The genome of wine yeast Dekkera bruxellensis provides a tool to explore its food-related properties

https://doi.org/10.1016/j.ijfoodmicro.2012.05.008Get rights and content

Abstract

The yeast Dekkera/Brettanomyces bruxellensis can cause enormous economic losses in wine industry due to production of phenolic off-flavor compounds. D. bruxellensis is a distant relative of baker's yeast Saccharomyces cerevisiae. Nevertheless, these two yeasts are often found in the same habitats and share several food-related traits, such as production of high ethanol levels and ability to grow without oxygen. In some food products, like lambic beer, D. bruxellensis can importantly contribute to flavor development. We determined the 13.4 Mb genome sequence of the D. bruxellensis strain Y879 (CBS2499) and deduced the genetic background of several “food-relevant” properties and evolutionary history of this yeast. Surprisingly, we find that this yeast is phylogenetically distant to other food-related yeasts and most related to Pichia (Komagataella) pastoris, which is an aerobic poor ethanol producer. We further show that the D. bruxellensis genome does not contain an excess of lineage specific duplicated genes nor a horizontally transferred URA1 gene, two crucial events that promoted the evolution of the food relevant traits in the S. cerevisiae lineage. However, D. bruxellensis has several independently duplicated ADH and ADH-like genes, which are likely responsible for metabolism of alcohols, including ethanol, and also a range of aromatic compounds.

Highlights

► Genome sequence of an important wine spoiling yeast ► Genomics of food related microorganisms ► Genetic background for aromatic compounds in wine ► Comparative genomics reveals evolutionary strategies.

Introduction

There is an enormous diversity among yeast species, including those that play important roles in traditional food processes, often in mixed cultures in spontaneous fermentations. One such yeast is Dekkera/Brettanomyces bruxellensis, associated with lambic beer fermentation and wine production, especially as a contributor, in a positive or negative manner, to flavor development (Du Toit and Pretorius, 2000). This yeast can produce phenolic compounds, such as 4-ethylguaiacol and 4-ethylphenol, which could lead to wine spoilage if present in high enough concentration (Heresztyn, 1986, Vigentini et al., 2008). In fact, D. bruxellensis represents a serious problem in wine industry, causing enormous economic losses as a consequence of wine spoilage (Wedral et al., 2010). However, in spite of the economic impact of D. bruxellensis, this yeast remains poorly studied.

D. bruxellensis is apparently not a close relative of baker's yeast Saccharomyces cerevisiae, but the phylogenetic position of the D. bruxellensis group has so far been rather impossible to determine (Woolfit et al., 2007). Both yeasts share several “peculiar” and rather “unusual” traits important for food-related properties, such as production of high ethanol levels, high tolerance towards ethanol, and the ability to grow without oxygen and in acidic environments (Rozpędowska et al., 2011). Apparently, given the lack of relatedness, these traits evolved in parallel in both groups, but it is unclear if the molecular mechanisms behind these properties are similar or different (Rozpędowska et al., 2011). In other words, these two yeasts represent an ideal model to study molecular processes involved in convergent and parallel evolutionary routes.

Ethanol production and capability to survive without oxygen are highly relevant in food fermentations. In S. cerevisiae, but not in D. bruxellensis, the corresponding genetic factors that underlie these traits have been relatively well studied. For example, the whole genome duplication (WGD), duplicated gene profiles, the horizontal transfer of the URA1 gene (coding for the DHODase, dihydroorotate dehydrogenase, catalyzing the fourth pyrimidine de novo pathway step), and lineage-specific duplication of the ADH genes (encoding alcohol dehydrogenases), have been shown to be at least partially responsible for the development of the S. cerevisiae high fermentation capacity and/or anaerobic properties (reviewed in Piskur and Langkjaer, 2004, Piskur et al., 2006). It is not known whether similar molecular strategies are responsible for the domination of the same environment by D. bruxellensis.

Recently, a partial genome sequence of one strain of D. bruxellensis has been reported, and the analysis estimated that this yeast has around 7.500 genes, of which many lack a homolog in the S. cerevisiae genome (Woolfit et al., 2007). Further analysis of the partial sequence has revealed that D. bruxellensis is not a simple haploid. Its genome contains approximately 1% polymorphic sites but the exact physical background for this heterozygocity is not known (Hellborg and Piskur, 2009).

