistance among nosocomial enterococci. However in1986, Zervos and his colleagues utilized plasmid DNA content as an epidemiologicalmarker and authenticated nosocomial transmission and exogenous acquisition ofgentamicin resistant S, jut~cu11.s for the first time. Thereafter, several studies startedutilizing plasmid DNA profiling as an epidemiological tool for typing enterococci, andwere able to provide molecular evidences fbr dissemination of multidntg resistantenterococci, withln, as well outside hospital settings [47,278. 284,2851.The plasmids encoding resistance to ampicill~n and gentamicin were isolated fromoutbreak strains, and checked for homogeneity by restriction digesting the plasmids,which authenticated that few strains of enterococci were responsible for disseminationwithin as well outside hospital over a wide geographical area [SI, 2861. Plasmid typinghas been used successfully by several authors in US, Japan and India, to study and provethe dissemination of gentamicin resistant enterococci in hospital settings [135, 287,2881.Few authors used plasmid typing for differentiating colonizers from pathogenicenterococci [289], as well to study the dissemination of MLS enterococci [290]. Whilemost of the recent studies have shown that plasmid typing, when used along with PFGEof chromosomal DNA gave more conclusive results for epidemiological typing ofenterococci [287, 2881.

REVIEW OF LITERATURE11.2.2. Chromosomal DNA Analysis by Pulsed-Field Gel Electrophoresis (PFGE)~FGE is presently considered as the "gold standard for the epidemiological analysis ofnosocomial enterococcal infections by many investigators. Although the PFGE techniqueIS more cumbersome and time-consuming, apart from high equipment costs, theenterococcal strain typing results they yield, makes them currently the single most usefuland reliable typing method. Further, some authors have experienced much more difficultyand inconsistency in the generation of clearly visible plasmid patterns with enterococcithan with other organisms such as E. coli and Shigelia species. In such instances PFGEproves to have an edge over other techniques, and generates clarity in polymorphisms ifany. among enterococcal strains studied [279].PFGE can be performed by two methods namely contour-clamped homogeneouselectric field (CHEF) electrophoresis, and field inversion gel electrophoresis (FIGE).Whilc CHEF electrophoresis reveals better hand resolution over 250 kb more reliably,FlGE reveals better band separation in the 50 to 200 kb range. Thus most studies useCHEF electrophoresis extensively for PFGE, since polymorphisms are seen more amonghands > 250 kb [279]. More than a decade several studies have used PFGE extensively innosocomial epidemiology for studying the clonal relatedness among nosocomial and, or,environmental enterococci, as well to track the dissemination of genetically related clones[75, 110. 148.272,281,282,291.292].Although PFGE has been more discriminatory than other typing methods asshown in several studies, epidemiological interpretation of PFGE profiles was not clearcutalways. The occurrence of genetic events, such as inversions, deletions and otherrearrangements of the chromosome. as well as insertion of transposons or mobileelements, where shown to be associated with substantial changes in the PFGE profiles,leading lo problems in clonality assessment and interpretations by several studies [293,2941. Hence, most researchers preferred applying PFGE along with other phenotypicand/or at least an additional genotypic technique, since the approach was highly helpfulfor appropriate epidemiological interpretation [27. llO.28 I].

RE VIEW OF LITERA WRESeveral authors have used PCR based genotyping techniques like RAPD-PCR orREP-PCR along with the gold standard techniqlle PFGE, for epidemiological typing ofnosocomial E. faecalis and E..faeciurn resistant to vancomycin, high-level gentamicin andand ampicillin [I 10, 281, 282, 291, 2931. Thal et al. [295] utilized plasmidcontent and hybridization analysis along with PFGE to trace the clonality ofglycopeptide-resistant E. ,faeciurn isolates collected from Michigan hospitals over a 6-year period. Most of these studies have concluded that PFGE was highly powerful inclustering the outbreak strains, which showed similarity in polymorphism profiles amongthe clustered isolates of enterococci. PFGE analysis although used mostly forepidemiological typing of nosocomial E, facco11.v and E, faecium, have also been shownto be useful for typing other enterococcal species like E. raffinosus, and other unusualspectes of enterococci [I481Some studtes have shown that multicenter PFGE studies with centralized serverfor interpretation will address epidemiological questions effectively. De Lencastre et al.[296] made an anempt to test the efficacy of a molecular surveillance network bysubjecting MRSA and vancomycin-resistant E, faccium genotypes for PFGE typing in sixhospitals in the metropolitan New York City area. The results of their study wereencouraging, and concluded that cooperative venture fnr molecular typing of enterococcihelps in rap~d tracking of drug resistant strain dissemination locally as well glohally.Furthennore, recent advances In image processing systems assist in effective analysis andinterpretation of PFGE banding profiles, and to store the PFGE banding profiles in adatabank for comparative analysis using computational methods. This would provide anefficient tracking system to assist hospitals, clinics, and chronic care facilities incontrolling the spread of multi-drug resistant enterococci locally, as well globally.11.2.3. Other Gcnotyping MethodsApan from plasmid DNA profiling and chromosomal DNA analysis by PFGE, severalother genotyping methods were used for epidemiological typing of enterococci. But mostof these methods were performed as an adjunct to the gold standard technique PFGE. Asreviewed above, several authors used RAPD-PCR for epidemiological typing of

REVIEW OF LITERA TUREenterococci [I 10, 281, 282, 2911. Most of these studies concluded that even thoughRAPD-PCR was simpler and easier to perform .when compared to PFGE, they were nobetter than PFGE in typing enterococcal strains. But the RAPD-PCR results were ableenough to track the epidemiological panern of enterococci, although with limitations.Ribotyping was used by Kuhn et al. [2721 for epidemiological typing of enterococci, asan adjunct with PFGE. Recently, Krawczyk et al. [297] evaluated a novel method basedon amplification of DNA fragments surrounding rare restrictron sites (ADSRRSfingerprinting) for typing strains of vancomycin-resistant E, fuecium. Another recenttechnique, amplified fragment length polymorphism (AFLP) is reported more frequentlyby several studies, for epidemiological typing of enterococci. Antonishyn et al. [55]evaluated fluorescence-based amplified fragment length polymorphism analysis-f-AFLPfor molecular typing of vancomycin-resistant E. /ueclum in a hospital epidemiologicalstudy and compared the results with PFGE. Jureen et al [283] conducted a comparativeanalysis of AFLP and PFGE among ampicillin-resistant E. fuecium in a hospital outbreakand subsequent endemicity. Most of these AFLP based studies produced patterns ofcomparable discriminatory power while possessing some advantages over PFGE, likeless-time-consuming and, internal standards for typing nosocomial enterococci.Although genotyping has an edge over other microbial typing methods, it can berlghtly concluded that no single "ideal" method can be used without clinicalepidemiologic investigation, but any of these techniques is helpful in providing focus toinfection control practitioners assessing possible outbreaks of nosocomial infection.12. ENTEROCOCCUS- THE INDIAN PERSPECTIVESurveillance studies conducted by various bodies across the globe have rankedenterococci among the top three pathogens causing nosocomial infections, leading tosubstantial morbidity end mortality [S, 11. 371. Although surplus data regardingnosocomial enterococci are available from United States and Europe, Asian countriesaccount for petite contribution regarding the significance and prevalence of this emergingnosocomial pathogen. The incidence of entemoccal infections and species prevalent in

RE VIEW OF LITERA TbREIndia has not been investigated extensively and thoroughly, and as evident from literaturethere is a paucity of information on the versatility of multidmg resistance amongenterococci from India. Hence, we will present a brief review regarding the significanceand prevalence of enterococci in lndia as evident from different stud~es conducted overthe years in the following section.12.1. Non-Human Reservoirs of EnterococciEven before the emergence of enterococci as a predominant nosocomial pathogen, theirsignificance outside the hospital setup was explored. The first Indian report regarding thesignificance of enterococci although outside hospital, dates back to 1970, when Thapliyalet al. studied the statistical correlations between total coliform and enterococcus counts inwater supply of Tarai region, and validated that enterococci can be used as a fecal~ndicator along with E. colt in potable water [298]. Thereafter in 2001, Grover andThakur further authenticated in their study that drinking water samples contaminated withE. ,Jueculi.s were unfit for consumption [299]. Saikia et al. [300] studied the prevalence ofEnren~coccus species and their antimicrobial susceptibility from the intestines of ducks in.Assam. They isolated E, Juc,colis and E. Jaecium in ducklings less than eight weeks old,and E. furcali.s. E.fiecium and E. yullrnarurn in duck more than eight weeks old. Further,all the Enterococctr.~ species isolated were resistant to the macrolide and l~ncosamideantibiotics, while chloramphenicol and gentamicin sulphate were the only antibiotics ofthose tested, which were moderately effective. While Shakuntala et al. [301] showed thatG~tc~rococcus species constituted 3 % of all bacteria isolated, in their study camed out tofind the prevalence of Streptococcus agaluctiae in milk from untreated healthy cows, andthose with mastitis in Manipur. But no studies from lndia have yet shown animal tohuman transmission of enterococci as a cause of infection.12.2. Significance of Nosocomial Enterococcal InfectionsAlthough few reports prevail on the non-human reservoirs of enterococci, most of theStudies conducted were focusing the significance of enterococci in human infections. In

RE VIEW OF LITER! WRE1974 Arora and Tyagi isolated enterococci from apparently healthy young adults withsignificant bacteriuria 13021. Thereafter, hardly.were any reports on enterococci fromhuman infections for more than a decade. In 1985 Mishra et al. [303] depicted thatenterococci plays a role in neonatal septicemia. Later in 1997 Bhat et al. 11741 showedthat neonatal bacteremia was caused by high-level aminoglycoside resistant-HLARenteroc~~ci in a tertiary care hospital in Mangalore. Of the total 41 strains of enterococcilsolated 35 (85.4%) were E. fuecalis and six (14.6%) were E. foecium. A total of three(8.6%) strains of E. /uecah.s and two (33.3%) E.,fueciurn strains showed HLAR, but nonewere vancomycin resistantNischal and Macaden [274] conducted a preliminary study on 50 isolates ofenterococci from various clin~cal samples. and performed biochemical speciation andchecked the hemolytic activity in enterococci using a panel of erythrocytes. They showedthat activity of the hemolysin was better against human A and 0 blood grouperythrocytes, and also against guinea pig and fowl erythrocytes at 37".Chanopadhyay etal. (3041 showed that E. {urculis appeared to be the commonest organism followed bypeptostreptococci and Stri~ptococcus i~iridon.5 in root canal infections. In 2001 Hiremathel al. [305] reported that Entc,rococc~.$ causes chronic suppurative Otitis media (CSOM),even though Pseudornonus uerupino.su and S. oureus were the predominant etiogensisolated In 2003 Parvath~ and Appalaraju [306] in their one-year study showed aprevalence rate of 3.5% enterococci from various clinical samples in a tertiary carehospital in Coimbatore. The highest incidence was from isolates of urine (43%), pus(40%), wound swabs (1 I%), and the least incidence were noted in discharges from fistula(2%), blood cultures (2%) and peritoneal aspiration fluid (2%) while E. faecullsaccounted to 88% of the isolates. Recently. Viswanathan 13071 reported the first case ofenterococcal conjunctivitis from Mumbai (Bombay) in a diabetic patient, caused by E.fuecu1i.s.Many studies have depicted that enterococci plays a major role in Infectiveendocarditis (IE). In 1996, Tripathi et al. 13081 in a prospective clinical and etiologicalstudy of IE showed that S. oureus was the predominating organism (24%), while

RE VIEW OF LITERA TUREEnterocOccUS was shown to contribute to 8%. While Khanal et al. [309] in their study ofIE showed that high level gentamicin resistant enterococcus was a major etiological agentof IE along with Staphylococci that added to the complexity of the problem. Garg et al.[3101 carried a study of IE and analyzed the clinical profiles and outcome in 192 Indianpatients over a period of ten-years (1992-2001) in a super specialty tertiary care referralhospital in Lucknow. Their results revealed that enterococci contributed to 8.1%, while~treptococcl was the commonest cause of IE accounting to 23.2% with high morbidityand mortality rates. Most of these studies concluded that Rheumatic heart disease-RHDwas the commonest underlying heart disease present in approximately 50% of patients,while blood cultures were positive in 50-70% of episodes approximately.Gupta et al. [3 1 I] studied the uropathogenic strains from inpatient and outpatientdepartments from April 1997 to March 1999 for their susceptibility profiles. Their resultsrevealed that E. farcalrs was the only gram-positive cocci causing UTI, apart frompredominant gram-negative bacilli. In a one and a half year study from a tertiary carehospital in Mumbai the incidence of Enr~rococcus species in urinary tract infections wasSound to be 7.3% with E. furculir as the predominant species isolated (87%) followed byE. fuecium (10.8%) and E. durans (2.14%) [312]. Desai et al. [I441 studied theprevalence and distribution of various species of entemcocci in another tertiary carehosp~tal in Mumbai, by processing various clinical specimens of catheterized patientswith UTI. Enterococci were prevalent in 22% of the total specimens with Foley'scatheters and bum wounds to be the major site of isolation. They ident~fied seven speciesofenterococci from a set of 202 isolates, with E.,/aecalis (49.5%) and E./aecium (35.6%)predominating, while E. swum (9.4%). E. hrrae (2.4%), E. rqflinosus (1.9%) and oneisolate each of E. gallinamm and E. cusselij7oi~us were the other members ofEnrerococcus species identified. Their study showed a high rate of colonization asopposed to infection. UTI by enterococci due to catheterization was found in 8.9% of thepatients, due to high rate of colonization of Foley's catheters, and use of broad- spectrumantibiotics.

REVIEW OF LITER? TURE12.3. Role of Enterococci in Polymicrobial infectionsThe role of enterococci in polynicrobial infections still remains debatable as reviewedalthough several experimental studies have shown that enterococci contributeto the severity of the disease in case of polymicrobial infections like intra-abdominal andwound infections, bums or abscesses, surgical site infections, and bacteremia. Sood et al.13131 conducted a retrospective analysis of positive blood cultures obtained during thepried of five years (from January 1991 to December 1995) in case of suspectedbacteremia from a tertiary care hospital in New Delhi. Their study showed that there wasan increase in bacteremia due to polynicrobial etiology, while the gram positive isolatesespec~ally that of E. faecalrs were multi-drug resistant which had prognostic andtherapeutic implications. Murthy et al. [314] showed that Enrcrococcus was isolated lessfrequently from postoperative wound infections, while S. aureus (32%) andP,\cudomonas species ( 2 1%) were the commonest organisms. Later Vijaya et al. [315]also allowed that En/erococcu.s was isolated less frequently from diabetic foot infections,uh~le S u1rrcw.s (33%) was found to be the most common isolate.Revathi et al. [316] showed that E.,faeculis ranked fifth contributing to 8.5% nextonly to P.\eudornonus species (36%), S. uureus (19%), Klehsiella species (15.5%) andI'rorc~~is species (I I%) in a five year (June 1993-June 1997) retrospective study ofbacteriological analysis of pus samples from bum patients at a Hospital in Delhi. WhileSingh et al. (3 17) in continuation of the previous study from the same hospital showedthat the bacteriological trend remained the same over the next five years (July 1997 andApril 2002) except for the emergence of Acinctoha~,!rr species (9%). But the incidence ofantimicrobial resistance among the nosocomial pathogens isolated was markedlyincreased, with concomitant resistance to penicillin and aminoglycosides exhibited by61% of E. faecalis resulting in limitation of therapeutic options. Sharma et al. [318]showed that enterococci play a role in peritoneal infection in patients undergoing acuteintermittent peritoneal dialysis (PD) in a prospective study carried (September 2000 toFebruary 2001). They concluded that in acute intermittent peritoneal dialysis theincidence of bacterial infection was 30% with preponderance of gram-negative over

RE VIEW OF LITER' TUREgam-positi~e organisms, and organisms of fecal origin were commoner than those oforigin reemphasizing the role of fecal streptococci (enterococci) in intra-abdominal~nfections.12.4. Hospital Reservoirs of EnterococciSince 1987 when Zervos et al. proved the exogenous acquisition of enterococci fromhospital environment [la], several studies started focusing in tracing the reservoirs ofenterococci in hospital environment for effective infection control. Fotedar et al. [319] intheir study showed that the housefly (Museo domestics) acted as a carrier of pathogenicmicroorganisms including enterococci in hospital environment in NewDelhi. In primary~solations, it was observed that the load of bacteria carried by the test group (surgicalward) of llies was significantly more (P < 0.001) than for the control flies (residentialarea). P. uenrgrnosu, E.,/ucculr.s and viridans streptococci were isolated only from the testflies. whereas the isolation rate of S uurcus was sign~ficantly higher in test housefliesthan in the control houseflies and concluded that houseflies may act as vectors ofpotentially pathogenic bacteria in a hospital environment. Chandrashekar et al. [I761conducted a microbiolog~cal surveillance by sampling the equipments, cradles, otherinanimate objects and environmental surfaces to trace the reservoirs of nosocomialpathogens In a neonatal intensive care unit-NICU. Their results showed a prevalence of4.7% for Enreroec)cca~ next only to A'leh.siellu pncumoniue (27.3%). E. coli (16.8%). S.uurcus (1 1.7%). and S. cpidi~rmidis and P. aeruginosu (10.2%). Thereafter no reportshave been published. to show the reservoirs of enterococci in hospital environment fromlnd~a.12.5. Antimicrobial Resistance among EnterococciThe property of antimicrobial resistance has enabled enterococci to emerge as aProminent nosocomial pathogen, with few studies from India highlighting this fact. In1995 Mathai et a]. showed high-level aminoglycoside resistance among enterococci13201. Later in 1997, Bhat et el. conducted a study to determine the drug resistance

REVIEW OF LITER4WREpattern of enterococci isolated from cases of neonatal bacteremia and showed that out of41 strains of enterococci isolated 8.6% and 33.3% of E. ,fueculis and E. fiecium strainsshowed high level aminoglycoside resistance-HLAR, while none exhibitedvancomycin resistance [174]. Although penicillin resistance has been quoted high amongenteroc~~~i in western literature, Jesudason el al. [321] showed that only 10.2% ofenterococci from urine cultures exhibited penicillin resistance in a study conducted in atertiary care hospital in Vellore, between January and June 1996. In 1999 Agarwal et al.[322] showed concomitant high-level resistance to penicillin (HLPR) andaminoglycosides in 16% enterococci at Nagpur. While low level vancomycin resistancewith MIC 516 pgiml was encountered only in 3.3% enterococci that were not associatedwith HLPR or HLAR. Parvathi and Appala Raju [323] screened 100 isolates of E.fircc,ulrs for beta-lactamase production and showed that 32% and 34% isolates wereposctive by acidometric method and cloverleaf technique respectively.In 1999 Purva et al. reported the first case of vancomycin-resistant E, fueciurnisolated from the blood culture of a patient with non-Hodgkins lymphoma from NewDelhl. India [324]. Thereafter, few more studies reported the prevalence of VRE in India,although the resistance rates were very low when compared to those of U.S and Europeanstudies. Parvathi and Appalaraju [323] in a one-year study showed that 95% of the 100~solates of enterococci were sensitive to vancomycin by disk-diffusion testing. Taneja etal. [I531 showed that 5.5% of enterococci isolated from urinary specimens betweenOctober, 2000 to April, 2001 in a tertiary care hospital exhibited vancomycin resistance.Among the eight (5.5%) VRE obtained. five were E..fuecium, and one each of E, fuecalis,E. cussellflu~~us and E. pseudou~~ium with MIC ranging from 8 to 32 kgiml.Mathur et al. [I 541 showed the emergence of multi drug resistance among 444 E.Jirecalis isolates in a one-year study at a tertiary care center of northern India. Theirresults revealed that 26% and 66% of the isolates exhibited HLAR and ampicillinresistance respectively, while 88% and 85% were resistant to ciprofloxacin anderythromycin respectively. Vancomycin resistance-was found in five (1%) isolates, ofwhich four had van-A phenotype and one had van-B phenotype as evident from the

REVIEW OF LITERATURE,,,~lts of MIC. Recently Karmarkar el al. [325] showed that 23% of the 52 enterococciisolated were resistant to vancomycin with. an MIC > 4 pgiml, but sensitive toteicoplanin, and underscored the importance of antimicrobial susceptibility testing ofenteroc~~~i. Although several studies have shown the molecular basis and epidemiologyof drug-resistant enterococci worldwide, no studies addressing this aspect are availablefrom lndia as evident from the available literature.12.6. Need of the Hour for Extensive Multifaceted studies on EnterococciMost of the lndian works published till August, 2004, have reported only the prevalenceand antimicrobial susceptibility pattern of enterococci over a small study period. Theemergence of resistance to penicillins, high-level aminoglycosides, and vancomycinamong enterococci in lndia is in high-rise as evident from recent literature. This diversityand, in some cases, species specificity of emerging antimicrobial resistance traits amongenterococcal isolates emphasizes the importance of accurate identification at the specieslevel, and continuous surveillance and characterization of the antimicrobial resistance incnterococci. Studies show that the genetics of antimicrobial resistant mechanisms inenterococcl are versatile and may vary geographically. which enables them todisseminate and transfer the determinants rapidly in hospital environments: the breedingground for most of the antimicrobial resistant nosocomial bacteria. Hence multifacetedstudies focusing on the molecular basis of antimicrobial resistance, and clonalrclationshlps among those drug resistant enterococci would help in projecting theversatility of this emerging nosocomial pathogen from Indian sub-continent. The resultsof such a study would be of immense value to set local. as well national standards fortherapeutic-prophylactic policies. and to strengthen the habits of appropriate andjud~cious antimicrobial use in health care settings. Further, the lack of geneticinformation regarding the antimicrobial resistance and epidemiology of enterococci fromIndian sub-continent has made it the "need of the hour" to probe into these aspects.Hence multifaceted studiff probing these issues from lndian sub-continent would be avaluable contribution, which can address the Indian perspective of Molecularcharacteristics of nosocomial enterococci to the world.

Chapter IEnterococci in ~knicac infections

CHAPTER ICHAPTER 1~ h c emergence of multidrug resistance among enterococci, including the intrinsicgly~opeptide resistance in E. casseli'avus and E. gallinarum, and the acquiredantimicrobial resistance by E. Jaecali.~ and E. ,faecium has triggered the microbiologylaboratories to identify enterococci appropriately to the species level [I 3, 3263. Although,E. /uecolis and E. faecium accounts up to 90% of clinical infections, there is an alarmingincrease in the incidence of other species of enterococci from various clinical sourceswith the properties of intrinsic resistance to several antibiotics including beta-lactams and&lycopeptides. The susceptibility of E. faecalis to the aminopenicillins (e.g., ampicillin)\bith the rare exception of those strains that produce p-lactamase, and the resistanceexhibited by E. faecium to most fi-lactams further underscores the importance of species~dentilication of enterococci, which helps in initiating appropriate antimicrobial therapyhased on the species isolated [ 13, 15-1 71.With the dissemination of enterococci in nosocomial settings, speciesidentification of enterococci tends to be of great use for epidemiological surveillancewithin hospitals. Further, it is imperative to differentiate VRE from other intrinsicallyvancomycin-resistant, catalase-negative, bile-esculin positive, gram-positive bacteriasuch as Pediococci, Leuconostocs and Lactobacilli, which are encountered although lessfrequently and may be misidentified as enterococci leading to complicating consequences[12, 3261. While conventional biochemical typing methods are useful in identifyingentemcocci, there are instances where only one phenotypic character differentiates onespecies from another. To further complicate, some strains of enterococci do not possesthe exact phenotypic character of the type strains, and there comes confusion over theirexact taxonomic status. In these instances molecular phenotyping, or genotypic methodsare essentiai for authentication of enterococcal species identity [97]. Hence properidentification of enterococci to species level appears quintessential for management,prevention and epidemiological~surveillance of these bacteria in any health care facility.

CHAPTER I. To determine the prevalence and species distribution of enterococci amongclinical specimens in JIPMER hospital, Pondicheny.MATERIALS AND METHODS1. Bacterial isolates['he study was conducted in JIPMER, a 900-bedded tertiary care hospital at I'c~ndlctlcrr\~.South India from July 2001 to June 200.1 Strains ofenterowcci were collcctcd dur~ng thrstudy period from various clinical specimens sent to the Microbiology labor at on^ such a3"blood. urine and exudate which includes wound swabs and pus (su~gical and non-surgical), catheters. wound tissues, ascitic fluld. synovial fluid and olher hody flu~ds" b!plating them on appropriate medium as per nature of the speCi~neri2. Presumptive identification of enterococciThe clinical specimens were plated on T~pticase Soy agar with 5% sheep blood, Mac-Conkey agar and Bile esculin azide agar (Hi-media laboratories, Mumbai, India) as pernature of the specimen and incubated overnight a[ 37°C without C0:. The morphology ofthe colonies were observed and looked for hemolysis on Blood agar plates. The unhowncatalase negative, bile esculin positive, gram-positive cocci occuning in pairs and shortchains were presumptively identified as enterococcus and subjected thereafter to a seriesofhiochemical and physiological tests for species identification [12.90].3. Lancefield groupingThe polysaccharide group antigens from the isolates presumptively identified asenterococcus were extracted by an enzyme extraction procedure, using a commercialLatex agglutination kit (The Binding site limited, Birmingham, B29 6AT) as permanufacturer's instruction. Briefly. 0.4ml of reconstituted extraction enzyme was takenin a test tube, and few colonies were emulsified in it and incubated in a water bath for 10-15 minutes by shaking after first 5 minutes. A drop of reagent (polystyrene latex particle

CHAPTER Isensitized with group-specific antibodies) was dispensed into a separatecircle on the reaction card supplied. Then one drop (50 11) of the extract (or positivecontrol) was placed to each drop of latex reagent, and the contents were mixed with a,nixing stick provided. The reaction card was gently rocked and rotated for a maximum"fane minute and examined for agglutination with naked eye without using a magnifyinglens.4. Conventional Phenotyping for species identification of enterococcusThe strains presumptively identified as enterococcus were tested for their phenotypic~haracteristics by performing an array of conventional biochemical and physiologicaltests devised by Facklarn and Collins as per standard procedures [12,90].4.1. Physiological testsA. Growth at IbC and 4PC: The test isolate was inoculated into Todd Hewin broth (Himedialaboratories. Mumbai. India) and incubated in water bath preset at 10°C and 45°Calong with an unlnoculated broth as negative control for 7 days and growth was judgedby turbidity every 24 hrs. Enterococcus grows at both temperatures.B. Growth in 65% NaCl brorh: The test isolate was inoculated into Brain heart infusion(BHI) broth (Hi-media laboratories, Mumbai, India) supplemented with 6.5% NaCl andincubated at 37°C along with an uninoculated broth as negative control for 3 days andgrowth was judged by turbidity every 24 hrs.4.2. Biochemical testsA. Carbohydrare fermentation rests: The tests were performed in hean infusion brothbase (Hi-media laboratories, Mumbai, India) using I% Andrade's indicator (Hi-medialaboratories, Mumbai, India) with pH adjusted to 7.1-7.2. The medium was sterilized bydispensing in 5ml amounts in test tubes to which commercially available sugar discs of1% concentration (Hi-media laboratories, Mumbai, India) were added using sterileforceps under aseptic conditions before inoculating the test isolates. The following sugarswere tested for fermentation using commercial discs: mannitol, sorbitol, inulin, arabinose,

CHAPTER Imelibiose, sucrose, raffinose, trehalose, lactose, glycerol, salicin, maltose, adonitol, andxylosc, while sohose and ribose were added to a final concentration of I% to the brothbase directly after sterilization. The single colony isolates to be tested for carbohydratefermentations were inoculated into 5ml of Todd-Hewin broth which was then incubatedovernight at 35°C. All sugar media were inoculated with lop1 of overnight Todd-Hewinbroth culture using a micropipette, and incubated at 35"C for up to 7 days. A positivesugar fermentation test was indicated by the medium turning pink in the presence of theindicator due to acid production. The results were read every 24 hours up to 7 days asrequired.B. Pyruwtc utilizlltion: Pyruvate broth u~th 1% Sodlum pyruvate was tnoculated w~thIOpl of overn~ght lndd-Hew111 broth culture to be tested and incubated at 35°C for 7days A posltrve test 1s lndlcated by the medlum turning from green to yellow In thepresence of lnd~cator bromothymol blue. due to acid product~on as a result of utlllzat~onof ppvate by the test sola ate4.3. Other teshA. Arffininc hydro/ysir: A suspension of the tesl isolate was inoculated onto Mollenargininr decarhoxylase test medlum (HI-med~a laboratories. Mumbai. India) with 1% L-Argin~ne hydmchlonde overlaid w~th sterile mineral oil and incubated at 37'C up to 7days. A positive action was shown by the development of a deep purplelviolet colorfrom alkalinimtion by ammonia production after an initial change to yellow. while anegative rcactlon is indicated by the color remaining yellow.B. Pigment production: The test isolate was inoculated on Trypticase soy agar (Hi-medialaboratories, Mumbai. India) with sheep blood and he plates were incubated at 37°Covernight. To check for pigment pmduction, a sterile cotton swab was used to pick upgrowth from culture plate. Isolates of E. mundrit and E. cawel1/7a1~u.~ produce a yellowpigment while other species of entemcocci do not.

CHAPTER IC. Md& dcicdiorr: The test isolates were stab inoculated through Motility m dum(Biomark laboratories. India) and the tubes were incubated at 30°C for 24hrs andobserved for motility.D. Hippurolc hydrolysis: A 1% solution of sodium hippurate was prepared and dispensedIn 0.4.ml volumes in test tubes, to which a suspension of the test isolate were inoculatedand emulsified thomughly and incubated for 2hours at 3PC. After incubation O.2mlNinhydrin solution (3.5~ Ninhydrin in 100ml of a mixture of equal parts of acetone andhutanol) was added and tncubated for a further lominutes at 37°C. A posit~ve reactionexhibited a purple coloration of the medlum, showing that glycine has been produced onhydrolys~s of the h~ppurateE. Vogcs-Prorhaucr tesl: A suspens~on of the test isolate groun overmght wastnoculated Into 2ml glucose phosphate peptone water and Incubated for 6 hours at 3S°C1 hen five drops each of 0 Meara reagent (Wg KOH and O 3g creatcne In IOOml dlstllledwater) and a-napthol wlutlon (5g u-napthol In 100ml ethanol) uere added and shake at~ntcnals for 30 mlnutes A pclsltl\e reactton \hou\ debclopment of a red color due toacetoln pmducl~onk. Pymlidonylarylomidasc- PYRosc rest: A snlut~on of pjml~donyl-0-naphyiam~de(PYR) (Stgma. St Lnu~s) uas prcparcd b! dl~sol~lng 25 mg of PYR ~n 1 ml methanol to\\hlrh 100ml d~st~lled uater ua\ added I or testlnp ?OpI of the substrate bas lnnculatedu ~th growth fmm blood agar plate to gl\ e turh~d suspension and tncubated at 37°C for 1hours Then to the incubated suspcnslon Dlmethyl am~noc~nnarnaldehyde 1070 (L L)conccnmted HCI was sddd A poslrl\e reactton was tnd~cated by de\elopment of redcolor w ~hn 30 modsG. Bile nmUn hylrdydv: The test isolate was inoculated on Bile esculin agar (Hi-media) .nd incubated ovarught at 37°C. The appearance of a bmwn black-blackcoloration of the med~um amund the inoculum indicates hydrolysis of the esculin.

CHAPTER IH. Tellurirc tolerance: Blood tellurite agar plates (Infusion agar with 5% sheep bloodand 0.04% final concentration of potassium tellurite) were inoculated and incubated at3PC for 24 hours, Isolates that reduce tellurite shows a heavy growth of jet-blackcolonies, and sensitive colonies fail to grow or show light growth.1. M&hyl-o-D-glucopyronosidc-MDC fernentation test: The substrate (MDG) broth(BHI bmth with 1% MDG and 0.006% bromocresol purple indicator) was inoculatedwith IOpl of fresh overnight culture and incubated in a water bath at 37°C for 24 hours. Apositive result was indicated by a color change from orange-red to yellow (Chen. et al,JCM, 2000)J. Tc~mr~lium reduction: Tetrazol~um (TTC) was ~ncorporated in Motility medium(Rlomark laboratories. India) to a final concentration of 0.01% and the test isolate wasstnh ~nnculatcd and Incubated at 35°C for 24 hrs. A pos~tive reaction was indicated if themcd~um turned red by the rcductlon of Tetrrv.ol~um. since TTC IS colorless In its oxid~zed(i~rm1 Slidr Caralarr feet: One or tuo ccilnn~es .sol the 1st ~wlate to be tested were plckedu~th a plat~num wlrc Icwp fmm a nutrlcnt agar plate and m~xed u~th a dmp of hydrogenperox~de placrd on a clean glass sl~de lmmed~ate prnduct~on of gas bubbles u ~th bnskellenescence IF the ~ndlcat~on of catalase product~on5. Mdrmbr Pbcootyplng.4. Whole ccllprorrin (H'CP) prrparafion: Preparation of WCP extracts and analys~s ofthc protein profiles by Sodlurn Dodayl sulfate-Polyac~lamide gel electmphoresis (SDS-PAGE) were performed as described pre~~ously 131. 981 with minor modifications.Bncfly. tat strains werc gmwn for 24 hours at 37°C on Tryticase soy agar with 590sh~cp blood. The samples were prepared by removing the bacterial powth from Ihesurface of agar plate wefully with a stcrilc disposable loop, and suspended In 5 rnl ofsterile saline solu~ion in order to obtain a turbidl~y equal to that of No.8 MacFarlanddcnsity struwld, ccntnifugad at 3000rpm for 5 minula and resuspended in O.5ml of an

CHAPTER 1aqueous lysozyme solution (I0 mgiml). The suspensions were incubated in a water bathpreset at 37'C for 2 hours. The Whole cell extracts were obtained by mixing one pan ofwhole cell extract to one pan of sample loading buffer and boiled for 5 minutes.B. SDS-PAGE: The WCP extracts were separated by SDS-PAGE along with a molecularweight marker (NEB) as per standard procedures [327]. The SDS-PAGE was performedusing 5% stacking gel and 10% separating gel at a constant current of 20 mA and stainedwith Coomassie brilliant blue R-250 (Appendix). Visual comparisons of the gels weremade and d(~umen1ation done using a Gel doc system (Vilber louben. France) for funheranalys~s of the image The detailed protocol is as described below.RequirementsMln~gel apparatus (Bangalore gene^. Ind~a)Pouter pack (capac~ty 2OOV. 5OOmA) (Bangalore gene^. Ind~a)Bolllng ua~er bathM~crnplpcrte (50- 100p1)Rtrhlng Rotar) shakerReagents/ Solutionsa. Stock solutions!M 'Tns.Hcl (pH R 8)I M Ins-Hcl (pH 6.8)IOa~o(u \) SDS (store at nnm tcm~rature)5000 Glycerol (\I\) ~n d~st~lled waterI"b(u \ ) Bromophenol blue- 100ml- l0mlTEMED (N.N.N',Na-taramclhylcne-ethylenod~am~ne)2-macaptoahmol or. Dlthlo~hrcitol(ilycinc

CHAPTER Ib. Working solutions30% Acrylamide stock- l OOmlAcrylamide - 29.28Bisacrylamide - 0.8gAdd distilled water to m&e 100ml and stir until completely dissolvedSeparating Gel Buffer (4x)- l OOml7Sml2M Tris-Hcl (pH 8.8)- I.5M4ml 10% SDS - 0.4%Add 2 l ml-dis~illed waterStacking Gel Bukr (4x)- l 0OmlSOml 1 M Tns-Hcl (pH 6.8)4ml 10% SDSAdd 46ml-d~stilled waterIOO,O Arnmonlum persuliale (APS)- 5ml0.58 APS d~ssolv~d In Sml distilled waterElectrophoresis Runnlng Bufl'er (I x)- 1 O(K)ml3s TnsI4 4g (ilyc~neIg SDSAdd dlstlllcd w atcr 10 make IOOOmlSample Buflir- I Om10 bml l M Tns-Hcl (pH 6 8)Sml SOab Glycerol21111 l0$0 SDSO.Sml 2-mercaplnerhanolI ml 140 Bmmophenol hlucAdd 0 Pml-dlst~lled waiuStarning mlut~on- l OOOml1.4 Coomassic Blue R-250450ml Mcthanol and Dlslilled water eachlOOml Glacial acetic aeid

CHAPTER IDeetaining solution- 1000mll00ml Methanol'l h l Glacial acetic acid800ml Distilled waterProtocolCalculation for X % Separating or, Stacking gelThe gel concentration percenrage, r e whether 7 5, 10, 12 5 or IS Oh, depends on onescxpenmental need, and the desrred quantlty of stock solutrons to cast the gel are~aiculated usrng he follow~ng formulaAcrylamrde stockStack~ng Separat~ngel bufferDistrlled waterXO/o 13 ml2 Sml10'0 APS 50 p1(7 5 - X% 3)mlTEMED 5 1 (10 pl ~f X% < 8%)Total \olumeI0 mlNc u.4 10% S~pilratron gel and 5% Stack~n gel as follous respectrtcly.Acrylamrde stock 3 4 mi Acnlamrde stock 1 6 mlSeparatrng gel buffer 1 5 rnl Stackrng gel buffer 1 5 mlDrst~lled u ater 4 1 mI Drstrlled uater 5 9 ml100.0 APS 50 p1 10°o 4PS 50 p1TEMED 5 PI TEMED 5Total \ olumc I0 ml Total holume I0 mlThe pmtein sample-WCP estract (20 PI)an cppendorf tubc and hoiled at 75- IWOC\\as mixed ~ith sample buffer (20 PI)for 3-5 minutes.Clem glass plates were assembled by placing h c spacers two on both sides. andone along the bottom edge. and hc whole assembly was fixed tightly withclampdgcl w ing stand.in

CHAPTER 1Subsequently, the glass plate assembly was sealed leak-proof with molten wax onall three side leaving the topside.The components of the separating gel mixture (without TEMED) were mixed andde-aerated and TEMED was added finally to the mixture and poured immediatelybetween glass plates upto 2cm below the notch.The gel was allowed to polymerize for 30-60 minutes at mom temperature, andthe acrylamide was overlaid with n-butanol, which helps to keep the gel surfaceflat ARer polymerization the butanol was removed and rinsed with water.Stacking gel mixture was prepared and cast over the separating gel as mentionedabove and comb was inserted carefully for formation of wells.The gel was allowed to polymerize for 15-30 minutes. and the comb was removedcarefully, and the wells were rinsed with distilled waler.The gel assembly was placed into the elatrophores~s chamber with the notchedplates fac~ng inside, and the upper and lower tanks were filled with runninghufTerThe wells wc7e loaded with the protein samples-WCP extracts (10-50 p1) dilutedIn sample bullkr. along with a molecular weigh1 marker (New England Biolabs).Elec~rophom~s wils carried out at a constant current of ZOmA till the dye frontn-ached the separattng gel and then increased to 25mA The elecuophoretic nmwas stoppd when the dye front reachcd the bottom of the gel. approximately after60-90-minutesThe gel was wmovcd carefully and stained using Cmmassie stain~ng solution for60-90-mlnutc% and subsequently dcsta~ned uslng Destainlng solution for overnightIn a mck~ng shaker.The gel was visually analyzed and documented appropriately for further analysis

CHAPTER I1. Prwalcnce of EntcrococciA total of 242 enterococci were isolated from various clinical samples submifled to theMicrobiology laboratory, JIPMER, Pondicherry, India during the study period from July2001 to June 2003. Enterococci isolated comprised of nine dinerent species as shown inTable 6. Two major species that topped the list were 171 isolates of E. fuerulr~ (71%)and 25 E. /uecrum(10%) contributing to 81% of all enterococcal species isolated.Remaining 19% of entemocci comprised seven different unusual species of enterococci,~,hlch included fifteen E gullinumm (6.2%), ten E u~ium (4.1%). six E. ruffinosu,(2 5Oj0), six E hrruc, (2 5%). four E, mundtrr (I .7%), three E. cu.r.scliflavu.s (1.2%) and twoI.; Jrrrun.c (0.8%).Table 6. Prevalence of dinerent species of enterococciEatcrococcus speeiesisolated (n=241)No. (%) of isolatesf. lot? rrrm 25 (10 7)L ~ullr~nrm If, (62)t utrum lO(4 I )t. ru/fino\ut h (2 5)The overall prcvalcnce of ratcrtxoccl amon!: various rl~n~cal spclniens tesrrdroutinely during the study pcmd \\as 0 4'0. \vhlch was dw\d frcvn a deno~r~nato~ of59.907 clinical specimens (total spcmens sim) as depicld in Table 7. 1 In. stat~sricela ~ l showal y ~ ~ that 95% Confidence Interval (('1) was 0 'ibO0-0 44% regard~nv the- -Predence of enraococo Our s~t~dv showed tha~ hldjpclnlen\ had h~gher prpawof 0 7%. while urlne and exudate ~ ~~trncna~hc~wcd a pre~alenw,ratc of0 Y"' each

Table 7. Prevalence of Enterococci among various clinical specimensNo. (%) ofSpcclmcn typc No. of rpcclmcnrcn~rococc~- - -- --- -. -- -- --Blocd 15,535 111 (07)CHAPTER ITOTAL 59,907 ?AZ (0.4)95% Confidence interval 036%-0.44%The distribution by site of isolation for the 242 enterococci as shown in Table 8,lncluded I l 1 isolates (46%) from bloodstream, 72 (30%) from urinary tract and 59 (24%)from exudate specimens Enlerococci isolated from exudate specimens sent to theMicrobiology laboratory included surgical and non-surgical wound swabs, peritonealfluids. pus, endotracheal aspirate t~ps, pleural fluids, CSF. ascitic flu~ds. catheter tips andmuscleu'ound tissues thn~ were sent for anaerobic culture testing The break-up of thetl~s~ribution of Enrc~roc.cxr.u.\ species among vanous clinlcal samples IS as depicted in'Table 8. E fut~cull.~ was isolated from 9656, 60% and 61% of unne specimens, bloodculture and exudate specimens respeclively, while the distribution of other enterococcal\pwlrs among various speamcns is as dep~cted. The infections were polqmicrobial in 46( IYa) of the 242 cases fmrn u hich enlenxcul were twlated The polqmicmbial culturesgenerall) yclded one or more of the follouing pathogens S airreux. Coagulase-negatne5taphylococc1. Lactcxoccc~. Leuconos~oc specles. Klebsiella species. Entembacter\pwies and P uc.ntgit~osuTable 8. Disvihut~on of Entt*roccn~c.u.\ spies among vanous clinical specimensha (%)of E a f ~ ~ tpcc~n u s isolated fmm \dour cllnicrl spccimcnsSpcclawm Mrrococci - --f furrullr & lurilum t pull~~E car\ t mm E muad E hltuc E ruff E durIYP (242)(171) (3) (15) (1) (10) (4) (6) (6) (2)

CHAPTER I2. Patient characteristicsEnterococci were i~~lated from various clinical specimens sent from 242 patients amongwhom 125 (52%) were males, 88 (36%) were females, and 29 (12%) did not have theirsex recorded, while 75% were inpatients among them. The median age of the patientswas 23 years (range, Neonates to 90 years).3. Conventional pbenotyping of enterococci,A total of 245 isolates were presumptively identified as enterococci, from which threeisolates upon subjection to preliminary phenoryping tests including PYRase test werelbund to be Lcuconmroc .ipenes and were excluded from further studies. Thus 242isolates of cnlerococci along with the set of CDC standard strains (courtesy of Dr.R~chard Facklam. CDC. Atlanta. U.S.A) were subjected to the conventional andextensive biochemical-~yping schema [l2. 901 for identification up to species level asshown in Table I and Table 9.The conventional phenotplng methods ~dentified nine different species ofcntenroccl among the 242 ~solates subjected for testing as mentioned in Table 6.Group-D-antigen wasextracted from 85% of entenrocci with least percent of extraction (50%) among unusualspcvlcs of m~emcocci. Thc complete phenotying results of all enterncocci isolated aretabulated in Table 9

CHAPTER IThe two major species isolated as confirmed by conventional and molecularphenotyping tests were 171 E. fuecalis (71%) and 25 E. foecium (10%). while theremaining 19% comprised seven unusual species of enterococci. All 242 enterococciwere identified appropriately by conventional phenotyping methods except 12 atypicalenterococcal strains (5%) that showed aberrant sugar reactions. The 12 atypical strainsInclude six rnannitol negative variant E. fuecu1i.s like species, one arginine negativevariant E. ,faeculis like species, three mannitol negative variant E. fuecium like speciesand, two arginine negative variant E. cu.vselr/7uvw like species. The validation of theexact taxonomic status of these atypical strains of enterococci was done by SDS-PAGE.Among the E. .fuecuhs lsolates some showed variable results for few tests. Sixstralns failed to ferment mamito1 and one strain did no! hydrolyze arginine, while 46%and 53% stnrins exhibited hlppurate hydrolysis and glycerol fermentation respectively.Three E jut~cul~s isolates could not utilize pyruvate, nor could they grow in the presence01'0.04% tellurite. Th~s was contrary to the typical blochernical characteristics (results) ofb.' :'urcuii.s lajlates. w~th the exception of some occasional asaccharolytic variantsfstrains.Most of the E /uc*cul~.s lsolates produced ac~ds w~thln 24-48 hours except some stralns (4-0°,0) that werc asacchm>lyric and falled to ferment mannitol. sorbitol and lactose.In case of E tc~trlum. three isolates were not able to fment mannilol even afterprolonged ~ncubatim for seven days, wh~le some stralns (1-3 strains) showed very weakor ahscncc of acid production fmm arahinose. melibiose and sucrose. Only 56% and 48%stralns hydmlymi hippura~e and tolerated !erm,ollum respectively, while 32%. 8% andX ~ of O the isolates fmcnted raffinose. xylose and MDG respectively.The remaining wven unusual entemcoccal species exhibited typical results of theskm.lard biochemical reacciow, although some strains failing occasionally. Two amongthree E co~scI~/Iavur isolated did no! hydmlyle arginine, urhich posed a difficulty inspecies identification md was overcome by Interpreting the results of other tes~s (likePlP~nnrl pmduction). Occasionally some species showed very weak or absence of acidproduction fmm various sugars as depicted in Table 9. Overall, the revised conventional

CHAPTER Iand extensive biochemical-typing schema of Facklam et al. that we followed throughoutwas able to speciate all entnococci isolated in our study, except few asaccharolyticstrains that posed difficulty in drawing conclusive results. Those strains along withstandard strains were subjected to WCP analysis by SDS-PAGE, and their exactcaxonornic status was confirmed as E. fueculi.~, E, jurciurn and E. cas.selfluvu~.4. Molecular pbenotyping of eoteroeocciH14 4 ~ 'The analysis of whole cell protein profiles of enterococcal isolates by SDS-PAGE.f'Funher, the gel Images were documented and analyzedcomputat~onally uslng gel analysis software. Bionumerics, version 2.5 (Applied Maths,Bclg~um). The slmilanty coeffic~ents of the whole densitometric curves of gel trackswen. calculated us~ng the paw-w~sc dlce correlat~on coeflicient. The cluster analysis ofthe similarity matrices was performed and a dendrogram constructed by the unweightedpair-group using arithmaic averages (LIPGMA). The dendrogram was constructed toshow thc slmilm!y of the atyp~cal asacchamlytlc variants of the E~rrr,rococcuspeciesalong w~th the CDC standard strams.A total of 36 entenwxcal ~solalc% were included in the molecular phenotypingsludy. along w~th 14 CDC standard strains The smins subjected for WCP fingerprinting~ncludcd: I? atypical strains: six rnannitol negative variant E. faecolr.~ like species, onearglnlnc negativc variant E. fo~~rulrs likc spies. three mannitol negative variant E.lut3i.iurn likc specie and. two arglnine negative \.ariant E. ca~.~cl~flurris like species, 14 E./ui*rolr.s straina which showed delay In famenling the sugars by more than two days, and10 biochemically active E. ,ficmli.r isnlatcs.

