EP2764086A1

EP2764086A1 - Yeast bioethanol production strains allowing rapid detection of wild contaminant strains

Info

Publication number: EP2764086A1
Application number: EP12768838.0A
Authority: EP
Inventors: Johan Thevelein; Françoise Dumortier
Original assignee: Katholieke Universiteit Leuven; Vlaams Instituut voor Biotechnologie VIB; Fermentec Tecnologias em Acucar e Alcool Ltda
Current assignee: Katholieke Universiteit Leuven; Vlaams Instituut voor Biotechnologie VIB; Fermentec Tecnologias em Acucar e Alcool Ltda
Priority date: 2011-10-05
Filing date: 2012-10-05
Publication date: 2014-08-13
Also published as: BR112014008030B1; WO2013050509A1; BR112014008030A2

Abstract

The present invention relates to industrial yeast strains for bioethanol production that are marked specifically by insertion of the AGT1gene in a stable, well-conserved position in the genome. The resulting strains are organisms obtained by self-cloning. The insertion not only allows rapid and specific detection by PCR directly in ferment or samples of the production strain and of all wild, contaminant Saccharomyces cerevisiae strains, even when present in low amounts, but also improves the fermentation performance of the strain on starch hydrolysates and other maltose and/or maltotriose containing media.

Description

YEAST BIOETHANOL PRODUCTION STRAINS ALLOWING RAPID DETECTION OF WILD CONTAMINANT STRAINS

The present invention relates to industrial yeast strains for bioethanol production that are marked specifically by insertion of the AGT1 gene in a stable, well-conserved position in the genome. The resulting strains are organisms obtained by self-cloning. The insertion not only allows rapid and specific detection by PCR directly in fermentor samples of the production strain and of all wild, contaminant Saccharomyces cerevisiae strains, even when present in low amounts, but also improves the fermentation performance of the strain on starch hydrolysates and other maltose and/or maltotriose containing media. Several specific yeast strains are in use in Brazil for the production of bioethanol with sugar cane derived sucrose as a substrate. In this bioethanol production process the yeast is reused as many times as possible in practice. It is recycled using continuous centrifugation and acid washing. This reduces the cost of the yeast very much so that large amounts of yeast can be used in the fermentation, which results in very rapid fermentation times (6-8h), a very high productivity of the plant and a low price of the ethanol.

The currently used production yeast strains are natural yeast strains which originally were present as contaminants in industrial sugar cane fermentations and have been selected because of their superior fermentation performance, lack of negative properties, such as foam formation and flocculation, and especially their potency to dominate (to be present in the highest percentage) and persist (to be maintained over many recycles) in the yeast recycling system over wild yeast strains that enter the plants as contaminants. The latter is one of the most important traits for the commercial strains in bioethanol production with yeast recycling, because a nondominating/nonpersisting strain has not enough influence on the fermentation and/or is rapidly replaced by wild contaminant yeast strains (Basso et al. 2008). Bioethanol fermentations in Brazil are initiated with a large amount of baker's yeast and a small amount of commercial production yeast. After a few recycles, the baker's yeast has completely disappeared and is replaced by the commercial production strain.

In spite of the positive properties of these yeast strains, they are far from ideal because they are just natural strains and selected in the first place on the basis of domination/persistence in the recycling system. The domination/persistence in the yeast recycling system, however, is not complete and depends very much on the conditions: origin and composition of the sugar cane, weather conditions during the harvest, sanitation practices in the plant, process conditions in the plant, etc. A major problem is the occurrence of contaminant wild strains that come with the sugar cane into the factory and infect the fermentations (da Silva-Filho et al. 2005).

