APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Dec. 1996, p. 4514–4520
Copyright ? 1996, American Society for Microbiology
Vol. 62, No. 12
PCR Differentiation of Commercial Yeast Strains Using Intron
Splice Site Primers
MIGUEL DE BARROS LOPES,1,2ALISON SODEN,1,3PAUL A. HENSCHKE,2AND PETER LANGRIDGE1*
Department of Plant Science1and Department of Horticulture, Viticulture and Oenology,3Waite Agricultural Research Institute,
The University of Adelaide, SA 5064, and The Australian Wine Research Institute,
Glen Osmond, SA 5064,2Australia
Received 26 June 1996/Accepted 22 September 1996
The increased use of pure starter cultures in the wine industry has made it necessary to develop a rapid and
simple identification system for yeast strains. A method based upon the PCR using oligonucleotide primers
that are complementary to intron splice sites has been developed. Since most introns are not essential for gene
function, introns have evolved with minimal constraint. By targeting these highly variable sequences, the PCR
has proved to be very effective in uncovering polymorphisms in commercial yeast strains. The speed of the
method and the ability to analyze many samples in a single day permit the monitoring of specific yeast strains
during fermentations. Furthermore, the simplicity of the technique, which does not require the isolation of
DNA, makes it accessible to industrial laboratories that have limited molecular expertise and resources.
In contrast to traditional wine-making in which the indige-
nous yeasts ferment the grape must, most modern wine-makers
inoculate with a pure culture of a selected yeast strain to
ensure a rapid, reliable, and predictable fermentation (29).
The availability of a range of commercial yeasts with different
wine-making characteristics and applications means that a num-
ber of yeast strains may be used in a single winery. It is gen-
erally assumed that indigenous yeasts are suppressed by the
competitive effect of addition of a high-density starter mo-
noculture, but studies show that indigenous yeast can still partic-
ipate in the fermentation (12, 26, 32). For these reasons, rapid
and simple methods for the routine verification of yeast strain
identity in starter cultures and fermentations would be useful.
Classical physiological methods are of limited use in the
identification of commercial wine-making yeasts. First, the vast
majority of the strains are of a single species, Saccharomyces
cerevisiae. A second difficulty is that many strains of industrial
importance are either diploid or aneuploid and therefore phe-
notypically wild type (3, 5). Furthermore, low sporulation fre-
quency, poor spore viability, and homozygosity for the ho-
mothallism gene HO leading to immediate diploidization of
meiotic segregants have made genetic analysis of these yeasts
difficult (2, 37). For these reasons, molecular methods have
been adopted for strain identification. Fatty acid composition
(22) and yeast protein fingerprinting (38) have been utilized
with some success, but these methods appear to be inadequate
for differentiating strains of the same species (27). In recent
years, the genetic diversity of commercial S. cerevisiae strains
has become apparent and has lead to a number of nucleic
acid-based identification techniques. The variability in the
chromosomal constitution of commercial yeast strains is con-
siderable, making chromosome karyotyping a useful method
(14, 27). Restriction fragment length polymorphism, which has
been an important development for mapping genes of agricul-
tural and medical importance, has also been successfully ap-
plied to yeast strain differentiation. Both the nuclear (7) and
mitochondrial (8) genomes have been analyzed by restriction
fragment length polymorphism. The potentials of these meth-
ods have been compared and discussed elsewhere (27).
The development of the PCR (31) has opened up new ave-
nues for yeast strain identification. The application of the PCR
for discrimination between wine yeast strains by the amplifi-
cation of random (28) or specific (16) target sequences has
recently been reported. In the present study, oligonucleotides
have been synthesized with sequences complementary to yeast
intron splice sites to target mutable regions of the genome.
With the use of these primers in the PCR, we are able to
differentiate between commercial yeasts commonly used in the
Australian wine industry. The method is rapid, reliable, and
simple, making the technique available to laboratories with
limited resources for verifying strain identity and monitoring
yeast growth during fermentations.
MATERIALS AND METHODS
Yeast strains and media. The yeast strains used in this study are listed in
Tables 1 and 2. All are strains of S. cerevisiae except AWRI 1034, which belongs
to the species Torulaspora delbrueckii. The 15Du strains (6) were obtained from
Steven Reed, Research Institute of Scripps Clinic, San Diego, Calif. The source
of yeast B431 is Brigalow Brewing Company, Slacks Creek, Queensland, Aus-
tralia. The source of K5088 is Cerebos Ltd., Seven Hills, New South Wales,
Australia. The other yeasts listed in Table 1 were obtained from the culture
collection at The Australian Wine Research Institute. In Table 2, AWRI 814 is
the original AWRI 729 yeast that was obtained from a dry yeast labelled Eper-
nay. Isolates of this yeast were sent to other culture collections and then returned
to The Australian Wine Research Institute at a later date (30). Yeasts were
grown on yeast extract-peptone-dextrose (YEPD) at 25?C by standard proce-
Preparation of DNA template for PCR. To obtain a DNA preparation method
suitable for the PCR, cells of strain AWRI 796 grown on YEPD plates and then
stored at 4?C were resuspended in either sterile water, 0.1% Triton X-100
(Sigma), or 500 ?g of Zymolyase 20T per ml (Seikagaku Corp., Tokyo, Japan).
The cells resuspended in water were then treated in one of the following ways:
(i) no treatment, (ii) boiled for 10 min, (iii) frozen in liquid nitrogen and then
boiled for 10 min, (iv) boiled for 10 min followed by freezing in liquid nitrogen,
or (v) same as method (iii) but followed by a second freeze-boil cycle. The cells
resuspended in Triton X-100 were boiled for 10 min, and the cells resuspended
in Zymolyase were incubated for 10 min at 37?C. A cell suspension of 2 ?l (5 ?
