Intron splice site PCR analysis as a tool to discriminate Dekkera bruxellensis strains

Ann Microbiol (2011) 61:153–157 DOI 10.1007/s13213-010-0110-8

SHORT COMMUNICATION

Intron splice site PCR analysis as a tool to discriminate Dekkera bruxellensis strains Ileana Vigentini & Claudia Picozzi & Roberto Foschino

Received: 10 June 2010 / Accepted: 26 July 2010 / Published online: 2 September 2010 # Springer-Verlag and the University of Milan 2010

Abstract Dekkera bruxellensis yeast species can develop off-flavours in wine through a specific reductive metabolism. In particular, volatile phenols are often produced in amounts that are higher than the perception threshold with a loss in product quality. Recent observations suggest that “brett spoilage” is strictly strain-dependent, and therefore, a rapid and reliable identification at strain level of D. bruxellensis becomes strategic for an efficient prevention. Among the techniques used to analyse DNA regions with high rate of sequence evolution, intron splice site PCR amplification (ISS-PCR) has allowed the detection of polymorphisms in commercial strains of S. cerevisiae. Recently, the genome of a D. bruxellensis isolated from wine has been sequenced and the results have shown that about 2% of the genes, a value similar to the ones found in other hemiascomycetes (1% in D. hansenii, 4% in S. cerevisiae) contain introns. Moreover, D. bruxellensis introns have 5′, 3′ and branch motifs that are very similar to the consensus motif in S. cerevisiae. Although the use of the 5′ intron–exon splice site as a target for ISS-PCR in D.

This paper is part of the special issue "Wine microbiology and safety: From the vineyard to the bottle (Microsafetywine)", 19-20 November, 2009, Martina Franca (Italy). I. Vigentini : C. Picozzi : R. Foschino DiSTAM, Università degli Studi di Milano, Milano, Italy R. Foschino (*) Dipartimento di Scienze e Tecnologie Alimentari e Microbiologiche, via Celoria 2, 20133 Milano, Italy e-mail: [email protected]

bruxellensis did not allow the discrimination at strain level, an optimisation of primers could permit the development of a consistent tool for the typing of the species. In the present study, 17 D. bruxellensis strains belonging to the international CBS collection have been investigated for the ISSs, employing specific oligonucleotides containing different 5’ consensus sequences: GTATGT (S. cerevisiae) and GTAAGT (D. bruxellensis). Results have shown that almost the whole yeast collection was discriminated at strain level using different combinations of primers. Therefore, to simplify the approach, a multiplex PCR protocol able to generate stable genetic profiles was developed. Keywords Dekkera bruxellensis . Yeast typing . Introns

Introduction Investigation on genome variability among individuals represents an important tool to assess their biodiversity and to study their environmental diffusion. The safeguard of microbial strains with technological interest is becoming a strategic activity in food and wine industries. Nevertheless, many species marginally implicated in productions are still poorly studied, and rapid and reliable protocols for their recognition and typing are not well described. This is the case of the yeast Dekkera bruxellensis that can cause spoilage and increasing economic losses in wine. During the last years, some studies on D. bruxellensis phenotypic and genetic aspects have been published (Conterno et al. 2006). At the present time, the physiological studies have clarified that

154

the spoilage activity, due to the production of ethylphenols, is a strain-dependent character, and the enzymes involved in this production have been isolated (Vigentini et al. 2008). Since no significant remedies to remove offflavours from wine are yet available, some applications to compare yeast development in wine have been studied (Lustrato et al. 2010; Goretti et al. 2009). However, only a few works have reported methods to distinguish D. bruxellensis at strain level (Martorell et al. 2006; Agnolucci et al. 2009; Oelofse et al. 2009); typing protocols appear useful but are time consuming or require sophisticated instrumentations. The setting up of molecular probes designed on genes that are responsible of the off-flavour production could be a partial solution for a rapid detection of dangerous strains since an ethylphenol production around 2 mg l−1 is considered a benefit for wine (Etiévant et al. 1989). About 30% of the D. bruxellensis genome have been sequenced (Woolfit et al. 2007). When the entire genome becomes available, more consistent possible correlations of genetic and physiological aspects to single strain protocols could be arranged. Thus, the availability of comparable results among laboratories throughout databases could represent an important goal to face the increasing problem represented by D. bruxellensis in the oenological field. Introns are DNA regions that are not linked to any genetic function, and can vary with low control; changes such as nucleotide substitutions, deletions, or insertions can occur in an intron structure. For this reason, they are considered to be good evolutionary indicators in studies of genome relatedness. However, short conserved sequences for the spliceosome assembling are required during the synthesis of mRNA: in S. cerevisiae, the lariat branch point TACTAAC and the 5′ motif GTATGT consensus sequences are strictly conserved (de Barros Lopes et al. 1996). A recent work identified introns in 40 D. bruxellensis genes that are S. cerevisiae orthologs; moreover, eight introns were specific to D. bruxellensis (Woolfit et al. 2007). In this work, intron splice site PCR amplification (ISS-PCR) was applied to a collection of 17 D. bruxellensis strains using primers similar to those employed for S. cerevisiae. The main goal was to test and improve the method on D. bruxellensis species, since it is a technique that is simple, rapid, and potentially accessible to industrial laboratories with limited molecular expertise and resources.