Wild yeast strains are much more difficult to suppress than bacteria, which can be controlled with low pH, high ethanol, and if required the use of antibiotics or other antibacterials. The wild yeast strains are very variable, many strains do not harm the fermentation and therefore do not require suppression or elimination. However, this is very difficult to predict. Some wild yeast strains cause flocculation, which strongly increases the fermentation time (sluggish fermentation) and can even lead to stuck fermentations. The flocks are usually made up of yeast and bacteria, and the latter have a strong effect on the type and strength of the flocculation. Some wild yeast strains can cause excessive foam formation, sucking a large part of the yeast into the foam. In this case as well, complexation with bacteria is a major factor determining the extent and the type of foam formation. In extreme cases, the foam can enter into the carbon dioxide exhaust and ethanol recuperation system and contaminate this whole system. This results in excessive costs for cleaning and for off-time of the plant.

Some of these strains have the capacity to dominate and persist in the recycling system and to expel completely the production strain. The response of the plant manager often depends on the consequences, if the fermentation continues without apparent problems often nothing is done, if the fermentation shows problems, action is taken. When infections with wild strains are detected and/or there are problems in the fermentation, one can try to displace the wild yeasts by infusion of a fresh batch of production yeast. However, this does not always work well in practice because of the huge size of the fermentors and the large amounts of yeast present compared to the amount of production yeast that is usually inoculated at the beginning of the fermentation. Complete displacement of the wild yeast in this way can be expensive because of large amounts of fresh production yeast required. In extreme cases, the fermentors have to be emptied and completely cleaned, causing expensive delays in the production process.

The occurrence of contamination is very erratic and unpredictable. It is certainly stimulated by rainy weather conditions and there are also suspicions that the initiation of contamination is aided by (accidental) changes in the process conditions and/or changes in the origin of the sugar cane. The latter has been impossible to verify because of the difficult and lengthy procedures currently used for contaminant yeast detection.

To control wild yeast infection, it first has to be monitored. Currently, monitoring of wild yeast infection is a very cumbersome, expensive and inefficient process. It is only done routinely in the best ethanol plants. Samples from the fermentations are isolated in the ethanol plants, and spread out on nutrient plates for isolation of single colonies. A small number of colonies (about 8-1 1 ) is isolated and grown up to obtain enough yeast for extraction of genomic DNA. The genomic DNA is used for chromosome separation on CHEF gels and the karyotype pattern obtained is used for strain identification. The whole process takes about 5-6 days. This is actually much too long because the fermentation time in the plant is only 6-8 h and therefore when the results of the contamination detection are available, the yeast has undergone several further rounds of fermentation and recycling, allowing further development or initiation of new contamination. Moreover, because of the high cost of the contaminant detection, it is only carried out four times in one season. As a result, it is impossible to identify any changes in the process conditions or in the origin of the sugar cane that may aid the initiation of infections with wild yeasts.

Therefore, there is an urgent need for a cheap and rapid detection system for contaminants. In principle, PCR detection is a cheap and reliable method, but the problem is to identify a gene that can be used to distinguish the production strains from the wild strains. The AGT1 gene encodes a transporter for oligosaccharides (Han et al. 1995), including maltotriose. In most or all cases where it has been tested, a yeast strain that is able to grow on maltose but not on maltotriose has a defective maltotriose transporter (Day et al. 2002; Stambuk et al. 2006; Smit et al. 2008). Introduction of a wild type AGT1 gene in this case restores growth on maltotriose. Surprisingly, we have found that many Saccharomyces cerevisiae strains, including bioethanol production strains and wild contaminants have a correct AGT1 open reading frame but an insertion in the promotor and/or terminator that compromises AGT1 expression and that as a result makes them maltotriose negative. This is especially unexpected, as has been demonstrated that the AGT1 gene in Saccharomyces cerevisiae is the predominant a- glucoside transporter, in contrast to the lager yeasts (Saccharomyces uvarum, Saccharomyces carlsbergiensis) where it is backed by the MAL gene cluster (Vidgren et al., 2005). Therefore, one would not expect that wild strains are affected in this important gene. Restoring the functional AGT1 gene in another genomic position in the production strain allows for rapid identification of the production strain and identification of all contaminating wild strains. Even more surprisingly, insertion of a functional AGT1 gene not only restores the growth on maltotriose, but it improves the fermentation of maltose, contrary to what was suggested in the literature (Smit et al., 2008; Alves et al., 2008), where it was reported that agtl deletion strains are fully competent for the efficient uptake of maltose in the medium (Alver st al, 2008). As several bioethanol production plants, especially outside Brazil, are using other substrates, like corn starch, the use of the AGT1 marked strain can not only help in the detection of contaminants, but also improve the fermentation rate and/or yield.