CHAPTER 1Flgure 5. Cluster analysis Of atypical strains of Enterococci wing Dlce coefficientand UPGMA method (Bionumerics, Applied Maths, Belgium).-Dr* c 4 c m30 a 80 70 80 IOOWCF'.YAOE.JIPDIER1i- -4Eraffnosus SSl278E durans SS1225E porc~nosusSS-1 M5E pseudoav SS1 Z77E rnundtn SS1233E avlurn SS-817E dfSDw SS-1295E gall~nwurn SS1228E ratb SS1494E hfree SS-12-27E rnabdoret SS-1226E feecalls MNVE faecahsMNVE faecahs MNVE faecahs MNVE faecalrs MNVE faecal~~ SS-1273-.-- E faecaks ANV- - E faecalrs MNV- ---E cassekfk ANV- -- E casshfk ANV- - E cassebfls SS-1299E faecorn SS-1274E faecurn MNVE faecurn MNVE faecrurn MNV. -- - - ---- 1L -SS. Destgnaton of CDC standard stratns.E pofc~r)osus IS currently destgnated as E ~11lOf~m.E pseudofiv - E psevdoevrum, E mafodoral - E malodoralusE cessdilla - E cesselillavus.MNV- Mannttol negatbe variant. ANV- Argln~ne negatlve vanant

CHAPTER 1The banding panm revealed almost similar protein profiles for the strains of thesame species studied (E. ./aecolis), although very minor quantitative differences wereobserved with no qualitative differences as shown in Figure 5. There were around 25-30clear bands visually detectable. The differences in the banding panem among differentatypical/biochemically variant enteroccccal species studied as compared with the CDCstandard strains were identified in the regions between 83-16 kDa regions. Computationalanalysis of the gel images and subsequent construction of a dendrogram validated theexact taxonomtc status of the atypicalibiochemically variant enterococcal species as Efi~c~~~~rrhs. E. furcrum and E. curselfluivs as shown in Figure 5 (Dendrogram). Sixmannitol negative and one arginine negative variant E /uecalrs like species, showed asimilar proteln profile of 195% which were un~que and indistinyishable from standard;IS well other biochemically normal strains as compared by pair-wise dice correlation~trflicient and authenticated as F /uecalrc species Wh~le, two argnme negative vanantsoft ta~~elrlluiwr were onf firmed as L tos~elmfla~~us after comparative analys~s of theWCP profiles w~th standard smlns The protein profiles of the mann~tol negative vanantI. laccrum stra~ns although unique d~d not exhrb~t sharp bands like the E fuecuLs strans.although slmilanty matrix was comparable with the standard L /uecrurn stran to validateit's taxonomic status Thc percent simllant~es of the clinical and standard stralns are asdepicted in the dendrogram shown in Figure 5DISCUSSIONtntcmoccus has emerged as a pmmlncnt nosocomial pathogen since last decadeworldwide (5. 13, 17, SX. 193. 2351 The results the present study showed E. fueculrs(71%) and E fotr.rum (10%) as the two predominant species in our clinical seNpaccounting for 81% of entemcocci isolated These findiworldwide renecting thc general trend In prevalace of en32x1 Pnvalcncc of unusual entcmcoccal species (199'0comparatively hi* than other studies that showed thenon-laeciwn cnttmcari as 2-1096 [ 16. 17. 3291 Some of

CHAPTER IFrom our perspective the prevalence tends to be higher which in part can be explained as,misidentification of species due to exhibition of aberrant sugar reactions by someenterococci or, due to lack of application of the complete range of tests to identify non-faecalis and non-fatcium enterncocci [94, 97). The prevalence rate (19%) of our studywas partly in accordance with another Indian study (1441 that showed 14.8% (excludingE. faecalis and E. fuecium) prevalence of unusual species of enterococci fromcatheterized patients with urinary tract infections. E. mundtii and E. duruns were notreported in their study, whose prevalence was 1.7% and 0.8% respectively in our study.k,'gall~narum (6.2%) and E. avium (4.1%) were the most commonly identified species,which markedly differs in isolation rate (0.3-1.2%) from other studies (16. 37, 3281.We have documented vaned distnbut~on by site of sola at ion for the 242cntemcocci ~solated rhrough our study lnterest~ngly as shown in Table 8. 46% ofenterococci were isolated fmm bloodstream, while 10% from unnary tract and 24% fromexudale specimens The highest prebalence rate of enterococci from blood stream thanfrom other sltn In our stud\. underscores the clin~cal s~gnificance of enterococcalinlntions Several studies carned in d11Tercnt parts of the world shows that unnary tractrrmains the most common site oi tsolat~on for enterococc~ [16. 18. 19. 18. 41. 142. 144.1 I!], although the result5 of our studv cc~ntrad~ct thic trend But the 96O0 pre\ alence of EIrrttulr~ as the pndomtnant species among unne specimens in our study is in~c~ncordance w ~th the results of most of these stud~esThe spauum oC entmoccal infections was diverse and occurred in a patientpopulat~on among uhom 125 (52%) were males. 88 (36%) were females and 29 (12%)did not have thelr sex recorded. whtle 7S0to were inpatients among them The median age01 the pat~enu uar 23 ycars (range. Neonate to 90 years) that was comparable with theabove ment~oned studies. although vanation In age o c c d among studies onentcrococcrl UTI where the mmal~an age was ,55-60 years modyThe highest ptzvalare rate (46%) of entemcocci fmm blood stream is of seriousconcern since majority of the isolates were from cases of septicemia without endocarditis

CHAPTER Iwith 44% of unusual enterococcal species contributing to this cause. The incidence ofenterococcal bloodstream infections in our setup was highest in the pediatrics specialty,since majority (60%) of the blood culture samples which yielded enterococcus were fromthe pediatrics age group, and sepsis was most commonly reported from neonates withfocal infections including meningitis, scalp abscess and pneumonia in this age group asshown by several studies [I 69. 170, 172, 174, 175, 177, 1781. The results of a recentpoint prevalence survey of nosocomial infections in 29 Pediatric Prevention NetworkNlCU conducted by CDC in U.S. has shown that enterococci ranks second only tocoagulase-negative staphylococci as the leading cause of nosocomial infections ofneonates in ICU (1681. On the other hand, the emergence of vancomycin resistantenterococci (VRE) in neonatal infections as shown by other recent studies poses atherapeutic challenge and emphasizes the urgency for more effective prevention~nterventions 1169-1711. Thus the sign~ficance of our findings is relevant in an era ofIncreasing rates of anr~m~crob~al resistance especially among the pediatric age group. Theremarning (40%) enterococcal bloodstream infections in our study were distributedevenly among other med~cal and surgical wards in an elderly patient population.~nclud~ng cam of septicemla with endocard~tis which is in concordance with the resultsof othcr stud~es (1 3, 162. 165-1671Apan from sept~cem~s'bacteremia, enterococci were isolated frequently (30%)Srom casa of unnq tract rnfect~ons, surgical and non-surgical wound infections andother miscellaneous ~nfcctions showing their versatility as depicted by several studiesworldwide [13. 17-21. 37. 38. 40. 143. 151. 153. 154. 1931. Some of the unusualInfections h t ylclded entemcoccl upon isolation in our study were as follows: aconjunctival swab and blood culture from two cases of endopthalmitis yielded E. arliumand E. .foecali.~ respectively. blood cultures of three ped~atric patients diagnosed withrespiratory pneumonia yielded two E, /orcola and one E. a~iurn respectively, swabs fromtwo cases of ostmmyelitis yielded E. foc.colis. while two E, faeelurn and one E. ,foecol~swere isolated from blood cultures of two pediatric patients and one elderly patientrespectively diagnosed with meningitis. Thus. as shown in our study the involvement of

ch!A PTER Ienter~~ci in these U~USUB! infections have been documented previously, althoughoccasionally [ 13, 188, 1891.The general risk factors for most of the enterococcal infections are a prolongedhospital stay by patients with severe underlying diseases and life threatening conditionswhich includes cancer, diabetes mellitus, chron~c renal failure, major trauma, surgicaland urinary and vascular catheterization as authenticated by several studies[22-24, 150, 155-1571, 68% of the patienls with the bloodstream infections in our studyhad a peripheral or central catheter in place within 48 hours preceding the blood culture.Further, patients who had previously received antimicrobial agents like expanded orhroad-spectrum cephalosporins that lack activity against enterococci, or a combinationtherapy, have three to five times greater risk of acquiring enterococcal bacteremiaexhib~ting resistance to antibiotics l~ke gentamicin. or vancomycin 136, 44. 47. 155. 158,I 59, 33 11. however some stud~es contradict this fact 124, 3321.Although entemcl plays a slgnlficant role in many nosocom~al infections theirrole In polymicrob~al infections still remalns debatable, but several experimental studieshave shown that entemcocc~ contribute to the severity of the disease in case ofp)l,micrnbial ~nfcctlons like intra-abdominal and wound infections, bums or abscesses,surg~cal site ~nfect~ons. and bacteremia [11. 200, 2281. Sood el al. (3131 conducted aretrospective analys~s of positive blood cultures obtained during the period of five yearsliom 1991 to 1995 and dep~cted a change In the trend of causative organisms among thepnlymicrobial culture positlve samples. Gram-negative bacteria were shown topredominate between 1991-94 w~th an average of 69% of cases. whereas in 1995 thegram-positive cocci ~solat~on was nearly the same as gram-negative bacteria (45% Vs.55%) with E. /an.alrs sola ales exhib~tlng multi-drug resistance, which had pmgnostic andtherapeutic implications. ~tthough the clinical significance of enterncocci in inm-abdominal and pclv~c infcclions remains murky. several repons depict their rnle inperllonitis, intn-rbdomiml or pelvic abscess. surgical site infection. suppurativethmbophlebitis, rcu~e dpingitis. and endometitis where eatemcoccus were isolatedalong with S, m-,CONS and gram negative bacilli which includes Pseudomonas.

CHAPTER IKlebsiclla E. coii and Acinetobacter commonly although at varying incidence [180-184,3 141. Most of these studies depict that 570% of enterococcal infections as polymicrobial,that varies according to the type of infection. But, only 19% of enterococcal infectionswere polymicrobial in our study. while majority were non-bloodstream isolates thatunderscores the clinical significance of enterococci isolated. The polymicrobial culturesgenerally yielded one or more of the following pathogens: S. aureus. Coagulase-negativeStaphylococci, Lactomi. Leuconostoc species. Klebsiella species. Enterobacterspecies and P. acruginosa.The necessity for rapid and precise identification of enterococcl to species levelhas increased dramatically over the years for ep~demiological studies.In our study we mls~dentified three gram-positive coccal isolateswhlch occurd In pals, shon cham and singly, and were catalase-negative, bile-esculinposltivc and salt role ran^ (6.5 Oh NaCI) as enterrxocci. Be~ng subjected to preliminaryphenotyping tests including PYRase lest these isolates were found to be Leuconosrocspecits and were excluded fmm huther characterization.Today, cvcn with the advent of molecular techniques. most clinical microbiologylabontoria worldwide follow the classical approach of Facklam and Collins [go] forspecies identification of entcmxwi. which involves assays combining morphologia1,

CHAPTER Ibiochemical and physiological characterization of enterococci. The grouping is based onkey phenotypic tests, and does not necessarily conform to the grouping by 16s rRNAsequencing or grouping by other molecular techniques. Facklam and his colleagues havefrequently updated the phenotyping schema to fit in all enterococcal species reported todate [12, 14. 311. Although we followed the updated phenotyping methods of Facklamand Collins 112, 14, 31, 901 for species identificatioh we did not use their groupingschema (to group enterococcus species according to the results of biochemical tests).since the grouping schema is not a mandate to he followed and depends on the choice of!he investigators to apply them in their setup [Facklam, Personal communication. 20041.However. wlh a rise in the number of enterococcal species there are instances whereonly one phenotypic character differentiates one species from another. To furthercomplicate the identification some strains do not exhib~t the exact phenotypiccharacteristics of the type stralns, causing confusion over their exact taxonomic status114. 31. 97). We tdentified nine different species of enlerococci among the 242 isolatessubjected [or testiny in our study (Table 6). but there were discrepancies and confusionregarding the taxonomic status of I2 (atypical) strains of enterococci that showedaherrant blochemica1 (sugar) reactions The I2 atyical strains include six rnannitolncgntl\e vanant E. /urculrs Itke species, one aq~nine negative variant E. ,fuecuhs likespec~es, h e mann~tol negat~vr vanan1 E /ui.crum like specles and two arginine negativevarlant E, cu.ssc~lr/lu~~c.s like species. %'hole cell protein (WCP) analysis by SDS-PAGEproved to assist in validating the species ident~ties as well, to identlfy stralns that do notexhibit phenotypic characterist~cs ~dentlcal to the type strains of each species [14. 31.98,091. We were able to val~date h e authentlc~ty of the unusual species, and the exactlaxonomlc slam of the atypical phenotypic bariant strains Identified by conventionalhlochemical testing. using WCP fingerpnntlng by SDS-PAGE as shown in Figure 5.Apart from thtse alypical strains. some stralns showed weak saccharolytic reactions andtook up 10 five to seven &p to cxhib~t a conclusive reaction.'he most prominent and common among all species of enterncocci are the E.lac8cabr md proper identification of this species is highly essential. Although mostlsolal~s exhibit the sun&d biochemical reactions. some strains were showing variable

CHAPTER Iresults for few tests in our study. Six strains failed to ferment mamitol, and one strain didnot hydrolyze arginine, both being key tests employed to identify E. faecalis, while 46%and 53% strains exhibited hippurate hydrolysis and glycerol fermentation respectively.Thm E. fieculis isolates strains could not utilize pyruvate, nor could they grow in thepresence of 0.04% tellurite, whose results was othenv~se to be the reverse for E. ,farealisIsolates usually. with the exception of some occasional asaccharolytic variantsistrains.Some E. fuecubs failed to ferment sorbitol and lactose. In case of E. faeciurn, threeisolates were not able to ferment mannitol even after a prolonged incubation of 7 days,while some strains (1-3 strains) showed very weak or absence of acid production fromarabinose. melibiose and sucrose. Only 56% and 48% strains hydrolyed hippurate andtolerated tetmmlium rrspatively, while 32% and 8% of the isolates fermented raffinoseand, xylose and MDG each respectively. The variable results for key tests (mannitol,arginine. tellurite reduction) used in identification~ditTerentiation of these two speciesposes a challenge and leads to m~sidentification of species, hence, to circumvent thisproblem we used WCP analysis for authenticating the exact taxonomic status of strainshelonglng to these two specles as shown by set era1 studies 114. 31,97-991.Although the seven unusual enterococcal species exhibited the expected results ofthe slandard biochem~cal reactions, some stralns faded occasionally. Two among three E.~~u.~sc.lr/]u~u\- isolated did not hydrolyze arglnlne. whlch posed a difficulty in speciesldentificatlon and was overcome by interpreting the resul~s of other tests like pigmentproduction and motility to difTerent~atc them from E, mu~ldrrr. Occasionally some speciesof unusual enteroctrci showed a very weak or absence of acid production from varioussugars as depicted in Table 9. Funher. the unique biochemical results of some strainsmay at times prove to be useful by acting as phenotpc markers, if any of these strainsare lnvolved in dissemination within the hosp~tal setup. U'hile se\~eral commercialphenotypic spaies identification systems (API Rapid ID 32 system). as well PCR based11 5, 105- 107. 109) and sequencmg based genot>pic techniques have been showed to beuseful in specie identification ofentemcocci [14, 3 3). we were successful in identifyingail the mterococci isolated in our study with the conventional biochemical phenotypingmethod and the WCP-SDS-PAGE based molecular phenotqping method [Id. 31.97-991.

CHAPTER ISUMMARYThe results of our study show that the prevalence of enterococci in our health care setupis highly significant.''. E. /ueculi.s (71%) and E. .furcium (10%) contributed to8 I % of all entemcoccal species isolated, while remaining 19% of enterococci comprisedseven different unusual species, which included E. gullinurum, E. uvium, E. raffino.vus. E.hirue, E, mundrii, E. cu.s.veliflu~w.~ and E, durun.,. The distribution by site of isolation forthe 242 entemcocci predominantly included l l l isolates (46%) from bloodstream, 72(30%) from urinary tract, and 59 (24%) from exudate specimens, while 75% of the all theisolates were from inpatient specimens. The conventional biochemical phenotyping tests~dentified and speciated majority of enterococcalspecies. while the molecularphenotyping using WCP fingerprinting by SDS-PAGE validated the authenticity of theunusual species, and the exact taxonomic status of the atypical phenotypic variant strainsand strains that showed weak saccharolytic reactions biochemically.Thus. appropriate species identification of enterococci plays a vital role inepidemiological surveillance wlthin hospitals, since the nosocomial transmission ofenterococcus is stead~ly increasing worldw~de. Howe\,er after initial identification.studying the antimicmh~al susceptibil~ty pattern of entemcocci IS required to initiateappropriate antlmicrob~al therapy. With the emergence of multidrug resistance includingam~noglycosides and glycopeptides. the necessity for sun.eillance of this emergingnosocomial pathogen has become necessap to minimize patient morbidity.

Chapter 11Jntimicro6ial@sistance in Enterococci

CHAPTER I1CHAPTER llt'nlerococcws has become an ascendant nosocomial pathogen by vinue of their versatilegenetic machinery enabling them to exhibit intrinsic, as well, acquired resistance toseveral antimicrobials. Hence, after the presumptive identification of enterwocci it isimperative to perform the antimicrobial susceptibil~ty testing of the isolate for rapidInitiation of appropriate antimicrobial therapy. Determination of susceptibility to selectedantimicrobial agents such as penicillin. ampicillin, and vancomycin is generally sufficientfor enterococci isolated from urinary tract and wound infections. However, for patientswith cndocardlris and pmbably other serious infections such as meningitis where asynergist~c therapy (aminoglycoside with a cell-wall active agent) is advocatedam~noglycoside activ~ty should be determtned. MIC of the aminoglycoside 2 2.000 @g/mlIS Indicative for HLR, and s~gn~fies reslstance to synergism also [13, 57, 2693. Thusqualitative and quantitative ant~microbial susceptibility testing of enterococci is highlycssent~al for cn'ect~ve management.Although the antlm~croblal \urceptlbllltj tests would depict the phenotype of thedntlmlcrobial resistance, the) do not dlrcrim~nate the reslstance markers encoded bycntemcoccl The am~noglycos~de resistant entcrcwcl may exhlblt \ersatlllty In theirgenetlc mechan~sm to encode resistance to a single antlm~crob~al In more than one way181, hence genolyplc analp~s helps In detc~t~ng the genetlcreslstanceof antlmlcrob~alOBJECTIVESTo dctcct ant~mjcrobial resistance and to determine the minimum inhibitoryconcentration of therapeutically active antibiotics against entemcocci.

cm PTER IIMATERIALS AND METH0,DSI. Antimicrobial Susceptibility testingAll isolates identified as enterococci were tested for their antibiotic susceptibility panemusing standard procedures and interpreted according to NCCLS guidelines as depicted inTable-4 and 5 [190]. The selection of antimicrobial agents for testing the susceptibility ofenterococci was according to the NCCLS guidelines, taking into consideration thecommon antib~otics used in JIPMER hospital.A. Disc diffusion method: The test was performed using Mueller-Hinton agar for thefollowing antibiotics, which include Penicillin (IOunits), ampicillin (10 pg), gentamicin-h~@ content (120 pg). streptomycin-high content (300 pg). ciprofloxac~n (5 pg),n~trofuranloin (300 pg). vancomycin (30 pg). teicoplanin (30 pg) and linezolid (30 pg).The inoculum was prepared as per NCCLS guidelines. Briefly. 2-3 colonies were pickedup from overnight growth on blood agar and suspended into 5 rnl saline and adjusted to0 5 MacFarland standard. The plates were inoculated and incubated at 3S°C upto 24 hrsIn ambtent air. and observed for the zone of inhibition and interpreted according toNCCLS guidelinesB. Agar screening method: The ~solates were screened for high-leb'el gentamicinresistance. high-level streptomycin resistance and vancomycin resistance and interpretedaccording to NCCLS gu~delines Briefly. 500 pgiml. 2,000 pg'ml and 6 pg'ml ofgentamicin. streptomycin and vancomyctn respectively were incorporated In BHI agar forthe screening method. The final inoculum was 10' CFU per spot, which was achieved byspotting the surface of agar with 10 pl of a bacterial suspension equivalent in turbidity toa 0.5 MacFuland standard. prepared frnm growth on an 18-24 hours agar plate. Theplates were incubated at 35°C for 24 hrs in amhient air. and growth of more than onecolony or r hw: of p>,wth was read as Indicative of resistance of the respectiveantibiotic. Swtomycin plates were re-incubated for an additional 24 hours if no growh

CHAPTER IIC. Agar dilution method: The Minimum inhibitory concentrations (MIC) of thefollowing antibiotics penicillin, ampicillin, vancomycin and teiwplanin were performedby agar dilution method and interpreted accordingly. The criteria for the range of MIC'stested were as follows: staning dilution was 2 fold less than the sensitivity breakpoint,and the final dilution was 2 fold more than the resistant breakpoint of the respectiveantibiotic tested, while an increment by two fold dilution of the antibiotic concentrationwas followed serially. The MIC for the isolates were performed with MHA platessupplemented with the respective antibiotic from their corresponding stock preparation.The final inoculum was 10' CFU per spot, nmhich was achieved by spotting the surface ofagar with 10 p1 of a bacterial suspension equivalent In turbidity to a 0.5 MacFarlandstandard, prepared from growth on an 18-24 hours agar plate. The plates were incubatedat 35°C for 24 hn in ambient air and interpreted according to NCCLS guidelines.2. Beta lactamarc detection b!. Chromogenic cephnlosporinnse methodThe entemcoccal ~solates exhibiting reststance to the beta lactam antibiotics: penicillinand, or, ampicillin were tested for 1%-lactamase productionby a chromogentccephalosporinase method using nttrocefin discs (BBL Microsystems) as permanubcturets Instructtons. Briefly. each nttrocefin disk moistened wtth one drop ofd~st~lled water was smeared with colonies from test straln and observed for a colorchange. H'hcn a bacterium produced p-lactamase enzyme in significant quantities. theyellow-colored disk turns red in the area where the isolate was smeared. A rapid colorchange from yellow to red occurs slnce the amide bond in the beta lactam ring ishydrolyred hy a beta-lactamase, while a negative result showed no color change on thedisk.3. Genotypic detection of rmiaoglgcoside resistance genesPCR hed lclcdbn of HUR genes: The aminoglycoside resistance is often encodedby aminoglycoside modifying enzymes (AMEs) among enterococci. The presence of thegenes that encode AAC(6')+APH(?") enzyme- HLR to gentamictn. and the ANT(@-1enzyme - HLR to streptomycin in the enterococcal isolates was confirmed by amplifyingspecific regions of those by multiplex PCR. The oligonucleotide primers chosen

CHAPTER !Ifor amplification of the aac(6')+aph(2") and ant(6)-l genes were selected from the~ublished sequences [255] as described previously [269].The primer sequences chosen to detect the aac(6 ')+aph(2") gene were:Forwardpnmer (primer 348) 5 '- TGA TGA TTT TCC TTT GAT GT -3 'Reverse prrmer @rimer 1723) 5 '- C M TCT TTA TAA GTC CTT TT -3 'The primer sequences chosen to detect the ant(6')-I gene were:Forward primer (primer 1268) 5'- ACT GGC TTA ATC AAT TTG GG -3'Rc,~~r.se primer /primc,r 1845) 5'- GCC TTT CCG CCA CCT CAC CG -3'ii) Oprimizarion of Multiplu-PCR: Initially the primer sequences for aac(6')+aph(ZW)and ant(6)-I genes were used separately fbr amplifying individual AME. Later primersfor both aac(6')+aph(2") and ant(6)-1 genes were used simultaneously for standardizing arapid mult~plex PCR for detecting HLGR and HLSR genes from HLAR enterococci. Thetemplate (bac~enal) DNA used for the PCR was standardized through various extractionprocedures Finally, the colony PCR protocol was followed throughout our study as theprocedure was simple and rapid. and the results were on par with those of reactions usingtemplate DNA extracted by conventional procedures. The volume of the PCR reactionmlxture was standardized hy scaling down to a mlnimal volume of 50 pi, without anycompmmlse in the qual~ty of amphfication reactlon and subsequent post PCR analysis.iii) Mubipler-PCR: The template DNA for the multiplex PCR was obtained as follows.Briefly, 4-5 bacterial colon~es from an overnight 5% TSA blood agar were suspendedusing a sterile macro t ~p in the 50 pl chilled reaction mixture containing 1X PCR buffer(with 1.5 mM MgCIZ). 4mM MgCIZ. 0.2mM denxynucleotide triphosphates (dATP.dC1'P. dGTP. d77'P). I pM primer (forward and reverse) of aac(6')+aph(T) gene and 0.5pM primer (forward and reverse) of ant(6)-I primer (Sigma Aldrich, Suffolk. UK), andSunits Taq DNA polymerase (Bangalore Genei, India). A gradient thermal cycler(Eppendorf. Gamany) was programmed with the following conditions: an initialdenaturation for10 minutes at 94"C, denaturation lor 1 minute at 94"C, primer annealingfor I minute at 56.Z°C and extmsion/polymerization for 3 minutes at 72°C for 30 cycles.A find exlension step was canied for 10 minutes at 72OC and held at 4°C till analysis.

CHAPTER IIiv) Post PCR onoiysis: The PCR products along with 100 basepair molecular weightmarker (Bangalore Genei, India) were subjected to electrophoresis in a 2% agarose gelusing 0.5X TBE buffer at 90 V, stained with ethidium bromide, visualized under an UVtransilluminatorand documented using a gel documentation system (Vilber-lourbet,France).RESULTSI. Antimicrobial Susceptibility testing.A. Disc diffusion methodThe results of antibiotic susceptibility pattern of the clinical enterococcal isolates testedhy d~sc-diffusion method are shown In Table 10 a. The results depicts that all 242enterocnccal isolates tested were susceptible to teicoplanin and linezolid. while 92% weresusceptible to vancornycin. For the plactarns tested against enterococci, 58% and 69%tsolates were susceptible to penlcillln and ampicillin respectively, while 42% and 54%~solates were susceptible to high-level gentarnicin and high-level streptomycinrespectively Only 38% of all enterwocci uere susceptible to ciprofloxacin, while theurinary isolates tested for nitrofurantoln and ciprofloxacin showed 78% and 32%susceptibil~ty respectively as depicted In Table 10 bThe break-up of the antihlotic susceptib~liry patterns of various enterococcalspates tested are deplcted In Tabla 10 c - c. E. /urc,uhs isolates showed highestresistance rates of 58% each to kth high-level pentamicin and ciprofloxacin. The high-level gentamicin mtstance was exhibited hy 86% (65 of 79 isolates) of high-levelstreptomycin mistant isolates. 83% (53 of 64 isolates) and 90% (37 of 41 isolates) ofpenicillin and ampicillin mistant isolates respectively. 79% (78 of 99 isolates) of thecipmfloxscin misant isolates, and 100°/~(I? of I? isolates) of the vancomycin resistantisolates. lnlmnediate level resistance was exhibited by 6% of the isolates to high-levelstreptomycin, 4% lo high-level gentamicin and vancomycin, and by 2% isolates tociprofloxacin.

CHAPTER 11Table 10 a. Antibiotic susceptibility pattern of enterococci byKirby-Bauer disc-diffision methodAntibiotic testedDisc Enterococci (242)stren%h, PI! No. (%) S No. (%) RPenicillin 10 units 140 (58) 102 (42)Ampicillin 10 167 (69) 75 (31)Gentamicin [HI.R]' 120 101 (42) 135 (56)Streptomycin [HI.R] 300 130 (54) 101 (42)Vancomycin 30 222 (92) 12 (5)Tetcoplanin 30 242 (100) 0 (0)'IiLR, high-level reslstanceThe E ju~v~rirnr isolates showed resistance of 64% and 60% to ciprofloxacin andh~gh-level gentamwin respectively. The high-level gentamicin resistance was exhibitedby 89% (8 of 9 isolates) of h~gh-level streptomycin resistant isolates. 86% (12 of 14isolates) and 859b (I Iof 13 isolates) of the penicillin and ampicillin resistant isolatesresprcrively. and 7890 (13 of 16 isolates) of the c~profloxacin reststant isolates.Intermediate level reslstance was exhibited hy 8O.0 of the isolates each to ciprofloxacinand vancomycin.Table 10 b. Antibiotic susceptibility pattern ofurinary enterococciEntrrorocrlE a c 6 / Ea c m (2) E. muftdii (I)I1'Cipo. Cipmflonscin; Nitrofur. Ni~rofurantotn (300 pg).

CHAPTER 11Table 10 c. Antibiotic susceptibility panern of E. fuecalis and E. faeciumAntibiotic testedPenicillinAmpicillinGcnram~cin [HLR]-NO. (%)S107 (63)130(76)65 (38)E. faecalk (171) E. faecium (25)-NO. (%) No. (%) No. (%) No. (%)I R S 1NANAh (4)64 (37)41 (24)100 (58)11 (44)Streptomycin [HLR]~ 84 (49) / I l (6) / 76 (44) / 16 (64) 1 0 (0) / 9 (36)NANo. (%)R14 (56)The unusual entemcoccal species showed 100% susceptibility for linezolid andtelcoplanln. wh~le E, c~us.\-c,lijlur~u.s and E. mundrii showed 100% susceptibility forpenicillin and ampicillin. Only 37% of unusual en\erococcal isolates were susceptible tociprofloxacin, with resistance exh~b~ted by nine E, ur,ium (n-LO), two E. durans (n-2),eleben E. ~ollrnanrm (n-IS), two E hirac, (n-6). two E. mundrii (n-4). and three E.~.u#ino.su.c (n-6). Only E. co.s.\c.liflat~r.~ (n-3) exhibited 100% susceptibilityciprofloxac~n. whtle vancomycin resistance was exhibited by six isolates (one E, durans,two I.;mundrii and three E gullrnurum). 4390 of the unusual entemcoccal speciesexhibited high-level gentamicin resistance, which included eight E. gallinontm, nine E.ursrirm, two E. ruflno.~us, and one E, dirrari.~. The high-level gentamicin resistance wasexhibited by 83% (20-nine E. or~,i~i, one E. durans, eight E. gullrnamm and two E.ru/linusus of 24 isolates) and XlQ,; (17-six E. ar~rum. one E, drrrans, eight E. gollitrammand two E. ru/iinr,.sus of 21 isolates) of the penicillin and ampicillin resistant isolatesrespectively. 56% (nine-four E. urium, one E, durans, two E. gallinarum and two E.ruDinorcrs, of 16 isolates) of the high-level streptomycin resistant isolates. 71% (20- nineE. u~.ium, one &. dumns, eight E pa//;narum and two E. ralfiriosus. of 28 isolates) of thecipmfloxacin mistant isolates, and by 67% (four-one E. htrans and three E. gallinarumof six isolates) of the vancomycin resistant isolates.to

CHAPTER IIB. MIC and their concordance with Disc dgfusion methodThe MIC for all enterococcal isolates tested and interpreted according to NCCLSguidelines 11901 by agar dilution and agar-screening method are depicted in Table 11 a.The results depicts that all 242 isolates tested were susceptible to vancomycin andteicoplanin by agar dilution method. while 19% of enterococci were presumptivelyresistant to vanwmycin by agar screening method. 60% and 43% of enterococci wereresistant to high-level gentamicin and high-level streptomycin respectively by agarscreening method, while 43% and 31% of enterococci were resistant to penicillin andampicillin respectively by agar dilution method.Table 11 a. MIC testing results of enterococci by agar dilutiordagar screening method'Interpre~al~ons based on NCCLS pldel~nes [190]. Susc- Susceprlbl~t)'I'm.Pm~clll~n. Amp. Amp~cllltn. \'an. \'ancom)r~n. Te. Te~coplanln. NA-notappl~cableVa Scr. Vancomycln reslslance (6 pp ml ) agar screenlngI-ll.Gm. HI&-level pntamlcln reslslancc (500 pg mL) agar screenlnpIlLSe. Hgh-level streptomycin reslalance (2000 kip ml.) agar screenlngThe distributions of MIC ranges for dilferent entemcoccal species tested are asdepicted in Table 11 b and r. The E ,fucculi.\ isolates showed resistance rates of 64%(1 10 of 171 isolates) to high-level gentamicin by agar screening method. which were indiscordance with disc diffusion results that cxhibitd only 58% (100 of 171 isolates)nsistance. The dime~nce is attributed to six E. ,/uc~colis isolates that showed intermediate

CHAPTER 11resistance by disc dihion, while another four isolates that were sensitive by discdiffision exhibited resistance by agar screening method. The agar screening methodshowed a 2% increase, i.e.: 46% resistance to high-level streptomycin when comparedwith the disc diffusion method (44%). The difference is attributed to eleven E. fuerulisisolates that showed intermediate resistance by disc diffusion, of which three exhibitedresistance by agar screening method while the remaining eight isolates exhibitedsensitivity. The 2% difference in the penicillin resistance between the results of MIC byagar dilution method and disc diffusion method, was due to three E. fueculis isolates thatshowed a lane size of lSmm and hence categorized as sensitive by disc diffision methodbut exhibited resistance by agar dilution. The results of ampicillin resistance were inconcordance with the disc diffusion method. The results of vancomycin resistance byagar dilution and agar screening methods were discordant with those of disc difisionresults. The vancomycln agar screening method depicted presumptive resistance among30 ( 1 7%) E. fbc~colrs isolates as per NCCLS bqidelines [190]. including one isolate whichshowed lntermed~ate resistance and five isolates that showed resistance by agar dilutionmethod, and six isolates which shoujed intermediate resistance to vancomycin by discdlfrusion method Funher. 70% (21 of 30 isolates) of E. ,/arculis isolates that wereprcsumptlvely mistant to vanwmycin by agar screenlng method also depicted resistanceto hlgh-level gentamicin.The MIC results of E. ,fac*crum ~solates showed concordance with the results ofd~sc diffusion method for all antibiotics tested, except for vancomycin as depicted inTable 11 b. The vancomycin agar screcnlng method depicted presumptive resistanceamong four (16%) E. jiuc~cium isolates. which included one sola ate that showed~ntenncdiate resistance by agar dilution method. one isolate which showed resistance andtwo isolates that showed intermediate resistance to vancomycin by disc diffusion method.Funher, 75% (three of four isolates) of E ,fuc.~.iurn isolates that were presumptivelyresistant to v-mycingentamicin.by agar screening method also depicted resistance to high-level

CHAPTER I1Speck tnted Antibiotic(no.of.ltdrtn)E Jueculir(I7''Table 11 b. MIC testing results of E. faccalis and E. faeciumtestedPenAmpTcNo. of isolates at ~peclfld MIC, in mg/mL2 4 8 16 32 -M107 76 69 67 51 3071 48 41 41 37 ISSurc' ,%S %I 0 0 0 9 9 0VaScr NA 1 NA NA 30 NA NAHLOm NA NA NA 110 NA NA NA 1 M 16176R */.3924The MlCs of the unusual entemc~cal species are shown in Table11 c. depictingthe ranges of MlCs for vonous antimicrobial agents tested by standard agar dilution andagar scwening methods. The results of MlCs for penicill~n. ampicillin. high-levelgentamicin and h~gh-level str~~lomycin resistance were in accordance with the discdilrus~on tesung results except for vancomycin The E. gullrnarum and E. cassd!fla~usrsolates showed reduced susceptibil~y to lesser concentrations of vancomycin ranging 2-8 p&'ml depicting their property of intrinsic resistance to vancomycin. The vancomyclnagar screening method depicted presumptive resrstance among I I unusual enterococcalrsolates (two E. rnunrlrir. one E. cusseIiflur~lt.v, one E, dorans and seven E. gallina~um),Including six isolates (one E, d1tran.v. two E. mrrndrir and three E. gallinarum) thatshowed vancomycin resisunce, and five isolates (four E. gallinarum and one E.c~sseltflu~u~) which showed susceptibili~y ro vancomycin by the disc diffision method.

CHAPTER 11'Spccin tutcd~no.of.holrta)E grll,namm(15)t cmselth,~\(3)ta~tumII1Ilf mundtn 141I hrrm 161t rafinost't(blt duwnt (?ITable 11 c - MIC lesttng results of unusual specles of enterococc!AnUMocLrlnldPaAmpVanTeVakrHLGmHLSlrPaAmpVanTcllCrHLLmHLSlrPmAmp\a"TrtrkrIilGmHlSuPrn41p\an1,\ 111Tr\nkHlGmti1SnPm\an1 c\,LC){lhHLSllPmNo.ollwlra~~t rpclfird MIC, In mglmLSlur' .%298134989a882lh88032870~ 6 4860S%466466NAR %533533NA0 0 0 0 U O l 0 0 0NANANA0ONANANA00hANANAOI178500NANANA01NANANA00533466bb6100IIN466513333003 3 0 O O O N A N A0NA4ANA90I*A4A4A90hAhAhiA00IOII90WANAhA70NANANA6I00666100IW10V 6 6 6 6 0 404I)44hA\420h444(1UhA\AhA(I0il95(I0hAhA4AO0\AhA4A100100IMI1050I11I1)f l O 0 100 0100h444\A440hA4AZAI02I1I1 il 0\ a h4 \4 \AHLhIilPm\A441h4441hA242* h ? -II II11 114AhA\A44(II4A"1(10hAh44444n0444A"IA1002I12:?11Ufl00440(1022(I001.4\4h4(10n04hA44440n4410010050l(KI50M66461cx1100100iWIM)I31333IMilcnlI(M666b033300906000090500000501150313333o000(166666600031333350\A h4hA: 1 l l 0 0 5 OA q ~ ~ ~ ~ 50 n 0\ m : ( 0 0 0 0 1 0 0(1TC 1 0 0 0 n 0 l w0VlktHLG~NANAAUANANA1I4ANANAhAJ0 5050h~ NA N4 ? NA NA O 100N4hAI1uh44A\400oU4AhANA4(10n444'

CHAPTER I12. Beta-lactamrse detection.None of the penicillin and ampicillin resistant enterococcal isolates tested for betalactamaseproduction using a nitrocefin disc yielded a positive result. This depictspenicillin binding protein modification based beta-lactam resistance among the betalactamresistant enterococci in our clinical set-up.3. Genotypic analysis of HLAR enterococci by Multiplex PCRThe 145 and 104 enterococcal isolates, which exhibited high-levelgentamlcin and h~gh-lebel streptomycin resistance by agar screening method, weresubjected to multiplex-PCR to study the genetic basls of their high-level aminoglycosideresistancc and the rtvults arc dep~cted In Table 12.Table 12. Multiplex-PCR results for HLAR enterococciThe bifunctional gentamicin resistance gene. aac(b')+aph(2") was present in 96%(106 of 110 isolates) of E frt~cul~r e sol ares which exhibited resistance to > 500pe/mlgentamicin. and lhe ant(6')-I gene (streptomycin resistant) in 94% (71 of 79 isolates) ofE. .fQecalis isolates which exhibited resistance to > 2000p&9nl streptomycin. Both thegenes were present together in 70 E ,Ct~cal~s isolates as shown in Figure 6.

CHAPTER 11Figure 6. Multiplex-PCR results of representative HLAR strains of EnterococcusM, 100 base palr marker, PC. Posltlve control. NC, Negatlve controlLanes I- 20 HI&-level amlnoglyco~lde resistant [HI-AR] test stralns of En/er~xoccusHI GR. HI&-level gentamlcln reslstance gene aac(h')+aph(2") - 1375 bpltl.SR, Hlgh-level streptomycin reslstance gene ant(6')-1 - 577 bpThe aac(b')+aph(2") gene was present in 87% (13 of 15 isolates) of E. ,/aeciumisolates which exhibited reslsunce to > 500pg/ml gentamicin and the ant(6')-I gene(streptomycin resistant) In 8906 (8 of 9 isolates) of E. /uc,crum lsolates which exhibitedresistance to > 2000p@rnl streptomycin. and both the genes were present together in eightE /uc,cium isolatesAmong the 46 isolates of unusual species of enterocwci tested foram~noglycoside resistant genotyty. only eight isolates (two E. pall~no~~rm and six E.~r~.i~rm) exhib~ted aac(b')+aph(2") gene, while four isolates (two E. gullinantm and two E.m*rum) exhibited the ant(6')-I gene, which also possessed aac(b')+aph(2") gene in them.Overall. !he isolates negative for these genes suggest alternate aminoglycoside resistancemechanisms among unusual enten~ixcnl species exhibiting HLAR

CHAPTER 11DISCUSSIONEnterococci have been become predominant nosocomial pathogen due to theirremarkable ability to acquire and disseminate antibiotic resistance by a variety of routesespecially in hospital settings. The emergence of resistance to aminoglycosides and P-lactamslglycopeptides among nosocomial enterococci is of clinical significance, sinceresistance to a combination of the abovementioned antimicrobials poses a greattherapeutic challenge (1 3, 17. 571. The trends in antimicrobial susceptibility vary within,as well, between countries and continents depending on various factors, which includesthe characteristics of the healthcare facility, infect~on control practices and antimicrobialuse. Hence, continuous surveillance of this nosocomial pathogen helps in tracking theemergence of newer resistance in any healthcare setup, which enables to initiate andexecute appropriate infection contml measures at the right time.We ~nvestigated the prevalence of resistance among enterococci to variousantibiotics. by different phenotypic and genotypic methods. Disc diffusion testing resultsdepicted that all 242 enterococci tested were highly susceptible to teiwplanin andlinezolid, which 1s consistent with results of some lndian studies (141, 3251 but incontrary with many studies from U.S. U.K. Europe [5. 15. 17. 57.264. 276,2951.Many independent and national surveillance studies conducted worldwide havedepicted 85 to 100% susceptibility for nitmfurantoin, and 45 to 85% susceptibility forcipmfloxacin (38-42. 69. 334-3371 A two-year Indian study (1997 to 1999) fmmChandigarh showed 85% and 92% susceptibility to cipmfloxacin by the urinaryenterococci from inpatients and outpatients respectively [31 I]. The present study showed38% of all enterococcal isolates to be sensitwe for ciprofloxacin, while another studyfmm New Delhi showed that only 12% of E. .fuc,calis were susceptible to ciprofloxacin(1541. These results arc consistent with the fact that ciprofloxacin are usehl only intreating UTI, and is less effective in treating other serious entemcoccal infections [13.161. The administration of fluoroquinolones has an impact on gastminrestinal flora, withan associatad increase in the carriage rate of E. ,fucciurn, apart fmm causing entemcoccal

CHAPTER I1superinfections [146]. Funher, studies have shown an increase in the isolation rate ofentcrococci from urine samples in the hospital correlated with the mounting consumptionof fluoroquinolones [I461 that may have a possible indirect impact on the selection andspread or vancomycin resistant enterococci in a hospital setup. Hence appropriate use offluoroquinolones by the clinicians based on the microbiological and antimicrobialsusceptibility will help minimize the adverse effects of fluoroquinolones on hospitalmicrobial ecology.The MIC testing of plactam antibiotics by agar dilution method depicted that43% and 31% of all enterococcl were resistant to penicillin and ampicillin respectively.which was concordant with the disc diffusion results. Our findings were similar to theresults of many recent studies that have depicted a gradual increase in the resistance ratesof penicillins over the years. But the resistance rates vary according to the country, withU.S isolates exhibiting comparatively higher resistance especially by E. ,fuecrum isolates.Boyce et al. [60] in their study showed the incidence of ampicillin-resistant enterococci(ARE) increased sevenfold at a university-afliliated hospital m U.S between 1986 and1988. Subsequently. many studies conducted in the early 90's showed beween 5 to 30%of entemcocci were ampicillin resistant and E. ,fu~,crum isolates exhibited higherresistance than E. /uet,uh.\ (62. 3381. Torell et al. (63, 641 showed the lncldence of AREamong enterococcal Isolates at a Un~\.ersity Hospital in Sweden increased from 0.5% to8.1 % between 1991 and 1995. The SENTRY antimicrobial surveillance programconducted during 1997 to 1999 in U.S. Canada. Lat~n America, Europe and Asia-Pacificthrough reference laboralones showed that 76% to 99% of enterococci were susceptibleto ampicillin, with highest resistance (24%) exhibited by U.S and Canadian isolateswhlch was increasing gradually over three yean of the study [32R].

CHAPTER IIresults were consistent with these facts, since 52% and 24% of E. faecium and E. faecalisrespectively exhibited resisthe to ampicillin. The prevalence of ampicillin resistant E.luecium (AREF) in several studies were in accordance to the results of ow study. Asurveillance of gram-positive bacteria isolated from patients with hospital-acquiredInfections or community acquired infections between 2000 to 2001 from China depictedthe rate of AREF was 73.8% (31142), that was significantly higher than ampicillinresistant E. .faecul~s(16.4%, 471286) (3391. Another study conducted in five Nordichospitals showed that all E. ,faecalis were susceptible to ampicillin, although half the E.fuecium (range 33 to 61%) were resistant to ampicillin [l03]. Some studies have depictednosocomial outbreaks with subsequent endemicity of AREF [61,283].Our results were concordant with many Indian studies depicting a gradualincrease in the resistance rates of penicillins over the years. while few studies contrastedour results. Only 10.2% of enterncocci from urine cultures exhibited penicillin resistanceIn a study conducted in a tertiary care hospital in Vellore, in 1996 [321], whichsubsequently increased to 32% as shown by another study from Nagpur [322]. A recentone-year study from a tentary care center in New Delhi showed emergence of multi drugresistance in E. ~u~culis isolates depict~ng 66% resistance to ampicillin (1541, which wasalmost double than the present stud) (31%). As none of the enterococci were klactamaseproducer, penicillin-binding protein appears to be the predominant mechanism ofresistance to klactams by the isolates In this study that was concordant with severalstudies worldw~de 162. 159. 335.3401, Marked diRerence was observed with other lndianstudies 1323. 3301. which have show up to 5@/0 $-lactarnase associated resistance. Theuse of nitmefin for detection of klactamase may have given a highly sensitive result inour study.The emergence of vancomycin resistance among entewci (VRE) brought thisnosocomial pathogen into limelight, although for a grave cause. Since their first detectionin 1986, there is a steady increase in vancomycin resistance among nosocomial isolates ofenterococci worldwide. From 1989 through 1993. the proponion of VRE reported toCDC's National Nosocomial Infections Surveillance (NNIS) system increased from 0.3%

CHAPTER I!to 7.9% [58] and subsequent studies showed upto 47% increase of VRE from 1994 to1998 (591. While a surveillance data reported by the NNIS system for 1993-1997compared with January-November 1998, showed a marked increase (55%) in VREassociated with nosocomial infections in ICU patients from U.S. hospitals [5]. But theof VRE varies among different countries and continents, which are governedby various factors including the use of glycopeptides in humans and animals (growthpromoters). Although several studies over the years show that U S and Europe top the listin VRE, it is yet to pose serious threat in some countries including India, althou& reportshave already depicted scanty prevalence of VRE [335].In our study all 242 enterococci tested were highly susceptible to vancornycin andteicoplanin bv agar dilution method. while 19% of enterowcci were presurnptivel\resistant to vancomycin by agar screening method as pcr kU'L S gu~delines I 1()0] Hulour disc diffusion testing results ~howed only 8% 04' eliterococcl a, resistan: 11)vancomvcin There were differences in our reallta between disc diffilslon testinp and agar~reeningagar dilution method that are shown to arise while testing cnterowcci forvancomvcin resistance [?70] The difference in the results ma\. be attributed to theIntrinsic low-level vancom\cin resistance (van C genotype) exhibited by I,..~1~11111irr1rrnand I.. c.cnsz/!fi~l.n\ isolates. which nlay go undetected hv disc dimusion teslin~ Rut. inour study /..,krzc trht and 1.. /trc~c,rrtnr isolates too showed contradicton results \\hen testedfor vancomycin resistance hv d~sc diffusion method as shown in Tables 10 and 11 Thebreakpoin~ concentration of 6 p$ml used in the vancomycin apat screening mcthtd basable to daect more number of entertuxxi which werc presuniptivclv resistant tobancomycin as per NCC1.S guidelines [190]. because wcral isolates toleratin$ thcconcentrations of 4-8 pcyiml could be detected by \.ancomvcin agar screenin$ 11lcth~ldwhich othemise goes undetected in the agar dilution method since only co~ice~llrat~oll.hetween 8-16 pgml are ~tidicated as intermediate resistance and corrcentrslion lesse~ thanthis breakpoint are considered as sensittve Iience. any1 enterwmci showing intermediateresistance/borderline scnsitivitv lo vancomycin by disc ditYilsion lnethod need 10tesred by agar screening method for presumptive resistance and by seal dilution methodbe

CHAPTER I1for determining MIC and confirming resistance as ev~dent by the results of our studv tlla!has also been recommended by other stud~es (93.?70. 27 11Although vancomycin resistance among enterococci is quite divergerrt whencompared with majority of U.S and European studies, our results were consistent withsome studies including those from India. The SENTRY antimicrobial surveillanceprogram conducted during 1997 to 1999 in U.S. Canada, Latin America, Europe andAsia-Pacific through reference laboratories showed that 83% to 100% of enterococciwere susceptible to vancomycin, with highest resistance (17%) exhibited by U.S isolateswhich increased gradually over three years of the study 13281 and in the preceding year(1997) 14% of the U.S isolates exhibited vancomycin resistance in the same study [37].These studies reflect the general trend of vancomycin resistance among enterococci inwestern counIries. as reviewed previously. A study from Nagpur, India showed that 3.3%of 150 enrerocmi ( 129 E. fueculrs and 2 1 E. ,juecilrm) exhibited low-level vancomycinresistance with MIC 516 pdml 13223. Subsequentl!. Purva et al. reported the first case ofvancomycin-resistant E fuc>c.rlmnr In 1999 from New Delhi isolated from the blond cultureof a patient with non-Hodgkins lymphoma [3?3]. Thereafter, studies showed I% to 5%vancomycin resistance among entcrococci [154. 3231 Recently Taneja et al. [I531showed that 5.5% of 144 urinary cnterococci isolated from urinary specimens exhibitedvancomycin resistance, which Included five E /uc,ci~rm. and one each of E. ,jueculrs. E.c~ussrliflo~ws and E pscwdou~'rum with MIC ranging from 8 to 32 pglml. Our study isconsistent with a study by Kamarkar et al. (3251 from Mumbai (Bombay) that showed23% of entemcocci (52- ten E. /ueculi.\. and fony-two E fuecflrm) isolated from clinicalspimens, were resistant to vancomycln w~th an MIC > 4 pp'ml, but sensitive tote~coplanin depicting a van-B resistance.In this context. the emergence of vancomycin resistance among enterococci inIndia is a cause lor concern in the near future. Hence. In hospitals where VRE is yet 10 bedetected. periodic culture surveys of stools or rectal swabs fmm patients at high risk ofVRE infection or colonization are recommended 13281. Fuflhermore, enterococci from allclinical specimens need to be tested by vancomycin agar screening method to detect them

CHAPTER 11early to initiate appropriate infection control measures, as well restructuring the antibioticpolicy if needed.One of the biggest therapeutic challenges is treating serious enterococcalinfections showing HLAR. A synergistic combination regimen is not possible even if theisolate is susceptible to either of the cell-wall active agent (P-lactamslglycopeptides)[13,140, 3411. Hence, studies on the prevalence of high-level aminoglycoside resistance areof prime importance to choose an alternate therapeutic combination. The prevalence ofhigh-level resistance to streptomycin was first reported in 1970, followed by severalreports of streptomycin and kanamycin resistance [249, 250, 1171 and the first report ofhigh-level plasmid-borne resistance to gentamicin came in 1979. It was reponed in threestrains of S~reprococcu.s fuecu11.s subspecies zymogenes that were also resistant to otheraminoglycosides [254]. Since then several studies have been conducted worldwide onprevalence of amtnoglycostde resistance in enterococci and the frequency of HLGR hasIncreased from 4.5% to 65% over the years 147. 342.3431.The present study showed that 60% and 43% of all enterococci tested wereresistant to high-level gentam~cin and streptomycin respectively by agar screeningmethod. which included 64% and 46% of E. /ueculis, 60% and 36% of E. ,faeciurn and.43% and 37% of the unusual enterococcal species exhibiting resistance to high-levelgentamicin and streptomycin respectively. In 1995, the first report of high-levelaminoglycoside resistance among enterococci from India was publtshed [320]. Later in1997, Bhat et al. showed that among 41 strains of enterococci isolated from cases ofneonatal bacterernia. 8.6% and 33.3% of E, fuecu1i.c and E. ./uecium strains respectivelyshowed high level gentam~cin resistance. while 6% and 50% of E. ,faecalis and E.lueclum strains respectively showed high level streptomycin resistance [174]. Anotherstudy from our hospital reponed endocarditis caused by high-level gentamicin resistantenterococci in 1998 [309]. since then the rate of HER enterococci was steadilyincreasing in our hospital as evident from the results of the present study. Our study isconsistent with another recent study by Randhawa et al. [344], which showed aPnvalntcc of 68% and 43% for HLGR and HLSR respectively, but their study did not

CHAPTER IIdifferentiate the species prevalence of enterococci. Likewise, another study by Karmarkaret al. [325] from Mumbai (Bombay) showed that 100% of E. fuecalis and 86% of E.lireciurn (52 enterococci- ten E. /ueculis and forty-two E. fuecium) isolated from clinicalspecimens were resistant to high-level gentamicin.The prevalence of HLGR was higher than HLSR in our study, which reflects thegeneral trend as evident through many studies from European and Asian countries thatincludes Spain, Croatia, Taiwan, Greece, Japan, and India (141. 325, 330, 335, 343-3491,But some studies contradict our findings as evident from their results, which shows ahigher prevalence of HLSR than HLGR among enterococci. A study from Canadashowed that 4% and 29% of E. /uc,cium strains exhibited HLGR, and HLSR respectively13381 that was consistent with the results of other studies from U.S. and Canada [329].Subsequently the mu111 centered SENTRY antimicrobial surveillance program conductedIn 1997 showed that 46% and 76% (Canada). 40% and 70% (US.), and 17% and 50%(Latin America) of E. /uc~c.rum strains exhibited HLGR and HLSR respectively, while46% and 52% (Canada). 27% and 32% (U.S ). and 43% and 33% (Latin America) of E.furculis strains exhibited HLGR and HLSR respectively [37]. Further. HLGR rates mayvary considerably between lahoratorles (1% to 28%) of different hospitals within thecountry as shown by a study from five Nordic hospitals [103].The excessive use of gentamic~n In our setup (as standard prophy1actic:therapeuticreg~mens) may be a reason for a hlgher prevalence (60%) of HLGR among enterococci, afact that has been shown by some studies [18, 47. 1611. For example, mostly for all casesof pre-operative prophylaxis in our hospital, a dose of gentamicln along withmetminidazole is glven wilhiw~thout ampic~llin (or) cephalosporins depending on thetype of surgay. Since the gentamicin prophylaxis is aimed to eradicate gramnegative(GN) organisms, the enterococci in the GI tract are tolerable to the dose regimen for GNbacteria. Hence they thrive and out compete other strains and creates a niche for itself.and under debilitating conditions gets disseminated through any possible foci leading tobactds or other associated infections (18. 47. 161). Subsequently, the re-operativePlophylsctic ngirncn is followed empirically for post-operative periods and remains

CHAPTER IIunaltaed, except on occasions of a culture report suggesting an alteration. In the pediatricspecialty of our hospital, exhpirically all cases of neonatal sepsis are treated withampicillin and amikacin or. gentamicin, and switched over to cefotaxime along withgentamicin if needed, or indicated otherwise. While, the cases of neonatal sepsis in theneonatal ICU are treated with cefotaxime and gentamicin initially, and switched over to acombination of Cefeperazone and Sulbactum with or without vancomycin if needed, orindicated otherwise.Although streptomycin is rarely used, 43% enterococci exhibited HLSR in thepresent study. One reason for the prevalence of HLSR may be the clinical isolates mightcany the HLSR and HLGR traits on the same genetic element [17]. The results of thepresent study depict concomitant resistance to different antimicrobials includingstreptomycin that is consistent with several studies. Agarwal et al. [322] showedconcomitant high-level resrstance to penicrllin and aminoglycosides in 16% ofenterococci from Nagpur. while another study showed concomitant resistance topenicilhn and amrnoglycosides in 61% of E ,fuecalis isolates [3 171. A recent study fromNew Delhi showed that HLGR and HLSR were present together in 43% of enterococciisolated fmm pediatric sept~cernic cases 134.41. Thus the concomitant resistance to otherantimicrobials along with HLAR narrows down the therapeut~c choice, thereby posingdificulty in treating serious enterococcal infectrons exhibiting mult~drug resistance.More than 90% of clinical HLGR enterncocci possess the bifunctional gentamicinresistance gene aac(6')+aph(2") that encodes resistance to virtually all therapeuticaminoglycosides, including gentamicin. tobramycin, amikacin, kanamycin, andnetilmicin, but not streptomycin [255]. while less than 10% possess otheraminoglycoside-modifying enzymes (AME) like aph(2")-lc, aph(2")-Id, and aph(2")-lb181. HLSR is most commonly encoded by ant(6')-l gene and can coexist with the gene(s)for HLR to other aminoglycosides [8. 171. Occasionally HLSR in enterococci may be dueto ant(3")-la gene [a], or due to ribosomal resistance [256]. Hence genetic analysis of theHLAR among entcrococci helps us to know the differences, if any, in the epidemiologyof the HLAR gcnas between different countries and continents (292,349,3501.