Material and methods Yeast strains and cultural media A total of 17 D. bruxellensis strains belonging to the international CBS collection were analysed in this study:

Ann Microbiol (2011) 61:153–157

CBS73, CBS1940, CBS1941, CBS1942, CBS1943, CBS2336, CBS2499, CBS2547, CBS2796, CBS2797, CBS4459, CBS4480, CBS4481, CBS4482, CBS4601, CBS4602 and CBS5006. All strains were cultivated on YPD medium (1% (w/v) Yeast Extract, 2% (w/v) Peptone and 2% (w/v) Glucose) at 25°C for 3-5 days. Cells were mantained at −80°C in YPD medium added with 20% (v/v) glycerol. DNA amplification Yeasts were grown overnight in liquid YPD medium with shaking. DNA from a 5-ml culture was extracted as described by Querol et al. (1992) using 500 μg ml−1 of Zymolyase 100T (USBiological, Massachusetts, USA) as lytic enzyme. The primers employed in the amplification reactions were EI1, EI2, LA1, LA2 (De Barros Lopes et al. 1996) and DbEI1 (5′-CTGGCTTGGTGTAAGT-3′); primer DbEI1 derives from EI1, where in position 14, the T was substituted with an A. For PCRs, 1 μl (80-100 ng) of genomic DNA was added to a 24-μl reaction mix consisting of 0.75 μM each primers, 200 μM dNTPs, 1x reaction buffer MgCl2 free, 2.5 mM MgCl2, 5% (v/v) dimethyl sulfoxide (DMSO), 1U Taq polymerase (Fermentas, Vilnius, Lithuania). Amplifications were carried on in a Mastercycler ep gradient S (Eppendorf, Hamburg, Germany) following the protocol described for S. cerevisiae, but modifiedby increasing the number of cycles to 40 (de Barros Lopes et al. 1996) and setting the annealing temperatures at 46 and 47°C when EI1/LA2 or DbEI1/ LA2 and EI2/LA1 pairs of primers were used, respectively. Multiplex amplifications were carried out as described above using EI1, DbEI1 and LA2 oligonucleotides. Electrophoresis was run in 1.5% (w/v) plus 0.4 μg ml−1 ethidium bromide agarose gels. PCR products were photographed under GelDoc UV transilluminator (Bio-rad, Hercules, CA, USA). ISS profiles elaboration The digitalized gel images were analysed using Quantity One version 4.6.2 (Bio-Rad). The software was used to detect bands, using a match tolerance of 2% (Thompson et al. 2008). A similarity matrix was constructed using Dice’s similarity coefficient. Dendrograms were constructed by the unweighted pair group method using arithmetic averages (UPGMA) and the horizontal axis indicated the coefficient of genetic similarity. The reproducibility was carried out using four D. bruxellensis strains: CBS1940, CBS1941, CBS1942 and CBS1943. PCR reactions were prepared in triplicate for each one of the four strains. Indexes of discrimination among ISS genetic profiles were calculated as described

Ann Microbiol (2011) 61:153–157

by Hunter and Gaston (1988) according to the repeatability percentage.