104cells) was used for each PCR amplification with primer EI1. The results were
compared with those obtained when purified DNA was used as a template.
Yeast DNA purification. Yeast DNA was isolated by standard procedures (1).
A cell suspension (10 ml) from an overnight culture grown in YEPD was resus-
pended in 200 ?l of breaking buffer (2% Triton X-100, 1% sodium dodecyl
sulfate, 100 mM NaCl, 10 mM Tris [pH 8], 1 mM EDTA [pH 8]). The yeast cells
were homogenized by vortexing for 3 min with glass beads (0.3 g) in the presence
of 200 ?l of phenol-chloroform-isoamyl alcohol. Tris-EDTA buffer (200 ?l) was
* Corresponding author. Phone: 61 8 303 7368. Fax: 61 8 303 7102.
Electronic mail address: firstname.lastname@example.org.
added, and the aqueous layer was collected after centrifugation. The DNA was
precipitated with ethanol. DNA concentrations were determined by measuring
the A260after incubation with RNase. For each PCR, 0.5 ?g of DNA was used.
Optimization of amplification. For all subsequent experiments, the DNA
template was prepared for PCR by freezing cells in liquid nitrogen followed by
10 min of boiling in a water bath.
The reproducibility of the method was tested with S. cerevisiae AWRI 796 and
the 15Du strains. Three isogenic 15Du strains that differ only at the mating locus
were analyzed. These strains were plated to single-colony density, and each
amplification reaction was performed with an individual colony.
For determining the effect of cell concentration on PCR, AWRI 796 cells were
resuspended in water at concentrations of 5 ? 102to 2.5 ? 105cells per ?l before
the freeze-boil treatment.
For comparing results of stationary-phase and dividing cells, yeast strain
AWRI 796 was grown for 24 h at 25?C on YEPD plates. These plates were then
stored at 4?C for either 0, 2, 4, or 8 h prior to preparing DNA template and PCR.
Intron splice site primers and PCR conditions. The following primers were
used in this study: EI1, CTGGCTTGGTGTATGT; EI2, CTGGCTTGCTA
CATAC; LA1, GCGACGGTGTACTAAC; LA2, CGTGCAGGTGTTAGTA.
The nucleotides that are conserved with the intron splice sites are underlined
(Fig. 1). Primers were used either singularly or in pairs. PCRs were performed in
50 ?l with 50 pmol of primer, 2 ?l of DNA template, 32 ?M of each deoxynucleo-
side triphosphate, 2.5 mM MgCl, and 0.2 U of Taq polymerase (Advanced
Biotech). The reactions were run for 33 cycles: denaturation at 94?C for 1 min,
annealing at 45?C for 2 min, and extension at 74?C for 1.5 min. An initial
denaturation for 3 min at 94?C and a final 5-min extension at 74?C were used.
Products of each amplification reaction were resolved on a 2% agarose gel
(Agarose NA; Pharmacia), stained with ethidium bromide, and visualized under
UV light. The gels were photographed with the Gel Cam Documentation System
(Sony), and the photographs were scanned to produce a computer image
(Hewlett-Packard Scan Jet 11CXIT).
Cloning and sequencing of amplified products. Amplified products from PCRs
were either gel isolated with a silica matrix (Geneclean II Kit; Bio 101) or
isolated by spin column chromatography (PCR Spinclean; Progen). The purified
fragments were then cloned into the pGEM-T vector (Promega), and the ligated
products were transformed into Escherichia coli DH5? (1). Inserts were se-
quenced with M13 forward and reverse primers with Taq polymerase (Prism
Ready Reaction Dye Primer Cycle Sequencing Kit; Applied Biosystems). The
sequences obtained were analyzed on SeqEd version 1.0.3 software (Applied
Biosystems) and compared with that of the yeast genome by use of a BLAST
search (Basic Local Alignment Search Tool) with the Saccharomyces Genome
Database (internet address, http://genome-www2.stanford.edu:5555/cgi-bin/
blastsgd). Predicted open reading frames were detected with DNA Strider 1.0
msoftware (Commissariat a l’Energie Atomique, Gif-sur-Yvette, France).
Design and use of intron splice site primers. Since introns
are not essential for gene function, they have evolved with
minimal constraint. There are conserved sequence motifs
within all introns, however, that are necessary for their removal
during the synthesis of mRNA. In the yeast S. cerevisiae, the
lariat branch point sequence TACTAAC is strictly conserved,
with mutations in this site preventing spliceosome assembly
and cleavage of the 5? intron junction. The sequence GTATGT
is also almost exclusively utilized at the 5? splice site. A third
sequence, (C/T)AG, defines the 3? end of the intron (40). To
detect polymorphisms in yeast strains by PCR, primers were
designed which contain sequences at their 3? end complemen-
tary to either the 5? splice site or lariat branch consensus
sequences. The primers were extended at their 5? end to pro-
duce 16-mer oligonucleotides by using random sequence. The
primer sequences were checked on the program Oligo4 (Na-
tional Biosciences) to test for possible secondary structure and
to obtain oligonucleotides with similar annealing tempera-
tures. Primers were used singularly or in pairs. When used
alone, the amplification is between two different intron splice
To develop a rapid PCR technique, several different meth-
ods of preparing DNA template were tested (Fig. 2). Cells
from the wine yeast AWRI 796, grown to the stationary phase
on YEPD plates, were resuspended in 100 ?l of either water,
0.1% Triton X-100, or Zymolyase. The cells resuspended in
water were either used for PCR directly or subjected to differ-
ent freezing and boiling treatments. Cells resuspended in Tri-
ton X-100 or Zymolyase were incubated at 100 or 37?C for 10
min, respectively. The cell suspension (2 ?l) was then used in
a 50-?l PCR with primer EI1 as described in Materials and
Methods. The different cell treatments were compared with a
reaction using purified DNA. The amplification products were
resolved on a 2% agarose gel and visualized by ethidium bro-
mide staining. The results in Fig. 2 demonstrate that each of
the different cell preparations generated a more complex am-
plification pattern than the cells that were resuspended in
water without any further treatment. Although there are sev-
eral differences in the amplification products with each of the
DNA preparations, all treatments generated patterns compa-
rable to that obtained with purified DNA.