A first aspect of the invention is an ethanol tolerant yeast strain, comprising a recombinant marker gene, whereby said marker gene improves the fermentation performance. Preferably, said marker gene is inserted in a region that is highly stable as shown by high sequence preservation between different Saccharomyces strains. Even more preferably, said region is a non-coding region. Most preferably, said region is situated on chromosome XIV, between position 335000 and 336000, even more preferably between position 335300 and 335600. Preferably, the improvement of the fermentation performance is an improvement of the fermentation performance on maltose, less preferably (also) on maltotriose. An ethanol tolerant strain, as used here, is a strain resistant to at least 12% ethanol, preferably 13% ethanol, more preferably 14% ethanol, even more preferably 15% ethanol, most preferably 16% ethanol or higher. Preferably, said ethanol tolerant strain is a strain used for the production of bioethanol, even more preferably, said strain is a diploid, polyploid or aneuploid strain. Preferably, said yeast is a Saccharomyces cerevisiae yeast, even more preferably, it is a yeast industrially used for bioethanol production, more preferably said yeast is derived from the group consisting of PE2, CAT1 , BG1 and SA1 (Basso et al., 2008), most preferably said strain is derived from PE2 or CAT1. Derived, as used here, means that the strain is a direct transformant from the strain of which it is derived. A recombinant marker gene, as used here, means that the gene is introduced using genetic transformation, thereby changing the DNA organisation, when compared to the (non-transformed) parental strain. Preferably, said recombinant marker gene is inserted in the genome. A gene, as used here, is not limited to the coding sequence, but includes the promoter region and the terminator sequence. Even more preferably, said recombinant marker gene is an AGT1 gene, preferably a functional AGT1 gene, or a homolog, ortholog or variant thereof. Preferably, said functional AGT1 gene comprises SEQ ID N° 1 . Homologs or orthologs, as used here, may differ in sequence, but do have the same biological function. Preferably, said homolog or ortholog has at least 50% identities at protein level, preferably at least 60% identities, even more preferably at least 70% identities, more preferably at least 75% identities, more preferably at least 80% identities, even more preferably at least 85% identities, even more preferably at least 90% identities, most preferably at least 95% identities, as measured in a BLASTp (Altschul et al., 1997; Altschul et al., 2005) on the whole protein. A variant as used here, is a gene comprising the AGT1 coding sequence, or a homolog or ortholog thereof, wherein said coding sequence is placed under control of another promoter and/or terminator sequence. Preferably, said promoter and/or terminator sequence is also a Saccharomyces cerevisiae promoter and/or terminator sequence. Said marker gene may be inserted at the place of the endogenous gene - in case of the AGT1 gene replacing the non-functional AGT1 copy, or, preferably, it is inserted at a region that is highly stable and thus preserved between different Saccharomyces strains Even more preferably, said region is a non-coding region. Most preferably, said region is situated on chromosome XIV, between position 335000 and 336000, even more preferably between position 335300 and 335600. In case of a diploid, polyploidy or aneuploidy strain, only one copy, or several copies may be inserted. In the preferred embodiment, all endogenous copies may be replaced.

In one preferred embodiment, the introduction of the marker gene may be accompanied by the introduction of one or more other mutations or additional genes. Preferably the introduction of the marker gene is combined with the inactivation of the IME1 or another essential sporulation gene, thereby ensuring that the newly introduced AGT1 gene cannot be crossed out by sporulation and mating. This prevents the transfer of the marker gene to natural yeast strains and in particular to contaminant strains appearing in the bioethanol production process.