CHAPTER 11The present study depicted that bifunctional gentamicin resistance geneaac(6')+aph(2") was present'in 96% (106 of 110 isolates), 87% (13 of 15 isolates) and40% (8 of 20 isolates) of HLGR E. foecalis. E. fuecium and unusual species ofentmci respectively. The ant(6')-l gene was present in 94% (74 of 79 isolates), 89%(8 of 9 isolates) and 25% (4 of 16 isolates) of HLSR E. foeculis, E. ,fuecium and unusualspecies of entemocci respectively. Both the genes were present concomitantly in seventyE. faecoli.~. eight E. jaecium, and four unusual species of entemocci. Our results wereconsistent with several studies, although few studies showed some differences in theprevalence of HLAR genes. The results of a Greek study showed that aac(6')+aph(2")gene was detected in 83% of E. /u~culis, while the ant(6)-I gene was detected in allHLSR isolates of E. fuecoli.~ and E. fuecium [292], which coincided with our results forthe prevalence of both aac(6')+aph(2") and ant(6)-I genes. The presence of otheraminoglycoside resistance genes like aph(2")-lcl aph(2")-Id/ aph(2")-lbl ant(3")-la. apartfrom aac(6')+aph(2") and ant(6')-I genes may be a contributing factor for the differences.Various studies from Germany [351]. U.K [350]. Greece [134], Netherlands [352], andKuwait [340] showed that the bifunctional gentamicin resistance gene aac(6')+aph(2")was present in all HLGR enterowccal isolates (1 00%) tested. Our results were similar tothese studles w~lh regard to the prebalence of aac(6')+aph(2") geneSome studies have indicated a vanery of distribution profiles of AME genesamong entemcocci. A recent study from a university hospital in Japan showed thataac(6')+aph(2") gene was present in 42.5% of E. furcalis and 4.394 of E.,faeci~trn, whilealmost half of E. fuc~cal~s and E. ,/ucJclum isolates were shown to possess ant(6)-la andaph(3')-llla genes The profile of AME gene(s) detected most frequently in individualstrains of E. ,fu~c.ul~.s was aac(6')aph(2") + ant(6)-la + aph(3')-llla. and isolates with thisprofile showed high level resistance to both gentamicin and streptomycin. In contrast,AME gene profiles of aac(6')-ti+ ant(6)-la+aph(3')-Ill& followed by aac(6')-li alone,were predominant in E. fat~cium. Only one AME gene profile of ant(6)-la+aph(3')-lIlawas found in E. uvium [349]. Anolher study from Spain depicted heterogeneity in thedistribution of AME genes like the previous study among HLAR strains, while more thanone AME gene was detected in 7 1 % of the strains [343]. While another study from U.S

CHAPTER 11showed that the aac(6')+aph(2"), aph(2")-lc, and aph(2")-Id genes were present inentemoccal isolates from ariimals, food, and humans, while aac(6')+aph(2") gene wasthe most common gene among the HLGR isolates evaluated in their study and wasdetected in various enterococcal species, including the E. faecalis. E. faecium. E.gallinarum, and E. cossel~flal~us isolates collected from human stool, chicken and porkpurchased in grocery stores and chickens, dairy cattle, swine, and turkeys on farms.These observations provide evidence of a large reservoir for these resistance genes inhuman, food and food-producing animals, indicating widespread dissemination of theseresistance determinants 13531.Funher, in the due process of our work we standardized the Multiplex PCR fordetecting the AME genes from bacterial colonies directly. The Colony PCR wascomparatively at par with the PCR performed using template DNA extracted by theted~ous and expensive conventional procedures. Hence, the colony PCR protocol can beadapted for routine diagnostics by clinical micmbiology laboratories to give a rapidantimicmhial susceptibility result within six hours of identification of enterococci.Although our study dep~cts a very high prevalence of aac(b')+aph(2") and ant(6')-1 genesamong the HLAR entcroccxcl. the prc%encc of other AME genes among these isolatescannot be ruled out, since we d ~d not test the isolates for presence of those genes whichhave been shown to be present along with the aac(6')+aph(2") and ant(6')-l genes amongHLAR entemcocct (343, 3491.SUMMARYThe results of our study showed that all the isolates tested were susceptible to teicoplaninand linezolid. and 92% were susceptible to vancomycin While 58% and 69% isolateswere susceptible to penicillin and ampicillin respectively, only 42% and 54% of theIsolates were susceptible to the aminoglycosides gentamicin (high-level) andstreptomycin (high-level) respectively. Minimal susceptibility against ciprofloxacin wasexhibi~ed by 38% of all enterococci, while the urinary isolates tested for nitrohrantoinand cipr~flo~acin showed 78% and 32Y0 susceptibility respectively. Of clinical

CHAPTER 11significance was the high-level gentamicin resistance exhibited by 58%, 60% and 43% ofE. ,fuecalis, E. Juecium and unusual enterocnccal species respectively. However, therewere differences in the results between disc diffusion testing and agar screeningagardilution method while testing enterococci for vancomycin resistance. None of thepenicillin and ampicillin resistant enterncoccal isolates tested positive for beta-lactamaseproduction.Genotypic detection of aminoglycoside resistance genes by Multiplex PCR,showed that the bifunctional gentamicln resistance gene aac(6')+aph(2") was present in96% of HLGR E. /ur*culia ~solates, while ant(6')-l gene (streptomycin resistant) wasdetected in 94% of HLSR E. fucculls. The aac(6')+aph(2") and the ant(6')-I gene wereprescnt together in 87% and 89% of E ,/urc.lrrrn isolates respectively. Among unusualspc~ies of enterococci tested for am~noglycoside resistant genotypes. only eight isolates(two E gullrnururn and SIX E. u~~virm) possessed aac(6')+aph(2") gene, while four isolates(two each of E. gulltnur~trn and E ui,~urn) possessed the ant(6')-l gene, which alsopossessed aac(6')+aph(2") gene In them. Overall. absence of these genes in somesugests alternate am~noglycoside resistance mechanisms among the enterncoccal speciesexhibiting HLARThe phenotypic and genotypic analysls of antimicrobial resistance in enterococcimay bc sufficient for rap~d initiation of appropriate anrlmicroblal therapy. However, forstudying the geographical trends of molecular basis of antimicrobial resistance and toundentond the epidmiology of antlmicmbial resistant enterococci many a times thisinformation alone 1s insufficient. Hence. further molecular characterization of the geneticdcterminants encoding antimicmbial resistance is required to investigate the magnitude ofthis multifaceted problem.

chapter 111Molkculhr Characterization oflFCih-helrninogbcosde qesktant Enterococci

CHAPTER IIICHAPTER 111The high-level aminoglycoside resistance exhibited by nosocomial enterococci apartfrom posing therapeutic challenge exacerbates the issue further, since their machineryhelps in dissemination or transfer of these determinants (drug resistance) via. plasmidsand transposons to other closely related species and genus more rapidly in nosocomialsettings [ 13. 12 1. 1381. The arninoglycoside resistant enterococci exhibit versatility intheir genetic mechanism to encode resistance to a single antimicrobial (aminoglycoside)in more than one way 181. Earlier studies have shown that there lies heterogeneity amongthe genetic determinants (plasmlds) encoding HLAR in enterococci [18, 135, 292, 350,354-3561 however, several recent studies have shown that a predominant type of plasmidwas present among many HLGR enterococci, which depicts their widespreaddlssernination In any given setting [IS. 251. 254. 354. 3571 Although a number ofplasmids have been associated with HLGR, the nature of the plasmids variesgeographically due to several reasons []?I. 1221. Hence after phenotypic and genotypicanalysis of HLAR enterococci, molecular characterization of the genetic determinantsencoding amlnoglycoside resistance helps In revealing the differences. if any, in thegeographical trends of molecular basis of aminoglycos~de resistance In enterococci, tounderstand the epidemiology of HLAR enterococcl.OBJECTIVETo detect and molecular characterize the genetic determinants encoding high-level aminoglycoside resistance in entemcocci.

CHAPTER I11MATERIALS AND METHODSI. Molecular characterization of HLAR enterococciA. Plasmid DNA profiling of HLAR enterococcii) Alkaline lysis method: The mini-preparations of plasmid DNA from HLARenterococci were obtained by the standard alkaline lysis method [327] with minormodifications as described below.Cell harvestingCell IyskA single bacterial colony was inoculated into Todd Hewin broth (THB 5ml) withgentamicin 500 pgiml and incubated at 37°C with vigorous shaking overnight.1.5 ml of THB culture was transferred to an eppendorf tube and centrifuged at13.000 rpm for 30 seconds at 4°C (This step was repeated if necessary, bydecanting the supernatant and adding 1.5 ml culture to the same eppendorf toincrease the cell mass)Then the medium was removed hy aspiration, and the bacterial pellet was lefi todry.The bacterial pellet was resuspended In a solution containing lysozyme 10 mg/mlin 10 mM Tris. I mM EDTA (pH 8 0). and 25% (w v) sucrose and incubated in awater bath at 3TT for I hourThen the mixture was centrifuged at 13.000 rpm for 30 seconds, and thesupernatant was decanted w~thout d~sturbing the cell pellet.The cell pellet was resuspended in 100 p1 of ~ce-cold Solution-l containingRNAse I00 p@ml by vigornus vonexing.Subsequently. 200 MI of freshly prepared Solution-I1 was added and mixed byinvaing the tube rapidly for S tlmes (without vortexing). and the tube was storedon ice.

CHAPTER IIIThen 150 p1 of ice-cold Solution-111 was added and vonexed gently for 10seconds to disperse Solution-Ill through viscous bacterial lysate and the tube wasstored on ice for 3-5 minutes.Then centrifuge at 13,000 rpm for 10 minutes at 4°C in a microfuge and transferthe supernatant to a fresh tube (If the supernatant was not clear re-centrifuge asmentioned above till the supernatant was clear).Recowry of Plasmid DNAEqual volumes of lsopropanol was added to the supernatant and mixed, andincubated at room temperature for 15 minutes (for better recovery of plasmidDNA the tube was stored at - 20°C for 1 hour).The mixture was centrifuged at 13.000 rpm for 20 minutes at 4'Csupernatant was carefully decanted leaving the (invisible) pellet undisturbed.and the1.5 ml of 70 % ethanol was m~xed to wash the (invisible) pellet thoroughly at 4OCand centrifuged at 13.000 rpm for I5 minutes at 4°C.The supernatant was removed by aspiration carefully and the pellet was allowedto air dry for 10 minutesFinally, the pellet was redissolved In 40~1 of Milli-0 water (by vortexing briefly)and stored at - 20°C till further analysisL~ais/E.nroc?ion buffers and SolutionsSolution-ISolulion-1150 mM glucose25 mM Tris-CI (pH 8.0)10 mM EDTA (pH 8.0)Solurion-l nu.^ prtyared rn hotc~hes o/ 100 ml, a~rrocla\~ed(or 15 minutes at 15 psion liquid qvclr and srorcd at 4°C.0.2 N NaOH (frtshly diluted form a 10 N stock)1 % SDSSolution-11 uosprcpured~rc.sh1~ hcforc us

CHAPTER 111Solution-1115 M Potassium acetate 60.0 mlGlacial acetic acid11.5 mlDouble distilled water28.5 mlSolution-Ill was stored a1 BC and [ransferred lo an ice bucker just before use.ii) Restriction endonuclease digestion of plasrnid DNA and Separation: The plasmidDNA isolated by alkaline lysis method was digested with restriction endonuclease EcoRl(Bangalore Genei, India) in accordance with the manufacturer's specifications asdescribed below.The lbllowrng mlxture was added to a microfuge tubeIOX restriction enzyme buffer 3 pl (to a final 1 X concentration)Plasm~d DNA10 p1koR1 restriction enzyme1 PIDouble distilled waterI6 p1(The final volume was made up to a 30 p1)The restrictton enzyme was added finally and gently mixed by spinning for 1-3seconds In mrcrofuge. The mixture was rncubated in a water bath set at 37°C for 5hours.The whole (undigested) plasmids and restriction d~gested plasmid DNA wereseparated on 0.8 YO agarose gel using 0.5 X TBE at 60 V stained with ethidiumbromide, visualized under an l!V-transilluminator and documented using a geldocumentation system (Vilber-lourbet. France).B. Southern blotting ojplasmid DA!4The experimental procedure was carried in Center for cellular and Molecular biology(C'CMB), Hydcrabad. India. The whole (undigested). and the restriction-digested plasmidDNA separated on agarose gels were transferred onto a nylon membrane by Vacuumblotting as described previously for Southern hybridization [327] with minormodifications.

CHAPTER IIIi) Vacuum transfer: The plasmid DNA separated on agarose gel was vacuum transferredonto nylon membranes by mi alkaline transfer method using an in-house Vacuum blottingapparatus (CCMB, Hyderabad, India) and Hybond-N+, which is a positively chargednylon membrane (Amersham biosciences) where nucleic acid samples may be fixed bysimple alkali treatment or alkali blotting. rather than UV exposure or baking. The briefstep-wise procedure followed was as below mentioned:The vacuum-blotting unit (apparatus) was kept in proper orientation to carry outthe blotting procedure.First, the pump inlet on the front panel was connected to the liquid trap, which inturn was connected lo the base of the vacuum blotter.Then a 3MM Whatman filter paper was placed on the vacuum plate after wettingit with distilled water, onto wh~ch the Nylon membrane (Hybond Nt membrane)cut according to the dimensions of the agarose gel, was placed after wetting itwith d~stilled water.A plastic mask (made from a polyethylene sheet) with a window in the center, cutaccording to the dimensions of the agarose gel was placed on the membrane insuch a way that it overlaps each side ofthe membrane by approximately 5 m.Then the frame was placed on top of the unit and the clamps were tightened.The agamse gel was placed onto to the membrane by gradually sliding the gelwithout entrapping air bubbles.It was made sure that the gel and mask overlapped by at least 2 m. while smallcracks or leakages in the agarose gel were sealed with lour melting point agarose.Depurination- The gel was covered fully w~th about 20 ml (depending on gelsize) of Solution-I (with a pipette) and immediately the vacuum pump wasswitched on and adjusted to exefl20 pounds pressure.The depurination step for plasmid DNA was carried until the bromophenol blueon the agarose gel turned yellow (about 20 minutes) while the depurination forchromosomal DNA from PFGE gels was canied for a longer duration (about 40minutes). Excess Solution-1 over the gel was removed after depurination bywiping the gel surface with a gloved finger or by using pipette.

CHAPTER 111wNucleic acid transfer- Immediately Solution-11 was poured onto the gel to coverit fully and transferred for about I - I .5 hours depending on the size of the DNA.It was made sure that the gel remains immersed all the time with Solution-l orSolution-Il whichever applicable during the depurination or transfer process.Once the transfer was completed, excess solution was removed from the gel andthe pump was turned off.The gel was removed carefully and stained with Ethidium bromide (0.5 ~lg/ml) for20 minutes and examined under UV-transilluminator for checking the efficiencyof nucleic acid transfer.Finally, the nylon membrane was removed (a lower right corner cut was made. tomark the orientation of the transfer of nucleic acid from agarose gel) and washedin Solution-Ill for 5-10 minutes w~th agitation, air-dried and stored at 4°C forsubsequent hybnd~zat~on experiments.Solutions and BuffersSolution-lSolution-llSolution-Ill0.22 M HCI (Depurination solution)0.44 N NaOH (Alkaline transfer solution)2 X SSCC. Preparation of DNA probes for HLAR genesThe DNA probes for high-level gentam~cin resistance [aac(b')+aph(2")] and high-levelstreptomycin resistance [ant(6)-I] genes were prepared as per standard procedures usingkits wherever applicable. Briefly, the aac(6')+aph(2") and ant(6)-1 genes were amplifiedfrom a standard straln by PCR and separated on a 2 % agarose gel. The gene specificfragments were gel purified using QIAGEN gel DNA extraction kit as per manufacturer'sinstructions (QIAGEN, Germany) and stored at-20°C till the radiolabelling of the probeswere done.i) Gel DNA utroction procedure: The gel DNA extraction protocol using amicrocentrifuge was followed as per manufacturers instructions. with all centrifugationsteps canid out at 13.000 rpm on a tabletop microcentrifuge (Biofuge. France).

CHAPTER IIIBriefly, the HER and HLSR gene specific DNA fragments were excised fromthe agarose gel with a clean, sharp scalpel by placing the gel on a UVtransilluminator using the reflector lights in the system.Then the gel slice was weighed in a colorless tube, and three volumes of BufferQG (containing guanidine thiocyanate) was added to one volume of gel andincubated at 50°C for 10 minutes with intermittent vortexing during incubation.After dissolving the gel slice completely, the color of the mixture was checked tobe yellow. If so, one gel volume of lsopropanol was added to the sample andmixed.Then the sample was applied to a QlAquick column to bind DNA, which wasplaced in the 2 rnl collection tube provided, centrifuged for 1 minute and the flowthroughwas discarded.Then 0 75 ml of Buffer PE was added to the column for washing. centrifugedtwice for a mlnute and the flow-through discarded.Finally, the QlAquick column was placed into a clean, sterile IS mlmicrncentrifuge tuhe and the DNA was eluted by adding 50 p1 of Milli-Q water tothe center of the column. and let the column stand For a minute beforecentrifuging for a minute The eluted DNA collected in the microcentrifuge tubewas stored at - ZO'T till the rad~olabelling of the probes were done.The eluted DNA was confirmed by running the sample on a 2 % agarose gel witha molecular weight marker used previously.ii) Radiolobeling of the DNA probe by Random priming method: The radiolabeled DNAprobes were generated by using a random primer kit and a radiolabeled dNTP- [a-32P]dATP (BARC. Mumbai, India) as per manufacturer's instructions.Briefly. 2 ~1 of template DNA was added (gel purified and eluted DNAcorresponding gentamicin and streptomycin genes. and lambda DNA [NewEngland Biolabs. UK]) in 20 PI of sterile water using a sterile 1.5 ml microfugetube

CHAPTER IllThe DNA was denatured at Y4'Cfor 2 minutes by placing the tube in a boilingwater bath, and the &be was removed and snap-freezed immediately by placingthe tube on ice.Then add appropriate volumes of reagents were added in the following order:Random primer buffer solution 5 plRandom primer solution 5dNTP mix (4 p1 each)Radiolabeled dNTP-[a-32PJdATPKlenow enzymeI2 pl4 pl2 PIThe final volume of the reaction mix was made up to 50 p1.The components were mixed gently by inverting the tube several times, andqu~ck-spun for 2 seconds In a microfuge at maximum speed.Then the tube was incubated with the reaction mixture at 37°C for 1 hour in awater bath.iii) Purification of radiolabelcd DM probe by Spun-Column chromotographj~: TheSpun column chromatography was used to separate the labeled DNA that passes throughthe gel-filtrat~on matnx, from lower molecular we~ght substances (viz, radioactiveprecursors) that are retained on thc column as F r standard procedures [327]Bnefly, a I-ml d~sposable synnge was plugged with a small amount of sterileglass wool (Supelco), whlch was accomplished by using the barrel of the syringeto tamp the glass wool in placeThen h e synnge was filled with Sephadex (3-50 (Amersham biosciences, U.S.A)quilibratd in iX TEN buffer (pH 8.0) and the buffer flown through by tappingthe side of the synnge barrel. The resln was added till the syringe was completelyfull.The synnge was inscned into a IS-ml disposable plastic tube and centrifuged at16og for 4 minutes at mom temperature in a swinging-bucket rotor in a bench-top centrifuge.

CHAPTER 111The resins were packed down and became partially dehydrated duringcentrifugation, and the steps were repeated till the volume of packed column was0.9 ml.The DNA sample was applied to the center of the column in a total volume of 0.1ml (50 p1 of Milli-Q water was added to the 50 p1 of the random primed mix tomake up the total volume to 0.1 ml), and the spun column was placed in a freshdisposable tube containing a decapped microfuge tube. The centrifugation wascarried out as rn previous step and the eflluent DNA was collected into thedecapped microfuge tube.The syringe was removed which contained unincorporated radiolabeled dNTPsand other small components, and disposed off safely in a radioactive waste.The decapped tube was removed carefully using forceps, recapped and labeledappropriately and stored at -20°C until needed.The mu& estimate of the proportion of radioactivity that has been incorporatedinto the template DNA was obtained by holding the tube with eluted DNA to ahand-held radloact~vtty mlnimonltor.BufferIOX TEN bufler0.1 M Tns-CI (pH 8.0)0.01 M EDTA (pH 8 0)1 M NaClD. DNA-DNA hybridization studiesThe nylon membranes transferred with the digested and undigested plasmid DNA, aswell the restriction digested chmmoson~al DNA (separated by PFGE) from theenlemcoccal test isolates and standard strains were hybridized with the DNA probes(HLGR gene probe, HLSR gene pmbe and lambda DNA marker probe) preparedpreviously, as per standard prixedures [327] with minor modifications.

CHAPTER 111i) PrehybtidizationThe nylon membranes were rolled into the shape of cylinder and placed inside theroller bonle together with the plastic mesh provided by the manufacturer (Hybaid,Therrno Hybaid, U.S.A).Approximately 0.1 ml of Pre-hybridization solution (containing equal volumes ofSDS and Na2P04, i.e: 7 % SDS and 0.5 M NalP04) was added for each squarecentimeter of the membrane (20 ml in total) and the bottles were closed t~ghtly.Then the hybridization tubes were placed inside the pre-warned hybridizationoven (Hybaid, Hybaid, U.S.A) at 6S°C for 15-20 minutes with agitation.ii) HybridizationThe Prehybridization solution was decanted from the hybridization bonle andreplaced with same volume of fresh solution contaming the radiolabeled DNAprobe.Then the bottles were closed tightly and replaced in hybridization oven quickly at6SUC, and the hybridization was canied out for 18 hours with agitation.Before adding the probe to the hybndlzatlon solution. it was denatured at l0O0Cfor 5 rninules by placing the tube In a boiling water bath (since the probe hasdouble stranded DNA) and snap-frozen by placing it on ice.iii) Pop1 HybridizationARer hybndlzatlon the membrane9 Here remo~ed from the hybndtzatlon bottles,and excess hybnd~zatlon solutlon uere briefly dralned from the membrane byholding the comer of the membrane u ~th forceps to the 11p of the bonle contamerThe hybridizat~on solution with the radiolabeled probe was decanted into a darkbottle. scald and stored at -20°C for reuse.The membrane was rinsed thrice for 2 minutes with Wash Solution-1 (2X SSCand 0.5 % SDS) using fresh solution for every rinse, and decanted into aradioactive disposal container.

CHAPTER 111Then 25 ml of Wash Solution-I (1 mlicm2 membrane) was added fresh to themembranes in the roller bottles, and kept in the hybridization oven at 65OC for 20minutes with agitation.The Wash Solution-I was decanted and replaced with 25 ml of Wash Solution-I1(0.1 X SSC and 0.5 % SDS) and kept in the hybridization oven at 65"C for another20 minutes with agitation.Finally, the membranes were removed from the bottles and the liquld drained offfrom them by placing it on a pad of towels.Then the damp membrane was placed on a sheet of Saran Wrap to cover it, andexposed appropriately to obtain an autoradiographic image.The radioactivity of the membrane after hybridization was measured using ahand-held radioactivity minimonitor, before AutoradiographySolutionsPrehybridizntion Solution (20 ml)14 'in SDS - I0 ml (7 ?a)I MNa2P04 - lOml(O5M)Wash Solution-1 ( 100 ml)2OX SSC I0 mi (ZX)I0 76 SDS 5 ml (O.5X)Make up the volume to 100 ml with double distilled waterWash Solution-I1 (200 ml)2OX SSC I ml(0. l X)IO%SDS IOmI(0.5X)Make up the volume to 200 ml with double distilled water.E. Autorudiogmphy/Phosphor imagingThe hybridized membranes were exposed and Images were developed using a phosphorimager (FUJIFILM, Kanagawa. Japan). The phosphor imaging has an edge over theconventional autoradiography since it 1s a rnhust and rapid method consuming very

CHAPTER IIlminimal time for exposing and developing the radiographic images, without any hazardfor the user.Briefly, the hybridized membranes covered with Saran wrap were exposed onto aPhosphor Imaging Plate- IP (FUJI FILM. Kanagawa. Japan) by placing the IP in aCassene - 20 X 25 (BAS Cassette- 2025, FUJIFILM, Kanagawa, Japan) for 1-2hours. After the exposure the IP was removed from the Cassette and scanned by alaser releasing photons that were collected to form an image by the PhosphorImaging device (Phosphor Imager. FUJIFILM, Kanagawa, Japan).The images were captured electronically, viewed and stored in the computerattached for further analysis.F. Srripping/Dqrobing membranesThe nylon membranes after phosphor imag~ng were reused for the subsequent experimentu.ith a dimerent probe, after successful removal,stripping off the probes from themembranes as per the manufacrurer's instructions (Hybond-N+, Amersham biosciences).Briefly. the nylon membrane was placed In a glass tray onto which 500 ml of0.1% hot boiling SDS solution was poured and kept for I hour with shaking.The procedure was repeated ~f the membrane exhlb~ted significant radioactivity asdetected by a hand held minimonitor.After stripping, the membrane uas esposed to a phosphor imager as describedpreviously to quant~fy the rad~oactivlty before reusing the membrane with anotherpmth.

CHAPTER 111RESULTS1. Molecular characterization of HLAR enterococciA. Plrrmid DNA profiling of HLAR enterococciThe genotypad enterococcal isolates were subjected for molecular characterization tostudy the genetic basis of their aminoglycoside resistance. Briefly, the plasmid DNA wasisolated fmm 66 isolates of enterococci, which included 57 randomly selected HLARclinical isolates and nine standard strains and transconjugants. The 57 HLAR (asconfirmed by PCR) isolates included forty-five E. ,fuecalis, six E. juecium, four E.gullinarum, one E. avrum and one E d~rrun.~. The whole plasmids, as well the restrictiondigested plasmids were separated, elecirophoresed and interpreted accordingly.The whole plasmid and the EtoRI d~gested plasmid profiles of the enterococcal~solates are depicted in Figures 7 and 8 Most of the HLAR enterococci yielded betweenone to five plasmids. Majority lsolates possessed at least two plasmids and the molecularweight of the plasmids ranged approximately 1 70 kb to 2 kb. The restriction-digestedplasmids were classified into groups with respect to the EcoRI profiles of their plasmidsas depicted in Table 133 and b. The groups were designated alphabetically (in caps) formore than one isolate exhibiting the same GoRI restriction plasmid profile. Four out ofthe fifty-seven tat isolates (two E, Jaeca1r.r. and one isolate each of E. gallinomm and E.durons) did not yield plasmids upon repeated extractions or were either refractory forrestriction digestion, hence excludal from further analysis and interpretation. 27 of 53isolates were classified into seven groups (groups A-G) as follows: thirteen isolates inpup-A. four isolates in gmupB and two ~solates each in group-C, D. E. F and G. GroupF and G included two isolates each of E /ut.crrrnr. while the isolates in other groups wereE ,faccalis. 26 isolata exhibited unique EcoRI restriction plasmid profile that could nothe clubbed with any groups and hence classified as 'bnique" restriction profiles. Whileonly t h among Ihe remaining nine isolates tested. which include three transconjugantsand the standardlcontml strains yielded plasmids. However, the three tmsconjugantsincluded in this pel failed to yield plasmid DNA upon multiple extractions.

CHAPTER 1111 Figure 7. Whole Plasmid DNA analysis ofIHLAR Enterococci.A. Lab number of Enterococcl from B. Lab number of Entarococcl fromlanes 1- 17 in respective order:lanes 1- 17 In mpective order:9357, 6,477,7108, 7181, 8756,6236, 6.641, 7868, 3896, 1670, 1396, An-1,4252, 5967, 10,564, 6,660, 7,107,831, 15,332,9953,1002,5342,9478,14S50,5099, 5318, 5797,4038, 11,122. 8670,14535,4343,3844,5969.C. Lsb number of Entorococci from 0. Lab numbor of Entomcocci from!me8 1- 18 in nspecttve order:lanos 1- 16 in nspocttvm ordor:4637. 10792. 15514-TC, S118- TC,6275,10,638,7132,881,271.6276, 10664- TC. 4028. 3757.8765.7918.57.6130.6265,8765.11660,11871,82577 CDC-SS-1273- E. hualis.891, 3846, 5298.15411. EF 1002- EspPaltlve control.E. faualls- PAM. JH2-SS. FA2-2 RF.E. faumllm- UTCC439.TC- Transconjugant; M- Hind Ill digested Lambda DNA marker; L- 1KB ladder.-- - - ---- - ----IN.8: Tha dotsilo of entlmicroblal mlstmnco of tito entorococul test ShrlnS an 1doplctmd In Table 13.----. . -

CHAPTER IIIFigure 8. Plasmid-Restriction endonuclease analysis andSouthern hybridization of HLAR enterococciA. Lab number of Enterococci from Ian- 1- 17 in respective order: 9357,6,477,7108,7181,8756,6236,4252,5967,10,564,6,660,7,107,831,5099,5348,5797,4038,11,122.II6. Lab number of Enterococci from Ian= 1- 17 in mpective order: 6,641,7868,3696,1670,1396, An-l,15,332,9953,1WZ, 5342,9478,14550,8670,14535,4343,3844,5969.IC. Lab number of Enterococci from lanes 1- 16 in respective order: 6275, 10,638,7132,881,271,6276,6130,6265,8765,11660,11871,8267,891, 3846,5298,15411.I0. Lab number of Enmrococcl (rom lanos 1- 16 In mp.ctfva 0rd.r: 4637. 10792.~WILTC,sire- TC. ow TC, 4028,3767.8766.7~18.67, cocss-127s- E. ~.C.IIS.EF 1002- EspPomlhra control. E. fw8c.H.- PAM. JHZSS. FA2-2 RF. E. t..CwCCUS.LY- Hind Ill dlg.IW Lambda DNA nurkw; L- 1 KB LwhhI

CHAPTER III8. Reaultc of DNA-DNA Hybridiutioo studiesThe bifunctional gentamicin and streptomycin resistance gene probes were used in crosshybridizationstudies with the PCR amplified AME gene products from the HLARenterococci used for plasmid DNA profiling. The DNA probes showed extensivehomology with the corresponding amplified AME gene product from all the HLARenterococcal isolates subjected for DNA Hybridization studies.ii. Plasmid DNA Hybridization of HLAR enterococciThe summary of the d~fferent restriction digested plasmid fragments hybridizing with theaac(b')+aph(2") and ant(6')-l gene probes are depicted in Table 13.a and b. Briefly, theplasmids classified into different groups with respect to the EcoRI profiles exhibited aunique DNA hybridization pattern for the respective gene probe with occasionalvariat~ons within the group. The sizes of hybridizing fragments depicted in Table 13.aand h are approximate measurements derived from the molecular weight standards runw~th every gel.The hybridiution patterns of the restriction digested plasmid DNA from HLARtest isolates with gentamicin and streptomycin DNA probes are shown in Figure 8.Thirteen E. ,faecalis categorized as Group-A based on their similarity in the EcoRlrcstnction plasmid profiles shouted approximately a I I-kb EcoRI fragment hybridizingwith gentamicin as well streptomycin gene probe. comprising the largest number ofisolates showing a similar hybridization pattern in a single group. The four E. ,faecalisGroup9 isolates showed a 15-kb EcoRI fragment hybridizing with gentamicin geneprobe, while the streptomycin gene probe hybridized to approximately 15-kb EcoRlframent In one isolate (the other three isolates being sensitive for streptomycin). Thetwo E. .faecalis Group-C isolates showed a 20-kb EcoRl fragment hybridizing withgentamkin gene pmbe, while the streptomycin gene probe hybridized to approximately

CHAPTER III12-kb EcoRl fragment in one isolate (the other isolate being sensitive for streptomycin).The two E. fueculis Group-D ~solateshowed a 7.5-kb EcoRI fragment hybridizing withgentamicin as well streptomycin gene probe. The two HLGR E, faeculis Group-E isolatesshowed an 8-kb EcoRI fragment hybridizing with gentamicin gene probe. The two E.{uec~um Group-F isolates showed a 13-kb Ec,oRI fragment hybridizing with gentamicingene probe, while the streptomycin gene probe hybridized with two EcoRl fragmentsapproximately of 20-kb and 70-kb in size in both the isolates. The two E, fuecium Group-G isolates showed a 13-kb EcoRI fragment hybridizing with gentamicin gene probe,while the streptomycin gene probe hybridized to approximately a 9-kb EcoRl fragment inboth the isolates.Only four of twenty-six isolates (15%) possessing "unique" EcoRl restrictionplasmld profiles showed Ec.oRI fragments of same size encoding resistance for both theam~noglycoside genes by hybridization. However, every isolate had a hybridizingfragment of dilferent size ranglng from 6 to 12-kb as depicted in Table 13.a and b. Forthe remaining 22 isolates. gentamlcin and streptomycin gene probes hybridized to EcoRlfragments of the dinerent sizes ranglng from 5 to 70-kb for the same isolate There wereno fragments showing hybridlzatron with the gene prohes for isolates that did not yieldplasrnids.iii. Chromosomal DN,4 Hy6ridi:ation studies of HLAR enterococciThc Smal restriction digested chromosomal DNA of the HLAR enterococci separated byPFGE d~d not show any fragments hybndlzlng with either of the DNA probes tested asshown In Figure 13 (Chapter VI) This confirms that none ofthe HLAR enterococci fromour hospital carricd Be gentamicin.strcptomycin resistance determinants on theirchromosome.

CHAPTER I11Table 13 a. Plasmid REA and DNA hybridization resultsLab no. Species HLARPlasmidSm.group Frap, Kb Frag, Kt8756 E..fuc~culi.c G+S Noplasmid NA NA5797 E. jurcu1i.c G+S A 11 11I670 E. fueculi.~ G+S Unique N11 NH9953 E. fueculis G+S A I I 111002 E. fuecali.\ G+S A I I 1114535 E ,Juc,culis G+S A 11 116275 E fuc.cu1i.s G+S A 11 116276 E .fuc~cul~.\ G+S A 11 116130 E. /ic,colr.\ G+S A 11 118765 E /uc~c~ulr.s A I1 1 II1 I I8257 E ,f(~i,culic G+S A 11 111891 i E /uc~ruli,s G+S 11 1111 11I0792 t: /uccuI. S A 11 1153 I Xiii /ui~culi.\ G+S Untque 24 I5434, ; E /ui,iuIi\ In~qui 0 1 XAi7.107 l E /UIY crlrzBI50W / t. /u~l~ulr.t ' G B831 6. lurruh 1 (i1028 j E /ui*cuirr j (1-sX765 E lui~c~ulr.\ / G+S DXX I 1 E /ii~ u/I.\ i1C 1 20 / NA791 X I E fuc~culr.\ / ti E ' 8i5342 E larc.~um j G+S 1 F 13 20.709478 E ,ui,rrun~ G+S F 1 I3 20.704038 1 E /ui.cirrnr G+S 9E Iuc.c.itmnt 9, G+S13514-TC E fucc~ulr.\ G+S NO105640-TC E /h~culi.v G+S No Plasmid NA NA9357 E, fut~rulr.\ G Unique I I NAE ,firzl.rum G+S Un~que NH 20. 7C5967 E ,fac~i~rrli.s G B 15 NA10.5M E.,faecalis G+S Uniquc 12 1 2

CHAPTER 1111S. NO./ 29/ 40I 4142I 47Table 13 b Plasmid REA and DNA hybridization resultsL P no. ~speciesFrag, Kb3536376,6601 1.1226,h41E forculr~t faetolrcE fu~tulr\G+SGG+S Un~que15I I9 57868 E fac,calrr G+S Unlque NH 8An-l E fuecuhc G+S Un~que 10 NH15,332 E /urcolrr G+S Unlque 10 5 10 514550 E faeculr~ G+S Unlque 8 168670 t gulltnunrm G+F, Unlque NH NH3844 E gullrnunirn G+S Un~que NH 2344 59691Efuc.culrr G Un~que NH NA45 10,638 1 E lurtult\ Un~que 6 NA/ G46 27 1 t /nc,tul~\ G+S Uque 4 5 947 6265 / F gullmunini (I Un~que 9 NA49HLARP1asmidgroupBUn~quellX71 f lut.cull\ (1+5 Un~que 6 550 / I541 1 I t lurctrrm ( CI+S / Umque51 4637 f /crctull\ C J + ~ Unlque II52 1 91 1 X- TC F u~llrrt! 5 No Plasm~d' , - ControlI- Control- Control- Convol--- ControlI I6NANASm.Fmg, Kb15N ANHG. High-level Gcntamicin resistance: S. High-level Strcptonlycin resistanceNH. No hybnd~zs~ion; NA. Not applicable: TC, transconjugant159I I6NANA51 I 7 7 f ,utcuit\ CJ+S Vo Plarm~d54 1 57 / L /ut.eu/r\ (J'S 1 ~~n~que 5 R55 1 6236 ~,uc.eulr~ I G+S 1 c 20 1 124252 I t fuc,culo 1 (l+S Unlque 146,477 L dururu 1 5 ho Piasm~dSR 7181 1 t u,rr,nl (r+S 12 59 1 71 32 115 gullrrmnmr G+S No Plasrnld NA NA IE /clreult\ CI UnlqueI I NA- , Control NHNHINHNHNHNHNHNHNHNH, NH

CHAPTER IIIEnterococci exhibiting aminoglycoside resistance pose one of the biggest therapeuticchallenges in treating serious infections. A synergistic combination regimen is impossibleeven if the isolate is susceptible to either of the cell-wall active agent (p-lactams/glycopeptides) [40, 131, 3411. Present study showed that 60% and 43% of allenterococci tested were resistant to high-level gentamicin and streptomycin respectively.The high-level plasmid-borne resistance to gentamicin was first reported in 1979 in threestrains of S. fuecalis [254] following which studies reported that HLGR conferringplasmids in E. foecalis isolated from diverse geographic locations were heterogeneous asdetermined by molecular genetic studies [18, 3541. Thus genotypic analysis andmolecular characterization of HLAR determinants is highly essential to know thed~fferences, if any, in the genetic basis of the HLAR in enterococci between differentcountries and continents 1292. 349. 3501.The results of whole plasmid and EroRl digested plasmid profiles of our studydepict heterogeneity among plasmids in HLAR enterococci, similar to several studiescarried out in different pans of the world ~ncluding U.S. France, U.K. Japan. and Greece1 18. 135. 292. 350. 354-3561, Most of these studies have also shown that a predominanttype of plasmid was present among many HLGR enterococcl depicting their widespreadd~ssemination. which is concordant with our study. The whole plasmid profiles depictedthat the molecular weigh1 of the plasmids ranged approximately between 70 kb to 2 kbamong the HLAR isolates in our study, similar to other studies showing the presence ofsame range of plasmids. while most of the smaller molecular weight plasmids werecryptic [18, 251. 254. 354. 3571. A number ofplasmids have been associated with HLGR,bul the nature of the plasmids harboring the resistance genes does not appear uniform.The role of the bifunctional gentamicin resistance gene aac(6')+aph(2") encoded byplasmids were depicted since late 90s by many studies. The cause for such diversity inplasmids conferring rhe same phenotype within a species is unclear. A possibleexplanation may & thal pmlonged prevalence of HLGR in clinical isolates of enterococcicould have allowed time for the aac(b')taph(2") gene to become associated with

CHAPTER IIItransposons and this would account for some degree of diversity observed in E. fueculisas evidenced [ 12 1, 1221.Several studies have depicted plasmid heterogeneity among HLGR E. ,fueculis,while some studies have shown homogeneity among HLGR plasmids in E. /ueciurnauthenticating the widespnad dissemination of a single gentamicin resistance plasmidand its derivatives through localization of the genetic determinants by hybridization[357]. Our study shows that two homogenous groups of plasmids were prsent among four}{LAR E. jaecium isolates, while other two E. /uecium isolates possessed a uniqueplasmid digestion profiles, which is concordant with the plasmid profiles of other studies(3571 But the chances of a single E. ,fuccium Isolate with HLGR plasmid gettingcirculated within the same hospital cannot be ruled out. since our study dealt only withsola ales from a single hospital unlike other studies which investigated isolates fromd~flerent hosp~tals [2137. 3.571. Other possible reason for this plasmid homogeneity may behecause HLGR among E. /uc,crtlrn is a relallvely new and infrequent event since early 90s[!X7. 357) It is postulated that given sunicient time the homogeneity displayed by E.krrcrrmr plasm~ds may be replaced as In E lorcolr.~ by a heterogeneous set of plasmids.thereby limiting the therapeutrc choices available 1121. 3581. The data from the presentstudy authenticates this fact w~th comparatively lesser HLGR among E. ,fuccrurn. but withthe rise of HLAR this may emerge as an imponant clinical problem in the near future.Although HLAR was pmposed to be encoded by plasm~ds initially [254],subsequent studies pmvided genetic evidence by DNA-DNA hlbndization experimentsto confirm that HLAR determinants are usually encoded on plasmid DNA in E. fuecolis.E larcrrrm, E. uurrm. E. hrrue. E, rolfinosus. E. gull~nonrrn and E. cuvselrflorvs [I 2 I.122. 251, 287, 350, 355-357. 359). Hence in the present study, DNA-DNA hybridizationcxperimcnts were carried to map the genetic locations of the HLAR determinants amongthe clinical isolates using bifunctional gentamicin (Gm) resistance gene probe. andstreptomycin (Sm) mistant gene probe.

CHAPTER 111The whole plasmid DNA was hybridized with Gm and Sm probes in our study,but the pattern of hybridization' was not clear enough to be conclusive about the role ofspecific plasmids encoding HLAR. Thus, for conclusive evidences the restrictiondigestedplasmids were southern transferred and hybridized with respective DNA probe,which showed different plasmid fragments hybridizing with the Gm and Sm gene probesas depicted in Table 13.11 and b. The plasmids classified into different groups (A-G) withrespect to the EcoRl profiles exhibited a unique DNA hybridization pattern for therespective gene probe with occasional variations within the group, while isolates having"unique" EcoRl restriction plasmid profiles showed heterogeneous hybridization patternsof different sizes ranging fmm 5 to 70-kh for the respective gene probe.Apan from 13 E lueclolts categorized as Group-A that showed approximately anI 1-kb EcoRI fragment hybridizing with gentamicin as well streptomycin gene probe, four~solates categorized as "Unique" based on the EcoRl restriction profiles too showed an11-kb EcoR1 fragment hybridizing with the Gm probe. But, none except one "unique"isolate hybndized with the Sm probe while other isolates were sensitive to streptomycin.Along with the two E. /ueculis Group-C isolates, one E. uiiurn isolate with "unique"rcslnction prolile too exhibited the same hybridization pattern showing a 20-kb and 12-kb EcoRI fragment hybridizing with Gm and Sm probes respect~vely. While another"unique" pmfiled t'. juc~culu Isolate showed an 8-kb EcoR1 fragment hybridizing withGm pmbe like the two E, faecalts Group-E isolates, however a 16 kb fragment of the"unique" Isolate hybridized with the Sm pmbe, while the GroupE isolates beingsensitive to streptomycin did not hybridize. The transferability of the HLAR determinantsbetween t'.jueculrs isolates [6, 1 16. 120, 1271 may be one reason for the hybridization ofidentical/near identical (molecular weight wise) fragments even among different plasmidtypes present in clinical entemcal isolates as shown in the present study.It is noteworthy to find that the four (2 each) GroupF and Group-G E. .foedurnisolates showed a 13-kb fragment hybridizing with Gm pmbe, but the Sm pmbehybridized with 20-kb and 70-kb fragments of the GmupF isolates, and with a 9-kbfragment of the GroupG isolates. Another interesting result to be pondered was the

CHAPTER IIIhybridization patterns of three E. faecium isolates (two Group-F, and one "unique"~rofiled isolate) that showed two fragments-20-kb and 70-kb hybridizing with the Smprobe. At the first instance we thought that the residual undigested plasmid DNA resultedin hybridization of Sm probe with two fragments (since other isolates depicted only asingle fragment hybridizing with Sm probe). Hence we repeated the experiment for thesethree E. faecium isolates by redigesting the plasmids with EcoRl and hybridized with Smprobe which yielded the same result and confirmed that ant(6')-I Sm probe hybridizedwith two plasmid fragments. A study by Ounissi et al [360] had previously depicted thistype of discrepancy with the ant(6')-I probe which hybridized with streptomycinsusceptible enterococci and staphylococci. They suggested that this could result from thepresence In the strain a silentlremnant ant-6 gene, or from hybridization to anotherportion of the genome. The gene for ANT(6) was nearly always (99.8% in staphylococciand 99.6% in enterococcr) associated with that encoding an APH(3'). This observation,combrned wrth the fact that the streptomycin-susceptible strains that hybridized with theant(6')-I probe were all kanamycin resistant. suggested a physical link between the tworesistance genes, resulting in cross-hybndrzatron. Further as a general estimate, probesurll hybndize to figments that have >80% complementarity and hybridization istherefore less affected by mrnor nucleotide changes and can detect related groups ofalleles 1327. Neil Woodford. Personal communication. 20051. Taking all these facts intoconsrdcration we tested the susceptibility of three E. fuec~um isolates to kanamycin andlound them resistant. Thus we concluded that these three E. fuecium isolates from owstudy might possess the aph(3')-llla gene conferring high-level kanamycin resistance andwere plasmid encoded (3611, which hybndized with the ant(6')-I Sm probe. althoughPC'R detection of aph(3')-llla gene was not done to validate further. Only four (includingone E. (ucciurn isolate compared with Group-A ~solates) among 26 isolates possessing"unique" EtaRl restrictron plasmid profiles showed hybridizing fragments of 6 kb, 10.5kb. I I kb, and I2 kb with Gm and Sm probes for each of four isolates. However for theremaining 22 isolates. Gm and Sm gene probes hybridized to EcoRl fragments of thedlffercnt sues ranging from 5 to 70-kb.

