12.07.2015 Views

Colletotrichum: complex species or species ... - CBS - KNAW

Colletotrichum: complex species or species ... - CBS - KNAW

Colletotrichum: complex species or species ... - CBS - KNAW

SHOW MORE
SHOW LESS
  • No tags were found...

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

The <strong>Colletotrichum</strong> gloeosp<strong>or</strong>ioides <strong>species</strong> <strong>complex</strong>RESULTSPhylogeneticsDNA sequences of five genes were obtained from all 158 isolatesincluded in the study and concatenated to f<strong>or</strong>m a supermatrix of2294 bp. The gene boundaries in the alignment were: ACT: 1–316,CAL: 317–1072, CHS-1: 1073–1371, GAPDH: 1372–1679, ITS:1680–2294. A BI analysis of the concatenated dataset is presentedin Fig.1. This tree is annotated with the <strong>species</strong> boundaries of thetaxa that we accept in the C. gloeosp<strong>or</strong>ioides <strong>complex</strong>, and theclades representing these taxa f<strong>or</strong>med the basis f<strong>or</strong> investigatingthe m<strong>or</strong>phological and biological diversity of our <strong>species</strong>. Ex-typeand authentic isolates are highlighted in bold and labelled with thenames under which they were <strong>or</strong>iginally described. The posteri<strong>or</strong>probability (PP) supp<strong>or</strong>t f<strong>or</strong> the grouping of most <strong>species</strong> rangesfrom 1 to 0.96, however supp<strong>or</strong>t f<strong>or</strong> deeper nodes is often lower,e.g. 0.53 f<strong>or</strong> the root of C. ti and C. aotearoa, indicating that thebranching may be uncertain f<strong>or</strong> the root of these <strong>species</strong>. Branchlengths and node PP are typically lower within a <strong>species</strong> thanbetween <strong>species</strong>.The large number of taxa in Fig. 1 makes it difficult to visualisethe interspecific genetic distance between the recognised <strong>species</strong>.The unrooted tree in Fig. 2 represents the results of a BI analysisbased on a concatenation of all eight genes, but restricted to theex-type <strong>or</strong> authentic cultures from each of the accepted taxa. Theanalysis was done without out-group taxa and clearly shows twoclusters of closely related <strong>species</strong> that we inf<strong>or</strong>mally label theMusae clade, and the Kahawae clade.To better resolve relationships within the Musae and Kahawaeclades a further set of BI analyses included eight genes and,wherever possible, several representative isolates of each of theaccepted <strong>species</strong>. All eight gene sequences were concatenatedto f<strong>or</strong>m a supermatrix f<strong>or</strong> each clade. F<strong>or</strong> the Musae clade of32 isolates the alignment was 4199 bp and the gene boundarieswere: ACT: 1–292, TUB2: 293–1008, CAL: 1009–1746, CHS-1:1747–2045, GAPDH: 2046–2331, GS: 2332–3238, ITS: 3239–3823, SOD2: 3824–4199. F<strong>or</strong> the Kahawae clade of 30 isolatesthe alignment was 4107 bp and the gene boundaries were: ACT:1–281, TUB2: 282–988, CAL: 989–1728, CHS-1: 1729–2027,GAPDH: 2028–2311, GS: 2312–3179, ITS: 3198–3733, SOD2:3734–4107. The additional genes sequenced provided additionalsupp<strong>or</strong>t f<strong>or</strong> our initial <strong>species</strong> delimitations with better resolutionf<strong>or</strong> some closely related <strong>species</strong>. F<strong>or</strong> example, the highlypathogenic coffee berry isolates (referred to here as C. kahawaesubsp. kahawae) were distinguished from other C. kahawaeisolates.Analyses based on concatenated data sets can maskincongruence between individual gene trees. The low levels ofsupp<strong>or</strong>t within some parts of the <strong>species</strong>-tree analysis (Fig. 3),in part reflects incongruence between gene trees. The levels ofsupp<strong>or</strong>t f<strong>or</strong> the Kahawae clade and f<strong>or</strong> the Musae clade are strong(PP=1) but the <strong>species</strong> that we accept within these clades havelower levels of supp<strong>or</strong>t than is shown between the other <strong>species</strong>outside of the clades. The scale bar in Fig. 3 represents a time scale,calibrated at zero f<strong>or</strong> the present day, and at 1 f<strong>or</strong> the last commonancest<strong>or</strong> (LCA) of the C. gloeosp<strong>or</strong>ioides and C. boninense <strong>species</strong><strong>complex</strong>es. The separate <strong>species</strong>-tree analyses f<strong>or</strong> the Musae andKahawae clades provide a finer resolution of evolutionary hist<strong>or</strong>ywithin