Another aspect of the invention is a method for detection of yeast contaminants during industrial bioethanol fermentation, comprising (a) isolation of the DNA of the yeast population, present in the fermentation (b) PCR amplification of the AGT1 gene (c) separating the PCR amplification products and (d) comparing the separation pattern obtained with a reference sample. Preferably, the primers used for PCR amplification are selected from the group consisting of CATGTTAC CTAC GTTAG GTAC (SEQ ID N°3), TCCAGGCACAAGTACCAGGT (SEQ ID N° 4), GATGGCAGATCAGGGGACG (SEQ ID N°5), TAGCACCTGGTACTTGTGCC (SEQ ID N° 6), TAACATGTCCACATAACATGCC (SEQ ID N° 7), ATTTGAACGACCACCTTTCCG (SEQ ID N° 8), TGTACCTAGTAACCTACAGGC (SEQ ID N° 9), ATAGTAAGTTGCATCTGTACCG (SEQ ID N° 10) and AGAAGTTTCGATTCGTAAAGGG (SEQ ID N° 1 1 ).

Still another aspect of the invention is the use of the AGT1 gene, or a homolog or ortholog or variant thereof for detection of contaminants in a Saccharomyces cerevisiae bioethanol production strain. A gene, as used here, is not limited to the coding sequence, but includes the promoter region and the terminator sequence. Homologs, orthologs and variants are as defined above. Preferably, said use is the use as a template for PCR amplification.

Yet another aspect of the invention is the use of the AGT1 gene, or a homolog or ortholog or variant thereof for improving the maltose fermentation in a Saccharomyces cerevisiae bioethanol production strain. Homologs, orthologs and variants are as defined above. Preferably, said use is the overexpression of the gene, by transforming one or more copies of the ATG1 gene or its homologs, orthologs or variants into a host strain wherein said overexpression means that the activity of the Agt1 protein in the strain according to the invention is higher than the host strain, either by a higher amount of protein, or by the production of a more efficient protein.

BRIEF DESCRIPTION OF THE FIGURES

Fig. 1 : Demonstration via PCR of the presence of the AGT1 cassettes in PE2-CD and CAT1 - CD: with primers F1 and R1 , a large PCR-product (> 3000bp) is occasionally obtained for the CD-strains, while an 867bp product is obtained with wild yeasts. With primers 11 and R1 , a 455bp product is obtained for PE2-CD and a 558bp product is obtained for CAT1 -CD, while no product is obtained with the wild yeasts. In this way, the PE2-CD and CAT1 -CD strains can be distinguished from each other and from other (wild) Saccharomyces cerevisiae yeasts.

Fig.2: Detection of contaminant Saccharomyces cerevisiae strains in a culture of PE2-CD. In the upper picture, the detection is directed towards the contaminant yeast. A contamination by 0.5% wild yeast can be detected with this method. In the lower picture, primers giving a product with the CD- strains are used.

Fig. 3: Detection of PE2-CD and CAT1 -CD in a mixture of strains. The mixture consists of equal amounts of PE2-CD and CAT1 -CD and different portions of a contaminant. It is clear that PE2-CD and CAT1 -CD can easily be distinguished from each other by this technique.

Fig.4: Detection of 37 different contaminant yeasts with primers 3F and 3Rv of set 3.

Fig.5: PE2-CD mixed with different amounts of "Crema de levedura"; this is the yeast cream that is used to re-inoculate a new fermentation. It contains the current production yeast and any wild yeast strains present. The PCR-primers used are those to detect the wild yeast; also the production yeast (which is not yet the "CD" form) is detected by those primers. It is clear that a 1 % presence of unmarked yeast from the "Crema de levedura" in the presence of 99% PE2-CD yeast can be detected. The PE2-CD strain is not detected at all, as should be the case with these primers.

Fig. 6: Glucose consumption (expressed as weight loss) during fermentation using PE2, PE2 1 AGT1 {one AGTI copy integrated), PE2-CD (two AGT1 copies integrated), CAT1 and CAT1 1 AGT1 (one AGT1 copy integrated) strains in a medium containing 20% glucose. Symbols are explained in the figure.