Results and discussion The use of primers that can anneal to specific conserved motifs, such as those designed for ISS-PCR, has been already well described for the inter- and intraspecific characterisation of wine yeasts (ITS, inter-delta elements, SSR, microsatellite markers, etc.). Although the ISS-PCR technique proved to be useful for the typing of S. cerevisiae species since it is able to detect the polymorphisms of highly mutable sequences as introns, no reliable results have been shown for the discrimination of D. bruxellensis species at strain level (de Barros Lopes et al. 1996, 1998). This has also been proved by the fact that unstable ISS profiles are obtained when yeasts are analysed with the primer EI1 alone (Oelofse et al. 2009). For this reason, primers EI1/LA2 and EI2/LA1 were preliminary employed in pairs to type the D. bruxellensis collection. Repeatability tests were performed with primers used singly and in pairs. Results confirmed what has already been reported for the employment of single primers (Oelofse et al. 2009), that the PCRs gave stable profiles with a 100% of genetic similarity among identical strains when primers were used in pairs (data not shown). Figure 1a shows that genetic profiles could be obtained through the amplification of the genomic DNAs of D. bruxellensis CBS

Fig. 1 ISS profiles of D. bruxellensis strains using a EI1/LA2 pairs of primers. Lanes: 1 CBS73, 2 CBS1940, 3 CBS1941, 4 CBS1942, 5 CBS1943, 6 CBS2336, 7 CBS2499, 8 CBS2547, 9 CBS2796, 10 CBS2797, 11 CBS4459, 12 CBS4480, 13 CBS4481, 14 CBS4482, 15

155

strains with the EI1+LA2 primers pair. Different fragment lengths (bp units) were generated from EI1+LA2 and EI2+LA1 amplifications: the longest amplicons that were separated onto gels measured about 1,200 bps for the former pair and 2,500 bps for the latter one. The cluster analysis generated by the application of UPGMA revealed that 15 and 10 different genotypes could be distinguished by EI1+LA2 and EI2+LA1, respectively. The genetic similarity between the strains ranged from 25 to 100% for EI1+LA2 (Fig. 1b) and from 28 to 100% for EI2+LA1 (data not shown); in both analyses, only D. bruxellensis CBS2797 showed a genotype significantly different from the other strains, which was separated from the rest of the collection by an individual branch. Similar indexes of discrimination were calculated for the two couple of primers even if EI1+LA2 generated a percentage slightly higher than EI2+LA1 (98.5 vs 93.0%). Considering that D. bruxellensis has an intron content similar to S. cerevisiae and that among D. bruxellensis genes the most represented 5′ motif seems to be GTAAGT instead of GTATGT as reported for S. cerevisiae (Woolfit et al. 2007), a new primer was tested. PCRs were carried out using DbEI1/LA2 pair since, as above described, EI1/LA2 resulted in a higher discriminating ability. Although the substitution of a single nucleotide in the EI1primer sequence led to a change in the molecular weight of fragments (Fig. 2a) and increased the range of genetic similarity between the strains (from 19 to 100%), the results indicated that DbEI1/LA2 showed a lower discriminatory

CBS4601, 16 CBS4602, 17 CBS5006, M 100-bp DNA ladder (Fermentas, Vilnius, Lithuania). b Dendrogram built by UPGMA analysis comparing ISS profiles obtained after EI1/LA2 amplifications

156

Ann Microbiol (2011) 61:153–157

Fig. 2 ISS profiles of D. bruxellensis strains using a DbEI1/LA2 pairs of primers. Lanes: 1, CBS73; 2, CBS1940; 3, CBS1941; 4, CBS1942; 5, CBS1943; 6, CBS2336; 7, CBS2499; 8, CBS2547; 9, CBS2796; 10, CBS2797; 11, CBS4459; 12, CBS4480; 13, CBS4481;

14, CBS4482; 15, CBS4601; 16, CBS4602; 17, CBS5006; M, 100 bp DNA ladder (Fermentas, Vilnius, Lithuania); b) Dendrogram built by UPGMA analysis comparing ISS profiles obtained after DbEI1/LA2 amplifications

index (92.0%) than EI1/LA2 (98.5%). Even in this case, D. bruxellensis CBS2797 generated an outgroup (Fig. 2b). Comparing the results obtained from the amplifications with these two oligonucleotides pairs on the same strain, it was evident that no bands with the same molecular weight were detected. This could suggest a really different genetic origin of the amplified fragments. Therefore, a multiplex PCR was applied on the whole collection using primers

DbEI1, EI1 and LA2 (Fig. 3), with an increase of the discriminatory power up to 99.3%. In conclusion, this work proved to be a simple and reliable method for strain typing of D. bruxellensis species. Moreover, further investigation of the fragment sequences could lead to interesting implications on genome structure, gene regulation and sequence evolution as already reported for S. cerevisiae (Mattick 1994).