The most important criterion for the usefulness of the tech-
nique was the reproducibility of the results. Five colonies of the
wine strain AWRI 796 and the laboratory yeast strains 15Du
(both haploid mating types and a diploid strain) were picked to
FIG. 1. Design of intron primers. The sequences of primers EI1 and EI2
were based on the consensus sequence GTATGT at the 5? exon-intron splice site.
The sequences of primers LA1 and LA2 were based on the lariat branch point
consensus sequence TACTAAC.
TABLE 1. Laboratory and commercial yeast strains
used in this study
15Dua ...........................................Laboratory strain
AWRI 939....................................Sake yeast
AWRI 350....................................Wine yeast
AWRI 729....................................Wine yeast
AWRI 834....................................Wine yeast (secondary fermentation)
AWRI 81......................................Wine yeast (flor yeast)
AWRI 1017..................................Wine yeast
AWRI 1034..................................Wine yeast (sweet wines)
AWRI 838....................................Wine yeast
AWRI 796....................................Wine yeast
aAll yeasts are S. cerevisiae strains except AWRI 1034, which is a strain of the
osmotolerant species T. delbrueckii.
TABLE 2. Possible isolates of AWRI 729
AWRI 729 ...................University of California, Davis
AWRI 814 ...................Australian Wine Research Institute
AWRI 825 ...................Department of Agriculture, Western Australia
AWRI 835 ...................Department of Agriculture, Western Australia
AWRI 925 ...................University of California, Davis
AWRI 947 ...................Australian Wine Research Institute
AWRI 1116 .................Epernay, France
AWRI 1117 .................Epernay, France
AWRI 1118 .................Epernay, France
VOL. 62, 1996PCR DIFFERENTIATION OF COMMERCIAL YEAST STRAINS 4515
determine whether reliable results could be obtained when
cells were prepared for PCR by simply freezing in liquid ni-
trogen and then boiling for 10 min. Figure 3A demonstrates
that although there are polymorphisms generated between
strains 15Du and AWRI 796, the amplification fingerprints
within each strain are very similar. Although cells prepared by
only boiling exhibited an amplification pattern comparable to
that of the freeze-boil samples (Fig. 2), this treatment gener-
ated less consistent results (data not shown). Since the freeze-
boil treatment of cells was simple and rapid, this method was
used for all subsequent experiments.
To determine the sensitivity of the reaction, a serial dilution
of cells was made before the freeze-boil treatment. The results
in Fig. 3B illustrate that when 5 ? 103to 5 ? 105cells were
used in the reaction, the amplification pattern remained con-
sistent, although some minor bands were variable. Further
dilution led to a significant change in the pattern, with only the
major fragments being amplified. These results demonstrate
that the amplification fingerprint was relatively insensitive to
the number of yeast cells used in the reaction, making it un-
necessary to accurately quantify the cell number.
In the initial stages of the study, there were problems in
obtaining uniform amplification fingerprints. Results indicated
that this was due to cells being at different stages of growth
prior to PCR. The reason for this is unclear. One explanation
is that the interaction of proteins with the genome in dividing
cells is either blocking the annealing of the primers to the
splice site sequences or inhibiting the Taq polymerase exten-
sion reaction. Since strains grow at different rates, this could be
a drawback for the technique, particularly when rapid identi-
fication is necessary. To avoid this problem, we have tested
whether transferring the yeast to 4?C to stop cell division was
sufficient to obtain reliable results (Fig. 3C). Cells that were
not incubated at 4?C gave poor and inconsistent amplification
(Fig. 3C, lane 4). Incubating cells at 4?C for 2 h or more,
however, ensured a dependable amplification. Although the
physiological state of the cells at the different time points was
uncertain, the result is important in that it permits PCR analysis
of strains soon after the formation of visible single colonies.
Differentiation of S. cerevisiae strains. Figure 4 shows the
amplification patterns obtained with the four intron primers
used separately for five S. cerevisiae strains, namely, a labora-
tory yeast and four commercial yeasts used in either the wine,
sake, baking, or brewing industry (Table 1). As well as ampli-
fying fragments shared by all of these yeasts, the method gen-
erated distinctive polymorphisms that permit unmistakable dif-
ferentiation of these strains. The brewing, baking, and sake
yeasts are most readily distinguished from each other and from
the two other strains by using primer LA2. In the baker’s yeast,
the LA2 primer amplified products of 280 and 320 bp. Only the
280-bp fragment was amplified with the brewing yeast, and
only the 320-bp fragment was amplified by the sake yeast. The
laboratory and wine strains did not produce either of these
fragments but can be distinguished by other changes in the
LA2 amplification pattern. Primer LA1 shows similar differ-
ences, but the amplification fingerprints of the baker’s and
brewer’s yeasts are very similar. The laboratory yeast 15Du is
most easily identified by the use of primer EI1, which gave an
amplification product of 1.55 kbp, which is not seen in any
other yeast analyzed to date. The wine yeast AWRI 796 is
differentiated by a product of 800 bp with primer EI1 that has
not been observed in other wine yeasts plus a 300-bp fragment
with primer EI2.