Fig.7: Sugar consumption (expressed as weight loss) during fermentation using PE2, PE2 1 AGT1 , PE2-CD, CAT1 and CAT1 1 AGT1 strains in a medium containing 10% glucose, 6% maltose and 4% maltotriose. The meaning of the strain names is as in Figure 5, except for PE2-CD which is indicated as PE2 2 AGT1 .

Fig.8: PCR-detection of PE2-CD-SD.

Lane 1 primer set F1 -R1 with PE2-CD-SD; this set gives only a band with "wild type" yeasts. As expected, no band is seen with PE2-CD-SD. Lane 2: the original PE2-CD; lane 3 the PE2- CD-SD; both strains were detected with primer set F3-I3.3 which is used to detect the CD- strains.

EXAMPLES

Strains used in the examples

- SAY 1403-7A (Alves et al. 2008) - PE2 (Basso et al. 2008)

- CAT1 (Basso et al. 2008)

- BG (Basso et al. 2008)

- SA1 (Basso et al. 2008)

Crema de levedura: mixtures of productions strains and contaminants (Fermentec) Example 1 : Generation of PE2-CD and CAT1 -CD

The AGT1 gene with its own promoter and terminator was amplified using DNA from the laboratory strain SAY 1403-7A as template. Primers 1 -AGT1 and 2-AGT1 (Tablel ) were used. PCR experiments were carried out following standard techniques as described by Sherman et al. (1992).

To PCR-amplify AGT1 from SAY 1403-7A |

1 - AGT1 lCATGTTACCTACGTTAGGTAC (SEQ ID NO:3)

2- AGT1 TCCAGGCACAAGTACCAGGT (SEQ ID NO:4) |

To distinguish PE2-CD, CAT1 -CD and contaminant yeast from each other

SET 1 F1 GATGGCAGATCAGGGGACG (SEQ ID NO:5)

11 TAGCACCTGGTACTTGTGCC (SEQ ID NO:6) |

The AGT1 gene was transformed into PE2 or CAT1 using the protocol of Gietz et al.(1995); The selection of the transformants was done on maltotriose-containing medium, as the original strains cannot grow on maltotriose, whereas the strain with a functional AGT1 gene can grow. PE2: AGT1 inserted in the two chromosomes XIV: nucleotides at position 335597 to 335600 are replaced by the AGT1 cassette CAT1 : AGT1 inserted in the two chromosomes XIV: nucleotides at position 335382 to 335497 are replaced by the AGT1 cassette. The resulting strains were called PE2-CD and CAT1 -CD.

The CAT1 -CD strain has been sporulated and tetrads analysed. From the tetrad analysis, the results clearly show that the AGT1 (with inactive PhiC31 integrase scar) has been inserted in both chromosomes XIV. This means that the AGT1 gene which was originally present in the strain [CAT1 +1AGT1] has been replaced by the new construct containing AGT1 plus the inactive PhiC31 integrase scar.

Example 2: Detection of the cassette in PE2-CD and CAT1 -CD Several primer sets have been developed to show the presence of the cassette in the CD- strains, and the absence in other (wild) yeasts. Each of these sets consists of three primers: one upstream of the cassette; one downstream of the cassette and one inside the cassette. The primers are listed in Table 1. An example of patterns obtained with different primer sets is shown Fig 1 , illustrating that the cassette is integrated, and that the pattern obtained for PE2- CD and CAT1 -CD differs from that of the contaminant strains.

Example 3: Detection of contaminant strains in a mixture with CD-strains

The presence of small quantities of wild yeast can be detected in the presence of a surplus of the CD-strains. This is done by PCR with the primers outside the AGT1 cassette. With primer F1 and R1 : only the wild yeasts will show the 867 bp product, no 867 band can be detected in the marked strains. Using the 11 - R1 combination, PE2-CD gives a band of 455 bp, and can not only be distinguished from the contaminant, but also from CAT1 -CD, which yields a band of 558 bp with this primer set. The example in Fig.2 shows that contaminations as low as 0.5 - 1 % can easily be detected with this method.