each clade, the time scale based on the same calibrationas Fig. 3.The UPGMA-based ITS gene tree (Fig 6). shows that C.theobromicola, C. h<strong>or</strong>ii, C. gloeosp<strong>or</strong>ioides, G. cingulata “f. sp.camelliae”, C. asianum, C. musae, C. alatae, C. xanth<strong>or</strong>rhoeae allf<strong>or</strong>m monophyletic clades and may be distinguished with ITS, butmany <strong>species</strong> are unable to be discriminated using this gene alone.Note that C. c<strong>or</strong>dylinicola and C. psidii are represented by a singleisolate, meaning that variation within ITS sequences across these<strong>species</strong> has not been tested.M<strong>or</strong>phology and biologyBrief m<strong>or</strong>phological descriptions, based on all specimens examined,are provided f<strong>or</strong> only those <strong>species</strong> with no recently publisheddescription. Conidial sizes f<strong>or</strong> all accepted <strong>species</strong> are summarisedin Fig. 7. Within a <strong>species</strong>, conidial sizes are reasonably consistentacross isolates, independent of geographic <strong>or</strong>igin <strong>or</strong> host. However,differences between <strong>species</strong> are often slight and size ranges oftenoverlap (Fig. 7). The shape of appress<strong>or</strong>ia is generally consistentwithin a <strong>species</strong>, some being simple in outline, others <strong>complex</strong> andhighly lobed.Several <strong>species</strong> are characterised in part by the developmentof perithecia in culture. These include four <strong>species</strong> in the Musaeclade (C. alienum, C. fructicola, C. queenslandicum, and C.salsolae) and three in the Kahawae clade (C. clidemiae, C.kahawae subsp. ciggaro, and C. ti). Ascosp<strong>or</strong>e size and shapecan be a useful <strong>species</strong>-level diagnostic feature (Fig. 8). In most<strong>species</strong> the ascosp<strong>or</strong>es are strongly curved and typically taperingtowards the ends, but in C. clidemiae and C. ti, they are m<strong>or</strong>e <strong>or</strong>less elliptic with broadly rounded ends and not, <strong>or</strong> only slightly,curved. Individual isolates within any of these <strong>species</strong> may lose theability to f<strong>or</strong>m perithecia, perhaps associated with cultural changesduring st<strong>or</strong>age.Large, dark-walled stromatic structures are present in thecultures of some <strong>species</strong> not known to f<strong>or</strong>m perithecia. Oftenembedded in agar, less commonly on the surface <strong>or</strong> amongstthe aerial mycelium, these structures differ from perithecia incomprising a compact tissue of tightly tangled hyphae rather thanthe pseudoparenchymatous, angular cells typical of perithecialwalls. They have a soft, leathery texture compared to the m<strong>or</strong>ebrittle perithecia. These stromatic structures sometimes developa conidiogenous layer internally, and following the productionof conidia they may split open irregularly, folding back to f<strong>or</strong>m astromatic, acervulus-like structure. These kind of structures arealso f<strong>or</strong>med by some <strong>species</strong> in the C. boninense <strong>species</strong> <strong>complex</strong>(Damm et al. 2012b, this issue).The macroscopic appearance of the cultures is often highlydivergent within a <strong>species</strong> (e.g. C. fructicola and C. theobromicola),in most cases probably reflecting different st<strong>or</strong>age hist<strong>or</strong>ies of theisolates examined. Prolonged st<strong>or</strong>age, especially with repeatedsubculturing, results in staling of the cultures, the aerial myceliumoften becoming dense and unif<strong>or</strong>m in appearance and colour, anda loss of conidial and perithecial production, and variable in growthrate (Fig. 9). In some <strong>species</strong>, individual single ascosp<strong>or</strong>e <strong>or</strong> singleconidial isolates show two markedly different cultural types, seenotes under C. kahawae subsp. ciggaro.Some <strong>species</strong> appear to be host specialised, e.g. C. h<strong>or</strong>ii,C. kahawae subsp. kahawae, C. nupharicola, C. salsolae, C. ti,and C. xanth<strong>or</strong>rhoeae, but those most commonly isolated havebroad host and geographic ranges, e.g. C. fructicola, C. kahawaesubsp. ciggaro, C. siamense, and C. theobromicola. <strong>Colletotrichum</strong>gloeosp<strong>or</strong>ioides s. str. is commonly isolated from Citrus in manywww.studiesinmycology.<strong>or</strong>g125

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!