CHAPTER I11As authenticated by preliminary cross-hybridization studies with PCR products ofHLAR enterococci, there were' no false-positivelfalse-negative results in the DNA-DNAhybridization experiments with plasmid DNA from majority of clinical enterococcalisolates, with the exception of ant(6')-l Sm probe with plasmid DNA of three E.faeciumlsolates that showed two fragments (20 and 70kb) hybridizing with the Sm probe. Therewere no fragments showing hybridization with the gene probes for negative controls, orlsolates that did not yield plasmids emphasizing the sensitivity of our hybridizationexperiments.The HLGR determinants apart being encoded by plasmids could also be carriedon the chromosome in E. juecu1i.s via dimerent transposons (123, 261, 355, 358, 3621.However In our study. Smul restriction digested chromosomal DNA of the HLARen~erococci separated by PFGE did not show any fragments hybridizing with either of theDNA probes tested. rhus confirming that HLAR determinants are encoded only byplasmid DNA and not by the chromosomes among the lsolates from our hospital.Thus the Lc.oRI restriction plasmid profiles and the hybridization panerns of theamlnoglycos~de resistant enterococcl especially those of E. fuccu1i.c from our hospital setup in South lnd~a shows heterogeneity among plasnilds. while some plasmids showedhomogeneity among the lsolates studied whlch may be due to dissemination of theplasmid determinants, or the plasmid possessing strain within our hospital. Although theaac(h')+aph(2") gene wnfemng the HLGR phenotype appears to be conserved, theremay be substantial diflerences in the flank~ng regions immediately adjacent to the fusedgene, which can be another cause for the diversity in plasm~ds confening the HLAR asshown in the prscnt study. The prolonged prevalence of HLGR in chnical isolates ofcnterococci that has enabled the aac(6')+aph(2") gene to become associated withlransposons accounting to some degree of plasmid diversity among E. ~/arcuhs, which isconcordant with other studies that have authenticated this fact [IZI. 1221. Although wedid not use any lnscnion Sequence (IS) probes (IS 256:257) to confirm the involvementof tmposons among the HLAR plasmids in our study. the diversity of plasmids and theplasmid hybridization panuns with Gm and Sm probes provide an indirect evidence to

CHAPTER I!!authenticate the involvement of Transposons. The restriction enzyme EcoRI used in ourstudy are not known to digest within the HKR gene or within the Transposon carryingthe gene. Thus the plasmid (hybridization) fragment size differences indicate that theDNA sequences flanking the gene were different. Hence different plasmid types may beInvolved in the dissemination of the strain, or, the HLAR determinants amongenterococci in our clinical setup in South India [363, Neil Woodford, Personalcommunication, 2005).SUMMARYThe results of our study depict the presence of plasmid DNA among most of the HLARb; /ueculrs isolates (53 of 60 isolates tested), which yielded between one to five plasmids,u hlle majonty isolates possessed at least two plasm~ds. The whole plasmid profilesdepicted that the molecular we~ght of the plasmids ranged approximately between 70 kb10 Z kb. The restriction-digested plasm~ds were classilied into seven groups (groups A-G)compnslng 27 isolates based on their homogeneity in d~gestion pattern, while 26 isolatesthat exhib~ted heterogeneous G.nRI restnctlon plasmid profiles could not be clubbed withany pups and hence class~fied as "unique" restriction profiles.To determ~ne the locatlon of the genetic elements confemng HLAR, thercstriction d~yestcd and the whole plasmid DNA from HLAR E. furtulis isolates wereSouthern transferred and subjected for DNA-DNA Hybridization using aac(6')+aph(2")and ant(6')-1 gene pmbes. The plasmids classified into different groups with respect tothe EtoRI profiles exhibited an identical DNA hybridization pattern for the respectivegene pmbe with occasional variations within the group Gentamicin and streptomycingene pmks hybridized to EcoRI fragments ol'rhe different sizes ranging from 5 to 70-kbk)r the same tsola~e. The sizes of hybridizing fragments were approximate measurementsderived from the molecular weight standards run with every gel after hybridizing with thelambda DNA probe. The Sml digested chmmosomal DNA of the HLAR enterococciseparated by PFGE did not show any fragments hybridizing with either of the DNAprobes tested confirming that none of the HLAR enterococci carried the gentamicin or,

CHAPTER IIIstreptomycin resistance determinants on their chromosome. Thus our study depicts thatHLAR determinants are encdded only by plasmid DNA among the isolates from ourhospital setup.The antimicrobial resistance although undoubtedly catapulted enterococci tobecome a prominent nosocomial pathogen since last decade, there are several otherfactors in enterococci that enhances the prospects of their pathogenicity even in thepresence or absence of antimicrobial resistance Hence, it is imperative to assess the roleof such (virulence) factors and their contribution to the pathogenicity of enterococci.

chapter IVVim knce (Factors in Enterococci

CHAPTER IVCHAPTER lVThe emergence of multidrug resistance in enterococci although contributes to itsnosocomial significance, resistance alone however does not explain the pervasiveness ofenterococci in nosocomial infections. Studies have shown that E. fueculr~ whileremaining sensitive to vancomycin and ampicillin continues to be the most frequentlyencountered enterococcal isolate, accounting for upto 80% of enterococcal infections [9].An allernative explanation is that the preponderance of E. ,fueculis infections isattrihutable to enhanced virulence. a prospect for which there are evidences. The factorsencodcng virulence traits have been shown to play a major role in transformation of thisInnocuous commensal Into an opportunistic nosocomial pathogen. Some lineages ofenterococci that have acqucred some virulence tracts are suspected to have an enhancedahility to cause d~sease as shown hy several studies These factors may enhance theah~lity of a nosocomcal strain to outcompete indcgenous commensal enterococci. Increasethe presence of nosocomlal strains in the gastrocntestcnal tract and as a result increase thestatistical likelihood of causlng d~sease due to breaches in containment [26. 1931.Although several virulence factors play a vctal role in establishing enterococcalcnl'ections. only few factors can exhibit enterococcal virulence at the level of tissuedestruction or tox~city, enhanccng cts abclcty to breach containment in the gastrointestinaltract or other commensal site and cause symptomatic disease. A factor that enhancesdcsease severity as opposcd to disease probability, would not necessarily appear in~ncreased numbers among clinical enterococcal isolates, but would be associated withmore severe clinical presentation or sequel [IZ. 1931. Hence studies on prevalence andassessment of the role of those significant virulence factors have become essential tounderstand the patho-biology of entemcoccal infections.

CHAPTER IVTo investigate the prewnce and clinlcal sigr~rticancc 01'vitrlo1is vllulcnc:cfactors in enterococciMATERIALS AND METHODSI. Screening for Bacteriocin and hemolysin productionThe production of Bacteriocin hy the enterococcal test isolates was detected qualitatively.hy the bacterincin assay performed using various standard indicator strains: E, faecalisJH? SS. E. faecalr\ FA2-2, E fueciurn BMRF & E ,faeciurn BMSS and S, aureus -NCTC 6571 as per standard procedures [l04]..1. Sofr-agar assay for Baaeriocin productionThe ~ndrca~or stralns uere rnoculated in N?GT broth filutrient broth No.2. Oxoid.U.S A) and incubated at 37°C for o\.ernight50 )rl of overn~ght indicator stratn was added to 5 ml of molten soft Tryptic Soybroth (HI medla. Mumhai. India) supplemented with O.So;~ yeast extract and 0.7%agar, mrxcd well and pnured onto a Tryptic soy agar plate supplemented with0.S0h yeast extractAfter solidification a single colony of each en\erncnccal isolate to be tested forbacteriocin prnduction was stahhed onto the semisolid media and the plates wereincubated at 37°C for 24 hours, and obscned for zones of inhibition (growthinhibition of the indicator strain) amund the test strains.6. Hemolysin assay: Pduction of Hemolys~n was determined by plating thecnterococcal isolates along with a positive contrnl (E, far~calis FA2-2) and negativecontrol (E.,facralis JH? SS) onto Todd-Hewit! agar (Oxoid. U.K) supplemented with 5%human blood. Plates war incubated at 37°C and obsened after 24 and 72 h. A clear zone

CHAPTER IVof p-hemolysis around the bacteria was considered indicative of the production ofHemolysin [2 1 I].2. Screening lor Cclatinase productionGelatinase production was detected by inoculating enterococcal isolates onto peptone-yeast extract agar containing gelatin (30 giL) and incubated at 35'C for 24 hrs and cooledto ambient temperature Ibr 2 hours. Appearance of a turbid halo or zone around thecolonies was cons~dered to be a positive indication of gelatinase production [I 521.3. Detection of "Enterococcal surface protein- espn gene by PCRThe prevalence of "esp" gene-a putative virulence factor for enterococci was detected byPCR uslng "esp" sequence specific primers as described previously with minormodifications [206]. The primer sequences chosen to detect the "esp" gene were as belowmentioned:Espll- F~>m,urdprrrncr (1217-1238)5'- TTG CTA .4TG CTA GTC CAC GAC C -3'Espl2- Rc~*t~r.st,prirnc,r (2171-2149)5'- GCG TC.4 ACA CTT GCA TTG CCG AA -3'.4. Optimization of "esp" Colony PCR: The template (bacterial) DNA used for the PCRwas standardized through various extraction procedures. Finally, the colony PCRprotocol was followed throughout our study since the procedwr was simple and rapid, aswell the results were on par with those reactions using template DNA extracted byconvent~onal procedures. The volume of the reaction mlxture used in PCR wasstandardized by scaling down to a minimal \.olume of 25 pl. without any compromise inthe quality of amplification reaction and subsequent post PCR analysis.B. "esp" Colony PCR: Bnefly. 2-3 bacterial colonies from an overnight 5% TSA bloodagar wnr suspended using s sterile micro tip in the 25 pl chilled reaction mixturecontaining IX PCR buffer (with 1.5 mM MgCI:) 2.5 mM MgCI?. O.2mMd~~ynucleotide viphosphates (dATP. dCTP, dGTP. dTTP). 0.2 pM esp primers

CHAPTER IV(forward and reverse) (Bangalore Genei, India) 2.5 U Taq DNA polymerase (BangaloreGenei. India). An Eppendorf gradient thermal cycler was programmed with the followingconditions: an initial denaturation for 2 minutes at 95°C. denaturation for 45 seconds at94°C. primer annealing for 45 seconds at 63°C and extension/ polymerization for 1minute at 72°C for 30 cycles. A final extension step was carried for 5 minutes at 72°C andheld at 40°C till analysis. DNA from a positive control strain- E. furculis EF-1002carrying "esp" gene (kindly gifted by Dr. Nathan Shankar, Univ. of Oklahoma HealthSciences Center. Oklahoma, USA) and a negative control strain- E. fuecolis FA2-2(kindly gifted by Dr. Yasuyoshi Ike, Gunma Univ. School of Medicine, Gunma, Japan)were included with each set of PCR amplification.C. Post PCR analysis: The PCR products along with 100 base pair molecular weightmarker (Bangalore Genei. India) were subjected to electrophoresis in a 1% agarose geluslng 0.5X TBE buffer. Subsequently the gel was stained with ethidium bromide,ilsualized under an lJV-trans~lluminator and documented using a gel documentationsystem (Vilkr-lour be^. France)4. Biofilm assayThe B~ofilm format~on by enterococcl was assayed as described previously (210. 2131.Br~efly, entemoccl that had been grown overnight were diluted 1: 100 in 200 p1 of trypticsoy broth with 0.25% glucose and inoculated onto polystyrene m~crotiter plates (Nunc,U.S.A). After 24 h of stauc ~ncubation at 37°C plates were processed by fixing withBouln's fixative for 30 min, stalned with 19.0 crystal violet (CV) for 30 min and rinsedwith dlst~lled water. CV was solubilized in ethanol-acetone (80:20 vol'vol) and opticaldensity at 570 nm (OD570) was determined. Each assay was performed in quadruplicateon at least three occasions. Biofilm formation by the isolates was determined based onthe OD570 readings after CV staining. The isolates were categorized based on theappmach of Toledo-Arana et al. (2101 and Mohamed et al. [213] as follows: strongbiofilm producers when OD570 was >2; as medium biofilm producers when OD570 wasbetween I to 2, as weak biofilm producers if the OD570 was > 0.5. but < 1, and as non-biofilm formers if the OD570 was < 0.5.

CHAPTER IY5. Statistical malysi~The categorical variables were compared using Chi-square test test) or Fisher's exacttest. Significance was defined as P

CHAPTER IVThe bihctional gentamicin resistance gene-aac(6')+aph(2") and thestreptomycin resistant gene-ant(6')-l were present together in sixty (83%) and one(100%)bacteriocin positive (Bat+) isolates of E. faeco1i.c (which included 5 and 17 E. faecalisisolates resistant to vancomycin and ampicillin respectively) and E. faecium respectively.The aac(6')+aph(2") gene alone was present in six (8%) bac+ E. faecalis isolates, whichincluded four and two isolates resistant to vancomycin and ampicillin respectively. Theremaining six (8%) Bac+ E. faecalis isolates were sensitive to other antibiotics tested,which included two (3%) isolates resistant to vancomycin.Hemolysin production was detected in 24 (14%) of the 171 isolates of E. faecalis,and two (100%) E. durans isolates, while none of other species isolated and testeddepicted production of hernolysin.2. Production of gelrtinase by enteroewci and correlation with antimicrobialresistance(ielarlnase production was detected In 99 (41%) of 242 isolates of enterococci as depictedIn Table 14. although 99 (58%) of 171 E, [orculr.\- Isolates was the only species among allen~erococci tested to produce gelatinase in the phenotypic assay. The bifunctionalgentamicin resistance gene-aac(b')+aph(2") and the streptomycin resistant gene-ant(b')-1were present together In 58 (58.5OI0) gelatinase positive (Gel+) isolates of E. ,faecalis,which included 5 and 17 isolates mistant to vancomycin and ampicillin respectively. Theaac(6')+aph(2") gene was present separately in 16 (16%) Gel+ E. faecalis isolates, which~ncluded 10 and 2 isolates resistant to vancomycin and ampicillin respectively, while 8(8%) of Gel+ E. Jaecalis isolates were streptomycin resistant. which included 3 isolatesresistant to vancomycin and the presence of ant(6')-1 gene in 4 isolates. The remaining 17(17%) Gel+ isolates were sensitive to other antibiotics tested, which included 6 (6%)Isolates mistant to vancomycin.

Figure 9. Screening representative strains of Enterococcus forBacteriocin production against E. faecalis JH2SS indicatorFigure 10. "Esp" PCR results of representativetest strains of EnterococcusEsp - Enterococcal surface protecn gene - 954 bpM - 100 base palr MarkerPC - Posl~ve contrd, NC - Negative controlLanes 1 - 24 - representat~ve Enteroms test stralns

CHAPTER 1V3. Detection of "Esp" gene by PCR and correlation with antimicrobial resistanceThe "Esp" gene was detected in 122 (50%) of 242 Isolates of enterococci as depicted inTable 14, which includes 1 12 (65%) of I7 1 isolates of E. faeculir, and I0 (40%) of 25 E.fuecium isolates, while none other species tested depicted the presence of "esp" gene.Overall, "esp" gene was present among 90 (82%) of the l I0 HLGR E. ,fu'arculis isolates,while only 22 (56%) of the 39 HLG susceptible E. fuecolis isolates possessed the "esp"gene (P=0.0001, 2 test value=34.34: OR=7.98; 95% CI-3.7-17.41). The results of the..espW PCR are shown in Figure 10. The bifunctional gentamicin resistance gene-aac(6')+aph(2") and the streptomycin resistant gene-ant(6')-l were present together in 68(61%) and 8 (80%) "esp" positive Isolates of E. /urculis (which included 6 and 22 E.fuc,culi.~ isolates resistant to vancomycin and ampicillin respectively) and E. fuecium(whlch included 3 and 5 E. fueciltm isolates resistant to vancomycin and ampicillinrespcxtively) respectively. The aac(6')+aph(2") gene was present alone in 22 (20%) and 2(20%) "esp" posit~ve ~solates of E, Jueculi.~ (which included 14 and 2 E. fueculis isolatesresistant to vancomycin and ampicillin respectively) and E. fuecium (the 2 E. fueciuml\cilates uere also reslstant to amplc~llln) respect~bely The ant(6')-I gene alone waspresent In 2 (2%) "esp" positive E, /uc~colis ~solates. while the remaining 20 (1 8%) "esp"pos~tive E /urc.ul~.~ isolates were sensttlve to other antibiotics tested, but included 4(3 540) lsolates reslstant to vancomycin. The "esp" gene was also present among 22 HLGsusceptible E. /i~-cuh.t ~solates. but absent among 20 HLGR and 39 HLG susceptible E.4. Production of combinations of bacteriocin, gelntinase, and "Esp"The prevalence of three virulence factors in combination was studied in E.,fac~culis and E.fui~cium isolates and their sratistical significance were analyzed. 60 (83%) of 72 E.luc~culis bact isolates carried "esp" gene, while in comparison 52 (52.5%) of the 99 E.fucrc,ulis isolates that did not pmduce bacteriocin (hac-) canied the "esp" gene (P=0.076,X'lest value= 3.13). 62 (86%) of 72 E. fuccoli.~ bac+ isolates also pmduced geiatinase ina phenotypic assay. while in comparison 37 (37%) of the 99 E. .faecalis isolates that were

CHAPTER IV,.bat-" produced gelatinase (P=0.001, 2 test value=9.76). 75 (67%) of l I2 E. faecalisisolates canying "esp" gene also produced gelatinase in a phenotypic assay, while incomparison 24 (41%) of the 59 E. fuecalis isolates that were not canying "esp" geneproduced gelatinase (P4.104, 2 test value= 2.64). The single E. fueciurn isolate thatproduced bacteriocin also carried "esp" gene on them, while none of the 9 esp+ E.lueciurn isolates produced bacteriocin.The prevalence of the virulence factors in E, faecalis and E. fueclum amongvarious clinical specimens are as depicted in Table 15 Among E. faeculis the highestprevalence of virulence factors were exhibited by the urinary isolates (46%) followed bybloodstream isolates (38%). while among E. furcium bloodstream isolates exhibited theh~ghest prevalence (50%) of virulence factor followed by urinary isolates.Table IS. Uls~nbur~on of V~rulence factors among various clin~cal specimensVirulrncr factorNo. (*A) of isolates pmitive for the(Total no. of itolatnvirulence factor-potitivc) Blood I'rinc ExudateE fut.culn - F.P~ gcnc ( I I?) 43 (38) 51 (46) lR(16)Hacrcnoccn (72) 22 (30) 30 (42) 20 (?R)Ciclal~na.~ (99) 37 (37) 42 (42) 20 (20)Blofilm(44) IR (41) 25 (57) 1 (2)f. lurc~um - tsp gcnc (lo) 5 (50) 2 (20) 3 (30)Hscvna-~n ( I ) . 1(10)5. Biofilrn formation by enterororci and their relation with other virulencedeterminantsThe ability of entcrococci to biofilm on abiotic surface was assayed usingPolystyrene microtiter plates. E. fucc~ola was the only species of enterncocci found toproduce biofilm albeit moderately. 44 (26%) of 171 E. fucculis isolates showed amodcrate or we& biofilm formation based on the approach of Toledo el al. 2001 andothers. Thirty-&me (75%) and eleven (25%) of44 E. faecalis isolates showed a moderate

CHAPTER IVor weak hiofilm formation respectively. The relationship between biofilm formation andthe presence of other virulence factors in E. /uc.culr.r was studied and their statisticalsignificance analyzed. Twenty-four (32%) of thc seventy-five "esp" and gelatlnasepositive E. /ueculi.\ isolates showed a moderate (2 1 isolates) or weak (3 isolates) biofilmformation respectively, while 10 (42%) of the 24 "esp" negative and gelatinase positiveisolates showed a moderate (8 isolates) or weak (2 isolates) hiofilm formationrespectively (P=0.713. X'test value=0.14), Individual or paired correlation analysis withdifferent virulence factors showed biofilm formallon by E, furculi.~ isolates as follows.la] 32 (29% of the I I2 "esp" pnsitlve isolates. and I2 (20%) of the 59 "esp" negative~solatcs (P-0.324. X'test value-0.97). Ibl 34 (34% of'the 99 gelatinase positlve ~solates.and 10 (14%) of the 72 gelatinase negative isolates (1'=0.004. test value=8.09). Icl 24(32%) of the 75 "esp" and gelatlnase posltlve isolates. and 2 (6%) of the 35 "esp" andcc1atin3~c ncgati\c i~~Iatc\ (1"O 005. tcst \aluc=7.74). Id] 8 (2??0) of the 37 "esp"posltlve and gelatlnase negarlve ~solates. and ? (6O)o) of the 35 "esp" and gelatinaseIncgatl\c ~solatcs (P:O086. F~sher's elact test). le] 10 (4?"a) of the 24 "esp" negative andgclaunasc poslt~vc ~scllatca. and 2 (6'0) of the 35 "csp" and gelatinase negatlve isolatesIP 0 001. F~shcr's exact tcst) The d~str~hut~ons of d~firent virulence facton among~\olatea from vanous clinical specimens arc as shown In Table IS. with 25 (574b) and 18(JI'o) of the hlc~lilm prtducen Irom unnaq and bloodstream E. (uccu1r.s isolateshntcructrc~. a nclrmal hunlun commensal though catapulted as a prominent nosocomialpathogen owlnp to their vcrsat~l~ty of ant~nilcroh~al resistance. it bas their property ofp;~thogeniclty addressed a centup hack-~n I X99 that first underscored their emergence as;I human pathogen 11 3. 261 . .

CHAPTER IVThus we focused on the prevalence ofsome prominent virulence factors In enterncocci by phenotypic and genotypic methods.Our results dep~cted that 65% of E /uc,c.ulls and 40% of E. fuecium isolatespossessed the "esp" gene. while none other species tested showed the presence of "esp"gene, whlch is concordant with other stud~es from U.S.. U.K. Italy. Netherlands. and(;enany 1206-210. 218. 2191. although some recent studies have shown the presence of"csp" addrt~onally among other species like E !'~fi?tl(l.\u.s. and E. duruns 1204. 3641. Thecell wall-asstr~ated protein-esp was inltlally shown to be assnc~ated only amonginfcct~on-der~ved E luc~tolrc Isolates and nor ~n clrn~cal E /utcrum isolates. or in fourother less pathogen~c enterococcal species tested 12061 But subsequently another studyshowed that a subpopulation of ep~demic vancomycin-resistant E /irr~

CHAPTER IVThe prevalence of "esp" varied between different studies depending on severalfactors. Our results matched that of a study from Italy, which revealed that Esp gene werefound in 60% and 72% of E. /uc,culi.c and E. luccium isolates respectively 12031. Anotherprospective study of 398 patients with E. luc~culis bacteremia screened for variousvltulence markers showed that 32% of E. ,fuc,culi.r isolates carried Esp gene, howeverthere was no significant association between 14-day mortality and the virulence markersstudied, singly or in cnmbinat~on [152]. In another study. 29 E. /ucculi~ isolates frompatients with endocarditis or bacteremia or from stools of healthy volunteers wereinvestigated for their ahility to adhere to lnt-407 and Girardi hean cell lines and for theprcsence of known enterococcal virulence factors. The incidence of Esp was shown to be72.4%. whlch remained the highest than other virulence factors studied. But the authorsconcluded that hacterial adherence was not s~gnificantly associated with any of thesevirulence factors I2171 However. thesc studies [203, 2171 investigated the prevalence of"esp" uslng a lesser sample slze than our study, thus making it difficult to make aconclusive statement regarding the s~gnificancr of "esp". The higher pre\alence (65%) of"esp" among E, iui~tulr.\ froni our setup depicts that this virulence trait may havepermeated more deeply into the species by horizontal transfer and would have acquired itcc~mpamtivcly earlier. thereby enhancing the ahil~ty of the organism to cause d~seasehcyond that intrinsic to the species hackground as hypothesized previously (1931. Thediversity of "esp" positi\e E Iucc.uI~.s isolates from different sources viz.. blood, urineand cxudate in our study authenticates thc cnhanced ahility of this organlsm to causediseaseAnother study demonstrated that Esp occurs more frequently among ampicillin-resistant and vancomycln-susccptihle E fuc~c.r~rni clones from hospital~zed patients, anohservat~on. which indicates that antibiotic-rcs~stant \.anants may frequently arise underant~hlotic selrctive pressure among esp-positive clones reaching ecological abundance inthe nosocomial habitat [?I91 Our results showed a very strong association betweengenlamicin resistance and ..esp"(P=O.OOOl. X.' test value=34.34: OR=7.98; 95% Ck3.7-17.41) among E. /iicculi,s isolates reflecting this fact. Statistically (OR-7.98) our studyshowed hat the isolates with "esp" gene could exhibit eight times greater resistance to

CHAPTER IVgentamicin, than those without "esp" gene. The aac(6')+aph(2") and ant(6')-I genes werepresent together among 61% and 80% of "esp" positive isolates of E. fueculis and E.fuecium isolates respectively, while many of these isolates exhibited ampicillinresistance, and some vancomycin resistance. The aac(6')+aph(2") gene alone was presentamong 20% each of "esp" positive E. fueculis and E. fuecium isolates respectively, whilethe ant(6')-i gene alone was present In only among 2% of E. luecu1i.s isolates. Theremaining 18% of "esp" positive E, /ueculr.s isolates were sensltlve to other antibioticstested, except 3.5% isolates that were resistant to vancomycin.Several studies have investigated the relationship between the presence of "Esp"and vancomycln reslstance In enterococcl (mostly in E. luccrlmi). slnce vancomycinreslstance 1s the most sign~ficant problem encountered by developed countries, althoughmany develop~ng countries are yet to experience the consequences of VRE. Our studydep~cted that X~",O and 75% of the vancomycln resistant E. fueculrs and E. fuecilrm~qolates respc~t~vely possessed "esp" gene. however it should be noted that only 17% and1646 of the E /u~,c.ulr\ and I:' /oc,c.~rmnl ~solates respectively were resistant to vancomycin.Ixav~s et a1 12201 showed that the presence of \.anant Esp gene in vancomycin resistantI:'/ot~c.ltrm(VREF) and VSEF was strongly associated w~th a specific epidemiologicsource becau.~ [he presence of Esp was higher in clinical infections and epidemicasstrlated ~solatcs. than In sun.eillance ~solates. Some stud~es have shown that Esp geneplays a major role in dissem~nation of vancomyc~n-resistant E, facr-turnsinceprcdomlnant epldcn~~c stralns harbored the Esp gene. whlle most of the non-epidem~cstrains were Esp gene negatlve [XU-2121. Funhermore. the genetic machinery thatenables dlssemlnat~on of' antimlcroh~al resistance deterniinants heween enterococci isalways of serious concern. since the same machinery is capahle of transferring virulencedeterm~nmts 12231.

CHAPTER IVAs shown in our study, the presence of "esp" in isolates susceptible and resistant10 different antibiotics indicates that this trait would have probably emerged prior to theacquisition of resistance not only to Vancomycinl gentamicin but also to other antibioticscommonly used in the hospital setting. The significant association between resistance to@arnicin and "esp" in the E. faecalis isolates suggests that antibiotic treatment selectsparticular clones among those that have reached ecological abundance in the nosocomialhabitat due to the presence of "esp". This is elucidated by the fact that the "esp"genotypes were frequent among VRE strains in hospitals from U.S and other developedcountries 1220. 2211. The higher occurrence of this trait in our study among HLAR~solate suppow this hypothesis.A variant espgene was recently found to be associated with nosocomialoulhreaks of vancomycin resistant E ,fueciurn (VREF) in three continents. although thisgene ewas not found in VREb strains isolated from healthy persons or animals [220].\uhscquently. other investigators reported the presence of the variant esp gene in clinical~sc~lates of vancomycin sensitive E ,fuec.iurn (VSEF) 1208, 209, 2191. Since we did notprohe for the presence of the variant "esp" gene in our study. we cannot rule out thepresence of this variant "csp" gene among the "esp negative isolates. Little is knownahout the function of the variant "esp" gene. although epidemiologic findings support itsrole as a virulence factor with an increased adherence capacity that explains itsahsociation with hospital outbreaks. Thus the findings of our study confirm the strongaswiation between the presence of the "esp" gene and the relation with hospitaloutbreaks and clinical infections among patients with high-level aminoglycoside resistanta5 well as aminoglycoside sensitive en~erococci. apart from other commonly usedar~tihiotics in any heath-care setting.Aner the initial adherence to the host tissues with the help of adhesins like "esp".cnlerococcus invades and cause systemic infections and modulates the host inflammatoryresponses. The ptentially toxic secreted products like hemolysin/c~.tolysini bacteriocinand gelatinasc causes direct tissue damage. which contribuies to the severity of

CHAPTER IVenterococcal infections [26. 227. 22x1. Bacteriocins are peptides or proteins produced bydifferent bactena, which can inhibit the growth of strains and species usually related tobacteriocin-producing bacteria, thus conferring an ecological advantage on the producerstrain (104. 2341. Studies have depicted that bacteriocin production is linked to the samegenetic determinant as hemolysinlcfiolysin synthesis in Enterococcus (100, 101, 128.2291, and its product~on has been depicted as a pathogenic marker by many studies (233-2351.Our study showed that 30% of all enterococci produced bacteriocin againstvarlous Indicator stralns tested. although the property was confined mostly to E. fui,cuI~~lsolatrs w~th the exception of one E furc.t~tm isolate. Our results reflect the generalpattern of hacterlocln production by E ,/uec~ult.\ and E. fuecrum only among medical~solatrs (233. 2351. Studles among food lsolates have shown that non-faecalis and non-l'acc~um enterococcl too can produce hacterincin (204. 2301 Our results showed that 42%01' L. /u(,c.u/I.Y produced hactrnoc~n wh~ch showed a namow spectrum of act~vity. activeonly ayalnst two ol' the spec1~5-spt~ific (E fuc~cultr) ind~cators and not against non-sp~~~cwnon-genus spec~fic (E /ucc,r~rn~. S ulrrc,u.\) indicator strains tested. while only 4%of E /'rrc.rtrnr (one sola ate) produced bacteriocin that showed a broad spectrum ofactlvlty, actlve agalnst all genus spsc~fic Indicator stralns tested (E, fuccult.s and E./oect~tnr). hut not apatnst nun-genus specific lndlcator stnln (S, uurc*us). The narrowspectrum of actlvlty exhibited b!E, /ucc.ul~.\ bactertoc~n(s), actwe only agalnst thespcvles-spcc~fic ind~cators 111 our study. IS In concordance w~th other studles that havedepicted the spcyles spscilicity of E /ctt.c.ulr.\ bacter~oc~ns [232. 1.141. As hpotheslzedprev~ously, thc spectrwn of the hacter~ocins In our study dep~cthat the BAC+ isolatescan outcompete ~ndlgenous enterocnccl and In turn under favorable conditions can eventransfer its genetlc elements encodlnp these v~rulence traits generally encoded byplasmids [26. 1041The d~versity of BACt E /uc~c~ulir lsolates from different sources viz.. blood(31%). urine (42%), and exudate (IXO,b) in our study authenticates the enhanced ability ofthis organism to cause disease. sincenumber of Independent studies using difl'erent

CHAPTER IVmodel systems have consistently found a role for the E. fueculis bacteriocinihemolysin inthe toxicity of enterococcal infections [200, 2281. However, most of the studies caniedout in clinical settings focused more on the significance of hemolysin1 cytolysin property,while only few studies explored the role of bacteriocins exclusively among clinicallsolates of enterococci. An initial study had shown that there was a relation betweenbacteriocin production and virulence of E. fuemlis [231], which was later confirmed bymany other studies. Galvez et al. 12321 found that among 90 enterococci strains of humanorlgin. 36 strains produced bacterioclns. Later Libertin et al. [233] screened the clinicalisolates for bacteriocin production, concluded that hemolysim'bacteriocin produced byenterocc~ci could be considered as a marker of pathogenicity.Our results showed that aac(b')+aph(?") and ant(6')-l genes were present togetheramong 83% and lOO?b of "BAC+ isolates of E, fueculr.\ and C', fuc~cirrrn (one ~solate)~solates respectively. wh~le many of these ~solates exhibited amplclllin resistance. andborne tancomycln resistance The aac(6')&aph(?") gene alone was present among 8%Dact E /uc~ulr\ and E fuc~cr~rrn isolates respectively. wh~le remaining 84b Bac+ E./uc,c.uIr\ ~solates were sensitlvc to other antih~otlcs tested. which included 3% isolatesresistant to vancomycln Slmllarlq. Del Camp et al [I041 had shown that >80% of VREprtduced bacteriocln. while only 60% of \ancomycln-susceptible isolates of din'erentorigins showed bactenocrn actl\lty. A preliminary study by our laboratory had shownthat antlmicrob~al rcsistant enterococcl posses bacterloc~n activity. than do susceptibleones 12341.Characterization of representative bacteriocins from E. /arculr.\ and E. facciurn inour prellm~nary study revealed that the E. /icctrlrs bacteriocin were sensitive (lostactiv~ty) to all temperatures tested. but E, {uc~c~ttrn bacteriocin showed activity at 60'~.but sensitive at 70°C and 80°C and all the bacteriocins were actwe only at pH 6.0 and 7.0.Proteinase K and Chymotrypsin inactivated E. fu~~culi.v bacteriocin. while only ProteinaseK inactivated E. /aeci~rm bacteriocin. The partially purified bacteriocins exhibitedmolecular weights ranging betwen 9 KD- 5 KD by SDS -PAGE analysis [234]. Theresults of our pilot study were concordant with another study from Spain. which showed

CHAPTER IVsimilar physico-chemical characteristics of the bacteriocin from clinical isolates, althoughtheir bacteriocins showed a broad spectrum of activity [104]. However many otherstudies that had characterized bacteriocins from non-clin~cal isolates (food/environmental) depict concordant as well discordant results, since the properties dependon the origin of the isolates [204,230]. Thus our results depict that the production and thedurability of bacteriocins would render enterococci an ecological advantage. tooutcompete the closely related non-producers in different environmental niches In humanbody, which may enhance the disease causlng ability ofthis nosocomial pathogen.Our study shoucd hemolysln prnductlon by 13'h of' E. ~ucculr.~ and 100% E.~111ru1i.\ (two ~solates) laolates. whlle none other specles Isolated and tested deplctedproduct~on of hemolysin Some studlea on enteroccxcl lsolated from endocarditis andhacterem~c patients showed that hcniolys~n cytolysin occurs at a frequency of 45-60?&1161. 235. 2361. while our results were highly concordant w~th other studies that showedmuch lesser prevalence ol'hemol.ys~n syolystn. A recent study from U.S. showed thatI I '%of 2 19 E, /uc,c.uir.~ ~solates frnm patients with enterc~occal bacteremia were pos~tivefor hemolysin [152]. wh~le Elsner et al (1601 reported the presence of cytolysln amongIb0/0 of E. /uc.culr,s blood culture isolates. Another study showed that cytolysin waspresent only In 17% of entemccai isolated from patients with endocardit~s or bacteremiaor fn,m stwls of healthy volunteers [2 171. Coque et al. [I291 showed that cyolysin wasmore common in non-end~ardills clinical isolates (3T0,b) and in hospital fecal isolates(31%) than among end~arditls (16%) and community fecal isolates (20%). However, thelesser incidence of hemolysin pmduction (14%) than bacteriocin production (42%) by E.

CHAPTER IVfucculi.~ isolates in the present study is of significance, since studies have shown that boththe properties are mediated by the same genetic determinants. Further genotypic analysisthese determinants would help us know the discordance in the concomitant expressionof both the properties. However from the clinlcal microbiological perspective, qualitativeanalysis of these virulence determinants would be sufficient enough to predict theprognosis of serious enterococcal infections.The gelatinase production in enteroctsci was depicted a century back, when thisproperty was used to classify S~reproc~occu.~ (uc,c.ull.\ into subspec~es. although theprocedure was stopped since the advent of Molecular taxonomical methods [13]. Butover the years, the virulence of enterococcal gelatlnase has been proven cn animal models1237. 2381. wh~ch was authent~cated by several stud~es depict~ng gelatinase production bycnterc~occi from human ~nfections Our study depicted gelatlnase prnductlon among 58%ul E /acc~ult.t ~wlates. wh~le none other species tested produced gelatinase in thephenotypic assay The aac(6')-aph(2") and ant(6')-I genes were present together in58 59.0 gelatinase pns~t~\c (Gel+) E ~uctulrr The aac(6')-aph(2") gene was presentheparatcly In 16'0 (;el+ t' /c~ei~ulrs isolates. while 8% of Gel+ E. ,/ucculis isolates were\trepton~ycin resistant, and the remaining 1790 Gel* isolates were sensitive to otherantlh~ot~cs tested. u.hlch Included 6Ba isolates reslstanl to vancomycinThe results of our study were concordant aith a study from U.S. that depictedgclatlnase product~on hy 54Oh of' E /(rc~crh\ Isolates from endocard~t~s. 5X0,0 of Isolateslroln other tnli~tions. 62% of hospital fecal ~solates. and in 27'0 of fecal lsolates fromhealthy volunteers, hut was ahsent In all the Xh non-E /oc.c,uli.\ Isolates studled [129]. Inmother study there was no s~gniticant association between 1.1-day mortality and any ofthe virulence markers studied among patlents with hacteremia due to E. faeculis, eventhough 649.0 of isolates produced gelatlnase [I 521. Several studles have shown that 45-()OD/o of the E /uc~culrs isolates from human infections produced gelatinase. whereas noneof the non- E. ,fuc~olis isolatcs did. But most of these studies did not address whethergelatinase affects the severity of disease. as it does In anlmal models [160. 2 17.239-2411.However, somc studies havc shown the role of the extracellular gelatlnases (pr~teases) in

CHAPTER IVhuman ~nfections is of clinical significance since they produce pathologic changes in thehost, by degrading lipids, deoxyribonucleic acid and hyaluronic acid (a component of theconnective tissues), or by disrupting the equilibrium by Inducing an inflammatoryresponse with the help of lipoteichoic acids, complement and neutrophils [26, 193,2041.Our results showed that majority of the HLGR enterococci possessed more thanone virulence factor. 83% of the BAC+ E. / u~~li.s ~solates carried "esp" gene, while 53%of the BAC- 1;. {ucculr.c ~solates camed the "esp" gene (P=0.076), which was statisticallyInsignificant slnce "esp" gene was also fbund among BAC- E. fu~t~ulis isolates. 86% of[he BAC+ E. /ucculrs isolates also produced gelatinase, while only 37% of the BAC- E./ucccrlrs lsolates could produce gelatlnase (P=0.001), which was highly significant sincegelat~nase productron was exh~hited by majority of BAC+ E. /ucculis isolates and lessernumher of BAC- 1;'. /rrcculrr ~solates. 67% of esp- E, /ucculis isolates producedgclat~nase, wh~le only 41°,0 of the esp- E /uc,rull.r lsolates produced gelatinase (P=0.104),u.hlch was statistically ~ns~gn~licant The slngle E facc,rrtm isolate that producedhncter~ocln also camed "esp" gcne on them. while none of the nine esp+ E. fuecir~misolates produccd bactenoc~n Thc prevalence rate of various virulence factors in thepresent study may not rellect the true inc~dencr. since the phenotypic assays (except for"csp") would not be ahle to rcvcal unexpressed virulence factors due to the presence of51lmt mutated genes as shown bq some stud~es [LS. 2411. although the probability for thisphenomcnon IS negl~gihle. Hence genotypic analys~s would be a better choice forcharacter~r~t~on (>I' the virulence detcnn~nants In enterococcl from "serious infections"that may a ~d In dererm~n~ng the outccimc ofthe d~seaseAlthough several v~rulence determinants have hecn depicted over the years Inenteroccxci. recently depicted pmpeny of hlofilm formation by enterococci has gainedmomentum due to thclr cl~nical signlficancc In nosocomial settings (210). Even thoughlnappropr~ate antibiotic use has contributed to the emergence of enterncocci In hospitals.studies have shown that acquisition of new virulence traits may have played an importantevolut~onary role [?b]. The capacity of entrrocnccl to cause infections is enhanced by theProPeny of biofilm formation especially on indurell~ng med~cal devices in hospitalized

CHAPTER IV~atients, since the proportion of enterococcal bacteremia associated with central venouscatheterslurinary catheters has increased dramatically over the years with significantmorbidity and mortality (13, 22, 261. Once a catheter has become colonized withm~croorganisms, invasion of the bloodstream can occur. Ilowever, for an effectivecolonization (of catheters), a microorganism must have the capacity to form a biofilm on(he dev~ce material, and studles have shown the biofilm-formlng characteristics ofhloodslream enterococcal isolates in vitro and the type of infection caused in vivo [365].Our resulls deplcted biofilrn formalron by 26% of E. foerulis, while none otherspccles tested depicted hlofilm formation on abiotic surface in a phenotyp~c assay uslngpolystyrene mlcrotlter plates based on the approach of Toledo et al. [210]. A moderate orueak hlofilm lormation was dep~cted by 75% and 2S0h oT t, furcolic respectively. Ourresults were concordant with the spectrum of blofilm formation as shown by Toledo et al.alncc none of the non-faecalis enterococcl produced biofilm in their study [210]. butd~acordant w~th another study tiom U.K. that showed 42O.0 of E. fuecirirn ~solates couldprtduce h~ofilm. apafl from 1:frrtculfi [?65] However the prevalence of h~ofilmlormatlon (26°,0) hy fi /crt~culr\ In our btudy was \erq less. when compared with other\rudles showlng I(lO0,o 13651. YJOO [213]. and 57'/0 [210] of the~r isolates producinghlclfilm. Funhcr. 57O.0 of' blotilm forming E lurc.ulr.\ were urinary ~solates. while 4I0,buere from bloodstream In our study w~th niajority of the patients catheterized. whichauthent~cates the s~gnificance 01' h~olilni fbnnat~on by these ~solates as shown by otheraludles [365].The genrtlc drlermlnants contrc,lllnp hlotilm formation in enterococcl are yet tohc unraveled completely, hut scveral studies have shown the in\.olvement of previouslycharactenzed vlrulencc determinants in hlofilm forn~ation [210. 2 12. ! 13. 3661. Theanalysis of the relationship between h~ofilm formation and the presence of other virulencefactors In t'.~o~cr1i.s depicted Interesting results through our study. Many studies havedcp~cted that "Esp" determinant 1s highly asscwiated w~th the abil~ty to form a biofilrn atahlotic surfaces [2 101, and "esp" induced blofilms had increased antimicrobial resistance13663. But the results of a recent study contradict the earlier hypothesis where the authors

CHAPTER IVdemonstrated that in vitro biofilm formation occurs not only in the absence of Esp, butalso in the absence of the entire pathogenicity island that harbors the Esp codingsequence. They concluded that E. fueculis forms complex hiofilms by a process that issensitive to environmental conditions and does not requirc the Esp surface protein [2 121.Our results showed that 29% of the "esp" posit~ve ~solates, as compared to 20% of the..esp" negative isolates (P-0.324.test value=0.97) showed biofilm format~on, while22% of the "esp" positive and gelatinase negative isolates, as compared to 6% of the"esp" and gelatinase negative ~solatcs (P=0.086, Fisher's exact test) showed biofilmfbrmation. which indicates that presence or absence of the "esp" gene does not greatly~nflucnce the biofilm firmation by E. /uccul~\ stat~stically as shown by other recentstudies. Another study showed that endocarditis ~solates of E. /ueculis produced biofilmsignificantly more often than nnn-endocarditis Isolates Furthermore. their results showedthat Esp was not required, hut Its presence was associated with higher amounts of biofilm[! 131 The absence of "esp" among 5 1% of blofilm formers in the same study motivatedthc Investigators to look Ibl. other genes that m~ght influence biofilm formation Theirrcbults showed that d~sruptions In other genes like epa (enterococcal polysaccharideant~gen). atn (autolys~n). gelE (gelatlnase). and Isr (quorum sensing locus) resulted infewer atlashed bacter~a and lesser biolilm formation as determined uslng phase-contrastmicroscopy. thereby emphas~r~ng the s~gnificant role of other determinants (apart "esp")In hiofilm Ibnnat~on In entertKocci [2 I?].However a recent study that dep~cted an esp-~ndependent hiofilm formation.suhsequently demonstrnt~d that CielE (gelatinase) enhances hiofilm formation by E.lu~~uli.\ (2121, which was funher authenticated by other stud~es [? 13. 3671. In our study3A00 of thc gelatinase positi\.e rsnlates, ah compared to 14°,~ of the gelatinase negative~holates (P=O.W.tea) showed blolil~n formarlon. while 42% of the "esp" negativeand gelatinase posit~ve ~%>lates, as comparrd to 640 of the "esp" and gelatinase negativeisolates (P=0.001, F1sher.s exact test) showed h~ofilm fonnation, which ind~cates thatPresence of yelatinase could enhance blofilm formation even In the absence of "esp"gene. On he other hand. 329/0the "csp" and gelatinase positive E. ./uccu~.~ isolatesd~~lcled biofilm fomatlon, while only 6°.0 of the "esp" and gelatinase negative E.

