Author:

Abstract:

Introduction - Brettanomyces yeasts, with the species Brettanomyces (Dekkera) bruxellensis as the most important one, are generally reported as spoilage yeasts in the beer and wine industry due to production of phenolic off-flavors. However, B./D. bruxellensis is also known as a beneficial or even crucial contributor in certain fermentation processes, such as the production of specialty beers. Nevertheless, despite its economic importance, Brettanomyces yeasts have remained poorly understood at the genetic and genomic level so far. In this study, the genetic relationship between 50 Brettanomyces strains belonging to all species presently identified within the genus and isolated from several sources was studied using a combination of DNA fingerprinting techniques. To further explore this relationship, we sequenced a (beneficial) beer isolate of B./D. bruxellensis (VIB X9085; ST05.12/22) and compared its genome sequence with that of a wine spoilage strain (AWRI 1499). Material and Methods – Strains were isolated from different food products and beverages (beer, soft drink, wine and isolates from lambic casks from the Cantillon brewery), representing the different Brettanomyces species, i.e. B. anomalus (D. anomala), B./D. bruxellensis (D. bruxellensis), B. custersianus, B. naardenensis and B. nanus. Isolates were clustered according to their LSU rRNA genes and patrons obtained by three DNA fingerprint techniques including RAPD-PCR, AP-PCR and rep-PCR. A paired-end library and two mate-pair libraries of the beer strain (VIB X9085; ST05.12/22) were sequenced on the Illumina HiSeq 2000 platform. Contigs were de novo assembled using SOAPdenovo v1.05, and scaffolded with SSPACE. S. cerevisiae gene models were used as reference for the prediction of coding sequences (CDS) by AUGUSTUS v2.5.5. Consensus SNPs and InDels were called using SAMtools and GATK with AWRI 1499 (published by curtin et al.; 2012) as a reference. GO Slim terms obtained with Blast2GO software were ranked by overall average variations to find functional differences between strains. Unique genes in both strains were determined by comparing both whole genome sequences using Blastn, and confirmed by PCR. Results and discussion - Cluster analysis of the fingerprints obtained in this study indicate a strong correlation between the genetic profiles and the niches where the respective strains were isolated, rather than with geographic origin or year of isolation, thus suggesting niche adaptation. Genomic comparison revealed that ST05.12/22 was found to be substantially different from AWRI 1499, especially at the level of single nucleotide polymorphisms (SNPs), which were present, on average, in every 300 bp. In addition, there were major differences in the genome structure between both strains, including the presence of large duplications (20-171 kb) and deletions (>1 kb). Gene content analysis revealed the presence of twenty-eigth genes which were predicted to encode proteins that are unique to AWRI 1499, including many genes involved in carbon and nitrogen metabolism. Most of these genes were clustered in five genomic regions, suggesting they may be acquired through horizontal gene transfer. Vice versa, two genes were found in ST05.12/22 that were missing in AWRI 1499. What’s innovative ¬¬- Together, these analyses describe tools to discriminate between Brettanomyces strains and provide a first glimpse at the exceptional genetic diversity and genome plasticity of B./D. bruxellensis yeasts. Practical revelance - By comparing sufficiently large sets of whole genomes, coupled with functional and phenotypic analyses, we hope to be able to answer the question whether there are distinct groups of B./D. bruxellensis which have a distinct impact on the production of beer and wine or other beverages, or, more generally, to further understand the behavior of this economically important yeast.