Differentiation of wine-making strains. All of the S. cerevi-
siae strains shown in Fig. 4 have very different physiological
characteristics that make them suitable for their specific appli-
cation. Figure 5 demonstrates that the primers are also useful
in differentiating commercial wine-making strains from each
other. The yeast strains analyzed are those commonly used in
the Australian wine industry as starter cultures for grape juice
fermentation and have been previously karyotyped by pulsed-
FIG. 2. Comparing methods of DNA template preparation. Lanes: 1 to 5,
cells resuspended in sterile water (1, no treatment; 2, boiled; 3, freeze-boiled;
lane 4, boil-frozen; 5, two cycles of freeze-boiling); 6, cells resuspended in Triton
X-100 and boiled; 7, cells resuspended in Zymolyase and incubated at 37?C; 8,
purified DNA; M, molecular size standards (sizes are indicated on the right in
base pairs). All reactions were with primer EI1 and strain AWRI 796.
FIG. 3. Optimization of PCR method. (A) Reproducibility of PCR method.
Lanes: 1 to 5, AWRI 796; 6, 15Dua; 7, 15Du?; 8 to 10, 15Dua/?. All DNA
templates were prepared by the freeze-boil method. All reactions were with
primer EI1. (B) Effect of DNA template concentration. Lanes: 1, 5 ? 105cells;
2, 1 ? 105cells; 3, 5 ? 104cells; 4, 1 ? 104cells; 5, 5 ? 103cells; 6, 1 ? 103cells.
All reactions were with primer EI1 and strain AWRI 796. (C) Stationary and
dividing cells. Strain AWRI 796 was grown on plates for 24 h at 25?C and then
stored at 4?C for 8 (lane 1), 4 (lane 2), 2 (lane 3), and 0 (lane 4) h. All reactions
were with primer EI1.
DE BARROS LOPES ET AL.APPL. ENVIRON. MICROBIOL.
field gel electrophoresis (25) (strains listed in Table 1). By
using primer EI1, seven of the eight wine yeasts can be differ-
entiated from each other and from the S. cerevisiae strains
tested in Fig. 4. Yeast AWRI 1034, shown in lane 7 of Fig. 5A,
produced an amplification pattern distinct from that of the
other yeasts. This yeast is T. delbrueckii and is the only non-
Saccharomyces yeast used in this study. An amplified product
at 1.33 kb separates the S. cerevisiae strains into two groups.
The 800-bp fragment is unique to AWRI 796 (Fig. 5A, lane 9).
The 200-bp fragment, which is detected in the laboratory yeast
15Du (lane 1) and was amplified in the sake yeast (Fig. 4),
permits the differentiation of wine yeast strain AWRI 1017
Although the differences between the strains are consistently
amplified with primer EI1, the polymorphisms can be minor.
The increased intensity of the 330-bp fragment distinguishes
AWRI 350 (Fig. 5B, lane 2) from AWRI 838 (lane 8). These
two strains, however, are more readily differentiated by use of
primers LA1 and LA2 (Fig. 5B). Primer LA1 amplifies 1,000-
and 1,100-bp fragments in yeast AWRI 838 but not in AWRI
350 (lanes 1 and 2). Primer LA2 (lanes 3 and 4) amplifies the
280-bp fragment in AWRI 838, which was also amplified in the
brewer’s and baker’s yeasts in Fig. 4 but not in AWRI 350.
Strain AWRI 81 (Fig. 5B, lane 7) is most easily differentiated
from strains AWRI 729 and AWRI 834 (lanes 6 and 7) by using
primer EI2. With this primer, the 1,070-bp fragment present in
the AWRI 729 and AWRI 834 fingerprints is not amplified for
AWRI 81. The use of each of the intron primers separately did
not allow the differentiation of AWRI 729 and AWRI 834. To
distinguish between these two yeasts, it was necessary to use a
primer pair. The use of primers EI1 and EI2 or LA1 and LA2
together gives poor amplification because of the complemen-
tarity of the primer sequences. However, when primers LA1
and EI1 (lanes 8 and 9) or LA1 and EI2 (lanes 10 and 11) are
used, the two strains could be differentiated. The first primer
combination produced a difference in the amplification of a
700-bp fragment in strains AWRI 834 and AWRI 729. When
primers LA1 and EI2 were used simultaneously, several poly-
morphisms were detected, including a 1,680-bp amplification
product for strain AWRI 834.
FIG. 4. Differentiation of S. cerevisiae strains with intron primers. Lanes: 1 to
5, primer LA1; 6 to 10, primer LA2; 11 to 15, primer EI1; 16 to 20, primer EI2.
L, laboratory strain 15Du; B, brewing strain B431; K, baking strain T5088; S, sake
strain AWRI 939; W, wine strain AWRI 796. The numbers indicate the sizes, in
base pairs, of the polymorphic fragments discussed in the text.
FIG. 5. Differentiation of commercial wine-making strains. (A) PCR using primer EI1. 15Du is a laboratory yeast. The AWRI yeasts are commercial wine-making
strains. Lanes 1, 15Du; 2, AWRI 350; 3, AWRI 729; 4, AWRI 834; 5, AWRI 81; 6, AWRI 1017; 7, AWRI 1034; 8, AWRI 838; 9, AWRI 796. (B) PCR using other
intron primers. Lanes: 1, primer LA1, strain AWRI 350; 2, primer LA1, strain AWRI 838; 3, primer LA2, strain 350; 4, primer LA2, strain AWRI 838; 5, primer EI2,
strain AWRI 729; 6, primer EI2, strain AWRI 834; 7, primer EI2, strain AWRI 81; 8, primers LA1 and EI1, strain AWRI 729; 9, primers LA1 and EI1, strain AWRI
834; 10, primers LA2 and EI1, strain AWRI 729; 11, primers LA2 and EI1, strain AWRI 834. The numbers indicate the sizes, in base pairs, of the polymorphic fragments
discussed in the text.