CHAPTER lV{aecalis isolates depicted biofilm formation (P=0.005), which was highly significantindicating that the presence of both the factors could influence the production of biofilm.However, it IS difftcult to predict the exact role of these virulence factors in biofilmformation unless the genetic machinery of the biofilm formation is unraveled completely.The prevalence of gelatinase production might have been more in our study if detectedgenotypically, since phenotypic assays are not 100% sensitive in depicting the presenceof the dcterm~nants encoding gelatinase production [25, 2411. Based on our results wespaulate that the gelatlnase production may enhance biofilm formation by cllnical E./urculi.c isolatesSUMMARYl'hc results of our study dep~ct the presence of various \,irulence factors amongcnterococcl Bacreriocln production was found In 42% of the E, /ueculi.\ isolates, and wasconfined mostly to this slngle species w~th the excepuon of one E. fuccrum isolate. The E./uc,c~Ir.\ bacler~ocln showed a narrou spcytrum of activity. active only agalnst tuo of thespucles-specific indicators and not agalnst non-specles specific indicator strains tested.~h~le the ti /uc*crrrm hacterlocln shoacd a broad spectrum of activity. against all genusrpec~fic lndlcator strains used In the study, hut not against non-genus spec~fic ~ndicatorstrain tested Hemolysin production wah depicted among 1-45; E /ucculis isolates and10O0~~ E diiru~~.\ ~solatcs. wh~le nonc other species ~solated and tested depicted productionof hcrnolysln Gelat~nasc product~on was detected among 58% of E, /uc.culi.\ isolates.N hlle nonc other spc~~cs ~solat~ul and testcd prnduced gelat~nase Most of the bacteriocinand gelatinase producrng E /ui,r.ulr.\ exh~bltcd HLARThe "Esp" gene was detected anlong 50% of all isolates of enterncocci. whichincluded 65% E. /ui~c~ulr.r and 4040 C'. /ci

CHAPTER IVproduced biofilm. The virulence traits depicted had permeated E. fucculis extensively inour clinical setup. Among E. /ucculi~ the highest prevalence of virulence factors wereexhibited by the urinary isolates (46%) followed by bloodstream isolates (38%), whileamong E. fuecium bloodstream isolates exhibited the highest prevalence (50%) ofvlmlence factor, followed by urinary isolates.The presence of antibiotic resistance and varlous virulence factors as shown In ourstudy although contributes to the emergence of enterncocci as a nosocomial pathogen.stud~es focusing on their genetlc machinery facilitating the exchange of thesedeterminants (plasmids) between strains is needed, and IS of immense value to preventand control the dissemination of \ ~rulence and antimicrobial resistance in any hospital\elup

Chapter 'I/Cell-CellCommunication andgene Transfer among JminoglycosideResistant Enterococci

CHAPTER VCHAPTER VThe versatile genetic machinery of enterococci has enabled them to become a prominentnosocomial pathogen. The pheromone induced aggregation substance (AS) mediatesefficient cell-cell (bacterium-bacterium) contact to facilitate plasmid exchange: mostlyencodlng antibiotic resistance and virulence traits, as part of a bacterial sex pheromone\).stem in entertrocci (17. 1161 However. the mere absence of this sex pheromone5ystcm among some clinical enterococcal Isolates doesn't wanant that plasmid exchangeuould not occur, slnce alternate sex-pheromone independent plasmid transfernlechan~sms have also heen dep~cted by several studies [IZO]. Hcnce. the property oftransferablu multldrug resistance including aminoglycosides. glycopept~des and beta-Iactams. is possibly one uxplanatlon Sor Intra and inter hospital dissemination ofcnterocc~-CI (6. 171 Funhcr. the .4S prtducrd by the pheromone respond~ng plasmids.ha\ e hcrn shown to enhance the ~nfccti\ ~ty of enterocncci by several means [I 36. I371Report:. have ahoun that there IICSdisparity in the gene transfer mechanisms ofenter(~occin various ga)graphical reglons Hence studies focusing on the gene transfer~nechanisms in entercxcxcl helps In understanding the d>namlcs of their geneticmach~nep. wh~ch In turn enahlcs to fi>miulate and linplement strategies that mlnlmizesthe problems of entercxcxcal ~nfect~cms. as uell the dissemination of Lirulence andant~microbial rcslstant traltb In any hospital setung.OBJECTIVESTo study the transf'ernhil~t\ and charactCrl/a~lon ol'tllr :ei~e!lc de~tv~~~i~vt:it,11)cnteroumi

CHAPTER VMATERIALS AND METHODS1. Pheromone response assayDetection of clumping (aggregation) by donor cells (test isolates) in response to thepheromone was carried out as described previously [I 16, 1271.A. Pheromone recovery5 ml N2GT broth (Oxoid Nutrient broth No.2 supplemented with 0.2% glucoseand buffered to pH 7.0 with O.lm Tris.HCI) was inoculated with 0.05 ml of anovernight culture of plasmid free E. faecolis recipient strain JH2-SSlFA2-2 RF.The cells were grown to mid log phase at 37'C with shaking.The cells were then pelleted by centrifugation at 12.000 rpm. and the culturefiltrate (supernatant) was collected and boiled for 10 minutes and used as"pheromone" for the clumping assays.B. Clumping assay0.5 ml of the culture liltrate (pheromone) was mixed with 0.5 ml of fresh N2GTbroth In I 1 ratio. to which 20 PI of overnight cultured cells (test isolates) wereadded to test for phcromone response or their abillty to clump.The mixture was incubated for 2-4 hours 1:10 with shaking and examined forclumps visually, as well microscoplcallg.(: DAPI staining: The mixture was also stained using 4'-6-Diarnidino-2-phenylindole(IIAPI) stain known to form fluorncent complexes with natural double-stranded DNA.and the clumping response was visualized under fluorescenl as well phase contrastmlcroscopc briefly as follows.DAPI stain 5 pgml (DAPI stain - Sigma Aldnch. US. Imgml stock made indouble distilled water) was added to a final concentration to the mixture (lml) andkept at room temperature for 10 minutes.10 pl of this suspension was placed on a glass slide and covered with a coverslipand visualized under fluomcent micmscope at excitation wavelength 350 nm. aswell under phase contrast.

CHAPTER V2. Conjugative in-vitro gene transfer assaysThe in-vitro gene transfer assays by conjugation were carried out by broth matings [I 161,or filter matings [368] whichever appropriate as described below:.4. Broth matingsOvernight cultures of donor (test isolates) and recipient strains (plasmid freerecipient strain E. furculis JH2SSl FA2-2 RF) were grown in Todd-Hewitt broth.0.5 ml of recipient cells and 0.05 ml of donor cells were mixed into 4.5 ml freshN2GT broth in a 1 : I0 donor recipient ratio.This I: I0 mixture was Incubated at 37°C with gentle agitation for 2-4 hours.After Incubation the mixture was vortexed to obtain a uniform suspension and 10fold serial d~lutions were plated (0.1 ml) on Todd Hewitt broth solid mediumsupplemented w~th 5% sheep blood and appropriate selective antibiotics andcolonies were counted after 48 hours of incubation at 37°C.The antibiotic concentrations used In the selective agar plates were as follows:streptomycin 250 pg and spectinomycin 250 pg (recipient markers), rifampicin 25pg and fusidic acld 25 pg (rec~pient markers), gentam~cin 500 pg (donor marker).Separate platings wcrc done from the mixturc to select donors (using donormarker alone) and rec~picnts (using recipient marker alone) separately. Thispmvldes a basis for estimating plasmid transfer frequency as transconjugants perdonor (or) recipient.Plasmid tmnsfer liequency was calculated by appropriate differencing betweentransconjugants and donor'reciplent count. i.e: transconjugants cfu ml I donor (or)recip~ent cfulml.B. Filter matingsOvernight cultum of donor and recipient strains (plasmid free recipient strain E.futsculis JHZSSI FA?-? RF) were grou8n in Todd-Hewn broth0.5 ml of recipient cells and 0.05 ml of donor cells were mixed into 4.5 mi freshN2GT hmth in a I: I0 donor recipient ratio

CHAPTER VThis I :I0 mixture was collected on a membrane filter (0.22 pm, Millipore, India),and the filter was placed on BHI agar plate supplemented with 5% sheep bloodand incubated at 37°C overnight.After incubation the cells were suspended in 1.0 ml of N2GT broth, and 0.1 ml ofthis cell suspension was spread on Todd Hewin broth solid medium supplementedwith 5% sheep blood and appropriate selective antibiotics, and colonies werecounted after 48 hours of incubation at 37°C.The antibiotic concentrations used in the selective agar plates were as follows:streptomycin 250 pg and spectinomycin 250 pg (recipient markers), rifampicin 25pg and fusidic acid 25 pg (recipient markers). gentamicin 500 pg (donor marker).Separate platlngs were done from the mixture to select donors (using donormarker alone) and recipients (using recipient marker alone) separately. Thisprovides a bass for estimating plasmid transfer frequency as transconjugants perdonor (or) recipient.Plasmid transfer frequency usas calculated by appropriate differencing betweentransconjugants and donor:recipient count. i.e transconjugants cfu/ml 1 donor (or)recipient cfulml3. Molecular characterization of TransconjugantsThc donors, recipients and transconjugants were molecular characterized to study theplasmid and genomic DNA profiles The whole (undigested) plasmid DNA and therestrict~on digested plasmid DNA were extracted as described in Chapter-Ill, Section-4.and separated on 0.8 % agamse gel using 0.5 X TBE at 60 V stained. visualized andd(xumented using a gel documentation system (Vilber-lourbet. France). The genomic!chromosomal DNA or the donors. recipients and transconjugants were Smal digested andseparated by PFGE as described in Chapter \'I and the gel patterns were subsequentlyanalyzed and interpreted appropriately

CHAPTER VRESULTSI. Pheromone response assayThe pheromone response of all 110 HLGR E. /aeaecalis and 20 HLG sensitive E. faecaliswere tested as per standard procedures [I 161. 68 (62%) HLGR E.,fuecalis and four (20%)of twenty HLG sensitive E. /uecalis exhibited clumping of cells that depicted apheromone response by enterococci tested as shown in Table 16.Table 16. Clumping response of Efuccu1i.tIsolam testedNO. tmtdNo. (%) of isolatesposilivc2. Conjugative in-vitro gene transfer assaysThe ~n-titrn gene transfer assays by conjugat~on were camed out by broth matings andfilter matings. Although transconjugants were obta~ned in broth matings, the transferfrequency was comparat~vely lesser than filter mating assay. Hence, our subsequentcxperiments were sw~tched to filter mating assay to check the plasm~d transfer frequencyfrom clinical donor isolates to a standard recipient E, fuecalis-FA2-2 (nfampin andfus~dic acid resistant). We subjected four groups of 30 randomly selected HLGR E.Iirc.c~u1i.s isnlates for the in-vitro gene transfer assay by filter lnatlng technique, the resultsol'which are depicted in Table 17. The isolates of four groups included were as follows:1. 'l'wenty aec(6')+aph(2")-HLGR. ant(6)-I-HLSR and pheromone response positive E.J~~tu.uli.v. 11. Four aac(6')+aph(2") and pheromone response positive E..farcalis, Ill. Fouraac(6')+aph(2") and ant(6)-I-HLSR positive but pheromone response negative E.,fuccolisand IV. Two aac(6')+aph(2") positive but pheromone response negative E. focculis. Thetransconjugants were assayad for their antimicrobial resistance by phenotpic andgenotypic assays. The transconjugants were assayed by PCR to check for the transfer ofthe virulence trait-"Esp".

~~CHAPTER V, 9953Lab no. ofdonor579716703846I 107926275145156276728014550Table 17. Filter-Mating experiments of HLGR-PCR positive E. faecalisDrug rcrialance pattern ofthe donor- - --- --Pcn ~mp, ~mPcn. Amp Gm, SmVan Gm SmPen Gm SmPen Amp Gm SmGm SmPcn Amp Gm 5mG~ smPen Amp ban Gm SmPcn Amp Gm Fm1 l6hO Pcn Amp (rm Sm825' 1 Pcn Amp Gm 5mhll(lPcn Amp Gm Sm02% I ~ m ., 4252509911.122I(. 1 ranwonjugant\. NT. No trmsonjuganh ohta~ncdI1 - I -sm .i 5 x 104 (3 sm G~ sm'I( gcmrvpcb. Pos~tlvc for lactaph ((rrn-H1 GR) an14 (Sm-HLSR) Esp gene h! PCRlnnsfer fq'. Tnnsfcr fqwncy- Tm*con~ugant\ donor ccllbi-.'-*i'ClumptngTCresponse Transfer freq.' TC phenotypeLIT ~m ~m ~ m ~m. , ~,pGm Sm I: i1Gm Sm34:;05I+ 45x10" Gm.Sm GiSm+ lRx10~ Gm Smt 70x 10' Gm Gm smI32 x 10'2 ox 104 ~m Gm sm ~m Gm sm Sm25x10~ GmGmNT42x10. I Gm 1 Gm35x10h0x 10'Gm Gm 1Gm Sm Gm Sm 1177' Pen Amp \an Gm SmI 42x 10'IW! Pcn Crm Sm ' + 19x10' Gm Sm Gm Sm 1I271 1 (im Sm 20x10" Gm Sm Gm SmX7M I Pcn Amp bm Sm3hx10' Gm Sm / Gm SmI,$298 I Pcn Om 5m 'Gm Sm Gm Sm 1I "i" II1871 , Gni Sm,I4018 1 Gm Sm5(lr10' 1 GmSm GmSm 15VhV I Pen. Van ~m 1 +1h.MI 1 Pcn. Amp. Van. Gm j . 34x10' / Gm Gm1 1 1h f i Pcnb.mp(1m5m . , 41rlo1 bmsm GmSm57 Pcn Amp Sm 5 Ox 10' Gm Sm 1 Gm Sm(im. Sm I . 30r loQPcn. Amp. GmGm3 2 ~ 1 0 'J O IO( ~Gm Sm 1 Gm Sm E%p 1Gm I Gm 1GmGm (3. 1Gm

CHAPTER VThe frequency of transfer of the gentamicin resistance marker ranged from 10" to10'' transconjugants per donor cell, while four donor strains did not yield transconjugantsin broth mating as well filter mating assays as depicted in Table 17. The transfer of "esp"gene-a putative virulence factor was demonstrated among transconjugants obtained fromtwo of the thirty donor E. fueculis as confirmed by PCR. However in both instances, onlyfew among the population of transconjugants showed the transfer of "esp" gene asevident by PCR, while the remaining transconjugants tested possessed theaminoglycoside resistance genes.3. Molecular characterization of TranseonjugantsThc donors, recipients and five transconjugants were characterized to study the plasmidand genomic DNA profiles. The plasmid DNA was present in all the donor strains used inthe ~n-vltro gene transfer assays, while the recipient strain and three of fivetransconjugants (which included two "esp" positive transconjugants) did not yield anyplasmid DNA upon multiple extractions The remaining two transconjugants showed anearlyden tical plasmid DNA profile of donor strains after restriction digestion,supponlng the transfer of plasmids liom the donor to recipient as shown in Figure 11.'The Smul macrorestriction d~gestlon of three transconjugants (including two "esp"pos~tlve transconjugants) and their corresponding donors and recipient were performedand separated by PFGE as shown in Figure 11 The macrorestriction PFGE pattern of thedonors were heterologous, when compared with FA?-2 recipient and esp-negative,nm~noglyccistde msis~ant gene positive transconjugants. which had an identical pattern.On contrary "esp" and aminoglycos~de resistance gene positive transconjugants showed a"closely related" pattern of the recipient with two band differences. leaving us tospeculate that chromosome-to-chromosome transmission of "esp" gene might haveoccurred. On he other hand the FA2-2 recipient and the esp-negative, HLGRtransconjugant showed an "iden~ical" PFGE pattern, authenticating the involvement ofplasmid DNA in transfer of gentamicin resistance.

CHAPTER YFigure 11. Molecular Characterization of E. faecalis donors,recipients and transconjugantsA. Piasmid DNA restriction [EcoRi] profiles of donors,recipients and transconjugants1.6275 yonor], 2- 6T30 [donor]; 1'and 2'- Co-pondinp Tnnsconjugant3- Rulpknl E I H C ~ I FAZ-2. 4- Rulp~ent E. h.ce11s JH2 SSM- Hlnd Ill dign1.d Lambda DNA marhar6. Chromosomal DNA restriction [Small profiles of donors.,.,,,2. (..m 2.. npnrg.tin. tun ~.nn palth. mnuonjwn*3.. WAR gmn pa* (nrpconjwn(R. R- E. FAZ-2: Y- ~ mbda bW.r PFQ m*.r.-

CHAPTER VDISCUSSIONEntmocci exhibit resistance to an array of antimicrobials including the most recentmolecules, a propeny that has catapulted them as one of the leading clinical challenges[193]. The sex-pheromone based gene transfer system has made them "unique" among allnosocornial pathogens. Since their discovery, sex-pheromone inducedlmediated genetransfer especially those of antimicrobial resistance, between E. faeculis isolates havebeen demonstrated [b. 1271. Small peptide sex pheromones that are specific for differenttypes of plasmids are secreted by plasmid free strains of E. ,faecalis (recipient cell) intothe culture medium. A potential donor cell containing a pheromone-responsive plasmid(generally encoding antibiotic resistance) comes into contact with its correspondingpheromone, following which the transcnption of a gene on the plasmid is turned on,resulting in the synthesis of a sticky substance called aggregation substance (AS) on itssurface Subsequently, formation of some sort of mating channels between the cells leadsto transfer of genetic determrnants encoding antibiotic resistance!vi~lence determinantsbetween donor and recipient strarns 16. 17. 1271. In addition to mediating bacterialbrnd~ng to other enterococci. AS also plays a role In binding of E. faecu1i.v to eukaryoticcells, including p ~g renal tubular cells and human rntestinal epithelial cells [199].Mostly, E, fueculis strains fmm hospitalized patients possessing pheromoneresponsive plasm~ds encoding antibiotic resrstance have shown a clumping response,leading to transfer ol genetic de~erminants (encoding antibiotic resistance) between thedonor and recipient strains [129-1.121. In our study 68 (62%) of the HLGR E. fac.culis,and 4 (20%) of 20 HLG sensitive E, fuec.olis exhibited clumping depicting production ofaggregation substance (AS) in response to a culture filtrate known to contain numerouspheromones, which suggests their potential to transfer plasmid DNA between clinicalisolates. although the degree of clumping varied between test strains.Our results matched a study fmm Japan, where 85% or the hemolytic strainsexhibited a clumping response and about 90% of these hemolytic strains were resistant toOne or more drugs, w hew this was true for only about 54% of non-hemolytic strains

CHAPTER Y[235]. While several studies from Greece, Japan, US, have shown that 95% to 100% ofpheromone responsive plasmids encoded gentamicin resistance among clinicalenterococci [133-135, 1611. These studies concluded that these pheromone responsiveplasmids might play a role in the spread of gentamicin resistance especially in a hospitalset up.Most of the studies including ours have used phenotypic assays for studying thepheromone responsiveness of the clinical isolates. While some of the recent studies haveused genotypic assays to detect the presence of Aggregation Substance (ASlagg) gene- aproduct of pheromone response by plasmids, which also contribute substantially to theincreased pathogenicity of enterococci. Coque et al. [I291 in a study among 192 Efaecalis showed that AS was found among 52% of isolates from endocarditis, 72% ofisolates from other infections, 56% of hospital fecal isolates and among 30% of fecalisolates from healthy volunteers, while none of the 86 non- E. faecalis isolates werepositive for AS. Another study from Italy revealed that AS gene was found among 33%of E. ,fuecu[is isolates, but none among the E. ,furcium isolates [203]. These studies alsoemphasize that the phenotypic assays may not reveal the exact prevalence of pheromoneresponse among enterococci, since many a times "silent AS genes" may not enableenterococci to exhibit a pheromone response that depicts a lesser prevalence of ASamong enterococci otherwise. Hence it is unwarranted that all the pheromone responsenegative strains may not be true negative, although the percentage of such strains lfprevalent would be very less.The clumping response exhibited by 20% of the HLG sensitive E. ,faeculis in ourstudy suggests, that apart gentamicin resistance plasmids, other plasmids encodinghemolysidbacteriocin, UV resistance and resistance to other antibiotics like tetracycline,kanamycin, erythromycin, vancomycin (but not gentamicin resistance) can also bepheromone responsive as shown by several studies [6, 26, 2351. Our finding isconcordant with the results of these studies since the four pheromone responsive E.fuecalis isolates (20%) in our study were sensitive to other antibiotics tested includingHLG, but were hemolysinibacteriocin positive. Thus our results depict that apart

CHAPTER Vfacilitating the transfer of antimicrobial resistance determinants, pheromone responsivestrains may facilitate transfer of virulence determinants between cells, which is of primeconcern in a hospital setting.Although aggregation substance (AS) is a pheromone-inducible surface proteinthat mediates binding of donor cells to plasmid free recipients, which is essential forhigh-efficiency conjugation of sex pheromone plasmids, it has also been shown to act asa virulence factor during host infection and AS remains to be the most well studiedadhesin of enterococci [199]. Several studies depict the significance of AS, as an~mportant virulence factor in the pathogenesis of enterococcal infections like UTI andcridocarditis. Our results shows that all the AS positive E. fueculis isolates possess one ormore of the virulence factors: esplgelatinaseibacteriocin studied, depicting these infectionderived isolates may represent a unique genetic Image capable of being disseminatedwithin the hospital setupThe other major function exhibited by AS is adherence of the E. juccalis isolatesto host tissues, since in vitro studies have shown that AS mediates adhesion to variouseukaryotic cell surfaces, such as cultured pig renal tubular cells and promotesinternalization by cultured human intestinal cells, suggesting that AS-expressing cellsmay likely form larger aggregates in vivo than cells not expressing this trait [195-1981.Furthermore, AS have been shown to exhibit resistance to killing by polymorphonuclearleukocytes and macrophages. thereby promoting intracellular survival of E. ,furt.ulisinside neutrophils [199]. While lsenmann et al. [201, 2021 have shown that interaction offibronectin and aggregation substance promotes adherence of E. ,fueculis to human colon.Thus studies focusing on the prevalence of AS, either by indirect phenotypic clumpingassays, or through genotypic assays to detect the presence of Aggregation Substancelagggene would be a good indicator to know the appalling characteristics of the clinicalenterococcal isolates.The transferability of various genetic determinants especially by conjugation hasenabled nosocomial strains of enterococci to outcompete indigenous commensal

CHAPTER Venter~~~~~i, thereby increasing the presence of nosocomial strains in the gastrointestinaltract, as well dissemination of these strains in a nosocomial setup. Our preliminary resultsdepicted that the donor isolates could transfer plasmids in both broth and filter matings[132], however in each case plasmids were transferred at a higher frequency to the E.fhecaiis recipient by filter mating than by broth matings, which has been shownpreviously by some studies [6, 7,2511. Hence, our subsequent experiments were switchedto filter mating assays unless otherwise specified (as broth matings). The four groups of30 randomly selected HLAR E. faecalis isolates subjected for the gene transfer assayswere having both pheromone response positive and pheromone response negative strains,the combinations of which are depicted in Table 17. The transconjugants were assayedfor their antimicrobial resistance as well for the transfer of the virulence trait "Esp" byphenotypic and genotypic assaysThe group 111 and IV HLAR E, fuecalis isolates in our conjugation studies werepheromone response negative as evident by clumping assay results. Hence we suggest theprevalence of a different plasmid system among these isolales, although the possibility ofnon-responsiveness to the corresponding pheromone due to a "silent gene" cannot beruled out as shown by some studies [25, 210, 2411. The mating experiments betweenplasmid free recipient E. faecalis (FA2-2 RF) and, group I11 and IV donor cells did notexhibit visible mating aggregates unlike the group I and 11 donors, although transfer ofplasmids occurred between donor and recipient cells. However, small mating aggregateswere observed when the broth mating mixture was stained using 4-6-Diamidino-2-

CHAPTER Vphenylindole (DAPI) and visualized under fluorescent as well phase contrast microscope,but aggregate of donor cells (self-clumping) were not even observable by microscopy inthe DAPI stained clumping assay after exposure to a culture filtrate of a standard plasmidfiee recipient strains E. faecalis JH2 SSI FA2-2 RF known to contain multiplep her om ones as shown in Figure 12. Further, these strains failed to transfer plasmidswhen E. ,faecium BM 4105 RFIBM 4105 SS were used as recipient strains, confirmingthat they do not possess broad-host-range plasmids like pAMPl or plP501 11251. Thus,our results predict that these isolates may possess non-pheromone responsive plasmids ora pheromone-independent plasmid system as described recently 1120, 1391, although wehave not done hybridization studies with DNA probes specific for these plasmids toconfirm the existence of these plasmid groups among enterococci in our hospital setup.Few studies including ours, have depicted the nature and mode of the plasmidtransfer [6, 7, 120, 132, 1391, while most other studies carried among nosocomialcnterococci worldwide have investigated only the ability of the enterococci to transferantimicrobial resistant plasmids between strains without probing the pheromoneresponsiveness of the HLGR plasmids [2S I . 350. 356-3581, Generally in a clinical setupthe demonstration of the transferability of genetic determinants alone would be sufficientenough to underscore the significance and their potential for dissemination ofantimicrobial resistance. However, investigating the nature and mode of gene transferwould further contribute in understanding the basics, which enable us to redefine currentstrategies to prevent and control the spread of multidrug resistance among enterococci.High-level plasmid mediated gentamicin resistance in E. ,faecali.c was firstreported in 1979 in France [254]. In 1983 a study from U.S showed that nine clinicalisolates of E. faecalis encoded high-level resistance to gentamicin as well otheraminoglycosides, and depicted that the genetic determinants for HLGR were canied onconjugative plasmids that transferred at a high frequency [258]. Later in 1990, the HLGRgenes were shown to be located on transposons, which increased the transferability of theHLAR isolates [369]. Thereafter, several studies showed the transferability of gentamicinresistance and the role of transposons mediating this process.

CHAPTER VFigure 12, Representative Phase contrast and Fluorescent Microscopy images ofClumping and Mating Assays of Entemoccus test strainsE faecalis JH2 SS pheromonesINo self-clumplng by E faecal~s - 6,660 Group-ll test strain agalnstE faecahs JH2 SS pheromonesE feecalls FA2 RF [recrplentj In a broth matlng rnlxture 1

CHAPTER VThe frequency of transfer of the gentamicin resistance marker ranged from 10" to10.' transconjugants per donor cell, which is concordant with several studies. A recentstudy from Greece [I 341 had depicted a similar plasmid transfer frequency of 10" to I om',But, all the HLGR isolates in their study have been shown to exhibit clumping response,proving that all HLGR plasmids were pheromone responsive, in concordance withseveral other studies [133, 135. 1611. However, the present study showed that all theHLGR plasmids were not pheromone responsive as described in the previous section.Some studies have shown a very high plasmid transfer frequency between 10" to 10.'among the pheromone responsive plasmids in broth matings [133, 1351, while otherstudies that did not differentiate the pheromone responsiveness of plasmids have shown a~lasmid transfer frequency (gentamicin resistance) between 10'~ to 10.'assays [251, 350, 356-3581, which matches the transfer frequency ofour study.by filter matingApart from the type of plasmid, the transfer frequency depends on several otherfactors, including the association of the plasmids with various transposable elements likeTn4001iTn5281, which are found commonly in enterococci [369]. Studies have shownthat variant forms of these transposons can be generated by molecular rearrangementsduring the process of transposition from the chromosome to a plasmid. or during in vivoconjugative transfer 13561. Interestingly, the frequencies of transfer of gentamicinresistance for the isolates containing truncated variants of Tn4001 (generally located onplasmids) are significantly higher than those isolates containing non-truncated variants(mostly canied on the chromosome) as shown by some studies from France and Greece(355. 3561.In our study few pheromone responsive HLGR E. ,/aecalis isolates were not ableto transfer the HLGRIHLSR marker both by filter and broth matings. The conjugativetransfer of gentamicin resistance may not be possible if the aac(6')+aph(2") gene iscartied on non-truncated transposons regardless of their location on a plasmid or on thechromosome (3561, which can be quoted as a possible reason for non-transfer of theHIXR markers by some of our isolates.

CHAPTER VAlthough majority of the donor isolates could transfer the HLSR marker alongwith HLGR marker, few donors failed as depicted by the profiles (phenotypiclgenotypic)of transconjugants (Table 17). The possible explanation for this phenomenon may be thatHLSR marker might be encoded by a different non-conjugativeinon-pheromoneresponsive plasmid, or else HLSR marker might be present on the chromosome of theclinical isolates that couldn't be mobilized. This fact was authenticated by ourhybridization studies, since different fragments of the EcoRl digested plasmidsh$ridized to the HLSR and HLGR gene probes respectively for the donor E. fueculisisolates that failed to transfer HLSR marker. The failure to transfer HLSR marker bysome E. jueculis isolates is in agreement with previous reports [287, 3571. which hadquoted similar reasons.Thus as depicted above. even though sex pheromones appear to play a significantrole in the evolut~on and transfer of certain plasmids in enterococci, II should be kept inmind that the importance of aggregation substance in enterococcal matings relates to itsahillty to initiate contact in liquid suspensions, and it is not an essential component inmatings that occur on solid surfaces [370]. In natural environments, interspecies contacton surfaces may be a common occurrence (e.g., within biofilms). and potential signalsthat may be present In such environments still influence the induction of multiplecomponents of a given mating system leadlny to transfer of genetic determinants betweencells. Further, the existence of an alternate conjugative plasmid transfer system that is"pheromone-independent", but capable of transferring the antibiotic resistant plasmidefficiently during broth matings [120, 125. 1391 is also of concern. Thus. the versatility ofthe genetic machinery of enterococci promotes dissemination of resistance and favorsstability of resistance genes even in the absence of exposure to antibiotics.Most interestingly, the transfer of "esp" gene-a putative virulence factor wasdemonstrated among transconjugants obtained from two of the 30 donor E. ,fueculis asconfirmed by PCR. However in both instances, only a subpopulation of transconjugantsacquired the "esp" gene as evident by PCR, i.e. only 2 among 18 transconjugants testedacquired the "esp" determinant from E. ,foeculis as depicted by PCR, while all the

CHAPTER Vtransconjugants tested possessed the aminoglycoside resistance genes, indicating that thetransferred antibiotic resistance determinants were not directly linked with the "esp"gene. This is not surprising, since antibiotic resistance determinants have not beenidentified in the Pathogenicity island (PAI) that encodes the "esp" gene in enterococci[2141. This fact was authenticated by our study, since the plasmid DNA isolated from thedonor strains could not show the presence of the "esp" gene by PCR, which otherwisewere positive for the HLGR determinants by PCR and DNA hybridizations.Since the time of discovery of this putative virulence factor-"esp" studies havetried unsuccessfully to demonstrate transfer of this gene [214]. Recently Oancea et al.[223] from Germany demonstrated the conjugative transfer of "esp" among E. ,fueculisand E. juecium isolates. During the same time period we demonstrated the transfer of"esp" between E. ,/aeculis isolates, however there were some differences between boththe studies. Our study could depict the conjugative transfer of "esp" gene between (two)cl~nical donor E ,fueculis (HLAR) and a standard E. ~uecalrs isolate (FA2-2 RF:rifampicin and fusidic acid resistant) with a transfer frequency of 4.2 x 10.'and 2.5 x lo4Ibr both the isolates respectively. while attempts with E. ,/uccium BM 4105 RF as arecipient stra~n failed to yield any transconjugant. In contrast, Oancea et al. 12231 showedthat oue E. fucculis, and none of the E. fueciurn when mated with recipients of the otherspecies (E. juecium) generated esp-positive subpopulation of transconjugants, i.e.: onlyone among founeen transconjugants acquired the esp determinant from E. jaecalis asdepicted by PCR.Molecular characterization of the plasmid DNA from 30 donors in our studyrevealed that they did not encode "esp" as evident by PCR (but encoded HLGRIHLSR),while the recipient strain and three (which included 2 "esp" positive transconjugants) ofthe five transconjugants tested did not yield any plasmid DNA upon multiple extractions.The remaining two HLGR transconjugants showed a nearly identical plasmid DNAProfile of their corresponding donor strains after restriction digestion, supporting thetransfer of plasmids (HLAR) from the donor to recipient as confirmed by hybridizationwith HLGR~HLsR probe. Thus these results led us to speculate a chromosome-to-

CHAPTER Vchromosome transfer of the "esp" gene from the donor to recipient E. faecalis strains.Hence, molecular characterization of the chromosomal DNA of three donors, recipientand corresponding transconjugants were performed by PFGE after Smal macrorestrictiondigestion. The macrorestriction PFGE panem of all three donors were heterologous,when compared with E. faecalis FA2-2 recipient and esp-negative, aminoglycosideresistant gene positive transconjugants, which had an identical pattern. These resultsauthenticate the involvement of plasmid DNA in transfer of gentamicin resistance, aswell non-transfer of the "esp" gene to the recipient. On contrary "esp" andaminoglycoside resistance gene positive transconjugants showed a "closely related"pattern of the recipient with two band differences (from recipient), leaving us to speculatethat chromosome-to-chromosome transmission of "esp" gene might have occurred.Although the chromosome-to-chromosome transmission of "esp" gene as shownIn the present study is concordant with another study by Oancea et al. [223], the plasmidlocalization of the "esp" gene shown by them was divergent from our results. They hadexplained that the "esp" gene was mobilized from the donor chromosome, integrated intoa conjugative plasmid, which was then transferred into the recipient, giving an esp-positive transconjugant. This could be of high clinical significance, since in our studyrnajonty of the donors possessed conjugative plasmids encoding gentamicin resistance, aswell "esp" positive, thus having every chance of mobilizing other determinants includingthat of "esp", although none tested in our study depicted the same. However, thepossibility of the "esp" gene transfer among other isolates is cnwarranted since they werenot tested. But, hybridization with an "esp" gene probe, which we did not perform on our~solates, would have further authenticated our results, although the molecularcharacterization of the plasmid and chromosomal DNA prove to be qualitative evidencesfor our study. To our knowledge, this is the second experimental evidence to prove thatchr~mosomall~ encoded virulence traits like "esp" is capable of being transferable byconjugation, which may give rise to newer genetic lineages.

CHAPTER VSUMMARYThe results of our study depict that the enterococci isolated from our setup were highlycapable of transferring genetic determinants as evident by conjugation assays. Thet he no typic assay depicted pheromone responsiveness by 62% of the all HLGR E./uc,colis isolates and 20% of HLG sensitive E. fuecalis isolates tested as evident by aclumping response. Four groups of 30 randomly selected HLGR E. ,fueculis isolatessubjected to in-vitro gene transfer assay showed that the gentamicin resistance markertransferred at frequencies ranging between I 0.' to 10" transconjugants per donor cell for26 donors. The transfer of "esp" gene-a putative virulence factor was demonstratedamong transconjugants obtained from 2 ot'the 30 donor E. ,fuecalis as confirmed by PCR.However in both instances, only few among the population of transconjugants showedthe transfer of "esp" gene as evident by PCR, while all the transconjugants testedpossessed the aminoglycoside resistance genes. The PFGE patterns of the donor strainswere heterologous, when compared with FA2-2 recipient and esp-negative.aminoglycoside resistant gene positive transconjugants. which had an identical pattern,authenticating the involvement of only the plasmid DNA in transfer of gentamicinresistance. On contrary "esp" and aminoglycoside resistance gene positivetransconjugants showed a "closely related" pattern of the recipient with two banddllferences, leaving us to speculate that chromosome-to-chromosome transmission of"esp" gene might have occurred.The gene-transfer mechanisms in enterococci facilitate the dissemination of thegenetic determinants encoding antimicrobial resistance and virulence in any hospitalsetup. However, only molecular typing of the isolates authenticates the relationshiphetween the isolates disseminated in a hospital setup. This would help us to know theclonality of the isolates and to prove the transmission of the clones, if any, betweenpatients in our hospital setup, during the study period.

chapter VIJminog~cosde qesktant Enterococci

CHAPTER VICHAPTER VIEnterococci rank among top three pathogens causing nosocomial infections worldwide,since last decade [I 3, 171. Although initially thought to have evolved from patient's ownflora, enterococci was later shown to be exogenously acqu~red from nosocomial settingsby Zervos and his colleagues in 1986, using molecular epidemiological tools [I 81. Sincethen, innumerable reports of nosocomial enterococcal infections were published and moststudies used molecular epidemiological tools for typing enterococci to study the clonalityof the isolates. In case of suspected outbreak conditions, the analysis of strain clonalityhelps in confirming the association between palients (hosts) and reservoirs forcnterococci, and to determine the possible modes of transmission. In some instances, thephenotyping and antibiotyping results of enterococci may help presumptively ininvestigating whether the isolates studied have, or lack clonal relationship, althoughunwarranted. However, most of the inconclusive results obtained by other typingmethods can be authenticated by the application of molecular epidemiologicaltechniques, which gives a clear picture regarding the clonality of the isolates.Although a variety of molecular epidemiological techniques have been applied forepidemiological typing of drug resistant enterococcl "Pulsed field gel electrophoresis"(PFGE) is considered to be the gold standard for molecular epiden~iological analysis ofgram-positive cocci, since last decade [279]. Other techniques like plasmid DNAanalysis-which was the first tool to be applied for epidemiological analysis of enterococci[I 81, PCR based typing and a more recent technique: AFLP have also been found to beeffective in epidemiological typing of en1erococci. Thus molecular typing of enterococciis inevitable to draw conclusive evidences regarding the epidemiology of drug resistantstrains in any health-care setting.

CHAPTER VIOBJECTIVESTo document the epidcmioiogical pattern of enteroa)cci bv Molecular typingof the isolatesMATERIALS AND METHODS1. Molecular typing of HLAR enterococciThe molecular typing of HLAR enterococci was performed by two methods namely,chromosomal DNA restriction endonuclease digestion hy PFGE and plasmid DNArestriction endonuclease digestion by agarose gel electrophoresis. Overall, we used thesame set of randomly selected HLAR enterococci for both the molecular typing methodswith occasional exceptions in few instances (where few strains were included/excluded ine~ther typing method, albeit occasionally). The results of both the methods werecompared and analyzed statistically to evaluate their significance.A. Pulsed field gel electrophoresis-PFGEThe HLAR enterococci were subjected to pulsed field gel electrophoresis to compare thepolporphisms in the genomic DNAs among test strains as described previously withminor modifications [279]. Briefly. the procedure involves preparation of agarose-DNAplugs, processing and restriction-digest~on of plugs as described below:i. Preparation and processing of DNA plugs0. Preparation of genomic DNA in agarose plugsEnterococci were grown overnight in 5 ml of Brain heart infusion broth-BH1 (HiMedia, Mumbai. India) at 37°C and the cells were harvested by centrihgation at6,000 rpm for 15 min at 4" cThe cells were suspended in an equal volume (5 ml) of PIV buffer (I M NaC1, 10mM Tris-HCI [pH 7.61) and mixed well.

CHAPTER VIA portion of this suspension (2.5 mi) was mixed with 2.5 ml of 1.6% Low meltingtemperature agarose (SEA-Plaque Agarose, FMC bioproducts, Rockland ME,USA) in sterile, distilled, deionized water at 40-50°C and stirred well to mix thesuspension.This mixture was then pipetted (100 pl) into a plug mold (Bio-Rad Laboratories,Hercules, California) and allowed to solidify.The agarose plugs were removed carefully from the plug mold without causingdamage to the plug and lysed by placing 1-4 plugs in 10 ml of fresh lysis solution(6 mM Tris-HCI [pH 7.61, 1 M NaCI. 100 mM EDTA [pH 7.51. 0.5% Brij 58,0.2% deoxycholate, 0.5% sodium lauroyl sarcosine, 20 mgiml of RNAse [DNAsefree] and I mdml of Lysozyme and incubated overnight at 37°C with gentleshaking (N.B: RA1Asc and Lv.voyme were uddedjresh(vju.stprior to lvsis step).Following incubation the lysis solution was replaced with 10 ml of ESP solution(0.5 M EDTA [pH 9-9.51, 1% sodium lauroyl sarcosine and 100 mdml ofprotelnase K) and incubated overnight at 50°C with gentle shaking.After incubation the ESP solution was decanted completely and the plugs werewashed thoroughly with dilute TE buffer (10 mM Tris-HCI [PH 7.51, 0.1 mMEDTA) thrice for 30 minutes with gentle shaking at 37"~.After washing, the plugs were transferred to a clean tube with fresh diluted TEbuffer (5 ml) and stored at 40°C.N.B. Optimization: L'arious incuhatiorr conditions, urrd ~~oncm~rulron of cn;,'meswere tried ,for g~womic DNA ugarose plug preparution during .standurdr:ution. andthe uhoi,e-menlioned conditions wcrc ,follon~ed throughout our studv since theyyii~lded thc best results comparutii~e~.b. Restriction digestion of DNA plugs: The restriction digestion of enterococcal DNAembedded in agarose plugs were performed using Smal (New England Biolabs, U.K)restriction enzyme following the recommendation of the manufacturer as follows:

CHAPTER VIaThe restriction digestion was performed by placing a DNA plug in a sterilemicrocentrifuge tube with 200 pl of distilled water, 25 pl of restriction buffer(supplied along with the restriction enzyme) and 20 Units of Smal restrictionenzyme and incubated at 2S°C in a water bath (Julabo, U.S.A) for 12 hours.After incubation, the restriction digested DNA plugs were washed with dilute TEbuffer (1 ml) for 1 hour at 37°C and electrophoresed using a CHEF apparatusunder optimal conditions.ii. Contour-clamped homogenous electric field-CHEF electrophoresisThe restriction digested DNA plugs were subjected for CHEF electrophoresis as belowmentioned.a. Loading rhe sample agarose plugsSingle plugs of agarose were placed on each tooth of the comb and thepreparation was left at room temperature for some time (to facilitate adhesion).The gel was cast around the comb and the plugs remained in place when the combwas removed after solidification of the gel. A lambda ladder PFG marker (NewEngland Biolabs. U.K) was loaded along with sample plugs in every mn.b. Casting the gelThe casting stand (14 cm x 12.7 cm) supplied with the CHEF-DR 11 apparatus(Bio-Rad laboratories, U.S.A) was used to cast the agarose gel for CHEFelectrophoresis as per manufacturers instructions.Briefly. 1.2% agarose gel was prepared (Low melting temperature agarose, SEA-Plaque Agarose, FMC bioproducts, Rockland ME, USA) in O.SX TBE buffer and80-100 ml of molten agarose was poured (for a 5 mm thick gel) and allowed tocool at room temperature for 1 hour.The comb was removed after solidification and the wells were overlaid with fewmilliliters of remaining molten agarose to fill the gap above the l lug, and allowedto cool for IS minutes before starting the elec~ophoresis.

CHAPTER VIr. CHEF electrophoresisThe CHEF-DR I1 electrophoresis chamber (Bio-Rad laboratories, USA) wasfilled with 2 liters of pre chilled O.SX TBE running buffer 30 minutes beforeelectrophoresis.A variable speed pump connected to the gel tank was turned on, to equilibrate thecirculating buffer to desired temperature. The temperature of the circulating TBEbuffer was maintained at 14°C with the pump tube coiled in a tcrnperaturecontrolled water bath (Julabo, U.S.A).The pump was turned off, and the gel was placed in the chamber so that thebottom rested against the two gel stops inside the chamber. It was checked toensure that about 2 mm buffer covered the gel and then the pump was turned on.The buffer flow rate was maintained by adjusting the variable speed pump knob.at 1 liter per minute without disturbing the gelThe lid was replaced onto the electrophoresis chamber, and the Pulsewave 760Switcher and Model 20012.0 power supply were turned onThe switching times were set using the Pulsewave 760 switcher with the pulsetime ramped from 1 second to 35 seconds and a start ratio of I .O over 27 hours at200 V. and the run was carried by maintaining the circulating buffer temperatureat 14°C.After electrophoresis, the gel was removed carefully and stained with ethidiumbromide (0.5 rngiml) for 20 minutes and destained in distilled water for 1 hour.Then the gel was transferred onto an UV transilluminator visualized anddocumented for further analysisB. Plasmid restriction endonuclease analysis-REA @pingi) Plasmid REA @ping: The plasmid DNA of the HLAR enterococci along with controlstrains were isolated by alkaline lysis method, digested with restriction endonucleaseEcoRI, separated by agmse gel electrophoresis and documented as described previouslyin Chapter 111.

CHAPTER VIC. Visual and Computational analysis of the gelsThe PFGE and Plasmid gels were visually interpreted based on the differences in bandingpattern as per the consensus guidelines of Tenover et al. [294]. The computerizedinterpretation of the banding patterns was performed, by analyzing the captured gelimages using Bionumerics software, Version 2.5 (Applied Maths, Kortrijk, Belgium).The dice coefficient of similarity was calculated and a dendrogram constructed forphylogenetic (cluster) analysis using the unweighted pair group method with arithmeticaverages (UPGMA).D. Statistical analysis of typing methodsi. Comporison of PFGE and Plasmid REA typing methodsThe concordance between PFGE and Plasmid REA typing was determined based on thesimilarity of clusters~groups obtained by visual interpretations matching the consensusguidelines of Tenover et al. [294], from the isolates for which both PFGE and PlasmidRE profiles were available.ii. Simpson 's index of diversify and Confidence intervalThc Simpson's index of diversity was used to test the discriminatory power (D) of thetyping methods. The calculation of single numerical index of discrimination (D), wasbased on the probability that two unrelated strains sampled from the test population willbe placed into different typing groups as described by Hunter and Gaston [371] Thisindex was derived from elementaly probability theory and is given by the followingcquation:where N is the total number of strains in the sample population, s is the total number oftypes described, and n, is the number of strains belonging to the jth type. This equationwas derived as follows. The probability that a single strain sampled at random willbelong to the jth group is nJN. The probability that two strains sampled consecutivelywill belong to that group is n, (n, - I)/ N(hf - I). These probabilities can be summed for all

CHAPTER VIthe described types to give the probability that any two consecutively sampled strains willbe the same type. This summation can be subtracted from 1 to give the equation above.No correcting factor for small populations has been made, as typing schemes are notvalidated with small samples. A Simpson's index close to zero indicaies that there is littlediversity as shown by the typing method (index= 0 indicates no diversity at all) whereas aSimpson's index approaching I indicates a high diversity as shown by the typingtechnique (index= 1 indicates maximum diversity where no two isolates are similar).An approximate 95% confidence interval was calculated as proposed byGmndmann et al. [372]. Briefly, the Simpson's index of diversity D. for the assessmentof the discriminatory power of typing techniques is an unbiased estimate of the truediversity A of a population based on a sample of n individuals. Inferences on the diversityol'the population involve a sampling process. Simply by chance, different samples willgive difrerent results, the difference being due to sample variation and by drawingrepeated samples the prec~sion of the mean estlmate for D will improve. If repeatedsamples of a fixed size n are drawn from the sample population, the values of D will bedistributed about >. ~ith the variance a'.where n , is the frequency n jn, n, is the number of strains belonging to the jth type, and n1s the total number of strains in the sample population. An estimate of the standarddeviation of. A is given by the square root of o', and we propose the following asapproximate 95% confidence interval (CI):-c-1 - l L~-~,~,v+:,o~JWe applied these equations for calculations using a simple in-house program writtenIn Microsott Excel to determine the diversity index (D) and confidence intervals tocompare the discriminatory power of PFGE typing with that of Plasmid REA typing for:aVisually interpreted PFGE gels Vs. Plasmid REA gels and.Computationally interpreted PFGE gels Vs. Plasmid REA gels.

CHAPTER VlRESULTS1. Molecular typing of HLAR enterococciThe molecular typing of HLAR enterococci was performed by two methods namely.PFGE and plasmid REA by agarose gel electrophoresis. The same set of randomlyselected HLAR enterococci for both the molecular typing methods were used withoccasional exceptions in few instances (where few strains were included/excluded ineither typing method, albeit occasionally). The comparative results of both the typingmethods are depicted in Table 18 a and b, and elaborated in the following sections.A. PFGE typingThe chromosomal DNA restriction endonuclease digestion of HLAR enterococci wasperformed using an infrequently cutting enzyme Smul and the digested DNA plugs wereseparated by pulsed field gel electrophoresis. The 76 isolates of enterococci subjected topulsed field gel electrophoresis included, 66 randomly selected HLAR clinical isolatesapan from ten standard strains and transconjugants. The 66 clinical isolates included fiftyfour E. ,fuecali.s, six E. Jurcium, four E. gullinarum, one E. avium and one E. duransexhibiting resistance to both or either of the aminoglycosides (gentamicini streptomycin)separately. The SmaI digestion yielded approximately 12-20 DNA fragments of varioussizes ranging from 50-450 kb in size by PFGE as shown in Figure 13. The PFGEpatterns of the isolates were analyzed visually and computationally as per standardconsensus guidelines [294]i. Visual interpretation of PFGE gelsThe PFGE gels were visually analyzed and interpreted based on the consensus guidelinesof Tenover et el. [294], although with more stringency with no band differences betweenisolates as one PFGE type. Four out of the sixty-six clinical isolates (which included oneisolate each of E. ,faecalis, E. gallinarum, E, uvium and E. durans) were excluded fromvisual analysis and interpretation, since these isolates were either refractory for restrictiondigestion, or did not yield a satisfactory separation upon repeated testing.

CHAPTER VlrFigure 13. Representative gel images of Smal macrorestriction digested chromosomal DNAof HLAR Enterococci by pulsed field gel electrophoresis [PFGE].Corrllpondlng Southern blot hybrldlzed Smdl dig-tad chrmcaomal DNA gel Conasponding Southsrn blot hyhdnsdwlth ant(6'l.l [HLSR] @na pro& by PFGE wlth aac(6')+aphlZ") [HLGR] gene prcbA. Lab number of enterococci from lanes 1- 9 in respective order:An-1, 15,332, 9953. 1002, 5342, 9478, 14,550, 8670, 14,535.8. Lab number of enterococci from lanes 1- 9 in respective order:6130, 6265, 8765, 11,660, 11,871, 11,881, 11,869, 8257, 891,C. Lab number of enterococci from lanes 1- 9 in respective order:4343,3844, 5969,6275,10.638,7132,8811 2713 6276.M- Lambda ladder PFG markerd

CHAPTER V/The enterococci were classified into groups with respect to the restrictionendonuclease (Smal) digestion profiles of their genomic DNA as depicted in Table 18 aand b. The groups were designated alphabetically (in caps) for more than one isolateexhibiting the same1 indistinguishable Smal restriction profile. The 33 of 62 clinicalisolates along with two transconjugants studied were classified into nine groups (A-I),while the remaining 29 isolates exhibited unique SmaI restriction profiles that couldn't beclubbed with any groups and hence classified as "unique" restriction profiles. The 35isolates were classified as follows; 16 isolates in group-A, two subtypes of group-Awhich showed a nearly identical PFGE pattern as group A isolates, three isolates ingroup-B and two isolates each in group-C, D, E, F, G, H and I. The group F and Gincluded two isolates each of E. fuecium. and group H isolates included twotransconjugants, while the isolates in other groups were E.,faecalis.ii. Computational analysis of PFGE gelsThe visually interpreted PFGE gels were subjected hrther for computational analysisusing the Bionumerics Software 2.5, Applied Maths, Belgium. The PFGE gels werescanned and the cluster analysis was performed, by measuring the Dice coefficientsimilarity of the bands. The dendrogram was constructed using Unweighted Pair groupmethod using Arithmetic averages (UPGMA) method with different position (band)tolerance and optimization settings (0.5-3.0%). Four out of the sixty-six clinical isolates(which included one isolate each of E. /ireculis, E. gullinarum, E. ui~iurn and E, durun.~)were excluded from computational analysis and interpretation, since these isolates wereeither refractory for restriction digestion, or didn't yield a satisfactory separation uponrepeated testing. The enterococci were classified into groupslclusters based on ahomology of >85% with respect to the Smul digestion profiles of their genomic DNA asshown by the dendrogram in Figure 14 and as depicted in Table 18 a and b. Thegroupsiclusters were designated in Roman numerals for more than one isolate exhibiting'85% homology in their Smal restriction profile at different position (band) tolerancesettings. A 2.0% position tolerance and 1.0% optimization setting of the gel, which wasconsidered an optimal setting (standardized after trying different combinations of bandsettings), yielded a total of thirteen clusters (I-XIII).

