Comparison of RAPDs, AFLPs and SSR markers for the genetic ...

Biosystems,Foster City,California,USA). PCR1
reaction mix contained 5 pmol of ScAAT2 primer pair
fluorescent-dye labeled with HEX (yellow),5 pmol of
ScAAT3 primer pair fluorescent-dye labeled with 6-
FAM (blue),7.5 pmol of ScAAT5 primer pair fluorescent-dye
labeled with TET (green),and 1 U of Taq
DNA polymerase (Biotools,Biotechnological and Medical
Laboratories,S.A.,Madrid,Spain) in 1 reaction
buffer (75 mM Tris HCl pH 9.0,50 mM KCl,20 mM
(NH4)2SO4). PCR2 reaction mix contained 2.5 pmol of
ScAAT4 primer pair fluorescent-dye labeled with TET,
2.5 pmol of ScAAT6 primer pair fluorescent-dye labeled
with HEX,10 pmol of ScAAT1 primer pair fluorescentdye
labeled with 6-FAM,and 1.5 U of Taq DNA
polymerase.
The amplification reactions were carried out using a
DNA Thermal Cycler 9600 (Applied Biosystems) under
the following conditions: a preliminary step of 5 min at
94 1C; 10 cycles each of 15 s at 94 1C,30 s at 58 1C
(decreasing 1 1C per cycle to reach a final temperature of
47 1C),and 30 s at 72 1C; 25 cycles each of 15 s at 94 1C,
30 s at 48 1C,and 30 s at 72 1C; and a final step of 5 min
at 72 1C. Amplifications were confirmed by running
10 mL of the PCR product on 2.5% agarose gel.
Aliquots (1–2 mL) of the PCR product were mixed
with 12 mL of formamide and 0.5 mL of a red DNA size
standard (Genescan-500 ROX,Applied Biosystems).
Samples were denatured at 94 1C for 3 min prior to
separation by capillary electrophoresis at 15 KV for
24 min in an ABI Prism 310 DNA Sequencer (Applied
Biosystems),and subsequently analysed using Genescan
software (Applied Biosystems).
The reproducibility of the three analysis techniques
was assessed by comparing results from two additional
independent DNA extractions from half of the study
strains that were chosen randomly.
2.6. Data analysis
Data matrices were built based on the presence or
absence of amplification products. Genetic distances
were estimated from these matrices using Dice’s algorithm
(Dice,1945). This coefficient was used for
clustering data according to the UPGMA method.
Bootstrap resampling of 1000 replicates was performed
to test the robustness of the topology of the dendrograms.
Differences between dendrograms were tested by
generating cophenetic values for each dendrogram and
assembling a cophenetic matrix for each marker type.
The Mantel matrix correspondence test was then used to
compare cophenetic matrices (Mantel,1967). Computing
was performed using NTSYSpc software version 2.0
(Rohlf,1993) and TFPGA software version 1.3 (Miller,
1997).
In order to evaluate the usefulness of each marker
system,all of the following were calculated: the
ARTICLE IN PRESS
F. Javier Gallego et al. / Food Microbiology 22 (2005) 561–568 563
arithmetic mean of Diversity Index per polymorphic
locus (DIavp ¼ [1–Spi 2 ]/np,where p i is the allele
frequency for the i allele and np is the number of
polymorphic loci),the arithmetic mean of Effective
Number of Alleles per polymorphic locus (ENA avp ¼ [1/
Sp i 2 ]/np),the total number of effective alleles per
polymorphic locus (Ne ¼ sumat ENAavp),and the
Assay Efficiency Index (Ai ¼ Ne/P). Ai combines the
ENA identified per locus and the number of polymorphic
bands detected in each assay (P). For dominant
markers (RAPDs and AFLPs) where one of only two
states (+,present or ,absent) can be distinguished at
each position,we assumed that each band position
corresponded to a locus with two alleles: presence and
absence of the band.
3. Results
3.1. Efficiency of polymorphism detection
The level of polymorphism detected with each marker
system is summarized in Table 2,as well as a
comparison between systems.
A total of 155 bands were amplified by the 32
primers used in the RAPD analysis (Fig. 1). Eight
primers—OPB (01,06,07,12,13,15) and OPC (01,
03)—allowed the intraspecific differentiation of the
yeasts,with a total of 13 polymorphic bands. By
combining the electrophoretic profiles of seven primers
(OPB01,OPB06,OPB07,OPB13,OPB15,OPC01 and
OPC03),we were able to use these polymorphisms to
differentiate 13 of the 27 S. cerevisiae strains in this
study. The number of amplified bands varied between
primers,ranging between two and 10,with sizes from
400 up to 2000 bp. Duplicate analysis of random strains
revealed no significant differences in banding patterns,
although some bands varied in intensity.
Analysis of AFLP banding patterns (Fig. 2) yielded
137 DNA fragments,with sizes ranging from 130 to
410 bp. A total of 15 of these bands were polymorphic,
allowing the differentiation of 19 of the 27 strains in this
study. Information from four additional primer combinations
(MseI-CTT/EcoRI-AC, MseI-CTT/EcoRI-AT,
MseI-CAA/EcoRI-AC and MseI-CAA/EcoRI-AT) was
also used in the differentiation process. The repetition of
this AFLP analysis obtained similar results.
SSR analysis (Fig. 3) detected a number of alleles for
each locus,ranging from four (loci ScAAT5 and
ScAAT6) to 10 (loci ScAAT1 and ScAAT3),with 39
total alleles identified. Twenty of the 27 S. cerevisiae
strains in this study were differentiated. To complete the
SSR analysis,information from three microsatellites
was combined: ScAAT1,ScAAT3 and ScAAT4. Amplification
products varied in size from 165 to 445 bp.
Despite this allelic diversity,percent heterozygosity