VOL. 62, 1996PCR DIFFERENTIATION OF COMMERCIAL YEAST STRAINS4517
Verification of wine yeast identity. A number of yeasts be-
lieved to be isolates of strain AWRI 729 exist (13, 27, 30)
(Table 2). To assess whether this method is effective for veri-
fying the identity of S. cerevisiae strains, the putative 729 strains
were analyzed. The results with primer EI1 are shown in Fig. 6
and clearly indicate that several of the strains are unrelated to
AWRI 729. The original yeast strain 729 (now named AWRI
814) was isolated in 1964 from a sample of dried yeast labelled
Epernay. The current AWRI 729 yeast was obtained from The
University of California, Davis, and shows an amplification
fingerprint identical to that of the original isolate (Fig. 6, lanes
1 and 2). A second isolate of strain 729 was retrieved from The
University of California, Davis, at a later date (and named
AWRI 925), and interestingly, this yeast has an amplification
fingerprint completely different from that of AWRI 729 (lane
5). Two isolates of AWRI 729 acquired from The University of
Western Australia (AWRI 825 and AWRI 835) are indistin-
guishable from the original strain (lanes 3 and 4). However,
three other yeasts from Epernay (AWRI 1116, AWRI 1117,
and AWRI 1118) are distinct from strain AWRI 729 (lanes 7 to
9). The results were confirmed by using the other intron prim-
ers and demonstrate the usefulness of the technique for yeast
strain authentication in culture collections.
Sequence analysis of PCR products. The specificity of the
PCR has been determined by sequencing several of the ampli-
fied fragments (Table 3). Generally, the analysis was per-
formed with products of primers EI1 and LA2 since PCR with
these primers amplifies towards the other conserved splicing
motif, assisting in the recognition of intron sequences. These
primers were used singularly in PCRs with strains 15Du and
AWRI 796. The results demonstrate that the conditions used
in the experiments are effective in annealing to the conserved
sequence motifs of introns. Of the 11 primer binding sites that
could be established by a comparison with the Saccharomyces
genome database, seven are conserved in the intron splice site
motifs of the primer (Table 3). In all sequences, at least one of
the primers has annealed to an exact match of the conserved
motif. As expected, the homology with the target DNA is less
specific at the 5? end of the primers.
The conserved splicing motifs are obviously present in both
intron and nonintron sequences, and it is expected that both
regions are being amplified. Analysis of the sequences supports
this expectation. Sequence LA2-25 is a 592-bp fragment. A
search of the Saccharomyces genome database indicates that it
is amplified from a region of chromosome V (cosmids 9747,
8198, and 9781 and lambda clones 3612 and 6052; nucleotides
38192 to 38784). This has not previously been identified as an
open reading frame (ORF) since there is a stop codon in all six
reading frames. The region, however, meets all the criteria of
a spliced gene (15, 40). The TACTAACA lariat branch point
motif is at position 38208. A potential 5? splice site, GTATGT,
exists 44 nucleotides upstream from the lariat branch point.
This distance agrees with that found in other yeast introns.
There are three possible ATG start codons within 127 bp of the
5? splice site. This is consistent with introns being located at the
5? end of genes, always in the first 150 bp, downstream of the
ATG. The region also contains a 3? splice site motif 15 bp from
the lariat branch point. This would confer an intron of 75
nucleotides, which is well within the common maximum size of
550 bp. Importantly, splicing of this sequence would lead to the
excision of a stop codon near the lariat branch point, producing
an ORF of 132, 146, or 161 amino acids, depending upon the
ATG utilized in translation. The predicted protein sequences
have no significant homologs in other organisms.
Sequences obtained from other amplified products also met
the criteria of spliced genes, but in these cases, the significance
is unclear. Two of these sequences were in coding regions of
previously identified genes (sequences EI1-15 and LA2-24). A
third sequence (LA2-34) was in a region of the yeast genome
that is not yet in the genome database; therefore, there is not
FIG. 6. Verification of S. cerevisiae strain identity. Lanes: 1, AWRI 729; 2,
AWRI 814; 3, AWRI 825; 4, AWRI 835; 5, AWRI 925; 6, AWRI 947; 7, AWRI
1116; 8, AWRI 1117; 9, AWRI 1118; 10, AWRI 796. All reactions were with
primer EI1. The unlabelled arrows indicate polymorphisms between the strains
with fingerprints distinct from that of AWRI 729.
TABLE 3. Sequence analysis of PCR fragments
Chromosome XVI se-
Second primer not in yeast
Chromosome XVI and un-
Encoding hypothetical Ser-
Predicted intron containing
LA2-34Not in genome databaseUnidentified sequence
aThe nucleotides which are identical to the primer are shown in bold. The
sequences that correspond to the intron splice site motif are underlined. Se-
quences for first and second primers are given.
bDesignations with the prefix SC are Saccharomyces Genome Database locus
DE BARROS LOPES ET AL.APPL. ENVIRON. MICROBIOL.
sufficient information to establish whether an ORF would be
created. Interestingly, an amplified fragment from AWRI 796
(EI1-17) is conserved at one end with a region of chromosome
XVI but at the other end has no homology to sequences in the
yeast genome database. This suggests that the primer has ex-
posed a change in chromosome structure of the wine yeast.