CHAPTER 1'1Figure 14. Computational Cluster analysis* of PFGE gels of HLAR Enterococci usingDice coefficientand UPGMA method (Bionumerics, Applied Maths, Belgium)..'V* CI85% Homology IClumler - cut on -.-aE CC VC LNI. ,Ie **I Y .>.I.OCI l.'.Y S C rI *YI".,I,,:.O.!Ia The settings used for computational cluster analysis were 2.0% (band) Tolerance,1.0% Optimization and a Homology/Cluster cut off of 85%.CC, Computational Clusters; VC, Visual Clusters; LN, Lab Number of test strainsN.B: The phenotypic and genotypic details of the test strainsare depicted in Table 18.21 14

CHAPTER VlThe 47 of 62 clinical isolates analyzed were classified into 13 clusters (I-XIII),while the remaining 15 isolates exhibiting < 85% homology in their SmaI restriction~rofile were classified as "unique" profiles. The 47 (clustered) isolates were classified asfollows; 13 isolates in Cluster 1, five isolates in Cluster 11, four isolates each in Clusters111 and IV, three isolates each in Clusters V, VI and VI1, and two isolates each in ClustersVIII-XIII. There were one each E. ./aecalis and E. gallinarum in Cluster IX and XIII,while Cluster XI had one each of E. faecalis and E. ,faecium. Cluster XI1 had two isolatesof h ,faecium, while the isolates in other clusters were E. ,faecalis, and the "unique"isolates had all three species among them. The clusters were subsequently compared withthe visual interpretation results based on the consensus guidelines of Tenover et al. [294].B. Plasmid REA typingThe plasmid DNA restriction endonuclease (RE) digestion of HLAR enterococci wasperformed using EcoRl a restriction enzyme that digests the gentamicin resistanceplasmids but not the transposons flanking these plasmids if any, thereby enabling us totype the plasmids to assess their clonal relatedness. The 66 isolates of enterococcisubjected to plasmid restriction endonuclease analysis included, 57 randomly selectedHLAR clinical isolates apart from 9 standard strains and transconjugants. The 57 clinicalisolates included forty five E. ,faecalis, six E, faecium, four E, gallinarum, one E, aviumand one E. durans exhibiting resistance to both or either of the aminoglycosides(gentarnicid streptomycin) separately. The EcoRl digestion yielded approximately 5-15DNA fragments of various sizes ranging from 1.5-65 kb in size by agarose gelelectrophoresis (Figure 8- Chapter IV). The plasmid RE patterns of the isolates wereanalyzed visually and computationally as per standard guidelines.i Visual interpretation of plasmid RE gelsThe plasmid RE gels were visually analyzed and interpreted based on the molecularweight standards run along with each gel as described previously [120, 2861. Four out ofthe 57 clinical isolates (which included two E. faecalis and one isolate each of E.gallinarum and E, durans) were excluded from visual analysis and interpretation, sincethese isolates didn't yield plasmids upon repeated extractions or were either refractory for

CHAPTER VIrestriction digestion. The enterococci were classified into groups with respect to therestriction endonuclease (EcoRI) digestion profiles of their plasmid DNA as depicted inTable 18 a and b. The groups were designated alphabetically (in caps) for more than oneisolate exhibiting the same1 indistinguishable EcoRI restriction profile. 27 of 53 clinicalisolates studied were classified into seven groups (A-G), while the remaining 26 isolatesexhibited unique EcoRI restriction plasmid profile that couldn't be clubbed with anygroups and hence classified as "unique" restriction profiles. The 27 isolates wereclassified as follows; 13 isolates in group-A, four isolates in group-B and two isolateseach in group-C, D, E. F and G. The group F and G included two isolates each of E.farciurn, while the isolates in other groups were E. faecalis. The "unique" isolates hadfour different enterococcal species among them as depicted in Table 18 a and bii. Computational analysis of plasmid RE gebThe visually interpreted plasmid RE gels were subjected further for computationalanalysis using the Bionumerics Software 2.5, Applied Maths, Belgium. The plasmid REgels were scanned and the cluster analysis was performed, by measuring the Dicecoefficient similarity of the bands. The dendrogram was constructed using UnweightedPair group method using Arithmetic averages (UPGMA) method with different position(band) tolerance and optimization settings (0.5-3.0%). Four out of the 57 clinical isolates(which included two E. faeculis and one isolate each of E. gallinarum and E. duruns)were excluded from computational analysis and interpretation, since these isolates wereeither refractory for restriction digestion, or didn't yield a satisfactory separation uponrepeated testing. The enterococci were classified into groupslclusters based on ahomology of >80% with respect to the ECoRI digestion profiles of their genomic DNAas shown by the dendrogram in Figure 15 and as depicted in Table 18 a and b. Thegroupslclusters were designated in Roman numerals for more than one isolate exhibiting>80% homology in their ECoRl restriction profile at different position (band) tolerancesettings. A 2.0% position tolerance and 1.0% optimization setting of the gel, which wasconsidered an optimal setting, yielded a total of eleven clusters (I-XI).

CHAPTER VIFigure 15. Computational Cluster analysis' of Plasmid RE gels of HLAR Enterococci usingDice coefficientand UPGMA method (Bionumerics, Applied laths, Belgium).The settings used for computational cluster analysis were 2.0% (band) Tolerance,1.0% Optimization and a HomologylCluster cut off of 80%.CC, Computational Clusters; VC, Visual Clusters; LN, Lab Number of test strainsN.B: The phenotypic and genotypic details of the test strainsare depicted in Table 18.

D. Simpson's index of diversity and Confidence intervalThe Simpson's index of diversity as shown in l'able 19, dcplcts that hoth typitiy,methods Plasmid REA and PFGE were haviny high and equal discriminatory power l'hevisually interpreted results showed a marginally lesser divcrsity index (D) of' 0 93 arid0 91 for Plasmid REA and PFGE respectively when compared with thc comli~rtationall~interpreted resulls. whicli sllowed a diversitv index (I)) oSO 07 and 0 01 fhr Plil>r~lid RI'Aand PFGE respect~vely Since the 95O6 confidence tntervals (('I I wc1.e oic~ lal~l~~rrs aidepicted in the bar graph IGraph 11, the diSferericcs were riot s~atist~ci~llv si~n~licariiTable 19 Statistical analys~s-Simpson's index of diversity and ('otifiderice interval[)I. dt\crs~c! !ndc\. ( I.co~~fidctrcc ~rtlcn:~l\SE. S1:111d;ird urror. "Colnp. COIII~II~:I~IOII,IIGraph I. Kepresentation of Simpson's indcx ol'di\ersii\, (L)) for I'la??I- .-Plamvl V tsd PFGE VIMI~I Plamtd Camp PFGF Camp

CHAPTER VIM~crob~al ~dentrficat~onclln~cal outcome of any dlsease.undoubtedly plays a major role ~n determ~n~ng theThe results of the microbial typing along with other supporting Clinico-epidemiologicaldetails helps in initiating and executing appropriate infection control measures formultidrug resistant nosocomial pathogens like enterococci in any health care setting at theappropriate time contributing to a substantial decrease in morbidity and mortality.Although various species of enterococci were isolated at regular intervalsthroughout our study period as depicted in previous chapters. we could find clustering ofparticular species during various time periods from specific unitsiwards. E. ,fueculi.s wasthe commonest and predominant of all species and was isolated consistently throughoutour study period, but we were unable to cluster them merely by their antibiotype. Hencemolecular typing techniques like plasmid DNA analysis and chromosomal DNA analysisby PFGE were performed to determine the clonality of these isolates. The profiles ofplasmid REA and PFGE gels were analyzed both, visually as per consensus guidelines[294] and using the Bionumerics software in our study. However for practical purposes,the visual interpretation of plasmid REA and PFGE gel profiles (based on the consensuswidelines of Tenover et a].) was uncomplicated as well concordant for most isolates,when compared to the computational analysis of the gel images.

CHAPTER VISeveral reasons could be quoted for the higher significance of the visualinterpretation of the gels especially those of PFGE. A major reason is the non-availabilityof costly software (Bionumerics) for analysis of gels in most of the health-care settings.The visual interpretation can normalize the gels authentically rather then arithmetically(as done by software based on the programmed settings), thereby helping to distinguishvery minor differences that go undetected due to the (band) tolerance senings chosen bycomputer software. This was shown in our study during the standardization of the optimal(band) tolerance settings using our PFGE and Plasmid REA gel images. The optimalsettlngs of 2.0% (band) tolerance and 1 .O% optimization, and a homology cut off of 80%and 85% (for Plasmid REA and PFGE dendrograms respectively) followed in our study,was able to give clusters which helped us to discriminateldistinguish the HLARenterococci isolated in our study. But there were instances when a decreaselincrease inthe settings (between 0.5-3.0%) yielded concordance in clustering for some isolates, butresulted in a diminishedldiscordant clustering (as compared with the clustering by visual~nterpretation) for the remaining isolates. Thus an optimal setting of 2.0% band toleranceand 1.0% optimization and a homology cut off of 80% or 85% was followed. whichcould take care of justifiable, if not stringent clustering of the isolates by computalionalanalysis matching the consensus guidelines of categorizing isolates as indistinguishable,closely related, possibly related, or different. based on the differences in number of bands[294]. The stringent visual clustering we followed based on the consensus guidelinescould not cluster 29 isolates, hence classified them as "unique", while the computationalanalysis yielded only 15 non-clusterable "unique" isolates which had a homology level of>XO% for all the isolates in the cluster, and non-clusterable "unique" isolates wereexhibiting 60-85% of homology. When we decreased the homology cut off to 75% andincreased the band tolerance settings up to 3.0%, the number of clusters tbrmed wasreduced thereby grouping more number of isolates with one to six band differences intoone or two clusters. The densitometric curve based Pearson product moment correlationwhen used for constructing dendrograms using the optimized tolerance settings yieldedhighly discordant clustering, since the curves obtained from different PFGE gels were nothomologous due to differences in signal strength, and background noise (signals), which

CHAPTER VIwere visible as non-specific peaks even though they exhibited same banding patternsvisually.The major reason for the discordance of the typing results by either method usingcomputational analysis was due to minor inevitable differences in the separation patternsofthe gel (due to minor variationsidifferences in running conditions of the PFGE) whichcould be normalized only collectively (for all gels grouped together for analysis) and not~ndividually according to the gel panern, on the other hand visual normalization was doneas per individual gel panems which gave authentic clustering of the HLAR enterococci.Thus for funher comparisons and discussions we used the visual interpretation results ofmolecular typing as recommended and followed by several studies [I 10, 279-282, 286,291, 293, 2941, however some studies have used computational analysis alone, or inconjunction with visual interpretations for studying the clonality or epidemiology ofenterococci [282, 283, 292, 293, 3641. Even though some studies have depicted somestandard protocols for PFGE typing of enterococci [279, 2941 there is a lack of aharmonized protocol that could be followed worldwide, which would help in minimizingthe errors in the performance and interpretation of PFGE experiments.Overall, our molecular tping study depicted (Table 18 a and b) that the strainsbelonging to various groupsiclusters were isolated from Individual patients in differentwards. The wards along with their corresponding lCUs were located on different floors ofthe same building of the hospital block in JIPMER. The medicine wards (with their ICU)were located on the third floor. the surgical wards on the second floor, pediatrics wardson the first floor and the gynecology wards in the ground floor. The same nurses work ona single ward. Interestingly, ten of the fifteen isolates of h gullinarum were frompediatrics unit, while seven of the ten isolates exhibiting a similar antibiotype were fromthe same ward within a span of two months. The remaining three of the ten E. gullinarumwere isolated from the same ward in the preceding three months, one of which showed anantibiotype similar ro the cluster of seven isolates. The same was the case of three E.CassellfZavus isolated from the same pediatrics unit within a span of two months in thePreceding year. Most of these (eight of ten E. gallinorurn, and all three E. casseliflovus)

CHAPTER VIisolates were from cases of septicemia, and the clinical and microbiological testingresults of these isolates depict the nosocomial spread of these species. Although unusualenterococcal species were prevalent in our hospital, it was the emergence of HLARamong E. faecalis that was of serious concern and needed special focus. Hence moleculartyping of randomly selected HLAR enterococci were performed by PFGE and PlasmidREA for epidemiological investigations, and to track the dissemination and evolution ofmulti-drug resistant strains more efficiently in our hospital setting.The plasmid REA typing was performed in conjunction with the PFGE ~qping ofchromosomal DNA for many reasons. Firstly, the molecular characterization of HLARenterococci showed that gentamicin resistance was encoded by plasmids and theseplasmids were capable of conjugation, hence there lie every possibility of plasmidtransferidissemination leading to plasmid epidemicloutbreak in any hospital setup.Secondly, some studies have depicted that plasmids can influence the outcome of PFGEtyping, since plasmids if possessing the same restriction site of the enzyme used to digestthe genomic DNA for PFGE, then it can appear as a same or additional band in thechromosomal DNA pattern leading to erroneous interpretation of PFGE results [280,2931. The DNA hybridization studies performed with the PFGE gels using Gm and SmDNA probes showed that none of the bands separated encoded GdSm resistance, whileplasmid DNA from all the isolates encoded GdSm resistance as evident by DNAhybridization. Thus it was authenticated that the bands resolved by PFGE wereexclusively of chromosomal origin and none were of plasmid origin, thereby ruling outthat plasmids did not influence the results of PFGE typing in our study (Figure 13).Moreover, plasmid fingerprinting is most useful for epidemiological studies that arelimited both temporally and geographically.Although chromosomal DNA analysis by PFGE is considered a gold-standardtechnique for molecular epidemiological studies since last decade, the plasmid DNAfingerprinting was the first molecular method used as a bacterial typing tool. Zervos et al.[I 81 authenticated nosocomial transmission and exogenous acquisition of S. faecalis forthe first time using plasmid DNA content as an epidemiological marker. Since then,

CHAPTER VIseveral studies applied plasmid DNA fingerprinting for molecular epidemiology ofnosocomial enterococci [47, 284, 285l:The evaluation of whole plasmid DNA althoughshown to be helpful in plasmid epidemics [286] have some limitations, since theundigested profiles often show multiple forms of plasmids, with circular and linear formsof plasmids migrating at different rates than the covalently closed circular form, whichleads to erroneous interpretation of plasmid DNA fingerprinting results. Hence,restriction digestion of plasmid DNA is recommended and being followed for molecularepidemiological studies of drug resistant enterococci [47, 125, 132, 135. 284. 285, 3731.The visual interpretations of the Plasmid REA and PFGE gels based on thesimilarity of clusterslgroups obtained by matching the consensus guidelines of Tenover etal. showed concordance between both typing methods, barring few exceptions, which hadsome clinico-epidemiological significance when analyzed. Most of the clusters obtainedwere concordant for both the typing methods in our study, since isolates grouped inClusters A-F by either typing method exhibited highly homologous patterns resulting inidentical clusters, which depicts a possible "intrahospital strain dissemination" as shownby other studies [280, 2861, while the isolates exhibiting "unique" REA profile were alsopossessing "unique" PFGE profile. Two HLGR E. ,fueculi.\ isolates that exhibited anidentical plasmid REA pattern and grouped as cluster B showed "unique" heterologousPFGE patterns of their chromosomal DNA that were different from other isolates inPFGE group B cluster. The discordant plasmidREA typing results were ofepidemiological significance, since intrahospital and interhospital (HLGR) plasmiddissemination has been depicted by few studies [286, 3541. The versatile geneticmachinery of enterococci facilitates the transmissionidissemination of the conjugativeplasmids between closely related strains [17, 1381. Hence, the same plasmid type (groupB) present among two E. faecalis isolates with different chromosomal restrictionendonuclease pattern (unique) as evident by PFGE, depicts that the intrahospitaldissemination of (HLGR) plasmids from group B isolates to closely related strains (whichwould have been otherwise sensitive to aminoglycosides) could have occurred as shownby studies from U.S. and Japan [135,286]. Our results were further authenticated becauseboth the discordant isolates were pheromone responsive, and were able to transfer the

CHAPTER VIHLGR determinants in vitro at a high frequency. Thus our Plasmid REA typing resultsdepict that antimicrobial resistant determinants encoded by plasmids can be disseminatedseparately (intrahospital plasmid dissemination) leading to plasmid outbreakiepidemic,apart from the commonly occurring intrahospital strain dissemination.The Molecular typing results were instrumental in deriving the much neededclinical and epidemiological significance of the HLAR isolates. The PFGE typingshowed that approximately 50% (33) of the (62) HLAR enterococcal isolates werehomogeneous and formed clusterslgroups (A-I), while the remaining 50% of the isolateswere non-clusterable depicting heterogeneityldiversity among the HLAR isolates. ThePFGE genotypic Cluster-A isolates were found to be "endemic" in our hospital duringour study period, since they were present in different medical and surgical wards situatedin 1, I1 and Ill floors for more than a year from September 2001 to December 2002 asshown in Table-18 a and b. Incidentally, three consecutive E. faecalis isolates whichwere genotypically similar and grouped under Cluster-A were from a single patient withkver after Aortic valve replacement. Thus, our results authenticate that in majority of theinfective endocarditis cases when consecutive blood culture samples yield the sameorganism with a similar antibiotype it could be considered as empiric evidence of theetiogen as shown by several studies [162-1641. Interestingly, two E, faecalis isolatesshowed a "closely related" PFGE pattern of cluster-A isolates with 2-3 band differencesand classified as Al and A2, since we followed a stringent definition for strain clustering(2941. As depicted in the consensus guidelines, these A1 and A2 E, faecalis isolateswould have originated from cluster-A isolates after a possible genetic event, i.e., a pointmutation or an insertionldeletion of DNA, since these two strains were isolated fromsame wards during the same time period when cluster-A strains were prevalent [294].Our results depict that enterococcus has high possibilities for genomic rearrangements,which would be evident through molecular typing techniques like PFGE. This evidencewould be highly helpful to trace the outbreak related isolates, like those of Al and A2 asshown in the present study. Furthermore, the endemicity of Cluster-A isolatesdocumented by PFGE typing would be remarkably useful for complementing the clinicaland epidemiological analysis of HLAR enterococcal isolates or any nosocomial

CHAPTER VIThree E. .faecalis PFGE Cluster-B strains were isolated from pediatrics ward andthe neonatal ICU (NICU) within a span of three months during 2001, but not isolatedthere after. This suggests that routine sanitation measures (without the knowledge of there valence of E. ,fheculis isolates) practiced by the particular ward was sufficient enoughto prevent (or eradicate) the dissemination of antimicrobial resistant nosocomialpathogens. The E. faecalis strains from clusters C, D, E and I were isolated from the sameor different wards during 2002, suggesting the circulationidissemination of the HLAR E.frrvcalis in our hospital. The HLAR E, faecium isolates from clusters F and G wereisolated within a short span during 2001 and 2002 respectively, which were alsopossessing "esp" gene. The prevalence of the strains with same pulsotqpe all through our~tudy period indicates widespread dissemination of these HLAR enterococci in ourhospital setup (intrahospital strain dissemination) as shown by several studies [161, 220,280,2861.The predominance of strains from cluster A, along with strains from other clusters(B. C. D, E) in particular ward(s) during specific tlme periods suggest that these isolateswere derived from a common source and spread from patient to patient, however we donot know the method of transient carriage in the nosocomial transmission of thecommonly prevalent aminoglycoside resistant enterococci, since the reservoir and modeof transmission of the HLAR enterococci were not determined in our study. However,several researchers have studied the epidemiology of nosocomial environmentalreservoirs of multidrug resistant enterococci. Enterococci have been shown to be capableof prolonged survival on hands, gloves, thermometers, blood pressure handcuffs, IV fluidpumps, bedrails and linen and various hospital environmental surfaces [65, 70, 73, 741.Further, nosocomial enterococci were shown to be resistant to heat (upto 80'~for 1 min),and could withstand routine disinfection procedures (1SOppm chlorine) followed forinfected linen, which underscores the significance of enterococci to survive anddisseminate in the hospital environment [71]. Most of the studies have shownconcordance between hospital environmental strains and the patient isolates oftenresistant to vancomycin or high-level gentamicin, which were confirmed using molecularepidemiological tools like pFGE andlor Plasmid typing [54, 75-80]. Thus as suggested in

CHAPTER VIthese studies any of the above mentioned source(s) could have been the reason for"intrahospital dissemination" of enterococci in our hospital.The 50% (29) non-clusterable isolates with "unique" PFGE patterns were isolatedfrom various wards (including those wards where the predominant clusters of enterococciwere found) throughout our two-year study period. Although few of these "unique"strains were "possibly related with the isolates from different clusters based on theconsensus guidelines 12941, the genomic heterogeneity exhibited by majority of the nonclusterableisolates depicts the diversity of HLAR enterococci as shown by severalstudies from U.S, Netherlands, Noway, Greece, U.K, [52, 282, 283, 292, 3631. Thediversity (50%) of the isolates as shown by PFGE typing in our study suggests that, apartfrom patient-to-patient spread that was equally (50%) a major cause for dissemination ofclusterable (homogenous) HLAR isolates, other possible sources can be due to colonizerisolates or fecal contamination, or they could be community acquired isolates. Severalepidemiological studies conducted in human subjects from community have yieldedenterococci resistant to various antimicrobials like ampicillin, gentamicin andvancomycin [81-831. Thus screening the inpatient population for fecal carriage ofantimicrobial resistant enterococci, and conducting point-prevalence studies would behighly significant and necessary in the wake of HLAR and emergence of vancomycinresistance among enterococci. This would help in initiating appropriate infection controlmeasures and restructuring the hospital antibiotic policy, if needed.The discriminatory power of a typing method is its ability to distinguish betweenunrelated strains. As discussed above, it is determined by the number of types defined bythe test method, and the relative frequencies of these types. These two facets ofdiscrimination are not generally presented as a single numerical value and thereforecannot be used for a straightforward comparison of different methods. Hence, wefollowed a method previously described by Hunter and Gaston [371] to give a singlenumerical index of discrimination of the typing method. An approximate 95% confidenceinterval was calculated as proposed by Grundmam et a]. [372]. Our statistical resultsdepicted (Table 19) that both Plasmid and PFGE typing methods, as well visual and

CHAPTER VIcomputational analysis of these results had an equally high discriminatory index. Hence,the choice of method depends on the user's need and the resources they are providedwith. Our statistical typing results were highly concordant with another recent study,which used AFLP and PFGE for typing ampicillin resistant E. fuccium [283].Studies have shown that epidemic drug resistant enterococci may possess specificgenetic characteristics (encoding virulence determinants) resulting in a distinct lineage,which could facilitate enhanced colonization/infection of the host [193]. The permeationof these virulence genetic characteristics into different species differs according to thesetup, patient demographics and other extrinsic factors. The "esp" gene and associatedvirulence factors (Table 14- Chapter IV) have permeated deeply into the E. faecalis,since 65% of these isolates exhibited the presence of this putative virulence factor. OurPFGE typing results, which depicts the clonality of the HLAR isolates, reveals thatmajority (except one) of the clusterable isolates (A-1) possessed "esp" gene (includingtwo E. ,fuecium clusters), while they were absent among 41% of the uniquelnonclusterableisolates. The clustering of' these "esp" positive isolates proves that strains ofthis chromosomal lineage are related, and may have been derived from a commonancestral strain as shown previously 12161. However. the presence of the "esp" geneamong 60% of uniqueinon-clusterable E. fuecu1i.s isolates also depicts that the pathogenicpotential due to the presence of "esp" gene need not be confined to a single geneticlineage as shown by a recent study [220], since closely related lineageslgenotypes can beequally virulent resulting in epidemicity.Our findings suggest a strong association between the presence of "esp" gene. andhigh-level gentamicin resistance, although both have been shown to reside on entirelydifferent genetic determinants (chromosomal and extra-chromosomal [plasmid]respectively). The prevalence of this combination is of high clinical significance in anyhealth-care setup, since sevelal studies have shown that specific genetic lineages (asshown by PFGE) exhibited the presence of "esp" with vancomycin and/or ampicillinresistance among clinical E. ,fuecalis and E. faecium isolates 1207, 2 19, 220, 3641. Thesestudies suggests that prior treatmentiexposure to the antibiotics like vancomycin andior

CHAPTER VIampicillin would have selected these clones facilitating to reach ecological abundance inthe nosocomial habitat due to the presence of "esp" gene, although studies are yet toanalyze the significance of the combination of HLGR and "esp" gene. Thus, our findingssuggest that the higher prevalence of gentamicin resistance in our hospital would havecontributed to selection of those clonesilineages with "esp" gene resulting in"intrahospital dissemination". Since, majority of these clones possess conjugativegentamicin resistance plasmids and the "esp" gene was transferable in vitro, thepossibilities for "interhospital strain dissemination" remains very high. Thus appropriatemeasures to contain the antimicrobial resistance in any health care setup would by allmeans decrease the probability of selection and dissemination the "esp" positive clones,which poses a stiff challenge ahead.Although we performed PFGE in a single stretch using the isolates collectedduring our study period, our molecular typing results depict that concomitantperformance of molecular typing (by PFGE and/or Plasmid REA typing) during anysuspected outbreaklincrease in the prevalence of nosocomial pathogens would be highlyhelpful in tracking and preventing the dissemination of multi-drug resistant strainsiclones, or plasmid determinants (encoding resistance) more eficiently at a given point oftime in a hospital setup. Some studies have shown that multicenter PFGE studies with aharmonized protocol and centralized server for interpretation can address epidemiologicalquestions effectively [296]. Thus a cooperative venture for molecular typing ofenterococci would provide a rapid tracking system to assist hospitals, clinics and chroniccare facilities in controlling the spread of multidrug-resistant enterococci locally,nationally as well globally.SUMMARYOur molecular tqping results depict that PFGE in conjunction with Plasmid REA typingwould he highly helpiirl in tracking the intrahospital strain dissemination, as welldissemination of genetic determinants. The PFGE typing of the chromosomal DNA ofHLAR enterococci yielded approximately 12-20 DNA fragments of various sizes ranging

CHAPTER Vlfrom 50-450 kb in size. The PFGE patterns of the isolates were analyzed visually andcomputationally as per standard guidetines. The visual analysis classified 33 of 62clinical isolates along with 2 transconjugants into 9 groups (A-I) based on the PFGEpattern, while the remaining 29 isolates were classified as "unique" based on theirrestriction profiles, since their profiles could not be clubbed with any groups. Thecomputational analysis of PFGE gels depicted several clusters of isolates through thedendrogram constructed by the dice-coefficient and UPGMA method. They were partlyconcordant with the clusters formed by visual interpretation, although differencessurfaced between results of clustering with either interpretation methods.The plasmid REA typing of HLAR enterococci after EcoRl digestion yieldedapproximately 5-15 DNA fragments of various sizes ranging from 1.5-65 kb in size. Theplasmid RE patterns of the isolates were analyzed visually and computationally as perstandard guidelines. The visual analysis classified 27 of53 clinical isolates studied into 7groups (A-G) based on the plasmid REA pattern, while the remaining 26 isolates wereclassified as "unique" based on their plasmid restriction profiles, since thcir profilescould not be clubbed with any groups. The computational analysis of plasmid RE gelsdepicted several clusters of isolates through the dendrogram constructed by the dicecoefficientand UPGMA method. They were partly concordant with the clusters formedby visual interpretation, although differences surfaced between results of clustering witheither interpretation methods. The Simpson's index of diversity depicted that both typingmethods, Plasmid REA and PFGE used in our study were having a high and equaldiscriminatory power, while the interpretation of the molecular typing results, visually aswell computationally showed approximately the same diversity index and the differenceswere not statistically significant. Thus the choice of interpreting molecular typing resultsdepends on the users need and objective, although we prefer visual analysis of the gelsfor lesser sample size. Finally, the concordance between PFGE typing and Plasmid REAdepict that a combination of both the typing methods would help in better discriminationof HLAR enterococci, when we speculate an involvement of both chromosomal andextra-chromosomal elements in antimicrobial resistance

Summary

SUMMARYThe results of our study conducted between July 2001 to June 2003 in JawaharlalInstitute of Postgraduate Medical Education and Research (JIPMER), a 900-beddedtertiary care hospital at Pondicherry, South lndia are summarized as follows:Enterococci in Clinical infectionsThe clinical prevalence of enterococci in our health care setup was highly significant andalarming. Among the 242 enterococci isolated, E. faecalis (71%) and E. fuecitrm (10%)contributed to 8 1 %, while remaining 19% of enterococci comprised seven differentunusual species, which included E. gallinarum, E, avittm, E. raffinosu.~. E. hirac, E.rnundtii, E. casselij7al~us and E. duruns. The distribution by site of isolation for the 242enterococci predominantly included l I1 isolates (46%) from bloodstream. 72 (30%) fromurinary tract, and 59 (24?,,) from exudate specimens, while 75% of the all the isolateswere from inpatient specimens. The infections were polymicrobial only in 46 (19%) ofthe 242 cases from which enterococci were isolated. The conventional biochemicalphenotyping tests identified and speciated majority of enterococcal species, while themolecular phenotyping using WCP fingerprinting by SDS-PAGE validated theauthenticity of the unusual species. and the exact taxonomic status of the atypicalphenotypic variant stralns and strains that showed weak saccharolytic reactionsbiochemically.Antimicrobial resistance in enterococciThe antimicrobial susceptibility testing by disc-diffusion method showed that all theisolates tested were susceptible to teicoplanin and linezolid, and 92% were susceptible tovancomycin. While 58% and 69% isolates were susceptible to penicillin and ampicillinrespectively, only 42% and 54% of the isolates were susceptible to the aminoglycosidesgentamicin (high-level) and streptomycin (high-level) respectively. Minimalsusceptibility against ciprofloxacin was exhibited by 38% of all enterococci, while the

SUMMARYurinary isolates tested for nitrofurantoin and ciprofloxacin showed 78% and 32%susceptibility respectively. Of clinical. significance was the high-level gentamicinresistance exhibited by 58%. 60% and 43% of E. fueculis. E. fuecium and unusualenterococcal species respectively. However, there were differences in the results betweendisc diffusion testing and agar screeninglagar dilution method while testing enterococcifor vancomycin resistance. None of the penicillin and ampicillin resistant enterococcalisolates tested for beta-lactamase production using a nitrocefin disc yielded a positiveresult.The genotypic detection of aminoglycoside resistance genes by Multiplex PCR,showed that the bifunctional gentamicin resistance gene aac(6')+aph(2") was present in06% of HLGR E. fuecu1i.s isolates, while ant(6')-1 gene (streptomycin resistant) wasdetected in 94% of HLSR E. ,fueculis (including many HLGR ~solates). Theaac(6')+aph(2") and the ant(6')-1 gene were present in 87% and 89% of E. fucciurnlsolates respectively. Among unusual species of enterococci tested for aminoglycosideresistant genotypes, only eight isolates (two E. ~ullit~arurn and six E, ui'irrrn) possessedaac(6')+aph(2") gene, while four isolates (two E. gullinurtin~ and two E. uiiurn)possessed the ant(6')-I gene, which also possessed aac(6')+aph(2") gene in them. Theabsence of these genes in some suggests alternate aminoglycoside resistance mechanismsamong the enterococcal species exhibiting HLAR.Molecular Characterization of High-level arninoglycoside resistant enterococciThe plasmid DNA was present among most of the HLAR E. jueculis isolates (53 of 60isolates tested), which yielded between one to five plasmids, while majority isolatespossessed at least two plasmids. The whole plasmid profiles depicted that the molecularweight of the plasmids ranged approximately ?70 kb to 2 kb. The restriction-digestedplasmids were classified into seven groups (groups A-G) comprising 27 isolates, while 26isolates that exhibited unique EcoRl restriction plasmid profile could not be clubbed withany groups and hence classified as "unique" restriction profiles.

SUMMARYThe genetic elements conferring HLAR were located (localized) by DNA-DNAHybridization studies using aac(6')+aph(2")probe and ant(6')-1 gene probes. Therestriction-digested plasmid DNA from HLAR E. faecalis isolates was southerntransferred and hybridized with the respective DNA probes. The plasmids classified intodifferent groups with respect to the EcoRl profiles exhibited an identical DNAhybridization pattern for the respective gene probe with occasional variations within thegroup. Gentamicin and streptomycin gene probes hybridized to EcoRI fragments of thedifferent sizes ranging from 5 to 70-kb for the same isolate. The sizes of hybridizingfragments were approximate measurements derived from the molecular weight standardsrun with every gel after hybridizing with the lambda DNA probe. The Smal digestedchromosomal DNA of the HLAR enterococci separated by PFGE did not show anykagments hybridizing with either of the DNA probes tested confirming that none of theHLAR enterococci carried the gentamicin~streptomycin resistance determinants on theirchromosome among the isolates from our hospital setup.C'irulence factors in enterococciThe presence of various virulence factors was depicted among enterococci isolated in ourstudy. Bacteriocin production was found in 42% of the E./ucculis isolates and wasconfined mostly to this single species with the exception of one C:,fuccium isolate. The E.,lueculis bacteriocin showed a narrow spectrum of activity, actlve only against two of thespecies-specific indicators and not against non-species specific indicator strains tested.The E. faecium bacteriocin showed a broad spectrum of activity, against all genusspecificindicator strains used in the study, but not against non-genus specific indicatorstrain tested. Hemolysin production was depicted among 14% E. faecalis isolates and100% E. durans isolates, while none other species depicted production of hemolysin.Gelatinase production was detected among 58% of E, faecalis isolates, while none otherspecies produced gelatinase. Most of the bacteriocin and gelatinase producing E, fuecalisconcomitantly exhibited HLAR.

SUMMARYThe "Esp" gene was detected among 50% of all isolates of enterococci, whichincluded 65% E.fueculis and 40% E.,fuecium isolates, while none other species depictedthe presence of "esp" gene. The bihnctional gentamicin resistance gene-aac(6')+aph(2")and the streptomycin resistant gene-ant(6')-l were present together in 61 '10 and 80%"esp" positive isolates of E. faecalis and E. faecium respectively. Biofilm formation wasfound among 26% of E. fueculis isolates, while none other species produced biofilm.Thus various virulence traits depicted had permeated E. faecalis extensively in ourclinical setup. In E. fueculis the highest prevalence of virulence factors were exhibited bythe urinary isolates (46%) followed by bloodstream isolates (38%), while in E. faeciumbloodstream isolates exhibited the highest prevalence (50%) of virulence factor, followedby urinary isolates.Cell-Cell communication & Gene transfer among aminoglycoside resistant EnterococciThe enterococci isolated from our setup were highly capable of transferring geneticdeterminants as evident by conjugation assays. Pheromone responsiveness (for cell-cellcommunication) was depicted by 62% of the all HLGR E. ,/urca/is isolates and 20% ofHLG sensitive E. ,facculi.~ isolates tested as evident by a clumping response. Four groupsof 30 randomly selected HLGR E. ,furculis isolates subjected to in-vitro gene transferassay showed that the gentamicin resistance marker transferred at a frequencies rangingbetween 10" lo 10.'transconjugants per donor cell for 26 donors. The transfer of "esp"gene-a putative virulence factor was demonstrated among transconjugants obtained from2 of the 30 donor E. ,fuecalis as confirmed by PCR. However in both instances, only fewamong the population of transconjugants showed the transfer of "esp" gene as evident byPCR, while all the transconjugants tested possessed the aminoglycoside resistance genes.The PFGE patterns of the donor strains were heterologous, when compared with FA2-2recipient and esp-negative, aminoglycoside resistant gene positive transconjugants, whichhad an identical pattern. On contrary "esp" and aminoglycoside resistance gene positivetransconjugants showed a "closely related" pattern of the recipient with two banddifferences, leaving us to speculate that chromosome-to-chromosome transmission of"esp" gene might have occurred.

SUMMARYMolecular typing of High-level aminoglycoside resistant enterococciThe PFGE typing of the chromosomal DNA of HLAR enterococci yielded approximately12-20 DNA fragments of various sizes ranging from 50-450 KB in size. The PFGEpatterns were analyzed visually and computationally as per standard guidelines. Thevisual analysis classified 33 of 62 clinical isolates along with 2 transconjugants into ninegroups (A-I) based on the PFGE pattern, while the remaining 29 isolates were classifiedas "unique" based on their restriction profiles, since their profiles could not be clubbedwith any groups. The dendrogram constructed by computational analysis of PFGE gelsdepicted several clusters of isolates, which were partly concordant with the clustersformed by visual interpretation. However differences surfaced between results ofclustering with either interpretation methods.The plasmid REA typing of HLAR enterococci after &oRI digestion yieldedapproximately 5-15 DNA fragments of various sizes ranging from 1.5-65 kb in size, andthe plasmid RE patterns were analyzed visually and computationally as per standardguidelines. The visual analysis classified 27 of 53 clinical isolates studied into 7 groups(A-G) based on the plasmid REA pattern, while the remaining 26 isolates were classifiedas "unique" based on their plasmid restriction profiles, since their profiles could not beclubbed with any groups. The dendrogram constructed by computational analysis ofplasmid RE gels depicted several clusters of isolates, which were partly concordant withthe clusters formed by visual interpretation, although differences surfaced between resultsof clustering with either interpretation methods. The Simpson's index of diversitydepicted that both typing methods: Plasmid REA and PFGE used in our study werehaving a high and equal discriminatory power. The visual and computationalinterpretation of the molecular typing results showed approximately the same diversityindex and the differences were not statistically significant, although our experience andresults depict that visual interpretation can be preferred for lesser sample size. Finally, acombination of both typing methods: PFGE and Plasmid REA typing would help inbetter discrimination of HLAR enterococci, when we speculate an involvement of bothchromosomal and extra-chromosomal elements in antimicrobial resistance.

CONCLUSIONSThe purpose (objective) of our work was to determine the prevalence of enterococci, andto characterize the antimicrobial resistance and virulence among these isolates to studytheir clinical significance in our hospital setting. Based on the observations made fromour study several conclusions could be drawn.>. Although E. ,faecalis and E. .fueciurn were the two predominant species there wasan emergence in the prevalence (19%) of other (unusual) enterococcal species inour setup. The alarming increase in the incidence of different species ofenterococci with the properties of intrinsic resistance to several antibioticsincluding beta-lactams and glycopeptides underscores the "importance of speciesidentification of enterococci". which helps in initiating appropriate antimicrobialtherapy based on the species isolated, as well tbr epidemiological sun,eillance ofthese bacteria in any health care facility. Since, the phenotypic differences amongenterococci may misrepresent the phylogenetic relationships complicating theclassification of enterococcal species, as well genus at times. every microbiologylaboratory should indulge in applying the complete range of tests necessary viz.phenotypiclgenotypic, to identify and discriminate enterococcal species.2. The maximal susceptibility (92%) to vancomycin, and complete sensitivity toteicoplanin and linezolid by all enterococci in our study shows that vancomycinresistance is in a nascent stage and yet to pose a serious threat in our setup, as wellin developing countries like India due to restricted usage. Nevertheless, ourresults serve an indicator of the healthcare catastrophes that vancomycinresistance could cause in the near future. The higher rates of resistance to highlevelgentamicin and streptomycin (60% and 43% respectively), and to penicillinand ampicillin (43% and 31% of respectively) by enterococci in our study depictsthe therapeutic challenges in treatment of serious enterococcal infections exertingaminoglycoside resistance, since a synergistic combination regimen is impossibleeven if the isolate is susceptible to either of these agents. The excessive use of

CONCLUSIONSgentamicin in our setup as standard prophylacticltherapeutic regimens may be areason for a higher prevalence of gentamicin resistance among enterococci.Differences may arise when testing enterococci for vancomycin andaminoglycoside resistance by different methods as shown in our study. Hence, "itis quintessential to adapt phenotypic methods like agar screeningiagar dilution forroutine testing of vancomycin and aminoglycoside resistance" to detect themearly to initiate appropriate infection control measures, as well restructuring theantibiotic policy if needed. The genetic analysis of aminoglycoside resistance by amultiplex colony PCR, depicted that approximately 95% of enterococci possessedthe aac(6]+uph(2'j and ur1t(67-I genes encoding high-level yentamicin andstreptomycin resistance respectively. The colony PCR protocol can be adapted forroutine diagnostics by clinical microbiology laboratories to give a rapidantimicrobial susceptibility result within six hours of identification of enterococciat times of need.i The molecular characterization of the genetic determinants encodingaminoglycoside resistance helped in revealing the differences in the epidemiologyof HLAR genes. The EcoRl restriction plasmid profiles and the hybridizationpatterns of the aminoglycoside resistant enterococci especially those of E.,fueculisfrom our hospital setting in South India showed heterogeneity among plasmids,although some plasmids showed homogeneity among the isolates studied. Thismay be due to dissemination of the plasmid determinants, or the plasmidpossessing strain within our hospital. Although the bifunctional auc(6 )+aph/2gene confemng the HLGR phenotype appears to be conserved, there may besubstantial differences in the flanking regions immediately adjacent to the fusedgene, which can be another cause for the diversity in plasmids conferring theHLAR as shown in our study. Thus appropriate usage of antibiotics, andimplementation of stringent infection control measures would prevent or reducethe rate of transferidissemination of antimicrobial resistant enterococci.

CONCLUSIONS> Various virulence factors studied like "esp" gene, bacteriocin/hemolysin,gelatinase and biofilm formation had permeated E. jbecalis extensively in ourclinical setup although at varying degrees. While gelatinase production andbiofilm formation were found exclusively among E. fa~rali.~, other virulencefactors like "esp" gene and bacteriocinihemolysin production were also foundamong other species like E. ,fiecium and E. durans (apart from E. faeculis)although at a lesser incidence. Although aminoglycoside resistance wasconcomitantly present among virulent enterococci, several (antibiotic) sensitivestrains possessed one or more of these virulence factors. Thus "detection andcharacterization of enterococci for any of these virulence determinants bymicrobiology laboratories would be highly helpful in determining the outcome, aswell to understand the patho-biology of enterococcal infections".L The transferability of the genetic determinants encoding antimicrobial resistance(HLAR) and virulence (esp gene) in enterococci by various gene transfermechanisms facilitates the dissemination of these determinants to other strains ina hospital setting, which may be otherwise lacking these traits giving rise to newergenetic lineages as shown in our study. Hence. "it is of immense value to studythe mode of gene transfer facilitating the exchange of the genetic determinantsamong enterococci", since there lies difference in the gene transfer mechanisms ofenterococci, in various geographical regions. The results of these studies would inturn enable us to formulate and implement strategies to prevent'minimize thedissemination of antimicrobial resistance and virulence in any hospital setting.> The molecular typing is more powerful and discriminative than most of thephenotype-based typing systems, since it provides a finer level of epidemiologicaldiscrimination, differentiating both closely and distantly related independentisolates that may otherwise appear identical. The "observations from our studybased on Plasmid REA and PFGE typing has authenticated this fact, and wasinstrumental in understanding the epidemiology and clonal relationship betweenthe HLAR enterococcal isolates disseminated in our hospital setup". Furthermore,

CONCLUSIONSmolecular typing of HLAR enterococci along with other supporting Clinicoepidemiologicaldetails, helps in' initiating and executing appropriate infectioncontrol measures for multidrug resistant nosocomial pathogens like enterococci inany health care setting contributing to a substantial decrease in morbidity andmortality.Hence, as recommended by the American Society of Microbiology (ASM) tasktbrce on Antibiotic Resistance [374] multifaceted approaches like "(I) local and globalsurveillance networks of studying emerging resistances; (2) education of human andanimal health professionals as well patients and the public on the issues of antimicrobialresistances; and (3) basic research of mechanisms of resistance and virulence,identification of new drug targets, development of more rapid and reliable diagnostictests and vaccines, and an increase in the research appropriations at the federal level", iffollowed appropriately, would help in reducing the health care catastrophes caused bydrug resistant nosocomial pathogens like enterococci to several folds.

REFERENCESNlAlD Fact sheet, The Problem of Antibiotic Resistance. Retrieved May 29th2004 from http://www.niaid.nih.gov/factsheets/anti.World Health Report on Infectious Diseases 2000; Overcoming AntimicrobialResistance. Retrieved Aug 3rd 2004 from http:llwww.who.netl.Burke JP. Antibiotic resistance--squeezing the balloon?. JAMA 1998; 280:1270-1.Jones RN. Resistance patterns among nosocomial pathogens: trends over thepast few years. Chest 2001; 119 (2 Suppl): 397s-404s.National Nosocomial Infections Surveillance (NNIS) System report, datasummary from January 1990-May 1999, issued June 1999. Am J Infect Control.1999; 27: 520-32.Clewell DB. Plasmids, drug resistance, and gene transfer in the genusStreptococcus. Microbiol Rev 1981; 45: 409-36.Clewell DB. Movable genetic elements and antibiotic resistance in enterococci.Eur J Clin Microbiol Infect Dis 1990; 9: 90-102.Chow JW. Aminoglycoside resistance in enterococci. Clin Infect Dis 2000; 31:586-9.Huycke MM. Sahm DF, Ciilmore MS. Multiple-drug resistant enterococci: thenature of the problem and an agenda for the future. Emerg Infect Dis 1998; 4:239-49.Emori TG, Gapes RP. An overview of nosocomial infections, including therole of the microbiology laboratory. Clin Microbiol Rev 1993; 6: 428-42.Moellering RC Jr. Emergence of Ettterococcus as a significant pathogen. ClinInfect Dis 1992; 14: 1 173-6.Facklam RR, Sahm DF, Texeira LM. En~crococcus. In: Murray PR, Baron EJ,Pfaller MA, Tenover FC, Yolken RH, editors. Manual of Clinical Microbiology.7th ed. Washington, D.C: ASM press; 1999.297-30s.Murray BE. The life and times of the Entcrococcus. Clin Microbiol Rev 1990;3: 46-65.

REFERENCESCarvalho Mda G, Steigenualt AG, Morey RE, Shewmaker PL, Teixeira LM,Facklam RR. Characterizatiori of three new enterococcal species, Enterococcussp. nov. CDC PNS-El, Enterococcus sp. nov. CDC PNS-E2, and Enterococcussp. nov. CDC PNS-E3, isolated from human clinical specimens. J ClinMicrobiol 2004; 42: 1192-8.Dutka-Malen S, Evers S, Courvalin P. Detection of glycopeptide resistancegenotypes and identification to the species level of clinically relevantenterococci by PCR. J Clin Microbiol 1995; 33: 24-7.Gordon S, Swenson JM, Hill BC, Pigott NE. Facklam RR, Cooksey RC, et al.Antimicrobial susceptibility panems of common and unusual species ofenterococci causing infections in the United States. Enterococcal Study Group. JClin Microbiol 1992; 30: 2373-8.Murray BE. Diversity among multidrug-resistant enterococci Emerg Infect Dis1998; 4: 37-47.Ze~os MJ, Dembinski S. Mikesell T, Schaberg DR. High-level resistance togentamicin in Streprococcu.~ ,/ac,calis: risk factors and evidence for exogenousacquisition of infection. J Infect Dis 1986; 153: 1075-83.Morrison A3 Jr, Wenzel RP. Nosocomial urinary tract infections due toEnterococcus. Ten years' experience at a university hospital. Arch lntem Med1986; 146: 1549-5 1.Jones RN, Marshall SA, Pfaller MA, Wilke WW, Hollis RJ, Erwin ME,Edmond MB, Wenzel RP. Nosocomial enterococcal blood stream infections inthe SCOPE Program: antimicrobial resistance, species occurrence, moleculartesting results. and laboratory testing accuracy. SCOPE Hospital Study Group.Diagn Microbiol Infect Dis 1997; 29: 95-102.Pfaller MA, Jones RN, Doem GV, Kugler K. Bacterial pathogens isolated frompatients with bloodstream infection: frequencies of occurrence andantimicrobial susceptibility patterns from the SENTRY antimicrobialsurveillance program (United States and Canada, 1997). Antimicrob AgentsChemother 1998; 42: 1762-70.

REFERENCESPatterson JE, Sweeney AH, Simms M, Carley N, Mangi R, Sabetta J, et al. Ananalysis of 110 serious eriterococcal infections. Epidemiology, antibioticsusceptibility, and outcome. Medicine (Baltimore). 1995; 74: 191 -200.Shlaes DM, Lev J, Wolinsky E. Enterococcal bacteremia without endocarditis.Arch Intern Med 1981; 141: 578-81.Malone DA, Wagner RA, Myers JP, Watanakunakom C. Enterococcalbacteremia in two large community teaching hospitals. Am J Med 1986; 81:601 -6.Eaton TJ, Gasson MJ. Molecular screening of Enterococcu.v virulencedeterminants and potential for genetic exchange between food and medicalisolates. Appl Environ Microbiol 2001; 67: 1628-35.Jett BD, Huycke MM, Gilmore MS. Virulence of enterococci. Clin MicrobiolRev 1994; 7: 462-78.Domig KJ, Mayer HK, Kneifel W. Methods used for the isolation, enumeration,characterisation and identification of f?nterOcocCu.~ spp. 1. Media for isolationand enumeration. Int J Food Microbiol 2003; 88: 147-64.Domig KJ, Mayer HK, Kneifel W. Methods used for the isolation, enumeration,characterisation and identification of Entcrococcus spp. 2. Pheno- and genotypiccriteria. In( J Food Microbiol2003; 88: 165-88.Schleifer KH, Kilpper-Balz R. Transfer of Strcprococcu.~ faecalis andStreptococcus ,/aecium to the genus Entcrococcus nom. rev. as Enterococcusfaecalis comb. nov. and Entcrococclts fuircciurn como. nov. Int. J. Syst. Bacteriol1984; 34: 31-34.Collins MD, Jones D, Farrow JAE, Kilpper-Balz R, Schleifer KH. Entrrococcusavium nom. rev.,comb. nov.; E. casselifluvus nom. rev., comb. nov.; E. duransnom, rev., comb. nov.; E. gallinarum comb. nov.; and E. malodoratus sp. nov.,Int. J. Syst. Bacteriol. 1984; 34: 220-223.Tymell GJ, Turnbull L, Teixeira LM. Lefebvre J, Carvalho Mda G, FacklamRR, et al, Enterococcus gi1~~u.sp. nov. and Enterococc~rs pallens sP. nov.isolated from human clinical specimens. J Clin Microbiol 2002; 40: 1140-5.