Fig. 3 ISS profiles of D. bruxellensis strains using a) Multiplex PCR with DbEI1, EI1 and LA2 primers. Lanes: 1 CBS73, 2 CBS1940, 3 CBS1941, 4 CBS1942, 5 CBS1943, 6 CBS2336, 7 CBS2499, 8 CBS2547, 9 CBS2796, 10 CBS2797, 11 CBS4459, 12 CBS4480,

13 CBS4481, 14 CBS4482, 15 CBS4601, 16 CBS4602, 17 CBS5006, M 100 bp DNA ladder (Fermentas, Vilnius, Lithuania). b Dendrogram built by UPGMA analysis comparing ISS profiles obtained after DbEI1, EI1 and LA2 amplifications

Ann Microbiol (2011) 61:153–157 Acknowledgment MIUR.

This work was supported by PRIN 2007 from

References Agnolucci M, Vigentini I, Capurso G et al (2009) Genetic diversity and physiological traits of Dekkera/Brettanomyces bruxellensis strains isolated during fermentation of Sangiovese grapes. Int J Food Microbiol 130:238–244 Conterno L, Joseph LCM, Arvik TJ, Henick-Kling T, Bisson LF (2006) Genetic and physiological characterization of Brettanomyces bruxellensis strains isolated from wine. Am J Enol Vitic 57:139–147 de Barros Lopes M, Soden A, Henschke PA, Langridge P (1996) PCR differentiation of commercial yeast strains using intron splice site primers. App Env Microbiol 62:4514–4520 de Barros Lopes M, Alison S, Marten LA, Henschke PA, Langridge P (1998) Differentiation and species identification of yeasts using PCR. Int J Syst Bacteriol 48:279–286 Etiévant PX, Issanchou S, Marie S, Ducruet V, Flanzy C (1989) Sensory impact of volatile phenols on red wine aroma: influence of carbonic macerationand time of storage. Sci Aliments 9:19–33 Goretti M, Turchetti B, Buratta M, Branda E, Codazzi L, Vaughan-Martini A, Buzzini P (2009) In vitro antimycotic activity of a Williopsis saturnus killer protein against food spoilage yeasts. Int J Food Microbiol 131:178–182

157 Hunter PR, Gaston MA (1988) Numerical index of the discriminatory ability of typing systems: an application of Simpson’s index of diversity. J Clin Microbiol 26:2465–2466 Lustrato G, Vigentini I, De Leonardis A, Alfano G, Tirelli A, Foschino R, Ranalli G (2010) Inactivation of wine spoilage yeasts Dekkera bruxellensis using low electric current treatment (LEC). J Appl Microbiol 109:594–604 Martorell P, Barata A, Malfeito-Ferreira M, Fernández-Espinar MT, Loureiro V, Querol A (2006) Molecular typing of the yeast species Dekkera bruxellensis and Pichia guilliermondii recovered from wine related sources. Int J Food Microbiol 106:79–84 Mattick JS (1994) Introns: evolution and function. Curr Opin Genet Dev 4:823–831 Oelofse A, Lonvaud-Funel A, du Toitet M (2009) Molecular identification of Brettanomyces bruxellensis strains isolated from red wines and volatile phenol production. Food Microbiol 26:377–385 Querol A, Barrio E, Huerta T, Ramón D (1992) Molecular monitoring of wine fermentation conducted by active dry yeast system. Appl Environ Microbiol 58:2948–2953 Thompson CL, Wang B, Holmes AJ (2008) The immediate environment during postnatal development has long-term impact on gut community structure in pigs. ISME J 2:739–748 Vigentini I, Romano A, Compagno C et al (2008) Physiological and oenological traits of different Dekkera/Brettanomyces bruxellensis strains under wine-model conditions. FEMS Yeast Res 8:1087–1096 Woolfit M, Rozpędowska E, Piškur J, Wolfe KH (2007) Genome survey sequencing of the wine spoilage yeast Dekkera (Brettanomyces) bruxellensis. Euk Cell 6:721–733