Whether this is due to the presence of an intron or to chro-
mosome translocation cannot be determined with the se-
quence data obtained.
Many precursor mRNAs in eukaryotes contain intervening
sequences or introns. The introns are precisely spliced from
the pre-mRNA to form functional mature mRNAs in a process
that requires the spliceosome, a large ribonucleoprotein com-
plex (40). The reason for the existence and distribution of
introns is debated continuously, and whether these intervening
sequences have a function remains unclear (9, 19). It is known,
however, that many introns are very close to selectively neutral,
and except for the sites that have been conserved for recogni-
tion by the spliceosome, their sequences are highly variable.
These changes in intron structure include nucleotide substitu-
tions, deletions, or insertions or the presence or absence of
introns in a gene (21, 23, 36).
Since introns have potentially high rates of sequence evolu-
tion, their analysis has become an important tool in studies of
genome relatedness. Evolutionary relationships between spe-
cies have been analyzed by determining the divergence of in-
tron sequences (35). Similar analyses have been used to study
the population genetics of a species (23). Furthermore, by
comparing the intron sequences of X-linked and Y-linked
genes, the extent of male-driven evolution has been analyzed in
a number of mammals (34). Primers that anneal to intron
splice sequences have been used to identify and map genetic
polymorphisms in cereals with the PCR (39). Splice site prim-
ers have also been employed for identifying genes in complex
genomes by using both PCR and hybridization methods (20).
In the present study, introns have been targeted to detect
polymorphisms in commercial yeast strains. A PCR-based
method that does not require the purification of DNA, making
the technique simple and rapid, has been developed. Sequenc-
ing of the amplified fragments shows that the conditions used
for PCR are effective in targeting the splice site motifs, and
fragment LA2-25 reveals that intron-containing ORFs are tar-
geted. These conserved splice site motifs are also present in
nonintron DNA, however, and sequencing demonstrates that
these regions of the genome are similarly amplified. The re-
sults from the use of the intron primers have been compared
with that of random primers. Random primers have either
given no amplification or else have provided few polymor-
phisms between the wine-making strains studied (results not
shown). This has also been observed by Quesada and Cenis
(28), who used randomly amplified polymorphic DNA analysis
for yeast strain differentiation. The ability of the intron primers
to produce such high polymorphisms may reflect the yeast
genome structure. Recent models argue that introns in multi-
cellular organisms have evolved to have a function in providing
an extra level of gene regulation for differentiation (19). The
most apparent example is alternative splicing of genes in
higher eukaryotes. It has also been found that for several
introns in higher eukaryotes, the intron sequences are very
large and well conserved across species boundaries, indicating
a regulatory function. Apart from an unusually large yeast
intron of 1,001 nucleotides in the DBP2 gene of S. cerevisiae
that has recently been shown to possess a regulatory function
(4), the smaller yeast introns (10) are less likely to be evolu-
tionarily constrained. This potential for intron sequence evolu-
tion, and the sexual isolation of wine yeast as a result of their
inability to produce viable meiotic products, might explain the
high degree of polymorphism detected by using the intron
The use of specific primers that target the delta elements of
transposons as a method for S. cerevisiae strain identification
has been reported (16). The authors declare that these primers
permit the differentiation of more than 95% of their commer-
cial strains. Furthermore, the few examples shown indicate
striking polymorphisms, although previous use of the transpo-
son primers has shown a lesser degree of difference (18). The
intron primers provide an alternative strategy to transposon
and randomly amplified polymorphic DNA (28) analysis for
strain differentiation, and targeting both transposons and in-
trons may lead to an efficient S. cerevisiae identification system,
making the need for chromosome karyotyping unnecessary.
The intron primers have a significant advantage, however, for
characterization of non-Saccharomyces yeasts. The transposon
delta sequences are not present in non-Saccharomyces yeasts,
and therefore the transposon primers are limited to S. cerevi-
siae and very closely related species (11, 24). Intron splice sites
are conserved in all yeasts that have been studied to date. For
example, in the unrelated fission yeast Schizosaccharomyces
pombe, the 5? splice site is either GTAAGT or the S. cerevisiae
consensus sequence GTATGT. The branch point sequence is
also less stringent than for S. cerevisiae, but many contain the
TACTAAC site (41). The usefulness of this can be seen with
yeast AWRI 1034, which is a strain of T. delbrueckii (Fig. 5A).
T. delbrueckii is a highly osmotolerant fermenting yeast, capa-
ble of growth in 50% glucose. For this reason, it is occasionally
used in the production of some sweet wine styles. The ampli-
fication pattern of this yeast is unique when compared with
that of the S. cerevisiae yeasts used in the study, clearly iden-
tifying it as an unrelated yeast.
The intron splice site primers are effective in differentiating
between the main commercial wine yeasts used by the Austra-
lian wine industry. These wine yeasts (excluding AWRI 796)
have been previously subjected to chromosome karyotyping
using transverse alternating field electrophoresis (25). Trans-
verse alternating field electrophoresis, although sensitive in
detecting polymorphisms in yeast strains, is time-consuming
and technically difficult, and the results can be difficult to
interpret. For example, AWRI 350 and AWRI 838 are easily
separated by use of a single primer with the PCR technique
(Fig. 5), but the chromosome karyotypes are difficult to distin-
guish. The PCR technique is much quicker and less labored
than chromosome karyotyping. Apart from strains AWRI 729
and AWRI 834, the commercial strains could all be differen-
tiated by performing a single amplification reaction. For these
two strains, polymorphisms were detected only when two prim-
ers were used simultaneously. It is curious to note, however,
that these two yeasts have very different fermentation charac-
teristics and exhibit considerable chromosome polymorphism
when analyzed by pulsed-field gel electrophoresis (25).