Example 4: Distinction of PE2-CD and CAT1 -CD from each other As indicated in example 3, it is possible to distinguish PE2-CD and CAT1 -CD from each other and from contaminants. This is further illustrated in Fig.3, proving that PE2-CD and CAT1 -CD can easily be distinguished from each other. PCR with a primer located in the AGT1 cassette together with one outside the cassette will give products of different sizes for PE2-CD and CAT1 -CD. Example 5: Detection of different wild Saccharomyces cerevisiae strains

Fig. 4 shows 37 different Saccharomyces cerevisiae strains which have been found as contaminants in bioethanol plants in Brazil. With PCR primer set 3, all 37 different yeasts can be detected.

Example 6: Distinction of PE2-CD from crema de levedura PE2-CD was mixed with different amounts of "Crema de levedura"; this is the yeast cream as used by Fermentec to re-inoculate a new industrial fermentation. It contains the current production yeast and any wild yeast strains present. The PCR-primers used are those to detect the wild yeast; also the production yeast (which is not yet the "CD" form) is detected by those primers. The results in Fig.5 show that wild yeasts and production yeast which were present in the yeast creme ("creme de levedura" ^**), can be detected when only 1 % of the yeast mixture consisted of wild/production yeasts. The substances present in this "creme de levedura" do not inhibit the detection.

Example 7: Fermentations with PE2-CD and related strains One of the prerequisites for the use of the CD-strains is that their performance in fermentations is at least as good as that of the original strains. Fermentations were performed with the original PE2 strain, a PE2 strain in which one AGT1 was inserted, and a PE2 strain where AGT1 was inserted two times (indicated as PE2-CD). Also, CAT1 and CAT1 with one AGT1 inserted were used. The results are shown in Fig.6 and Fig.7. The fermentation shown in Fig.6 is performed with a medium containing 20% glucose. It is clear that the presence of AGT1 has no influence on the fermentation of glucose. The fermentation shown in Fig.7 is performed with a medium containing a mixture of glucose (10%), maltose (6%) and maltotriose (4%). From the figure, it is clear that AGT1 insertion has a positive influence on the fermentation performance of PE2 as well as CAT1. The course of the fermentation can be divided into three phases: a first phase in which there is rapid fermentation of the glucose; a second phase in which maltose is fermented; and a third phase in which there is a slow fermentation of maltotriose. The presence of AGT1 causes, unexpectedly, a significant increase in the rate of maltose fermentation, and with two copies the maltose fermentation rate is even higher than with one copy.

Example 8: inactivation of the IME1

The 2 copies of IME1 , essential for sporulation, have been deleted in the PE2-CD strain. This strain does not sporulate anymore. This prevents that the introduced genetic modification, the specifically inserted AGT1 gene, is transferred to natural yeast strains that can act as contaminant strains in bioethanol production. The resulting strain is superior in view of biosafety regulations, whereas the fermentation performance is not affected. This new strain is called PE2-CD-SD (CD: Contaminant Detection; SD: Sporulation Deficient).

The primer sets used for the contaminant detection gave the same results as obtained previously with PE2-CD (see Fig.8).

REFERENCES

Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z., Miller, W. and Lipman, D.L. (1997), Gapped BLAST and PSI-BLAST: a new generation of protein database search programs, Nucleic Acids Res. 25, 3389-3402.

Altschul, S.F., Wootton, J.C., Gertz, E.M., Agarwala, R., Morgulis, A., Schaffer, A.A. and Yu, Y.K. (2005). Protein database searches using compositionally adjusted substitution matrices, FEBS J. 272, 5101 -5109.