Here we determined the whole genome sequence of the D. bruxellensis strain Y879 (CBS2499) and used it to deduce several “food-relevant” properties and evolution pathways of this yeast.

Section snippets

Genome sequencing and assembly

The genome of D. bruxellensis strain Y879 (CBS2499) was sequenced using a combination of 454 and Illumina sequencing platforms (GYBS 454 standard rapid, GYHO 454 standard rapid, GYHG 454 titanium 4 kb, GYFW 454 titanium 4 kb, GXXW Illumina 2 × 76 300 bp, ICHI Illumna 2 × 150 270 bp, and ICCY Illumina 2 × 100 4 kb CLIP). All general aspects of library construction and sequencing can be found at the JGI website (http://www.jgi.doe.gov/). An initial assembly of GXXW was conducted for QC purposes using the

General genome parameters

The 13.4 Megabase genome of D. bruxellensis CBS2499 was sequenced using a combination of 454 and Illumina platforms, assembled with AllPaths assembler and annotated using JGI annotation pipeline to predict 5600 genes (Table 1A, B, C, D, E). The obtained genome size is significantly smaller from the one deduced from the previously determined partial sequence (Woolfit et al., 2007). The previous wrong prediction was likely due to the problems with ploidy because D. bruxellensis is not a simple

Acknowledgments

JP acknowledges a Slovenian ARRS grant and a Swedish Sörensen Foundation grant to support his work on wine yeasts. TG and MMH acknowledge a grant from the Spanish Ministry of Science (BFU09-09268). The sequencing and annotations were conducted by the U.S. Department of Energy Joint Genome Institute and was supported by the Office of Science of the U.S. Department of Energy under contract no. DE-AC02-05CH11231.

References (51)

  • M. Du Toit et al.

    Microbial spoilage and preservation of wine: using weapons from nature's own arsenal

    South African Journal of Enology and Viticulture

    (2000)
  • S.R. Eddy

    Accelerated profile HMM searches

    PLoS Computational Biology

    (2011)
  • R.C. Edgar

    MUSCLE: multiple sequence alignment with high accuracy and high throughput

    Nucleic Acids Research

    (2004)
  • A.J. Enright et al.

    An efficient algorithm for large-scale detection of protein families

    Nucleic Acids Research

    (2002)
  • J. Felsenstein

    PHYLIP (Phylogeny Inference Package) version 3.6

    (2005)
  • T. Gabaldón

    Large-scale assignment of orthology: back to phylogenetics?

    Genome Biology

    (2008)
  • S. Galafassi et al.

    Dekkera/Brettanomyces yeasts for ethanol production from renewable sources under oxygen-limited and low-pH conditions

    Journal of Industrial Microbiology & Biotechnology

    (2011)
  • O. Gascuel

    BIONJ: an improved version of the NJ algorithm based on a simple model of sequence data

    Molecular Biology and Evolution

    (1997)
  • Z. Gojković et al.

    Horizontal gene transfer promoted evolution of the ability to propagate under anaerobic conditions in yeasts

    Molecular Genetics and Genomics

    (2004)
  • I.V. Grigoriev et al.

    The Genome Portal of the Department of Energy Joint Genome Institute

    Nucleic Acids Research

    (2012)
  • S. Guindon et al.

    Estimating maximum likelihood phylogenies with PhyML

    Methods in Molecular Biology

    (2009)
  • L. Hellborg et al.

    Complex nature of the genome in a wine spoilage yeast, Dekkera bruxellensis

    Eukaryotic Cell

    (2009)
  • T. Heresztyn

    Metabolism of volatile phenolic compounds from hydroxycinnamic acids by Brettanomyces strains

    Archives of Microbiology

    (1986)
  • J. Huerta-Cepas et al.

    Assigning duplication events to relative temporal scales in genome-wide studies

    Bioinformatics

    (2011)
  • J. Huerta-Cepas et al.

    ETE: a python environment for tree exploration

    BMC Bioinformatics

    (2010)
  • Cited by (0)

    View full text