An interesting result was also observed with strains AWRI
729 and AWRI 814 (Fig. 6). These are putative isolates of the
same yeast that have been stored in different culture collec-
tions. The intron primers were unable to detect polymorphisms
between these two strains. The strains display similar growth
characteristics, amino acid uptake properties, and fermenta-
tion characteristics in the laboratory but have been reported to
produce different levels of volatile acidity by the wine industry.
Whether the reason for this difference is environmental or
genetic is not clear. Preliminary pulsed-field gel electrophore-
VOL. 62, 1996 PCR DIFFERENTIATION OF COMMERCIAL YEAST STRAINS 4519
sis studies were unable to discriminate between the strains
(25), but more recently, a minor polymorphism has been no-
ticed by increasing the resolution of the chromosomal finger-
print method (13). The explanation for this modification re-
mains unknown. The genetic composition of some yeasts has
been demonstrated to be quite unstable (17), and combined
with the freeze-drying method used for the long-term storage
of these strains, closely related yeasts could give differences in
karyotype. It is unlikely, however, that the PCR method will
detect these changes except in rare cases where the amplified
fragment spans the chromosome break. A more detailed analysis
will be required to resolve the evolutionary divergence of these
strains, so as to determine the sensitivity of the intron primers.
In summary, a PCR-based method using intron splice site
primers has been developed for identification of commercial
yeast strains. The main advantage of this technique is that it is
simple, rapid, and accessible to industrial laboratories with
limited molecular expertise and resources. The speed of the
technique and the ability to analyze many samples in a single
day permit its use in monitoring the growth of yeasts during
propagation and fermentation for routine quality control. The
method has been used for confirmation of the identity of active
dried strains used for wine-making (6a) and has been effective
in differentiating commercial winery strains from both the in-
digenous S. cerevisiae and non-Saccharomyces yeast flora
present in the juice and on winery equipment in one winery
We thank Neil Shirley and Jing Li for DNA sequencing, Ken Chalm-
ers, Angelo Karakousis, and Anna Martens for helpful discussions,
Sonia Dayan for critical reading of the manuscript, and Angelos Kalog-
eropoulos and Peter Phillipsen for communication of unpublished results.
Miguel de Barros Lopes acknowledges the receipt of a postdoctoral
fellowship from the CRC for Viticulture. Alison Soden acknowledges
the receipt of a APRA-Industry Award in conjunction with Lallemand
Australia. This work was funded by the Grape and Wine Research
Council of Australia and the CRC for Viticulture.
1. Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A.
Smith, and K. Struhl. 1994. Current protocols in molecular biology, vol. 2.
John Wiley & Sons, Inc., New York.
2. Bakalinsky, A. T., and R. Snow. 1990. Conversion of wine strains of Saccha-
romyces cerevisiae to heterothallism. Appl. Environ. Microbiol. 56:849–857.
3. Bakalinsky, A. T., and R. Snow. 1990. The chromosomal constitution of wine
strains of Saccharomyces cerevisiae. Yeast 6:367–382.
4. Barta, I., and R. Iggo. 1995. Autoregulation of expression of the yeast Dbp2p
‘DEAD-box’ protein is mediated by sequences in the conserved DBP2 in-
tron. EMBO J. 14:3800–3808.
5. Bidenne, C., B. Blondin, S. Dequin, and F. Vezinhet. 1992. Analysis of the
chromosomal DNA polymorphism of wine strains of Saccharomyces cerevi-
siae. Curr. Genet. 22:1–7.
6. Cole, G. M., D. E. Stone, and S. I. Reed. 1990. Stochiometry of G protein
subunits affects the Saccharomyces cerevisiae mating pheromone signal trans-
duction pathway. Mol. Cell. Biol. 10:510–517.
6a.de Barros Lopes, M., and P. A. Henschke. Unpublished results.
7. Degre ´, R., D. Y. Thomas, J. Ash, K. Mailhiot, A. Morin, and C. Dubord.
1989. Wine yeast strain identification. Am. J. Enol. Vitic. 40:309–315.
8. Dubourdieu, D., A. Sokol, J. Zucca, P. Thalouarn, A. Dattee, and M. Aigle.
1987. Identification des souches de levures isole ´es de vins par l’analyse de
leur ADN mitochondrial. Connaiss. Vigne Vin 21:267–278.
9. Fink, G. R. 1987. Pseudogenes in Yeast? Cell 49:5–6.
10. Hawkins, J. D. 1988. A survey on intron and exon lengths. Nucleic Acids Res.
11. Hawthorne, D., and P. Philippsen. 1994. Genetic and molecular analysis of
hybrids in the genus Saccharomyces involving S. cerevisiae, S. uvarum and a
new species, S. douglasii. Yeast 10:1285–1296.
12. Heard, G. M., and G. H. Fleet. 1985. Growth of natural yeast flora during the
fermentation of innoculated wine. Appl. Environ. Microbiol. 50:727–728.
13. Henschke, P. A. 1990. Evaluating wine yeasts for improved wine quality, p.
157–165. In P. J. Williams, D. M. Davidson, and T. H. Lee (ed.), Proceedings
of the Seventh Australian Wine Industry Technical Conference, 13–17 Au-
gust 1989, Adelaide, South Australia, Australian Industrial Publishers, Ad-
elaide, South Australia.
14. Johnson, J. R., and R. K. Mortimer. 1986. Electrophoretic karyotyping of
laboratory and commercial strains of Saccharomyces and other yeasts. Int. J.
Syst. Bacteriol. 36:569–572.