REFERENCESLmg MM, Ingham SC, lngham BH. Differentiation of Enterococcus spp. bycell membrane fatty acid methyl ester profiling, biotyping and ribotyping. LenAppl Microbiol 2001; 33: 65-70.Del Camp0 R. Ruiz-Garbajosa P, Sanchez-Moreno MP, B~quero F, Torres C,Canton R, Coque TM. Antimicrobial resistance in recent fecal enterococci fromhealthy volunteers and food handlers in Spain: genes and phenotpes. MicrobDrug Resist 2003; 9: 47-60.Noble CJ. Camage of group D streptococc~ in the human bowel. J Clin Pathol1978; 31: 1182-6.Marcinek H, Wirth R, Muscholl-Silberhom A, Gauer M. Enrerococcus fueculisgene transfer under natural conditions in municipal sewage water treatmentplants. Appl Environ Microbiol 1998; 64: 626-32.Fridkin SK, Edwards JR, Courval JM, Hill H, Tenover FC, Lawton R, et al.Intensive Care Antimicrobial Resistance Epidemiology (ICARE) Project andthe National Nosocomial Infections Sun~eillance (NNIS) System Hospitals. Theeffect of vancomycin and third-generation cephalosporins on prevalence ofvancomycin-resistant enterococci in 126 U.S. adult intensive care units. AnnIntern Med 2001; 135: 175-83.Pfaller MA, Jones RN, Doem GV. Sader HS, Kugler KC, Beach ML. Survey ofblood stream infections attributable to gram-positive cocci: frequency ofoccurrence and antimicrobial susceptibility of isolates collected in 1997 in theUnited States, Canada, and Latin America from rhe SENTRY AntimicrobialSurveillance Progam. SENTRY Participants Group. Diagn Microbiol InfectDis 1999; 33: 283-97.Bouza E, San Juan R, Munoz P, Voss A, Kluytmans J; Co-operative Group ofthe European Study Group on Nosocomial Infections. A European ~erspectiveon nosocomial urinary tract infections 1. Report on the microbiology workload,etiology and antimicrobial susceptibility (ESGN1-003 study). European Study~~~~p on Nosocomial Infections. Clin Microbiol Infect 2001 ; 7: 523-3 1.Bouza E, Sari Juan R, Munoz P, Voss A, Kluytmans 3; Co-operative Group ofthe European study Group on Nosocomial Infections. A European perspective

REFERENCESon nosocomial urinary tract infections 11. Report on incidence, clinicalcharacteristics and outcome {ESGNI-004 study). European Study Group onNosocomial Infection. Clin Microbiol Infect 2001; 7: 532-42.40. Gordon KA, Jones RN; SENTRY Participant Groups (Eurnpe, Latin America,North America). Susceptibility patterns of orally administered antimicrobialsamong urinary tract infection pathogens from hospitalized patients in NorthAmerica: comparison report to Europe and Latin America. Results from theSENTRY Antimicrobial Surveillance Program (2000). Diagn Microbiol lnfectDis 2003; 45: 295-301.41. Davies HD, Jones EL. Sheng RY, Leslie B, Matlow AG, Gold R. Nosocomialurinary tract infections at a pediatric hospital. Pediatr Infect Dis J 1992; 11:349-54.42. Moulin F, Quintart A. Sauvestre C. Mensah K, Bergeret M, Raymond J.Nosocomial urinary tract infections: retrospective study in a pediatric hospital.Arch Pediatr 1998; 5 (Suppl 3): 2748-278s.43. Richards MJ, Edwards JR, Culver DH, Gaynes RP. Nosocomial infections incombined medical-surgical intensive care units in the United States. InfectControl Hosp Epidemiol 2000; 2 1 : 5 10-5.44. Diekema DJ, Pfaller MA, Jones RN: SENTRY Participants Group. Age-relatedtrends in pathogen frequency and antimicrobial susceptibility of bloodstreamisolates in North America: SENTRY Antimicrobial Surveillance Program,1997-2000. Int J Antimicrob Agents 2002; 20: 412-8.45. Wisplinghoff H, Seifert H. Tallent SM, Bischoff T, Wenzel RP, Edmond MB.Nosocomial bloodstream infections in pediatric patients in United Stateshospitals: epidemiology, clinical features and susceptibilities. Pediatr Infect DisJ. 2003; 22: 686-91.46. Weiss CA 3rd, Statz CL, Dahms RA, Remucal MJ, Dunn DL, Beilman GJ. Sixyears of surgical wound infection surveillance at a tertiary care center: review ofthe microbiologic and epidemiological aspects of 20,007 wounds. Arch Surg1999; 134: 1041-8.

REFERENCESZervos MJ, Kauffman CA, Therasse PM, Bergman AG, Mikesell TS, SchabergDR. Nosocomial infection by gentamicin-resistant Streptococcus .faecalis. Anepidemiologic study. Ann Intern Med 1987; 106: 687-91Weems JJ Jr, Lowrance JH, Baddour LM, Simpson WA. Molecularepidemiology of nosocomial, multiply aminoglycoside resistant Enterococcus,foecolis. J Antimicrob Chemother 1989; 24: 12 1-30.Wells VD, Wong ES, Murray BE, Coudron PE, Williams DS. Markowitz SM.Infections due to beta-lactamase-producing, high-level gentamicin-resistantEnterococcu.sfuecalis. Ann Intern Med 1992: 116: 285-92.Patterson JE, Wanger A, Zscheck KK, Zewos MJ, Murray BE. Molecularepidemiology of beta-lactamase-producing enterococci. Antimicrob AgentsChemother 1990; 34: 302-5.Patterson JE, Singh KV, Murray BE. Epidemiology of an endemic strain ofbeta-lactamase-producing Enrerococcus faeca1i.s. J Clin Microbiol 1991; 29:25 13-6.Bopp LH, Schoonmaker DJ, Baltch AL, Smith RP, Ritz WJ. Molecularepidemiology of vancomycin-resistant enterococci from 6 hospitals in NewYork State. Am J Infect Control 1999; 27: 41 1-7.Bischoff WE, Reynolds TM, Hall GO, Wenzel RP, Edmond MB. Molecularepidemiology of' vancomycin-resistant Enterococcus faecium in a large urbanhospital over a 5-year period. J Clin Microbiol 1999: 37: 3912-6.Takahashi S, Hirose T, Matsukawa M, Shimizu T, Kunishima Y, Takeyama K,et al. Analysis of cross infection using genomic fingerprinting in nosocomialurinary tract infection caused by Enterococcus ,faecalis. J Infect Chemother1999; 5: 46-48.Antonishyn NA, McDonald RR, Chan EL, Horsman G, Woodmansee CE, FalkPS, et al. Evaluation of fluorescence-based amplified fragment lengthpolymorphism analysis for molecular typing in hospital epidemiology:comparison with pulsed-field gel electrophoresis for typing strains ofvancomycin-resistant Enterococcus,faecium. J Clin Microbiol 2000; 38: 4058-65.

REFERENCESNourse C, Murphy H, Byme C, O'Meara A, Breatnach F, Kaufmann M, et at.Control of a nosocomial outbreak of vancomycin resistant Enterococcus,fuecium in a paediatric oncology unit: risk factors for colonisation. Eur J Pediatr1998; 157: 20-7.Murray BE. Vancomycin-resistant enterococcal infections. N Engl J Med 2000;342: 71 0-2 1.Nosocomial enterococci resistant to vancomycin--United States, 1989-1993.MMWR Morb Mortal Wkly Rep 1993; 42: 597-9.McGeer AJ, Low DE. Vancomycin-resistant enterococci. Semin Respir lnfect2000; 15: 314-26.Boyce JM. Opal SM, Potter-Bynoe G, LaForge RG, Zewos MJ, et a!.Emergence and nosocomial transmission of ampicillin-resistant enterococci.Antimicrob Agents Chemother 1992; 36: 1032-9.Harthug S, Digranes A, Hope 0, Kristiansen BE, Allum AG, Langeland N.Vancomycin resistance emerging in a clonal outbreak caused by ampicillinresistantEn~erococcrts/aecium. Clin Microhiol Infect 2000; 6: 19-28.Chirurgi VA, Oster SE, Goldberg AA, McCahe RE. Nosocomial acquisition ofbeta-lactamase negative, ampicillin-resistant Enterococcus. Arch Intern Med1992; 152: 1457-61.Torell E. Cars 0. Olsson-Liljequist B, Hoffman BM, Lindback J, Burman LG.Near absence of vancomycin-resistant enterococci but high carriage rates ofquinolone-resistant ampicillin-resistant enterococci among hospitalized patientsand nonhospitalized individuals in Sweden. J Clin Microbiol 1999; 37: 3509-13.Torell E, Cars 0, Hambraeus A. Ampicillin-resistant enterococci in a Swedishuniversity hospital: nosocomial spread and risk factors for infection. Scand JInfect Dis 2001; 33: 182-7.Bonten MJ, Hayden MK, Nathan C, van Voorhis J, Matushek M, et al.Epidemiology of colonisation of patients and environment with vancomycinresistantenterococci. Lancet 1996; 348: 1615-9.VII

REFERENCESOon LL, Ling MM. Chiew YF. Gastrointestinal colonisation of vancomycinresistantEn!erococcus in a Singapore teaching hospital. Pathology. 2001; 33:216-21.Roghmann MC, Qaiyumi S, Johnson JA, Schwalbe R, Morris JG Jr. Recurrentvancomycin-resistant Entrrococcus ,faecium bacteremia in a leukemia patientwho was persistel~tiy colonized with vancomycin-resistant enterococci for twoyears. Clin Infect Dis 1997; 24: 514-5.Moellering RC Jr. Enter0coccu.s species, S!rcptococcus hovis, and Lruconoslocspecies. In: Mandell GL, Bennett JE, Dolin R, editors. Principles and Practice ofInfectious Diseases. 5th ed. Philadelphia: Churchill Livingstone; 2000. 2147-2155.Zervos MJ, Bacon AE 3rd, Patterson JE. Schaberg DR. Kauffman CA.Enterococcal superinfection in patients treated with ciprofloxacin. J AntimicrobChemother 1988; 21: 113-5.Noskin GA, Stosor V, Cooper I, Peterson LR. Recovery of vancomycinresistantenterococci on fingertips and environmental surfaces. lnfect ControlHosp Epidemiol 1995; 16: 577-8 1.Freeman R, Keams AM, Lightfoot SF. Heat resistance of nosocomialenterococci. Lancet 1994; 344: 64-5.Freeman R, Gould FK, Ryan DW, Chamherlain J, Sisson PR. Nosocomialinfection due to enterococci attributed to a fluidized microsphere bed. The valueof pyrolysis mass spectrometry. J Hosp infect 1994; 17: 187-93.Livornese LL Jr, Dias S, Same1 C, Romanowski B, Taylor S, May P, et al.Hospital-acquired infection with vancomycin-resistant Enterococctt.~ faeci~rmtransmitted by electronic thermometers. Ann intern Med 1992; 117: 112-6.Brooks S, Khan A, Stoica D, Griffith J, Friedeman L, Mukherji R, et al.Reduction in vancomycin-resistant Enterococcus and Clostridium difficileinfections following change to tympanic thermometers. infect Control HO~PEpidemiol 1998; 19: 333-6.

REFERENCESDicuonzo G, Gherardi G, Lorino G, Angeleni S, Battistoni F, Bertuccini L, etal. Antibiotic resistance and gknotypic characterization by PFGE of clinical andenvironmental isolates of enterococci. FEMS Microbiol Len 2001 ; 201 : 205-1 I.Kuriyama T, Williams DW, Patel M, Lewis MA, Jenkins LE, Hill DW, et al.Molecular characterization of clinical and environmental isolates ofvancomycin-resistant En(erococcus faecium and Enterococcu.~ facjculis from ateaching hospital in Wales. J Med Microbiol 2003; 52: 82 1-7.Martinez JA, Ruthazer R, Hansjosten K, Barefoot L. Snydman DR. Role ofenvironmental contamination as a risk factor for acquisition of vancomycin-resistant enterococci in patients treated in a medical intensive care unit. ArchIntern Med 2003; 163: 1905-12.Mayer RA, Geha RC. Helfand MS. Hoyen CK, Salata RA. Donskey CJ. Role offecal incontinence in contamination of the environment with vancomycin-resistant enterococci. Am J lnfect Control 2003; 3 1: 221-5.Reisner BS, Shaw S, Huber ME, Woodmansee CE. Costa S. Falk PS, et al.Comparison of three methods to recover vancomycin-resistant enterococci(VRE) from perianal and environmental samples collected during a hospitaloutbreak of VRE. lnfect Control Hosp Epidemiol2000; 21 : 775-9.Smith TL, lwen PC, Olson SB, Rupp ME. Environmental contamination withvancomycin-resistant enterococci in an outpatient setting. lnfect Control HospEpidemiol 1998; 19: 5 15-8.D'Agata EM, Jirjis J, Gouldin C, Tang YW. Community dissemination ofvancomycin-resistant Enterococcus faecium. Am J lnfect Control 2001; 29:3 16-20.Endtz HP, van den Braak N, van Belkum A, Kluytmans JA, Koeleman JG,Spanjaard L, et al. Fecal carriage of vancomycin-resistant enterococci inhospitalized patients and those living in the community in The Netherlands. JClin Microbiol 1997; 35: 3026-3 1.Nourse C, Byme C, Murphy H, Kaufmann ME, Clarke A, Butler K. Eradicationof vancomycin resistant Enlerococcus faecium from a paediatric oncology unit

REFERENCESand prevalence of colonization in hospitalized and community-based children.Epidemiol Infect 2000; 124: 53-9.Aarestmp FM, Agerso Y, Gerner-Smidt P, Madsen M, Jensen LB. Comparisonof antimicrobial resistance phenotypes and resistance genes in Enterococcusfaecalis and Enterococcus faecium from humans in the community, broilers,and pigs in Denmark. Diagn Microbiol lnfect Dis 2000; 37: 127-37.Coque TM, Tomayko JF, Ricke SC, Okhyusen PC, Murray BE. Vancomycin-resistant enterococci from nosocomial, community, and animal sources in theUnited States. Antimicrob Agents Chemother 1996; 40: 2605-9.Coque TM, Arduino RC, Murray BE. High-level resistance to arninoglycosides:comparison of community and nosocomial fecal isolates or enterococci. ClinInfect Dis 1995; 20: 1048-51.Klare I, Heier H, Claus H, Bohrne G, Mar~n S. Seltmann G, et al. Entcrococcus/aecium strains with vanA-mediated high-level glycopeptide resistance isolatedfrom animal foodstuffs and fecal samples of humans in the community. MicrobDrug Resist 1995; I: 265-72.Klare I, Badstubner D, Konstabel C. Bohme G. Claus H, Witte W. Decreasedincidence of VanA-tpe vancomycin-resistant enterococci isolated from poultrymeat and from fecal samples of humans in the community after discontinuationof avoparcin usage in animal husbandry. Microb Drug Resist 1999; 5: 45-52.Petts DN. Evaluation of a modified nitrous acid extraction latex agglutinationkit for grouping beta-hemolytic streptococci and enterococci. J Clin Microbiol1995; 33: 1016-8.Facklam RR, Collins MD. Identification of Envrococcu.v species isolated fromhuman infections by a conventional test scheme. J Clin Microbiol 1989; 27:73 1-4.Ford M, p e JD, ~ Gould FK. Use of cephalexin-aztreonam-arabinose agar forselective isolation of En~erococcus faecium. J Clin Microbiol 1994; 32: 2999-3001.

REFERENCESWilley BM, Kreiswirth BN, Simor AE, Willaims G, Scriver SR, Phillips A, etal. Detection of vancomycin resistance in Enterococcus species. J ClinMicrobiol 1992; 30: 1621-4.Swenson JM, Clark NC, Ferraro MJ, Sahm DF, Doern G, Pfaller MA, et al.Development of a standardized screening method for detection of vancomycin-resistant enterococci. J Clin Microbiol 1994; 32: 1700-4.Carvalho MG, Teixeira LM, Facklam RR. Use of tests for acidification ofmethyl-alpha-D-glucopyranoside and susceptibility to efrotomycin fordifferentiation of strains of Enterucoccus and some related genera. J ClinMicrobiol 1998; 36: 1584-7.Chen DK, Pearce L, McGeer A, Low DE, Willey BM. Evaluation of D-xyloseand 1% methyl-alpha-D-glucopyranoside fermentation tests for distinguishingEnterococcus gullinarum from Enterococ.c~ts ,fuccium. J Clin Microbiol 2000;38: 3652-5.Manero A, Blanch AR. Identification of Enterococcrr.v spp. with a biochemicalkey. Appl Environ Microbiol 1999; 65: 4425-30.Teixeira LM, Facklam RR, Stelgenvalt AG, Pigott NE, Merquior VL, BrennerDJ. Correlation between phenotypic characteristics and DNA relatedness withinEnterococcus fuecium strains. J Clin Microbiol 1995; 33: 1520-3.Merquior VLC, Peralta JM, Facklam RR, Teixeira LM. Analysis ofelectrophoretic whole-cell protein profiles as a tool for characterization ofEnterococcus species. Curr. Microbiol 1994; 28: 149- 153.Vittal Prakash.P, Sarnbasiva Ra0.R. Parija.SC. Molecular Phenotyping ofAtypical Enierococcus species. In: Abstract book of the XXVII NationalConference of Indian Association of Medical Microbiologists, November 5th-9th, 2003, Mumbai, India.Brock TD, Davie JM. Probable identity of a Group D Hemolysin with abacteriocine. J Bacteriol 1963; 86: 708-12.Brock TD, Peacher B, pierson D. Survey of the bacteriocines of enterococci. JBacteriol. 1963; 86: 702-7.

REFERENCES102. Berlutti F, Thaller MC, Schippa S, Pantanella F, Pompei R. A new approach touse of bacteriolflic enzymes' as a tool for species identification: selection ofspecies-specific indicator strains with bacteriolytic activity towardsEnlerococcus strains. Int J Syst Bacterial 1993; 43: 63-8,103. Simonsen GS, Smabrekke L, Monnet DL, Sorensen TL, Moiler JK, KristinssonKG, et al. Prevalence of resistance to ampicillin, gentamicin and vancomycin inEnterococcus faecalisand En~erococcu.\ ,fuecium isolates from clinicalspecimens and use of antimicrobials in five Nordic hospitals. J AntimicrobChemother 2003; 51 : 323-31.104. Del Campo R, Tenorio C, Jimenez-Diaz R, Rubio C, Gomez-Lus R, Baquero F,Torres C. Bacteriocin production in vancomycin-resistant and vancomycin-susceptible Entrrococcus isolates of different origins. Antimicrob AgentsChemother 2001; 45: 905-12.105. Kariyama R, Mitsuhata R, Chow JW, Clewell DB, Kumon H. Simple andreliable multiplex PCR assay for surveillance isolates of vancomycin-resistantenterococci. J Clin Microbiol 2000; 38: 3092-5.106. Singh KV, Coque TM, Weinstock GM, Murray BE. In vivo testing of anEnterococcus ,faeculis efaA mutant and use of efaA homologs for speciesidentification. FEMS Immunol Med Microbiol 1998; 21: 323-31.107. Cheng S, McCleskey FK, Gress MJ, Petroziello JM, Liu R. Namdari H, et al. APCR assay for identification of Ent~rorocc~rs,fu~t.,um. J Clin Microbiol. 1997;35: 1248-50.108. Knijff E, Dellaglio F, Lombardi A, Andrighetto C, Torriani S. Rapididentification of Enterococc~ts durans and Enterococcus hiroe by PCR withprimers targeted to the ddl genes. J Microbiol Methods 2001; 47: 35-40.109. Monstein HJ, Quednau M, Samuelsson A, Ahme S, lsaksson B, Jonass~~~ J.Division of the genus Enlerococnrs into species groups using PCR-basedmolecular typing methods. Microbiology 1998; 144: 1171-9.1 10, Chiew YF, Hall LM. Comparison of three methods for the molecular typingsingapore isolates of enterococci with high-level aminoglycoside resistances.Hosp Infect 1998; 38: 223-30.

REFERENCES111. T~rrell GJ, Bethune RN, Willey B, Low DE. Species identification ofenterncocci via intergenic ribosomal PCR. J Clin Microbial 1997; 35: 1054-60.112. Ozawa Y, Courvalin P, Gaiimand M. Identification of enterococci at the specieslevel by sequencing of the genes for D-a1anine:D-alanine ligases. syst ~pplMicrobiol 2000; 23: 230-7.113. Friesenegger A. Fiedler S, Devriese LA. Wirth R. Genetic transformation ofvarious species of Enterococcus by electroporation. FEMS Microbiol Lett 1991 ;63: 323-7.114. Jacob AE. Hobbs SJ. Conjugal transfer of plasmid-home multiple antibioticresistance in Streptococcus faewlis var, zymogenes. J Bacteriol 1974; I 17: 360-72.I IS. Clewell DB, Yagi Y, Dunny GM, Schultz SK. Characterization of three plasmiddeoxyribonucleic acid molecules in a strain of Streproc,or,cu.s /ueculi.s:identification of a plasmid determining erythromycin resistance. J Bacteriol1974; 1 17: 283-9.116. Dunny GM, Craig RA, Carron RL. Clewell DB. Plasmid transfer inStreptoc.occus faecalis: production of multiple sex pheromones by recipients.Plasmid 1979; 2: 454-65.117. Courvalin PM, Shaw WV, Jacob AE. Plasmid-mediated mechanisms ofresistance to aminoglycoside-aminocyclitol antibiotics and to chloramphenicolin group D streptococci. Antimicrob Agents Chemother 1978; 13: 716-25.118. De Boever EH. Clewell DB. Fraser CM. Entc~rococt~u.\. faeculis conjugativeplasmid pAM373: complete nucleotide sequence and genetic analyses of sexpheromone response. Mol Microbiol 2000; 37: 1327-41.119. Wu K, An FY, Clewell DB. Enterococcus ,fuecalis pheromone-respondingplasmid pADI gives rise to an aggregation (clumping) response when cells areexposed to subinhibitory concentrations of chloramphenicol, erythromycin, ortetracycline. Plasmid 1999; 41: 82-8.120. Ike y, Tanimoto K, Tomita H. Takeuchi K, Fujimoto S. Efficient transfer of thepheromone-independent Enterococcus ,faecium plasmid PMGl (Gmr) (65.1

REFERENCESkilobases) to Enterococcus strains during broth mating. J Bacteriol 1998; 180:4886-92.121. Simjee S, Gill MJ. Gene transfer, gentamicin resistance and enterococci. J Hosplnfect 1997; 36: 249-59.122. Hodel-Christian SL, Murray BE. Comparison of the gentamicin resistancetransposon Tn5281 with regions encoding gentamicin resistance inEnterococc~ts ,faecalis isolates from diverse geographic locations. AntimicrobAgents Chemother 1992; 36: 2259-64.123. Rice LB, Carias LL, Marshall SH. Tn5384, a composite enterococcal mobileelement conferring resistance to erythromycin and gentamicin whose ends aredirectly repeated copies of 18256. Antimicrob Agents Chemother 1995; 39:1147-53.124. Quintiliani R Jr, Courvalin P. Characterization of Tn1547, a compositetransposon flanked by the IS16 and lS256-like elements, that confersvancomycin resistance in Enterococcu.r,faecalis BM428 1 . Gene 1996; 172: 1-8.125. Tomita H, Tanimoto K, Hayakawa S, Morinaga K, Ezaki K, Oshima H, Ike Y.Highly conjugative pMG1-like plasmids canying Tn1546-like transposons thatencode vancomycin resistance in Enterococcu.~ faeci~im. J Bacteriol 2003; 185:7024-8.126. Gamier F, Taourit S, Glaser P, Courvalin P, Galimand M. Characterization oftransposon Tn1549, conferring VanB-type resistance in Enterococcus spp.Microbiology 2000; 146: 148 1-9.127. Dunny GM, Brown BL, Clewell DB. Induced cell aggregation and mating inStreptococcus faecalis: evidence for a bacterial sex pheromone. Proc Natl AcadSci U S A 1978; 75: 3479-83.128. Dunny GM, Clewell DB. Transmissible toxin (hemolysin) plasmid inStreptococcus faecalis and its mobilization of a noninfectious drug resistanceplasmid. J Bacterial 1975; 124: 784-90.129. Coque TM, Patterson JE, Steckelberg JM, Murray BE, lncidence of hemolysin,gelatinase, and aggregation substance among enterococci isolated from patients

REFERENCESwith endocarditis and other infections and from feces of hospitalized andcommunity-based persons. J lnfect Dis 1995; 171 : 1223-9.130. Valdivia E, Martin-Sanchez 1, Quirantes R, Maninez-Bueno M, Galvez A.Maqueda M. Incidence of antibiotic resistance and sex pheromone responseamong enterococci isolated from clinical human samples and from municipalwaste water. J Appl Bacterial 1996; 8 1 : 538-44.13 1. Heaton MP, Discotto LF, Pucci MJ, Handwerger S. Mobilization ofvancomycin resistance by transposon-mediated fusion of a VanA plasmid withan Enterococcus faccium sex pheromone-response plasmid. Gene 1996; 171: 9-17.132. Vittal Prakash.P, Sambasiva Rao.R, Malay.K.Ray, Parija.SC. Plasmids and SexPheromone response of High Level Aminoglycoside Resistant enterococci. In:Abstracts of the 111 Winter Symposia on Infection and Immunity, ChristianMedical College, December 16th-17th, 2004, Vellore, India.133. Shiojima M, Tomita H. Tanimoto K, Fujimoto S, Ike Y. High-level plasmid-mediated gentamicin resistance and pheromone response of plasmids present inclinical isolates of Ei~terococcus,/aecali.v. Antimicrob Agents Chemother 1997;41: 702-5.134. Pournaras S, Tsakris A, Palepou MF, Papa A. Douboyas J, Antoniadis A. et al.Pheromone responses and high-level aminoglycoside resistance of conjugativeplasmids of En~erococcu.r,/uecali.r from Greece. J Antimicrob Chemother 2000;46: 101 3-6.135. Ma X, Kudo M, Takahashi A, Tanimoto K, Ike Y. Evidence of nosocomialinfection in Japan caused by high-level gentamicin-resistant Enlerococcusfaecalis and identification of the pheromone-responsive conjugative plasmidencoding gentamicin resistance. J Clin Microbiol 1998; 36: 2460-4.136. Jett BD, Atkuri RV, Gilmore MS. Enterococcus faecalis localization inexperimental endophthalmitis: role of plasmid-encoded aggregation substance.infect lmmun 1998; 66: 843-8.137. Hirt H, &landsen SL, Dunny GM. Heterologous inducible expression ofEnterococcus faecalis pCFI 0 aggregation substance asclo in Lactococcus lactis

REFERENCESand Streptococcus gordonii contributes to cell hydrophobicity and adhesion tofibrin. J Bacteriol 2000; 182: 2299-306.138. Clewell DB. Bacterial sex pheromone-induced plasnlid transfer. Cell 1993; 73:9-12.139. Tanimoto K, Ike Y. Analysis of the conjugal transfer system of the pheromoneindependenthighly transferable Enterococcus plasmid pMG 1 , identification ofa tra gene (traA) up-regulated during conjugation. J Bacteriol 2002; 184: 5800-4.140. Moellering RC Jr. The enterococci: an enigma and a continuing therapeuticchallenge. Eur J Clin Microbiol Infect Dis 1990; 9: 73-4.141. Prakash VP, Rao SR, Parija SC. Emergence of unusual species of enterococcicausing infections, South India. BMC Infect Dis 2005; 5: 14.142. Mathai D, Jones RN. Pfaller MA; SENTRY Participant Group North America.Epidemioloby and frequency of resistance among pathogens causing urinarytract infections in 1,510 hospitalized patients: a report from the SENTRYAntimicrobial Surveillance Program (North America). Diabm Microbiol InfectDis 200 1 ; 40: 129-36.143. Wazait HD, Patel HR. Veer V, Kelsey M, Van Der Meulen JH, Miller RA, et al.Catheter-associated urinary tract infections: prevalence of uropathogens andpattern of antimicrobial resistance in a UK hospital (1996-2001). BJU int 2003;91: 806-9.144. Desai PJ, Pandit D, Mathur M, Gogate A. Prevalence, identification anddistribution of various species of enterococci isolated from clinical specimenswith special reference to urinary tract infection in catheterized patients IndianJournal of Medical Microbiology 2001; 19: 132-7.145. Shankar N, Lockatell CV, Baghdayan AS, Drachenberg C, ~ilmore MS,Johnson DE. Role of Enterococcus ,/aecu[is surface protein Esp in thepathogenesis of ascending urinary tract infection. Infect hmun 2001 ; 69: 4366-72.

REFERENCES146. Dan M, Heshkovits A, Mirsky L, Gutman R. Increase in the isolation rate ofenterococci from Urine since the introduction of Fluoroquinolones. Int J ofinf.dis 1997; 2: 110-1 12.147. Fraimow HS, Jungkind DL, Lander DW, Delso DR, Dean JL. Urinary tractinfection with an Enrerococcus faecalis isolate that requires vancomycin forgrowth. Ann Intern Med 1994; 12 1 : 22-6.148. Wilke WW, Marshall SA, Coffman SL, Pfaller MA, Edmund MB, Wenzel RP,et al. Vancomycin-resistant Enterococcu.~ raflinosus: molecular epidemiology.species identification error, and frequency of occurrence in a national resistancesurveillance program. Diagn Microbiol lnfect Dis 1997; 29: 43-9.149. Watanakunakom C, Patel R. Comparison of patients with enterococcalbacteremia due to strains with and without high-level resistance to gentamicin.Clin lnfect Dis 1993; 17: 74-8.150. Reid KC, Cockerill I11 FR, Patel R. Clinical and epidemiological features ofEn~erococcrts cas.srli/la~~u.s//lo~~escc~~~.r and Enterococcus gallinarirm bacteremia:a report of 20 cases. Clin Infect Dis 2001; 32: 1540-6.15 1. Pappas G, Liberopoulos E, Tsianos E, Elisaf M. Entc,rococcu.s casscli/luvusbacteremia. Case report and literature review. J Infect 2004; 48: 206-8.152. Vergis EN, Shankar N, Chow JW, Hayden MK, Snydman DR, Zervos MJ, et al.Association between the presence of enterococcal virulence factors gelatinase,hemolysin, and enterococcal surface protein and mortality among patients withbacteremia due to Enterococcus fueculis. Clin Infect Dis 2002; 35: 570-5.153. Taneja N, Rani P, Emmanuel R. Sharma M. Significance of vancomycinresistant enterococci from urinary specimens at a tertiary care centre in northernIndia. Indian J Med Res 2004; 119: 72-4.154. Mathur P, Kapil A, Chandra R, Sharma P, Das B. Antimicrobial resistance inEnterococcus ,faecalis at a tertiary care centre of northern India. Indian J MedRes 2003; 11 8: 25-8.155. Noskin GA, Till M, Patterson BK, Clarke JT, Warren JR. High-level gentamicinresistance in Enterococcus,faecalis bacteremia. J Infect Dis 1991 ; 164: 12 12-5.

REFERENCES156. Graninger W, Ragette R. Nosocomial bacteremia due to Enterococcus fueca1i.swithout endocarditis. Clin lnfecf Dis 1992; 15: 49-57.157. Zaas AK, Song X, Tucker P, Perl TM. Risk factors for development ofvancomycin-resistant enterococcal bloodstream infection in patients with cancerwho are colonized with vancomycin-resistant enterococci. Clin lnfect Dis 2002;35: 1139-46.158. Safdar N, Maki DG. The commonality of risk factors for nosocomialcolonization and infection with antimicrobial-resistant Staphylococcus aureus,Enterococcus, gram-negative bacilli, Clostridium difficile, and Candida. AnnIntern Med 2002: 136: 834-44.159. McCarthy AE, V~ctor G. Ramotar K, Toye B. Risk factors for acquiringampicillin-resistant enterococci and clinical outcomes at a Canadian tertiarycarehospital. J Clin Microbiol 1994; 32: 2671-6.160. Elsner HA, Sobottka I, Mack D, Claussen M, Laufs R, Wlrth R. Virulencefactors of En!erococcus faecalis and Entrvococcus /aecium blood cultureisolates. Eur .I Clin Micmbiol lnfect Dis 2000; 19: 39-42.161. Huycke MM, Spiegel CA, Gilmore MS. Bacteremia caused by hemolytic. highlevelgentamicin-resistant Entcrococcus,/aeculis. Antimicrob Agents Chemother1991 ; 35: 1626-34.162. Megran DW. Enterococcal endocarditis. Clin Infect Dis 1992; 15: 63-71.163. Lewis CM, Zervos MJ. Clinical manifestations of enterococcal infection. Eur JClin Microbiol Infect Dis 1990; 9: 1 1 1-7.164. Rice LB, Calderwood SB, Eliopoulos GM, Farber BF, Karchmer AW.Enterococcal endocarditis: a comparison of prosthetic and native valve disease.Rev Infect Dis 1991; 13: 1-7.165. Femandez-Guerrero ML, Herrero L, Bellver M, Gadea 1, Roblas RF, deGorgolas M. Nosocomial enterococcal endocarditis: a serious hazard forhospitalized patients with enterococcal bacteraemia. J Intern Med 2002; 252:510-5.166. Kholeif MA, Kinsara AJ, George BO, Mowallad AW, Bugis FA, Osoba AO.Enterococcus faecalis endocarditis. Saudi Med J 2002; 23: 1120-3.

REFERENCES167. Femandez-Guerrero ML, Verdejo C, Azofra J, de Gorgolas M. Hospitalacquiredinfectious endocarditis not associated with cardiac surgery: anemerging problem. Clin Infect Dis 1995; 20: 16-23.168. Sohn AH, Garrett DO, Sinkowitz-Cochran RL, Grohskopf LA, Levine GL,Stover BH, et al; Pediatric Prevention Network. Prevalence of nosocomialinfections in neonatal intensive care unit patients: Results from the first nationalpoint-prevalence survey. J Pediatr 200 1 ; 139: 821-7.169. Yuce A, Karaman M, Gulay 2, Yulug N. Vancomycin-resistant enterococci inneonates. Scand J Infect Dis 2001; 33: 803-5.170. Lee HK, Lee WG, Cho SR. Clinical and molecular biological analysis of anosocomial outbreak of vancomycin-resistant enrerococci in a neonatalintensive care unit. Acta Paediatr 1999; 88: 651-4.171. Malik RK, Montecalvo MA. Reale MR. Li K, Maw M, Munoz JL, et al.Epidemiology and control of vancomycin-resistant enterococci in a regionalneonatal intensive care unit. Pediatr Infect Dis J 1999; 18: 352-6.172. Bilikova E, Babela R, Krcmery V. Nosocomial enterococcal bacteremia inchildren. Pediatrics 2003; 111: 445-6.173. Dobson SR, Baker CJ. Enterococcal sepsis in neonates: features by age at onsetand occurrence of focal infection. Pediatrics 1990; 85: 165-7 I .174. Bhat KG, Paul C, Bhat MG. Neonatal bacteremia due to high levelaminoglycoside resistant (HLAR) enterococci. lndian J Pediatr 1997; 64: 537-9.175. McNeeley DF, Saint-Louis F, Noel GJ. Neonatal enterococcal bacteremia: anincreasingly frequent event with potentially untreatable pathogens. PediatrInfect Dis J 1996; 15: 800-5.176. Chandrashekar MR, Rathish KC, Nagesha CN. Reservoirs of nosocomialpathogens in neonatal intensive care unit. J Indian Med Assoc 1997; 95: 72-4.177. Breton JR, Peset V, Morcillo F, Cano J, Samon A, Perez-Belles C, GobemadoM. Neonatal meningitis due to Enterococcus spp.: presentation of four cases.Enferm lnfecc Microbial Clin 2002; 20: 443-7.XIX

REFERENCES178. Chang Chien HY, Chiu NC, Li WC, Huang FY. Characteristics of neonatalbacterial meningitis in a teaching hospital in Taiwan from 1984-1997. JMicrobiol Immunol Infect 2000; 33: 100-4.179. Barie PS, Christou NV, Dellinger EP. Rout WR, Stone HH, Waymack JP.Pathogenicity of the Enlerococcus in surgical infections. Ann Surg 1990; 212:155-9.180. Johnson RA, Zajac RA, Evans ME. Suppurative thrombophlebitis: correlationbetween pathogen and underlying disease. lnfect Control 1986; 7: 582-5.181. Harbarth S, Uckay I. Are there patients with peritonitis who require empirictherapy for Enterococcus? Eur J Clin Microbiol lnfect Dis 2004; 23: 73-7.182. Sitges-Serra A, Lopez MJ, Girvent M, Almirall S, Sancho JJ. Postoperativeenterococcal infection after treatment of complicated intra-abdominal sepsis. BrJ Surg 2002; 89: 361-7.183. Garcia-Rodriguez JA, Garcia Sanchez JE, Garcia Sanchez E. Role ofenterococci in intra-abdominal infection and possible implications in thetherapeutic choice. Rev Clin Esp 1995; 195 (Suppl4): 28-34.184 Dougherty SH. Role of Enterococclrs in intraabdominal sepsis Am J Surg1984; 148: 308-12.185. Jones ME, Karlowsky JA, Draghi DC. Thornsbeny C. Sahm DF, Nathwani D.Epidemiology and antibiotic susceptibility of bacteria causing skin and softtissue infections in the USA and Europe: a guide to appropriate antimicrobialtherapy. lnt J Antimicrob Agents 2003; 22: 406-19.186. Sader HS, Jones RN, Silva JB; SENTRY Participants Group (Latin America).Skin and soft tissue infections in Latin American medical centers: four-yearassessment of the pathogen frequency and antimicrobial susceptibility patterns.Diagn Microbiol lnfect Dis 2002; 44: 281-8.187. Rennie RP, Jones RN, Mutnick AH; SENTRY Program Study Group (NorthAmerica). Occurrence and antimicrobial susceptibility patterns of pathogensisolated from skin and soft tissue infections: report from the SENTRYAntimicrobial Surveillance Program (United States and Canada, 2000). DiapMicrobiol lnfect Dis 2003; 45: 287-93.

REFERENCES188. Faden H, Dryja D. Recovery of a unique bacterial organism in human middleear fluid and its possible role in chronic otitis media. J Clin Microbial 1989; 27:2488-91.189. Sandoe JA, Witherden IR, Settle C. Vertebral osteomyelitis caused byEnterococcus ra/finosus. J Clin Microbiol 2001 ; 39: 1678-9.190. Methods for dilution antimicrobial susceptibility tests for bacteria that growaerobically. National Committee for Clinical Laboratory Standards (NCCLS).5th ed. Approved standard M7-A5. Wayne (PA): 2000.191. Zeana C, Kubin CJ, Della-Latta P, Hammer SM. Vancomycin-resistantEnterococcus ,fuecium meningitis successfully managed with linezolid: casereport and review of the literature. Clin lnfect Dis 2001; 33: 477-82.192. Wong MT, Kauffrnan CA, Standiford HC. Linden P, Fort G, Fuchs HJ, et al;Ramoplanin VRE2 Clinical Study Group. Effective suppression of vancomycin-resistant Entet'oCOCc~~ species in asymptomatic gastrointestinal carriers by anovel glycolipodepsipeptide, ramoplanin. Clin Infecl Dis 2001; 33: 1476-82.193. Mundy LM, Sahm DF, Gilmore M. Relationships between enterococcalvirulence and antimicrobial resistance. Clin Microbiol Rev 2000; 13: 513-22.194. Guzman CA, Pruzzo C, LiPira G, Calegari L. Role of adherence in pathogenesisof Enterococcu,s,faecali.s urinary tract infection and endocarditis. lnfect lrnmun1989; 57: 1834-8.195. Kreft B, Mane R, Schramm U, Wirth R. Aggregation substance ofEnterococcrts ,faecalis mediates adhesion to cultured renal tubular cells. lnfectImmun 1992; 60: 25-30.196. Olmsted SB, Dunny GM, Erlandsen SL, Wells CL. A plasmid-encoded surfaceprotein on Enterococcus faecalis augments its internalization by culturedintestinal epithelial cells. J lnfect Dis 1994; 170: 1549-56.197. Sartingen S, Rozdzinski E, Muscholl-Silberhorn A, Marre R. Aggregationsubstance increases adherence and internalization, but not translocation, ofEnterococcus faecalis through different intestinal epithelial cells in vitro. Infectlmmun 2000; 68: 6044-7.

REFERENCES198. Sussmuth SD, Muscholl-Silberhom A, Wirth R, Susa M. Marre R, RozdzinskiE. Aggregation substance promotes adherence, phagocytosis, and intracellularsurvival of Enrrrococcus faecalis within human macrophages and suppressesrespiratory burst. lnfect lmmun 2000; 68: 4900-6.199. Rakita RM, Vanek NN, Jacques-Palaz K, Mee M, Mariscalco MM, Dunny GM,et a]. Enrerococcus fuccalis bearing aggregation substance is resistant to killingby human neutrophils despite phagocytosis and neutrophil activation. lnfectlmmun 1999; 67: 6067-75.200. Chow JW, Thal LA, Perri MB, Vazquez JA. Donabedian SM, Clewell DB, et al.Plasmid-associated hemolysin and aggregation substance production contributeto virulence in experimental enterococcal endocarditis. Antimicrob AgentsChemother 1993; 37: 2474-7.201. lsenmann R, Schwarz M, Rozdzinski E. Marre R, Beger HG. Aggregationsubstance promotes colonic mucosal invasion of Enterococcus faccalis in an exvivo model. J Surg Res 2000; 89: 132-8.202, lsenmann R, Schwarz M, Rozdzinski E, Christ C, Schmidt E, Augat P, et al.Interaction of fibronectin and aggregation substance promotes adherence ofEn~erococcusfuecalis to human colon. Dig Dis Sci 2002; 47: 462-8.203. Dupre 1, Zanetti S, Schito AM, Fadda G. Sechi LA. Incidence of virulencedeterminants in clinical Enterococcus ,faec,iurn and Enterococcus ,furcali.sisolates collected in Sardinia (Italy). J Med Microbiol2003; 52: 491-8.204. Semedo T, Santos MA, Lopes MF, Figueiredo Marques JJ, Barreto Crespo MT,Tenreiro R. Virulence factors in food, clinical and reference Enterococci: Acommon trait in the genus? Syst Appl Microbial 2003; 26: 13-22.205. Johnson JR, Clabots C, Hirt H, Waters C, Dunny G. Enterococcal aggregationsubstance and binding substance are not major contributors to urinary tractcolonization by Enterococcus farcalis in a mouse model of ascendingunobstructed urinary tract infection. lnfect lmmun 2004; 72: 2445-8.206. Shankar V. Baghdayan AS, Huycke MM, Lindahl G, Gilmore MS. Infection-derived Enterococcus faecalis strains are enriched in esp, a gene encoding anovel surface protein. Infect Immun 1999; 67: 193-200.

REFERENCES207. Willems RJ, Homan W, Top J, van Santen-Verheuvel M, Tribe D, ManziorosX, et al. Variant esp gene as a marker of a distinct genetic lineage ofvancomycin-resistant Enterococcus faecium spreading in hospitals. Lancet2001; 357: 853-5.208. Woodford N, Soltani M, Hardy KJ. Frequency of esp in Entcrococcrrs,faeciumisolates. Lancet 2001; 358: 584.209. Baldassam L, Bertuccini L, Ammendolia MG. Gherardi G, Creti R. Variant espgene in vancomycin-sensitive E~~terococcu.~.fuecium. Lancet 2001; 357: 1802.210. Toledo-Arana A, Valle J. Solano C, Arrizubieta MJ, Cucarella C, Lamata M, etal. The enterococcal surface protein, Esp, is involved in Enteroroccus,faecali.~biofilm formation. Appl Environ Microbiol 2001; 67: 4538-45.211. Waar K, Muscholl-Silberhom AB, Willems RJ, Slooff MJ, Harmsen HJ,Degener JE. Genogrouping and incidence of virulence factors of Enrerococcusfaecalis in liver transplant patients differ from blood culture and fecal isolates. JInfect Dis 2002; 185: 1 121 -7.212. Kristich CJ, Li YH. Cvitkovitch DG. Dunny GM. Esp-independent biofilmformation by Enferococrus,fuec.alis. J Bacteriol 2004; 186: 154-63.213. Mohamed JA, Huang W, Nallapareddy SR, Teng F, Murray BE. Influence oforigin of isolates, especially endocarditis isolates, and various genes on biofilmformation by Enterococcus.faecalis. Infect lmmun 2004; 72: 3658-63.214. Shankar N, Baghdayan AS. Gilmore MS. Modulation of virulence within apathogenicity island in vancomycin-resistant Enterococcus ,facculis. Nature2002; 41 7: 746-50.21 5. Leavis H, Top J, Shankar N, Borgen K, Bonten M, van Embden J, et al. A novelputative enterococcal pathogenicity island linked to the esp virulence gene ofEnterococcus fuecium and associated with epidemicity. J Bacteriol 2004; 186:672-82.216. Lund B, Edlund C. Bloodstream isolates of Enterococcus,faecium enriched withthe enterococcal surface protein gene, esp, show increased adhesion toeukaryotic cells. J Clin Microbiol. 2003 Nov;41(11):5183-5.

REFERENCES217. Archimbaud C, Shankar N, Forestier C, Baghdayan A, Gilmore MS. CharbonneF, Joly B. In vitro adhesive properties and virulence factors ofEnterococcu.~ecalis strains. Res Microbiol2002; 153: 75-80.218. Hammerum AM, Jensen LB. Prevalence of esp, encoding the enterococcalsurface protein, in Enterococcus ,faecalis and Enterococcus faecium isolatesfrom hospital patients. poultry, and pigs in Denmark. J Clin Microbiol 2002; 40:4396.219. Coque TM, Willems R, Canton R, Del Campo R, Baquero F. High occurrenceof esp among ampicillin-resistant and vancomycin-susceptible Enterococcusjbecium clones from hospitalized patients. J Antimicrob Chemother 2002; 50:1035-8.220. Leavis HL, Willems RJ, Top J. Spalburg E. Mascini EM, Fluit AC, et al.Epidemic and nonepidemic multidrug-resistant Enterococcus faecium. EmergInfect Dis 2003; 9: 1108-15.221. Mascini EM, lalink KP, Kamp-Hopmans TE, Blok HE, Verhoef J. Bonten MJ.et al. Acquisition and duration of vancomycin-resistant enterococcal camage inrelation to strain type. J Clin Microbiol 2003; 41: 5377-83.222. Routsi C, Platsouka E, Willems RJ, Bonten MJ, Paniara 0, Saroglou G, et al.Detection of enterococcal surface protein gene (esp) and amplified fragmentlength polymorphism typing of glycopeptide-resistant Enterococcus ,/aeciumduring its emergence in a Greek intensive care unit. J Clin Microbiol 2003; 41:5742-6.223. Oancea C, Klare 1. Witte W, Werner G. Conjugative transfer of the virulencegene, esp, among isolates of Enterococcus,faecium and Enterococcus,faecalis. JAntimicrob Chemother 2004; 54: 232-5.224. Rich RL, Kreikemeyer B, Owens RT, LaBrenz S, Narayana SV, WeinstockGM, Murray BE, Hook M. Ace is a collagen-binding MSCRAMM fromEnterococcusfaecalis. J Biol Chem 1999; 274: 26939-45.225. Nallapareddy SR, Singh KV, Duh RW, Weinstock GM, Murray BE. Diversityof ace, a gene encoding a microbial surface component recognizing adhesive

REFERENCESmatrix molecules, from different strains of Enterococcus faecalis and evidencefor production of ace during human ~nfections. Infect lrnmun 2000; 68: 52 10-7.226. Nallapareddy SR, Weinstock GM, Murray BE. Clinical isolates of En~erococcusfaecium exhibit strain-specific collagen binding mediated by Acm, a newmember of the MSCRAMM family. Mol Microbiol 2003; 47: 1733-47.227. Johnson AP. The pathogenicity of enterococci. J Antimicrob Chemother 1994;33: 1083-9.228. Jen BD, Jensen HG, Nordquist RE, Gilmore MS. Contribution of the pADI-encoded cytolysin to the severity of experimental Enteroco~.cus ,faecolisendophthalmitis. Infect Immun 1992; 60: 2445-52.229. Ike Y, Clewell DB. Evidence that the hemolysinibacteriocin phenotype ofEnrerococcus ,faecalis subsp. zymogenes can be determined by plasmids indifferent incompatibility groups as well as by the chromosome. J Bacterial1992; 174: 8172-7.230. Franz CM, Holzapfel WH, Stiles ME. Enterococci at the crossroads of foodsafety? Int J Food Microbiol 1999; 47: 1-24.231. Montgomerie JZ, Kalmanson GM, Harwick HJ. Guze LB. Relation betweenbacteriocin production and virulence of Streplococcus fuecalis var. liquefaciens.Proc Soc Exp Biol Med 1973; 144: 868-70.232. Galvez A, Valdivia E, Maqueda M, Montoya E. Production of bacteriocin-likesubstances by group D streptococci of human origin. Microbios 1985; 43: 223-32.233. Libertin CR, Dumitm R, Stein DS. The hemolysinibacteriocin produced byenterococci is a marker of pathogenicity. Diagn Microbiol Infect Dis 1992; 15:1 15-20.234. Vittal Prakash.P, Sambasiva Rao.R, Parija.SC. Incidence and Characterizationof Hemolysin/Bacteriocin as a determinant of Virulence among Enterococci. In:Abstract book of the XXVIl National Conference of Indian Association ofMedical Microbiologists, November 5th-9th, 2003, Mumbai, India.xxv

REFERENCES235. Ike Y, Hashimoto H, Clewell DB. High incidence of hemolysin production byEnterococcus (Streptococcus),faecalistrains associated with human parenteralinfections. J Clin Microbiol 1987; 25: 1524-8.236. Huycke MM, Gilmore MS. Frequency of aggregation substance and cytolysingenes among enterococcal endocarditis isolates. Plasmid 1995; 34: 152-6.237. Dupont H, Montravers P, Mohler J, Carbon C. Disparate findings on the role ofvirulence factors of Enterococcus faecalis in mouse and rat models ofperitonitis. Infect lmmun 1998; 66: 2570-5.238. Engelben M, Mylonakis E, Ausubel FM, Calderwood SB, Gilmore MS.Contribution of gelatinase, serine protease, and fsr to the pathogenesis ofEnterococczt.s,faeculis endophthalmitis. Infect Immun 2004; 72: 3628-33239. Roberts JC, Singh KV, Okhuysen PC, Murray BE. Molecular epidemiology ofthe fsr locus and of gelatinase production among different subsets ofEnterococcus,faecalis isolates. J Clin Microbiol 2004; 42: 2317-20.240. Kanemitsu K, Nishino T. Kunishima H, Okamura N, Takemura H, YamamotoH, et al. Quantitative determination of gelatinase activity among enterococci. JMicrobiol Methods 2001 ; 47: 1 1-6.241. Creti R, lmperi M, Bertuccini L, Fabretti F. Orefici G, Di Rosa R, et al. Surveyfor virulence determinants among Enterococcus faecalis isolated from differentsources. J Med Microbiol 2004; 53: 13-20.242. Huebner J, Quaas A, Krueger WA, Goldmann DA, Pier GB. Prophylactic andtherapeutic efficacy of antibodies to a capsular polysaccharide shared amongvancomycin-sensitive and -resistant enterococci. Infect lmmun 2000; 68: 4631 -6.243. Arduino RC, Murray BE, Rakita RM. Roles of antibodies and complement inphagocytic killing of enterococci. Infect Immun 1994; 62: 987-93.244. Gentry-Weeks CR, Karkhoff-Schweizer R, Pikis A, Estay M, Keith JM.Survival of Enterococcus ,faeculis in mouse peritoneal macrophages. InfectImmun 1999; 67: 2 160-5.