Alves, S.L.-Jr, Herberts, R.A., Hollatz, C, Trichez, D., Miletti, L.C., de Araujo, P.S. and Stambuk, B.U. (2008) Molecular analysis of maltotriose active transport and fermentation by Saccharomyces cerevisiae reveals a determinant role for the AGT1 permease. AppI Environ Microbiol 74, 1494-14501 .

Basso LC, Amorim HV, de Oliveira, AJ, Lopes, ML. 2008. Yeast selection for fuel ethanol production in Brazil. FEMS Yeast Res 8, 1 155-1 163. da Silva-Filho, E.A., Brito dos Santos, S.K., Resende, A.M., de Morais, J.O., de Morais, M.A.-Jr, Simoes, D.A. (2005) Yeast population dynamics of industrial fuel-ethanol fermentation process assessed by PCR-fingerprinting. Antonie van Leeuwenhoek 88, 13- 23.

Day, R.E., Rogers, P.J., Dawes, I.W. and Higgins, V.J. (2002) Molecular analysis of maltotriose transport and utilization by Saccharomyces cerevisiae. AppI Environ Microbiol 68, 5326-5335.

Gietz, R.D., R.H.Schiestl, A.R.Willems & R.A.Woods (1995) Studies on the transformation of intact yeast cells by the LiAc/SS-DNA/PEG procedure. Yeast 11 : 355-360.

Han, E.K., Cotty, F., Sottas, C, Jiang, H. and Michels, C.A. (1995) Characterization of AGT1 encoding a general a-glucoside transporter from Saccharomyces. Mol Microbiol 17, 1093-1 107.

Sherman, F., G.R.Fink & J. B. Hicks (1992) Methods in yeast genetics. Cold Spring Harbor Laboratory press. Cold Spring Harbor, New York.

Smit, A., Moses, S.G., Pretorius, I.S. and Cordero-Otero, R.R. (2008) The Thr505 and Ser557 residues of the /\GT7-encoded α-glucoside transporter are critical for maltotriose transport in Saccharomyces cerevisiae. J AppI Microbiol 104, 1 103-1 1 1 1 . Stambuk, B.U., Alves-Jr, S.L., Hollatz, C, and Zastrow, C.R. (2006) Improvement of maltotriose fermentation by Saccharomyces cerevisiae. LettAppI Microbiol 43, 370-376.

Vidgren, V., Ruohonen, L, and Londesborough, J. (2005). Characterization and functional analysis of the MAL and MPH loci for maltose utilization in some ale and lager yeast strains. Appl Environm Microbiol 71 , 7846-7857.

Claims

1 . An ethanol tolerant yeast strain, comprising a recombinant marker gene, whereby said marker gene improves the fermentation performance.

2. An ethanol tolerant yeast strain according to claim 1 , wherein the improvement of fermentation performance is an improvement of maltose fermentation.

3. An ethanol tolerant yeast strain according to claim 1 or 2 wherein said maker gene is an AGT1 gene.

4. An ethanol tolerant yeast strain according to any of the previous claims, wherein said yeast strain is a Saccharomyces cerevisiae strain.

5. An ethanol tolerant yeast strain according to claim 4, wherein said yeast strain is a PE2 or CAT1 strain.

6. An ethanol tolerant yeast strain, according to any of the previous claims, further comprising a disruption of an essential sporulation gene.

7. An ethanol tolerant yeast strain, according to any of the previous claims, further comprising an IME1 disruption.

8. A method for detection of yeast contaminants during an industrial bioethanol fermentation, comprising (a) isolation of the DNA of the yeast population, present in the fermentation (b) PCR amplification of the AGT1 gene (c) separating the PCR amplification products and (d) comparing the separation pattern obtained with a reference sample.

9. The method according to claim 8, wherein said amplification is carried out using primers selected from the group consisting of SEQ ID N°3-SEQ ID N°1 1.

10. The use of the AGT1 gene for detection of contaminants in a Saccharomyces cerevisiae bioethanol production strain.

1 1 . The use of the AGT1 gene for improving the maltose fermentation in a Saccharomyces cerevisiae bioethanol production strain.