15. Kalogeropoulos, A. 1995. Automatic intron detection in nuclear DNA se-
quences of Saccharomyces cerevisiae. Yeast 11:555–565.
16. Lavalle ´e, F., Y. Salvas, S. Lamy, D. Y. Thomas, R. Degre ´, and L. Dulau. 1994.
PCR and DNA fingerprinting used as quality control in the production of
wine yeast strains. Am. J. Enol. Vitic. 45:86–91.
17. Longo, E., and F. Vezinhet. 1993. Chromosomal rearrangements during
vegetative growth of a wild strain of Saccharomyces cerevisiae. Appl. Environ.
17a.Martens, A. Unpublished results.
18. Masneuf, I., and D. Dubourdieu. 1994. Comparaison de deux techniques
d’identification des souches de levures de vinification base ´es sur le polymor-
phisme de l’ADN ge ´nomique: re ´action de polyme ´risation en chaine (PCR)
et analyse des caryotypes (e ´lectrphore `se en champ pulse ´). J. Int. Sci. Vigne
19. Mattick, J. S. 1994. Introns: evolution and function. Curr. Opin. Genet. Dev.
20. Melmer, G., and M. Buchwald. 1992. Identification of genes using oligonu-
cleotides corresponding to splice site consensus sequences. Hum. Mol.
21. Nellen, W., C. Donath, M. Moos, and D. Gallwitz. 1981. The nucleotide
sequences of the actin genes from Saccharomyces carlsbergensis and Saccha-
romyces cerevisiae are identical except for their introns. J. Mol. Appl. Genet.
22. Oosthuizen, A., J. L. F. Koch, B. C. Viljoen, H. B. Muller, and P. M. Lategan.
1987. The value of long-chain fatty acid composition in the identification of
some brewery yeasts. J. Inst. Brew. 93:174–176.
23. Palumbi, S. R., and C. S. Baker. 1996. Contrasting population structure from
nuclear intron sequences and mtDNA of humpback whales. Mol. Biol. Evol.
24. Pearson, B. M., A. T. Carter, J. M. Furze, and I. N. Roberts. 1995. A novel
approach for discovering retrotransposons: characterization of a long termi-
nal repeat element in the spoilage yeast Pichia membranaefaciens and its use
in strain identification. Int. J. Syst. Bacteriol. 45:386–389.
25. Petering, J., P. Langridge, and P. A. Henschke. 1988. Fingerprinting wine
yeasts: the application of chromosome electrophoresis. Aust. N. Z. Wine
Ind. J. 3(3):48–52.
26. Querol, A., E. Barrio, T. Huerta, and D. Ramo ´n. 1992. Molecular monitoring
of wine fermentations conducted by active dry yeast strains. Appl. Environ.
27. Querol, A., E. Barrio, and D. Ramo ´n. 1992. A comparative study of different
methods of yeast strain characterisation. Syst. Appl. Microbiol. 15:439–446.
28. Quesada, M. P., and J. L. Cenis. 1995. Use of random amplified polymorphic
DNA (RAPD-PCR) in the characterization of wine yeasts. Am. J. Enol Vitic.
29. Rankine, B. C. 1968. The importance of yeasts in determining the compo-
sition and quality of wines. Vitis 7:22–49.
30. Rankine, B. C. 1977. Modern developments in selection and use of pure
yeast cultures for winemaking. Aust. Wine Brew. Spirit Rev. 96(8):31–33 and
31. Saiki, R. K., S. Scarf, F. Faloona, K. B. Mullis, G. T. Horn, H. A. Erlich, and
N. Arnheim. 1985. Enzymatic amplification of beta-globin genomic se-
quences and restriction site analysis for diagnosis of sickle-cell anemia.
32. Schu ¨tz, M., and J. Gafner. 1993. Analysis of yeast diversity during sponta-
neous and induced alcoholic fermenation. J. Appl. Bacteriol. 75:551–558.
33. Sherman, F., G. R. Fink, and J. B. Hicks. 1986. Laboratory course manual
for methods in yeast genetics. Cold Spring Harbor Laboratory Press, Cold
Spring Harbor, N.Y.
34. Shimmin, L. C., B. H.-J. Chang, and W.-H. Li. 1993. Male-driven evolution
of DNA sequences. Nature (London) 362:745–747.
35. Slade, R. W., C. Moritz, A. Heideman, and P. T. Hale. 1993. Rapid assess-
ment of single-copy nuclear DNA variation in diverse species. Mol. Ecol.
36. Tarlow, J. K., A. I. F. Blakemore, A. Lennard, R. Solari, H. N. Hughes, A.
Steinkasserer, and G. W. Duff. 1993. Polymorphism in human IL-1 receptor
antagonist gene intron 2 is caused by variable numbers of an 86-bp tandem
repeat. Hum. Genet. 91:403–404.
37. Thornton, R. J., and R. Eschenbruch. 1976. Homothallism in wine yeasts.
Antonie Leeuwenhoek 42:503–509.
38. Van Vuuren, H. J. J., and L. van der Meer. 1987. Fingerprinting of yeasts by
protein electrophoresis. Am. J. Enol. Vitic. 38:49–53.
39. Weining, S., and P. Langridge. 1991. Identification and mapping of poly-
morphisms in cereals based on the polymerase chain reaction. Theor. Appl.
40. Woolford, J. L., Jr. 1989. Nuclear pre-mRNA splicing in yeast. Yeast 5:439–457.
41. Zhang, M. Q., and T. G. Marr. 1994. Fission yeast gene structure and
recognition. Nucleic Acids Res. 22:1750–1759.
DE BARROS LOPES ET AL. APPL. ENVIRON. MICROBIOL.