REFERENCES245. Marshall SH, Donskey CJ, Hutton-Thomas R. Salata RA, Rice LB. Genedosage and linezolid resistance in Enterococcus ,faecium and Enterococcus.faeculis. Antimicrob Agents Chemother 2002; 46: 3334-6.246. Williamson R, le Bouguenec C, Gutmann L, Horaud T. One or two low affinitypenicillin-binding proteins may be responsible for the range of susceptibility ofEnterococcusfaecium to benzylpenicillin. J Gen Microbiol 1985; 131 : 1933-40.247. Williamson R, Gutmann L, Horaud T, Delbos F, Acar JF. Use of penicillinbindingproteins for the identification of enterococci. J Gen Microbiol 1986:132: 1929-37.248. Fontana R, Ligozzi M, Pittaluga F, Satta G. Intrinsic penicillin resistance inenterococci. Microb Drug Resist 1996; 2: 209-13.249. Moellering RC Jr, Wennersten C, Weinberg AN. Synergy of penicillin andgentamicin against Enterococci. J Infect Dis 1971; 124(S 124): 207.250. Calderwood SA, Wennersten C. Moellering RC Jr, Kunz LJ, Krogstad DJ.Resistance to six aminoglycosidic aminocyclitol antibiotics among enterococci:prevalence, evolution, and relationship to synergism with penicillin. AntimicrobAgents Chemother 1977; 12: 401 -5.251. Pepper K, Horaud T, Le Bouguenec C, de Cespedes G. Location of antibioticresistance markers in clinical isolates of Enterococcus fueca1i.s with similarantibiotypes. Antimicrob Agents Chemother 1987; 31 : 1394-402.252. Pepper K, Le Bouguenec C, de Cespedes G, Horaud T. Dispersal of a plasmidbornechloramphenicol resistance gene in streptococcal and enterococcalplasmids. Plasmid 1986; 16: 195-203.253. Korten V, Huang WM, Murray BE. Analysis by PCR and direct DNAsequencing of gyrA mutations associated with fluoroquinolone resistance inEnterococcus faecalis. Antimicrob Agents Chemother 1994; 38: 2091 -4.254. Horodniceanu T, Bouyeleret L, El-Solh N, Bieth G, Delbos F. High-level,plasmid-borne resistance to gentamicin in Streptococcus faeculis subsp.zymogenes. Antimicrob Agents Chemother 1979; 16: 686-9.255. Ferretti JJ, Gilmore KS, Courvalin P. Nucleotide sequence analysis of the genespecifying the bifunctional 6'-aminoglycoside acetyltransferase 2"-

REFERENCESaminoglycoside phosphotransferase enzyme in Streptococcus faecalisidentification and cloning of gene regions specifying the two activities. JBacterial. 1986; 167: 631-8.256. Eliopoulos GM, Farber BF, Murray BE, Wennersten C, Moellering RC Jr.Ribosomalandresistance of clinical enterococcal to streptomycin isolates.Antimicrob Agents Chemother 1984; 25: 398-9.257. Chow JW, Zervos MJ, Lerner SA, Thal LA, Donabedian SM, Jaworski DD, etal. A novel gentamicin resistance gene in Enterococcus. Antimicrob AgentsChemother 1997; 41: 51 1-4.258. Murray BE, An FY, Clewell DB. Plasmids and pheromone response of the beta-lactamase producer Streptococc~~s (Enterococcus),fuecalis HH22. AntimicrobAgents Chemother 1988; 32: 547-5 1.259. Murray BE, Singh KV, Markowitz SM, Lopardo HA, Patterson JE. Zervos MJ,et al. Evidence for clonal spread of' a single strain of beta-lactamase-producingEnterococeu~ (Streptococcus) fuccalis to six hospitals in five states. J Infect Dis1991; 163: 780-5.260. Murray BE, Lopardo HA. Rubeglio EA, Frosolono M, Singh KV. lntrahospitalspread of a single gentamicin-resistant, beta-lactamase-producing strain ofEnlerococcus .foeculis in Argentina. Antimicrob Agents Chemother 1992; 36:230-2.261. Rice LB, Eliopoulos GM, Wemersten C, Goldmann D. Jacoby GA, MoelleringRC Jr. Chromosomally mediated beta-lactamase production and gentamicinresistance in Enterococcus fuecalis. Antimicrob Agents Chemother 1991; 5:272-6.262. Mainardi JL, Legrand R, Arthur M, Schoot B, van Heijenoon J, Gutmann L.Novel mechanism of beta-lactam resistance due to bypass of DD-transpeptidation in Enterococcusjaecium. J Biol Chem 2000; 275: 16490-6.263. Uttley AH, Collins CH, Naidoo J, George RC. Vancomycin-resistantenterococci. Lancet 1988; I: 57-8.

REFERENCES264. Leclercq R, Derlot E, Duval J, Courvalin P. Plasmid-mediated resistance tovancomycin and teicoplanin in Entrrococcus faecium. N Engl J Med 1988; 319:157-61.265. Arthur M, Courvalin P. Genetics and mechanisms of glycopeptide resistance inenterococci. Antimicrob Agents Chemother 1993; 37: 1563-71.266. McKessar SJ, Berry AM, Bell JM, Tumidge JD, Paton JC. Geneticcharacterization of vane, a novel vancomycin resistance locus of Enterococcu.~,faecalis. Antimicrob Agents Chemother 2000; 44: 3224-8.267. Noble WC, Virani 2, Cree RG. Co-transfer of vancomycin and other resistancegenes from Enlerococcus jaeca1i.s NCTC 12201 to S~aphy1ococcu.v uurclrs.FEMS Microbiol Lett 1992; 72: 195-8.268. Stuphy1ococcu.s aureus resistant to vancomycin--United States, 2002. MMWRMorb Mortal Wkly Rep 2002; 51: 565-7.269. Swenson JM, Ferraro MJ, Sahm DF, Clark NC, Culver DH, Tenover FC.Multilaboratory evaluation of screening methods for detection of high-levelaminoglycoside resistance in enterococci. National Committee for ClinicalLaboratory Standards Study Group on Enterococci. J Clin Microbiol 1995: 33:3008-1 8.270. Swenson JM, Hill BC, Thornsbeny C. Problems with the disk diffusion test fordetection of vancomycin resistance in enterococci. J Clin Microbiol 1989; 27:2140-2.271. Swenson JM, Ferraro MJ, Sahm DF, Charache P, Tenover FC. Newvancomycin disk diffusion breakpoints for enterococci. The NationalCommittee for Clinical Laboratory Standards Working Group on Enterococci. JClin Microbiol 1992; 30: 2525-8.272. Kuhn I, Burman LO, Haeggman S, Tullus K, Murray BE. Biochemicalfingerprinting compared with ribotyping and pulsed-field gel electrophoresis ofDNA for epidemiological typing of enterococci. J Clin Microbiol 1995; 33:2812-7.

REFERENCES273. Coudron PE, Mayhall CG, Facklam RR, Spadora AC, Lamb VA, Lybrand MR.Dalton HP. Streptococcusfuecium outbreak in a neonatal intensive care unit. JClin Microbiol 1984; 20: 1044-8.274. Nischal M; Macaden R. Biochemical speciation and haemolytic activity inenterococci. lndian Journal of Medical Microbiology 1996; 14: 205-8.275. Gulati V, Agganval A, Khanna S, Narang VK. Biochemical speciation ofenterococci causing human infections. Indian J Med Sci 1997; 51 : 310-2.276. Handwerger S, Raucher B, Altarac D, Monka J, Marchione S, Singh KV, et al.Nosocomial outbreak due to Enterococcus faecilrm highly resistant tovancomycin, penicillin, and gentamicin. Clin Infect Dis 1993; 16: 750-5.277. Tomayko JF, Zscheck KK, Singh KV, Murray BE. Comparison of the betalactamasegene cluster in clonally distinct strains of Enierococcus fueculis.Antimicrob Agents Chemother 1996; 40: 1170-4.278. Zervos MJ, Mikesell TS, Schaberg DR. Heterogeneity of plasmids determininghigh-level resistance to gentamicin in clinical isolates of Streptococcus faecalis.Antimicrob Agents Chemother 1986; 30: 78-81.279. Murray BE. Singh KV, Heath JD, Sharma BR, Weinstock GM. Comparison ofgenomic DNAs of different enterococcal isolates using restrictionendonucleases with infrequent recognition sites. J Clin Microbiol 1990; 28:2059-63.280. Seetulsingh PS, Tomayko JF, Coudron PE, Markowitz SM, Skinner C. SinghKV, et al. Chromosomal DNA restriction endonuclease digestion patterns ofbeta-lactamase-producing Enterococcu,s,faecuIis isolates collected from a singlehospital over a 7-year period. J Clin Microbiol 1996; 34: 1892-6.281. Malathum K, Singh KV, Weinstock GM, Murray BE. Repetitive sequencebasedPCR versus pulsed-field gel electrophoresis for typing of Enterococcusfaecalis at the subspecies level. J Clin Microbiol 1998; 36: 21 1-5.282. van den Braak N, Power E, Anthony R, Endtz HP, Verbmgh HA, van BelkumA. Random amplification of polymorphic DNA versus pulsed field gelelectrophoresis of Smal DNA macrorestriction fragments for typing strains ofvancomycin-resistant enterococci. FEMS Microbiol Len 2000; 192: 45-52.

REFERENCES283. Jureen R, Harthug S, Sornes S, Digranes A, Willems RJ, Langeland N.Comparative analysis of amplified fragment length polymorphism and pulsedfield gel electrophoresis in a hosp~tal outbreak and subsequent endemicity ofampicillin-resistant Enterococcus fuecium. FEMS Immunol Med Microbiol2004; 40: 33-9.284. Mayer LW. Use of plasmid profiles in epidemiologic surveillance of diseaseoutbreaks and in tracing the transmission of antibiotic resistance. Clin MicrobiolRev 1988; 1: 228-43.285. Wanger AR, Murray BE. Comparison of enterococcal and staphylococcal betalactamaseplasmids. J Infect Dis 1990: 161: 54-8.286. Donabedian SM, Chow JW, Boyce JM, McCabe RE, Markowitz SM, CoudronPE, et al. Molecular typing of ampicillin-resistant, non-beta-lactamaseproducingEnterococcus,fuecium isolates from diverse geographic areas. .I ClinMicrobiol 1992; 30: 2757-61.287. Simjee S, Fraise AP, Gill MJ. Plasmid heterogeneity and identification of aTn5281-like element in clinical isolates of high-level gentamicin-resistantEnterococctis ,fuccium isolated in the LJK. .I Antimicrob Chemother 1999; 43:625-35.288. Miranda G, Lee L, Kelly C, Solorzano F, Leanos B, Munoz 0, et al.Antimicrobial resistance from enterococci in a pediatric hospital. Plasmids inEnterococcus faeculi.~ isolates with high-level gentamicin and streptomycinresistance. Arch Med Res 2001; 32: 159-63.289. Watanabe T, Kumata H, Sasamoto M, Shimizu-Kodota M. The distribution ofhomologous enterococcal plasmid DNA sequences in human faecal isolates. JAppl Bacterial 1992; 73: 131-5.290. Werner G, Hildebrandt B, Klare I, Witte W. Linkage of determinants forstreptogramin A, macrolide-lincosamide-streptogramin B, and chloramphenicolresistance on a conjugative plasmid in Enterococcusfaecium and disseminationof this cluster among streptogramin-resistant enterococci. Int J Med Microbiol2000; 290: 543-8.

REFERENCES291. Bedendo J, Pignatari AC. Typing of Enterococcus,fbecium by polymerase chainreaction and pulsed field gel electrophoresis. Braz J Med Biol Res 2000; 33:1269-74.292. Papaparaskevas J, Vatopoulos A, Tassios PT, Avlami A, Legakis NJ,Kalapothaki V. Diversity among high-level aminoglycoside-resistantenterococci. J Antimicrob Chemother 2000; 45: 277-83.293. Werner G, Willems RJ, Hildebrandt B, Klare I. Witte W. Influence oftransferable genetic deteninants on the outcome of typing methods commonlyused for Enterococcus,faecium. J Clin Microbiol 2003; 41: 1499-506.294. Tenover FC, Arbeit RD, Goering RV, Mickelsen PA, Murray BE, Persing DH,et al. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol1995; 33: 2233-9.295. Thal L. Donabedian S, Robinson-Dunn B. Chow JW, Dembry L, Clewell DB, etal. Molecular analysis of glycopeptide-resistant Enterococcu.~ ,faecium isolatescollected from Michigan hospitals over a 6-year period. J Clin Microbiol 1998;36: 3303-8.296, de Lencastre H. Severina EP, Roberts RB, Kreiswinh BN. Tomasz A. Testingthe efficacy of a molecular surveillance network: methicillin-resistantSraphvlococcus aurrus (MRSA) and vancomycin-resistant Ettterococcrrs,faecium (VREF) genotypes in six hospitals in the metropolitan New York Cityarea. The BARG Initiative Pilot Study Group. Bacterial Antibiotic ResistanceGroup. Microb Drug Resist 1996; 2: 343-5 1.297. Krawczyk B, Lewandowski K, Bronk M, Samet A, Myjak P, Kur J. Evaluationof a novel method based on amplification of DNA fragments surrounding rarerestriction sites (ADSRRS fingerprinting) for typing strains of vancomycin-resistant Enterococcus faecium. J Microbiol Methods 2003; 52: 341-51.298. Thapliyal DC, Ahluwalia SS, Sethi MS. Negi SK. Statistical correlationsbetween total coliform and Enterococcus counts in water supply of Tarai region.Indian J Med Res 1970; 58: 840-4.

REFERENCES299. Grover PS, Thakur K. Shima drinking water--a bacteriological analysis. JCommun Dis 2001; 33: 44-52.300. Saikia PK, Duna GN, Devriese LA, Kalita CC. Characterization andantimicrobial susceptibility of Enterococcus species from the intestines of ducksin Assam. Res Vet Sci 1995; 58: 288-9.301. Shakuntala I, Singh HL, Singh NG B, Singh Y1. Prevalence of Streptococcusagalactiae in milk from untreated healthy cows and those with mastitis in andaround Imphal, Manipur. Indian Journal of Medical Microbiology 1999; 17: 34-6.302. Arora S, Tyagi SC. Significant bacteriuria in apparently healthy young adults.Indian J Pathol Bacterial 1974; 17: 39-44.303. Mishra JN, Rai MG, Chakraborty S, Prasad S. Study of neonatal septicemia.lndian Pediatr 1985; 22: 281 -5.304. Chattopadhyay UK, Deb Ray AK, Chakraborty P. Pal D. Study of the efficacyof different root canal disinfectants in controlling microbes in non vital singlerooted teeth after biomechanical preparation. lndian Joumal of MedicalMicrobiology 2000; 18: 119-21.305. Hiremath S L, Kanta R C, Yeshwanthrao M, Vasantha Kumar C M. Aerobicbacterial isolates of CSOM and their antibiotic Sensitivity Pattern. lndianPractitioner 2001; 54: 486-9.306. Parvathi S, Appalaraju B. Isolation, characterisation and antibiogram ofEnterococci from clinical samples. lndian J Pathol Microbiol2003; 46: 501-3.307. Viswanathan R. Enterococcus faecalis in conjunctival swab from a diabeticpatient. Bombay Hospital Journal 2003; 45: 473-4.308. Tripathi N, Anupurba S, Aganval BV, Nityanand Tripathi, Shampa Anupurba.A clinical and aetiological study of infective endocarditis. Annals of theNational Academy of Medical Sciences (India) 1996; 32: 193-8.309. Khanal B, Harish BN, Sethuraman KR. Endocarditis caused by high levelgentamicin resistant enterococci: a case report. lndian Journal of MedicalMicrobiology 1998; 16: 41 -2.

REFERENCES310. Garg N, Kandpal B, Garg N, Tewan S, Kapoor A, Goel P, et al. Characteristicsof infective endocarditis in a developing country-clinical profile and outcome in192 lndian patients, 1992-2001. Int J Cardiol 2005; 98: 253-60.31 1. Gupta V, Yadav A, Joshi RM. Antibiotic resistance pattern in Uropathogens.lndian Journal of Medical Microbiology 2002; 20: 96-85.312. Miskeen PA, Deodhar L. Studies on the incidence of Enterococcrts species inurinary tract infections, and their identification by a test scheme. BombayHospital Joumal 2001; 43: 124-7.313. Snod S, Kapil A, Das B, Rattan A. A five-year study of positive blood cultures:with emphasis on polymicrobial hacteraemia. lndian Practitioner 1997; 50: 575-0.314. Murthy R, Sengupta S. Maya N, Shivananda PG. Incidence of post operativewound infection and their antibiogram in a teaching and referral hospital. lndianJ Med Sci 1998; 52: 553-5.315. Vijaya D, Lakshmikanth. Sheshadri. Bacteriology of diabetic foot infection.Biomedicine 2000; 20: 176-9.316. Revathi G. Pun J, Jain BK. Bacteriology of bums. Bums 1998; 24: 347-9.317. Singh NP, Coyal R, Manchanda V. Das S, Kaur I, Talwar V. Changing trends inbacteriology of bums in the bums unit, Delhi, India. Bums 2003; 29: 129-32.318. Sharma RK. Kumar J, Gupta A, Gulati S. Peritoneal infection in acuteintermittent peritoneal dialysis. Ren Fail 2003; 25: 975-80.319. Fotedar R, Banejee U. Singh S, Shriniwas, Verma AK. The housefly (Muscadomesticu) as a canier of pathogenic microorganisms in a hospital environment.J Hosp Infect 1992; 20: 209- 15.320. Cherian BP, Mathai E. Chandy S. Determination of high-level resistance toaminoglycosides among enterococci. Indian J Med Res 1995; 102: 255-7.321. Jesudason MV, Pratima VL, Pandian R, Abigail S. Characterisation of penicillinresistant enterococci. Indian Joumal of Medical Microbiology 1998; 16: 16-8.322. Aganval VA, Jain Y1, Pathak AA. Concomitant high-level resistance topenicillin and aminoglycosides in enterococci at Nagpur, Central India. IndianJournal of Medical Microbiology 1999; 17: 85-7.XXXIV

REFERENCES323. Parvathi S, Appala Raja B. Comparative evaluation of beta lactarnaseproduction in enterococci by acidometric method and clover leaf technique.Indian Journal of Medical Microbiology 2000; 18: 122-4.324. Purva M, Chaudhary R, Dhawan B, Sharma N, Kumar L. Vancomycin-resistantEnterococcus bacteremia in a lymphoma patient. Indian Journal of MedicalMicrobiology 1999; 17: 194-5.325. Karmarkar MG, Gershom ES, Mehta PR. Enterococcal infections with specialreference to phenotypic characterization & drug resistance. Indian J Med Res2004; 1 19 (S): 22-5.326. Willey BM, Jones RN, McGeer A, Witte W, French G, Roberts RB, et al.Practical approach to the identification of' clinically relevant Enterococcrisspecies. Diagn Microbiol Infect Dis 1999; 34: 165-71.327. Sambrook J, Russell D. Molecular Cloning: A Laboratory Manual. 3rd ed. ColdSpring Harbor, NY: Cold Spring Harbor Laboratory Press; 2000.328. Low DE, Keller N, Barth A, Jones RN. Clinical prevalence, antimicrobialsusceptibility,. and geographic resistance patterns of enterococci: results fromthe SENTRY Antimicrobial Sun,eillance Program, 1997-1999. Clin lnfect Dis2001; 32 (S 2): S133-45.329. Jones RN, Sader HS, Erwin ME, Anderson SC. Emerging multiply resistantenterococci among clinical isolates. I. Prevalence data from 97 medical centersurveillance study in the United States. Enterococcus Study Group. DiagnMicrobiol Infect Dis 1995; 21: 85-93.330. Devi PS, Rao PS, Shivananda PG. Characterization, antibiotic susceptibilitypattern and detection of beta-lactamases in Enterococci. Indian J PatholMicrobiol 2002; 45: 79-82.33 1. Jean SS, Fang CT, Wang HK, Hsue PR, Chang SC, Luh KT. lnvasive infectionsdue to vancomycin-resistant enterococci in adult patients. J Microbiol ImmunolInfect 2001; 34: 281-6.332. Jones WG, Barie PS, Yurt RW, Goodwin CW. Enterococcal bum sepsis. Ahighly lethal complication in severely burned patients. Arch Surg 1986; 121:649-53.xxxv

REFERENCES333. Tsiodras S, Gold HS, Coakley EP, Wennersten C, Moellering RC Jr, EliopoulosGM. Diversity of domain V of 23s rRNA gene sequence in differentEnterococcus species. J Clin Microbiol 2000; 38: 3991-3.334. Gales AC, Sader HS, Jones RN; SENTRY Participants Group (Latin America).Urinary tract infection trends in Latin American hospitals: report from theSENTRY antimicrobial surveillance program (1997-2000). Diagn Microbiollnfect Dis 2002; 44: 289-99.335. Barisic 2, Punda-Polic V. Antibiotic resistance among enterococcal strainsisolated from clinical specimens. Lnt 3 Antimicrob Agents 2000; 16: 65-8.336. Hsueh PR, Chen ML. Sun CC, Chen WH, Pan HJ, Yang LS, et al.Antimicrobial drug resistance in pathogens causing nosocomial infections at auniversity hospital in Taiwan, 1981-1999. Ernerg lnfect Dis 2002; 8: 63-8.337. Mutnick AH, Biedenbach DJ, Jones RN. Geographic variations and trends inantimicrobial resistance among Enterococcus faecalis and Enlcrococcusfaecium in the SENTRY Antimicrobial Surveillance Program (1997-2000).Diagn Microbiol lnfect Dis 2003; 46: 63-8.338. Louie M, Simor AE, Szeto S, Patel M. Kreiswirth B, Low DE. Susceptibilitytesting of clinical isolates of Enterococcu.sfaecium and Enterococcus,faecalis. JClin Microbiol 1992; 30: 41-5.339. Li JT, Li Y, Wang J; China Bacterial Resistance Surveillance Study Group.Surveillance on gram-positive bacteria isolated from patients with hospitalacquired infections or community acquired infections. Zhonghua Yi Xue Za Zhi2003: 83: 365-74.340. Udo EE, Al-Sweih N, John P, Chugh TD. Antibiotic resistance of enterococciisolated at a teaching hospital in Kuwait. Diabm Microbiol Infect Dis 2002; 43:233-8.341. Levison ME, Mallela S. Increasing Antimicrobial Resistance: Therapeuticimplications for Enterococcal Infections. Curr lnfect Dis Rep 2000; 2: 417-423.342. Mederski-Samoraj BD, Murray BE. High-level resistance to gentamicin inclinical isolates of enterococci. J lnfect Dis 1983; 147: 751-7.XXXVI

REFERENCES343. Del Campo R, Tenorio C, Rubio C, Castillo J, Torres C, Gomez-Lus R.Aminoglycoside-modifying enzymes in high-level streptomycin and gentamicinresistant Enterococcus spp. in Spain. Int J Antimicrob Agents 2000; 15: 221-6.344. Randhawa VS, Kapoor L, Singh V, Mehta G. Aminoglycoside resistance inenterococci isolated from paediatric septicaemia in a tertiary care hospital innorth India. Indian J Med Res 2004; 119 (S): 77-9.345. Schouten MA, Voss A, Hoogkamp-Korstanje JA. Antimicrobial susceptibilitypatterns of enterococci causing infections in Europe. The European VRE StudyGroup. Antimicrob Agents Chemother 1999; 43: 2542-6.346. Tsakris A, Poumaras S, Maniatis AN, Douboyas J, Antoniadis A. Increasingprevalence of high-level gentamicin resistance among Enterococci isolated inGreece. Chemotherapy 2001; 47: 86-9.347. Schmitz FJ, Verhoef J, Fluit AC. Prevalence of aminoglycoside resistance in 20European university hospitals participating in the European SENTRYAntimicrobial Surveillance Programme. Eur J Clin Microbial Infect Dis 1999;18: 414-21.348. Teng LJ, Liaw SJ, Hsueh PR, Ho SW, Luh KT. Heterogeneity of resistanceelements in clinlcal lsolates of enterococci with high-level gentamicinresistance. J Fonnos Med Assoc 1998; 97: 855-9.349. Kobayashi N, Alam M, Nishimoto Y, Urasawa S, Uehara N, Watanabe N.Distribution of arninoglycoside resistance genes in recent clinical isolates ofEnrerococcns .faecalis, Enterococcus ,faecium and En~~.rococcus avium.Epidemiol Infect 2001; 126: 197-204.350. Simjee S. Manzoor SE, Fraise AP, Gill MJ. Nature of transposon-mediatedhigh-level gentamicin resistance in Enterococcus facculis isolated in the UnitedKingdom. J Antimicrob Chemother 2000; 45: 565-75.351. Kaufhold A, Podbielski A, Horaud T, Ferrieri P. Identical genes confer high-level resistance to gentamicin upon Enterococcus ,faecalis, Enterococcusfaecium, and Streptococcus agalactiae. Antimicrob Agents Chemother 1992;36: 1215-8.

REFERENCES352. van Den Braak N, van Belkum A, Kreft D, te Win R, Verbrugh HA, Endtz HP.The prevalence and clonal expansion of high-level gentamicin-resistantenterococci isolated from blood cultures in a Dutch university hospital. JAntimicrob Chemother 1999; 44: 795-8.353. Donabedian SM, Thal LA, Hershberger E, Perri MB, Chow JW, Barllett P, et al.Molecular characterization of gentamicin-resistant Enterococci in the UnitedStates: evidence of spread from animals to humans through food. J ClinMicrobiol2003; 41: 1109-13.354. Patterson JE, Masecar BL, Kauffman CA. Schaberg DR, Hierholzer WJ Jr,Zervos MJ. Gentarnicin resistance plasmids of enterococci from diversegeographic areas are heterogeneous. J Infect Dis 1988; 158: 212-6.355. Casetta A, Hoi AB, de Cespedes G, Horaud T. Diversity of structures carryingthe high-level gentamicin resistance gene (uac6-uph2) in En'nterococcus,/ueculisstrains isolated in France. Antimicrob Agents Chemother 1998; 42: 2889-92.356. Daikos GL, Bamias G, Kattamis C, Zervos MJ, Chow JW, Christakis G, et al.Structures, locations, and transfer frequencies of genetic elements conferringhigh-level gentamicin resistance in En'nterococcus ,/ueculis isolates in Greece.Antimicrob Agents Chemother 2003; 47: 3950-3.357. Woodford N, Morrison D. Cookson B, George RC. Comparison of high-levelgentamicin-resistant En~erococcus faecium isolates from different continents.Antimicrob Agents Chemother 1993; 37: 68 1-4.358. Thal LA, Chow JW, Patterson JE, Perri MB, Donabedian S, Clewell DB, et al.Molecular characterization of highly gentamicin-resistant Enterocvccus fuecalisisolates lacking high-level streptomycin resistance. Antimicrob AgentsChemother 1993; 37: 134-7.359. Straut M, de Cespedes G, Horaud T. Plasmid-borne high-level resistance togentamicin in Enterococcus hirae, Enterococcus avium, and Enterococcusraffinosus. Antimicrob Agents Chemother 1996; 40: 1263-5.360. Ounissi H, Derlot E, Carlier C, Countalin P. Gene homogeneity foraminoglycoside-modifying enzymes in gram-positive cocci. Antimicrob AgentsChemother 1990; 34: 2 164-8.

REFERENCES361. Trieu-Cuot P, Courvalin P. Nucleotide sequence of the Streptococcus /aecalisplasmid gene encoding the 3'5"-aminoglycoside phosphotransferase type 111.Gene 1983; 23: 331-41.362. Thal LA, Chow JW, Clewell DB, Zervos MJ. Tn924, a chromosome-bornetransposon encoding high-level gentamicin resistance in Enterocuccusfaecolis.Antimicrob Agents Chemother 1994; 38: 1152-6.363. Hall LM, Duke B, Urwin G, Guiney M. Epidemiology of Enterororcus fuecalisurinary tract infection in a teaching hospital in London, United Kingdom. J ClinMicrobiol 1992: 30: 1953-7.364. Harrington SM, Ross TL. Gebo KA, Merz WG. Vancomycin resistance. esp.strain relatedness: a I-year study of enterococcal bacteremia. J Clin Microbiol2004; 42: 5895-8.365. Sandoe JA, Witherden IR, Cove JH, Heritage J, Wilcox MH. Correlationbetween enterococcal biofilm formation in vitro and medical-device-relatedinfection potential in vivo. J Med Microbiol 2003; 52: 547-50.366. Waar K, van der Mei HC, Harmsen HJ. Degener JE. Busscher HJ. Enterococcus,faecoli.s surface proteins determine its adhesion mechanism to bile drainmaterials. Microbiology 2002; 148: 1863-70367. Pillai SK, Sakoulas G, Eliopoulos GM, Moellering RC Jr, Murray BE, InouyeRT. Effects of glucose on fsr-mediated biofilm formation in Enrerococcus.faccalis. J Infect Dis 3004; 190: 967-70.368. Franke AE, Clewell DB. Evidence for a chromosome-borne resistancetransposon (Tn916) in S. faecali,~ that is capable of "conjugal" transfer in theabsence of a conjugative plasmid. J Bacterial 1981; 145: 494-502.369. Hodel-Christian SL, Murray BE. Mobilization of the gentamicin resistance genein Enlerococcusfaecalis. Antimicrob Agents Chemother 1990; 34: 1278-80.370. Clewell DB, Francia MV, Flannagan SE, An FY. Enterococcal plasmid transfer:sex pheromones, transfer origins, relaxases, and the Staphylococcus aureusissue. Plasmid 2002; 48: 193-201.XXXIX

REFERENCES371. Hunter PR, Gaston MA. Numerical index of the discriminatory ability of typingsystems: an application of Simpson's index of diversity. J Clin Microbiol 1988;26: 2465-6.372. Gmndmann H, Hori S, Tanner G. Determining confidence intervals whenmeasuring genetic diversity and the discriminatory abilities of typing methodsfor microorganisms. J Clin Microbiol 2001; 39: 4190-2.373. Tomita H, Pierson C, Lim SK, Clewell DB, Ike Y. Possible connection betweena widely disseminated conjugative gentamicin resistance (pMG1-like) plasmidand the emergence of vancomycin resistance in Enterococcus fuecium. J ClinMicrobiol 2002: 40: 3326-33.374. Jones RN. The emergent needs for basic research, education, and surveillance ofantimicrobial resistance. Problems facing the report from the American Societyfor Microbiology Task Force on Antibiotic Resistance. Diagn Microbiol InfectDis 1996; 25: 153-61.

ADDENDUMBased on this thesis work a paper was published jn a peer-reviewed international journal,and five papers were presented in National and International conferences.Prakash VP, Rao SR, Parija SC. Emergence of unusual species of enterococcicausing infections, South India. BMC Infectious diseases. 2005; 5(1): 14.Vittal Prakash.P, Sambasiva Rao.R, Malay.K.Ray, Parija.SC. Plasmids and SexPheromone response of High Level Aminoglycoside Resistant enterococci. In:Abstracts of the I11 Winter Symposia on Infection and Immunity, ChristianMedical College, December 16th-17th, 2004, Vellore, India.Vittal Prakash.P, Sambasiva Rao.R, Malay.K.Ray, Parija.SC. Heterogeneity ofPlasmids among High Level Aminoglycoside Resistant Enterococci. In: Abstractsof the XXVIll National conference of lndian Association of MedicalMicrobiologists, November 25th-2Rth, 2004, Lucknow, India.Vittal Prakash.P, Sambasiva Rao.R, Parija.SC. Sex Pheromone Response ofPlasmids in High Level Aminoglycoside Resistant E.jlcculis. In: Abstracts of theXXVIIl National conference of Indian Association of Medical Microbiologists,November 25th-28th, 2004, Lucknow, India.Vittal Prakash.P, Sambasiva Rao.R, Parija.SC. Molecular Phenotyping ofAtypical Enterococcus species. In: Abstracts of the XXVII National conference ofIndian Association of Medical Microbiologists, November 5th-9th, 2003,Mumbai, India.Vittal Prakash.P, Sambasiva Rao.R, Parija.SC, Incidence and Characterization ofHemolysidBacteriocin as a determinant of Virulence among Enterococci. In:Abstracts of the XXVII National conference of Indian Association of MedicalMicrobiologists, November 5th-9th, 2003, Mumbai, India.N.B: Reprint ofthe Publication enclosed

1BbMd CentralResearch articleEmergence of unusual species of enterococci causing infections,South lndiaVittal P Prakash'l, Sambasiva R Rao'p2 and Subash C Parija'Addrs Ilepanmm~ ulMarobloiogy, iawahariai inaltuleof Poatlpaduaie Medliai tduralian and Rocarch (iI12MtRJ. Pond~chrm. indin and'Vae-Chancelior. NIR Ilnnrrnrj of Heahh Srlmra. V~~ayawada, lndtatmail Visni P Prakash' v~i~aiprakah@rsd~ilmail corn, Sambaa~va R Rnu- r.rarnhnn~ra$hn~ma~i rom, Suharh ( Paryn panlax@v$nl cornGirrmpandmg aulhorPublnhd I7 March lWSBMCInfmur D,swscr 2WS 5 I4 b, 10 li861i47i.13l4.5.14Rcte~rd 11 Dcrembr 1W4lWSThis tnlcie n anliable from http:lfwr* bromedrenml remlll7l-1334151140 1WS Pnkuh et 1. licrnrrr &oMrd Central Lcd'?I ran Ope- Zcnr ante a11,.o.tea .Wrr mM lens ol or Cma .c Cc-me* Am D.lo* . renrt*n :- p+r-'~ .nrair ((LC all DJI on 110 .eprcI.~! 3- n an! mM .F pro. P C cle 018 08 *o"L PIODP~.) (re4AbstractBackground: Enterococci tend to be one of the leading causes of nosocom~al infections, wiih E.foerde and E. foec~um accounting up 10 90% of the clin~cal solaces Nevertheless, the Incidence ofother speoes of enterococci from cilnical sources shows an aiarming increase with the propwilerof intrins~c resistance to several anub~otics including beta-lactams and glycopeptides. Thus properidentificat~on of enterococci to species level a qu~ntessenual for management and preventJon ofthese bacteria In any healthcare facility Hence rh~s work was undertaken to study the prevalenceof unusual species of enterococci causlng human ~nfecrrons, In a tenlary care hospital m South lndiaMethods: The study was conducted In a tertiary care hospid In South Ind~afromJuly 2001 w June2003, lwlates of enterococci were collected from various cllnicai specimens and speciated uslngextensive phenotyp~c and phpiolog~ul tesu. Antimlcrob~ai suscepdbilm/ tesung were performedand interpreted as per NCCLS guldellnes. Whole cell protein (WCP) (ingerprinung of enterococclwere done for specles validation by sodium dodecyl sulfate-polyacrylam~de gel eiecuophoresls(SDS-PAGE) and analyzd computauonally.Rasultr: Our study showed the prevalence of unusual (non.faecalis and non.faeclurn enterococcl)and atypical (biochemical variant) species of enterococcl as 19% (46 ~solates) and 5% (I 2 ~solates)respecuvely. The 7 unusual species (46 isolates) isolated and confirmed by phenotypiccharacterization includer. IS E, galliponrm (6.1%), I0 L anum (4.1%). 6 E rofftnosus (2 5%). 6 E hrme(2.5%), 4 E. mun1(1.7%), 3 E cosseliiovus~including the two aclpical isolates (1.1%) and 1 E. durons(0.R). The 12 atypical enterococcal species (5%) that showed aberrant sugar reactions inconvenuonal phenotyp~ng were confirmed as E foecolii E. foec~um and E. tosseliflaws respectivelyby WCP fingerprint~ng. The andmicrobial susceptibil~ty testlng depicted the emergence of high-leveiaminoglycoside and beta-lactam resistance among different species apart from inrrinslc nncomycinrerlstmce by some species, while all the species tested were susceptible for Ilnewltd andte~copbnin.Cmciucion: Our study reveals the mergence of multl.drug resistance moq unusual species ofenrtrococci posing a serious rhenpwtic challenge, Precise idenfiation of enterococci to specieslevel enables us w accur the spec~es-specific anumicrob~i reslmnce chantristics. apn fromknowing the epidemiological plmm md their clinical significance in huminhcdons.

Tabk I: An8lv1h d MC rbnp d umnual lpcln d nu-I.Spuesurud Anublmr No of trohtes a rpclfird MlC on lymL lute.%(no of ~saktu Tested1E mum (lo) Pan 9 9 9 9 7 6 10Amp 9 6 6 6 6 0 QVan 4 2 0 0 0 0 I0TI 0 0 0 0 0 0 IWVa kr N A N A N A N A N A N A IWHLGm N A NA N A N A N A N A 10HLSu N A N A N A N A N A N A SO--- - - - - ---E cmrd&vur Pen 0 0 0 0 0 0 1000)Amp o o o o o o I wVan 3 3 0 0 0 0 N ATe 0 0 0 0 0 0 I WVa ki N A N A N A N A N A NA 66 6HLGm N A NA N A N A N A N A I WHLSu N A N A N A N A N A----.- - - - - - ----N A 100t dumns (2) Pm 2 I I I 0 0 SOAmp I I I I o o sovan 2 I 0 0 0 0 100Te I 0 0 0 0 0 I WVa Scr NA N A N A N A N A N A SOHLGm N A N A NA N A N A NA SOHLStr N A N A N A NA NA N A 0---- - - --- - - - - - - -- ----- - - - -Epkiwrvm Rn 9 9 8 8 8 8 46 6(15)Amp 8 8 8 8 7 6 46 6Van I3 9 2 0 0 0 NATe 0 0 0 0 0 0 I WVa Scr N A N A N A N A NA N A 53 3HLGm N A N A NA N A N A N A 46 6HLSw N A NA N A N A N A N A 666---- -- -- --- - - - ---- - - - -E hlmt (6) Pen 3 3 2 2 0 0 66 6Amp 6 1 2 1 0 0 666vm 0 0 0 0 0 0 100Te 0 0 0 0 0 0 I WVI Srr N A N A N A N A N A NA IWHLGm N A N A N A N A N A NA IWHLStr N A NA N A N A N A N A I WE mundui (4) Pen I I I 0 0 0 100Amp. 0 0 0 0 0 0 100Van. 2 2 0 0 0 0 IWTe. 0 0 0 0 0 0 100Va.Scr. N A N A N A N A N A NA SOHLGm. N A N A N A N A N A N A 100HLSV. N A N A N A N A NA N A SO

T~bk I Anal@# of MlC ranfll of unwval species olmtemc~cl iConrm@- - - - - - - - - - - - --L mfFPsur Pen 4 4 4 4 4 4 33 I(61Amp 4 4 4 4 4 0 33 3Van 0 0 0 0 0 0 IM)TI 0 0 0 0 0 0 IM)Va Scr N A N A N A N A N A N A 100HLGm N A N A N A N A N A N A 66 6HLSv N A N A N A N A N A N A 66 6- - -- .- - .- - - - - - ---- - - - .-NOTE: '21ntrpmruonr had on NtCLS gu~deitoes.NA.m rpplsabla Susc Surrcptlblap.Pen.Psa

BMC lnfediws Diseases 2005.5:14h~p:llmvw.biaMd~.~1471-233(IY14Dice coefficientWCP-PAGE JPMEraffinosus SS- 1278durans SS- 1225porclnosusSS-1505pseudoav. 88-1277mundtii 85.1233aviumdisparSS-817SS-1295gallinarum SS-1228ranihiraeSs-1494SS.1227malodorat. 85-1 226faecalisfaecalisfaecalisfaecalisfaecalisMNVMNVMNVMNVMNVfaecalis 85-1 273faecalisfaecalisANVMNVcasselifla. ANVcasselif/a. ANVcasselifla. SS-1299faecium SS- 1274faeciumfaeciumfaeciumMNVMNVMNVFigure ICluster analysis of aypiul strains of Enceracacu uslng Dlce coefficient and UPGMA method (B~onumaics, Applld Matha. &I.giurn). Note : 55- Designation of CDC sundard smins. L porcinosus is currently designated as L vUlowm. L preuh.. L pcewdwnum: E maMarm- E malodamtus. L cmsdia.. E casse~ws. MNV Mannkol negative varianc ANV- Arginine negativevarianr

BMC Infectious Diseases 2005,6:14Mtp:llwww.biomedcenhal.mm/l471-233415114ited 100% susceptibility to ciprofloxacin None of the 46isolates was positive for B-laaamase, but resistance forblactamagents were prevalent variably among differentspec~es 7he results of MlCs for penicillin, ampicillin.h~gh-level gentamic~n and high-level streptomycin resist.ante were tn accordance with the disk diffusion testingresults except for vancomycin Dlsk diffusion testingshowed vancomyrin resistance for 6 isolates (I E. durans,2 E mundrii, 3 E gallinarum), but the agar screeningmethod exhibtted vancomycin resistance for I I isolates (2E mundti~, I f casse1,flovus. 1 E durans. 7 E, galli.narum)(including 5 isolates4 E. gall~narum and 1 E. car.selifltvur, which showed susceptibility to vancomycln bythe disk diffusion method)DiscussionOur study reveals that the prevalence of unusual species ofenterococci as 19% in our clinical setup in South IndiaMany studies and reviews show the prevalence of non-faecalisandnon-faeciumenterococcias 2-10% 13.6,121. Previousstudies from lnd~a have reponed E, faecalis and E.faecium as the only prevalent species 113.161, which maynot reflect the true Incidence rate From our perspectivethe real incidence tends to be higher whlch in pan can beexplained as. m~sidentificat~on of species due to exhibltlonof aberrant sugar reactions by some enterococci or,due to lack of applicat~on of the complete range of tests toidentify non-faecalis and non-faecium enterococci 17.171The prevalence rate (19%) of our study was panly inaccordance with another Indian study 1181 showing14.8% (exciuding t foeralis and E, faec~um) prevalence ofunusual species of enterococci from catheterized patientswlth unnarj tract ~nfections. E, mundt~~ and E, durans wennot reponed in their study, whose prwalence was 1 7%and 0.8% respectively in our study. E. gall~narum (6 2%)and E, amurn (4 1%) were the most commonly identifiedspecies, which markedly dlffers In isolation rate (0.3-1 2%) from other studies 16,19,201. The incidence ofinfections caused by unusual enterococcal species is ofsenous concern, since 43.5% of the isolates were fromcases of septicemia without endocarditis. Apan from septicemia,the unusual species of enterococci were isolatedfrequently from cases of urinaly tract infections, surgicaland non-surgical wound infections and peritonitis. Mostof the patients with the bloodstream infertions had aperipheral or central catheter Further, only 13% of ente.rococcal infections were polymicrobial, with majorityfrom non-bloodsueam isolates that underscores the clini.cal significance of the8e unusual enterococcal spedes.Although the unusual species of enterococci were isolatedat regular intervals throughout our study period, we couldfind clustering of specific species during a specific timeperiod from specific unitslwards. Interestingly, 10 amongthe 15 isolates of E. gallmarum isolated during our studyperiod were from pediatrics unit, while 7 ofthe 10 isolatesexhibiting a similar antibiotype were isolated from thesame ward w~thin a span of 2 months The remaining 3 ofthe 10 E gallinarum were isolated from the same ward inthe preceding 3 months, one of which showed an antibi.otype similar to the cluster of7 isolates. fie same was thecase of 3 E. casreltflavur isolated from the same pediatric5un~t within a span of2 months in the preceding year. Mostof these (8 of 10 E gnlltnarum, and all 3 E, tnssel~flt~w)isolates were from cases of septicemia, Although molecu.lar epidem~ological studies have not been done to comparethe genetic similarities of these isolates, the datadepicts the nosocomial spread ofthese species WCP analysisby SDS.PAGE had been proven to asslst in validatingthe species tdcntities as well, to identify strains that do notexhibit phenotypic characteristics identical to the typestrains of each species 14,10.2 1 ) We were able to valtdatethe authent~rity of the unusual species, and the exact taxonomicstatus of the atypical phenotypic variant strainsidentified by convent~onal btochemical testing as shownin Figure.1, using WCP fingerprinting by SDS-PAGE.Ciprofloxacin resistance was 63% among isolates (except.ing E, carrel~falvus) which proves that it may be successfulonly in treatingenterococcal urinary tract lnfect~ons I IIj,stnce most of our ~solates were from bloodstream andother related specimens. None of the isolates were p-lacta.mas? producer, but pen~cill~n and amp~cill~n resistancewere exhibited by 54.3% and 45 7% isolates. We suggestpeniclllln bind~ng protein modification based resistancefor our isolates as a basis lor p.lactam resinance, asdepicted previously 122.231, but markedly differs fromother Indian studies 115.241 showing up to 50% 8-laaa.mase assoctated resistance. The prevalence of htgh-levelgentamicin resistance (43 4%) and high-level streptomy.tin (37%) among unusual enterococcal isolates from ourstudy partially correlates with studies from japan 1251 andUntted States [12,261. In our study. most stralns withhigh-level gentaniicin resistance lacked high-level streptomycinresistance, and vice versa, thus facilitating the com.bination therapy (cell wall inhibitor plusaminoglycos~de) treaunent options for serious enterococ.cal bloodstream infections Il,l] The prevalenceofvanco.mycin resistance was 24% by agar screening /agar dilutionmethod and 13% by disk diffusion Ihe difference may beattributed to the intrinsic low level vancomycln resistance(van C genotype), exhibited by 4 E galltnanrm and. 1 Ecardiflaws isolates, which may go undetected by disk dif.fusion testing 1271. Of vrious concern was the low-levelvancomydn resistance exhibited by one E. duranr and twoE. mundrii (MIC 5 6 pglml) The genotypic basis ofvancomycinresistance for these 3 isolates yet to be studied, willgive us a definitive picture regarding its clinical significance,since studies have reponed the prevalence of vancomycinresistance in these two species, and itstransferable nature from E. durans to E. fuctum 128-301.

Mon mt (or &teedon b( vmcomvdn rnhmea inencemcord.JUm M M 1989.7.1:2110-118 Green M. &&don K, Mrhclr M: Recovw of nncomvdn-mdpbno., &Mzmbo 199.. 2):1501.6I9 CWCemW E JN I. hopouol CT Ran .G. lrmmq nD, Ma.lr ng RC Jr E monor GM Chruhrlutksn of nrromydnmmnce In ~ O C dmns., M 4nDmoob Cnanwrr 1995.1b:Bl.-5Pre-publication historyThe pre-publ~cation history for this paper can be accessedhere:Publlrh with BloMedCenhal and everyI scientist can read your nark h e of charge'BmMCdCenvll wdl b themor1 ngnlficant dmlopment fordrucm~nrtingthe reule of b,omd~~rlmrrrth~n wr Bfetlme 'Sir PIYI Num Umer RCWU~ UKYour rerearch papers will be- av81labk he ofchugcta the mtts b~omrd~olcommunltypm r~lmrd andpubl~shadlmmad~ately upon Kcapnrecried In PubMadmdrrchW rn PubMdCemnlpun - p u kw tlw cc$y!qhtlubmeyeur manuxnpt h.nBloMcdcentralIPage 8 of 8(w nunan nor w dam WUII