Faculty of Science
DIFFERENTIAL CONTRIBUTIONS OF SACCHAROMYCES CEREVISIAE
BRETTANOMYCES CLAUSSENII TO A BELGIAN STRONG BEER
2019 | BREANNE MARIE MCAMMOND

B.Sc. Honours thesis – Biology

AND

DIFFERENTIAL CONTRIBUTIONS OF Saccharomyces cerevisiae AND Brettanomyces
claussenii TO A BELGIAN STRONG BEER
by

BREANNE MARIE MCAMMOND

A THESIS SUBMITED IN PARTIAL FULFILLMENT
OF THE REQUIREMENT FOR THE DEGREE OF

BACHELOR OF SCIENCE (HONS.)
in the
DEPARTMENT OF BIOLOGICAL SCIENCES
(General Biology)

This thesis has been accepted as conforming to the required standards by:
Jonathan D. Van Hamme (Ph.D.), Thesis Supervisor, Dept. Biological Sciences
Eric M. Bottos (Ph.D.), Co-supervisor, Dept. Biological Sciences
Mark Rakobowchuk (Ph.D.), External Committee Member, Dept. Biological Sciences

Dated this 20th day of August, 2019, in Kamloops, British Columbia, Canada

© Breanne Marie McAmmond, 2019

i

ABSTRACT
Parallel 49 Brewing Company has become an award-winning microbrewery at the heart of
British Columbia’s growing beer culture. Wild Ride, one of their most successful beers, is brewed
with a co-culture of Saccharomyces cerevisiae and Brettanomyces claussenii (synonym B.
anomalus). While the use of co-cultures in brewing represents a profitable niche market, Parallel
49 has ceased production of Wild Ride as these fermentations are technically challenging and
difficult to reproduce. In order to support the development of alternative production methods,
Parallel 49 needs to gain an understanding of the genomic profiles of the two yeast strains, profile
yeast metabolites in relation to gene expression, and understand the genetic and metabolic
interactions during co-culture fermentation. I hypothesize that, during co-culture fermentation, the
“omic” profiles of the two yeast strains will be altered, and that there will be detectable interactions
between the two strains. In order to test this hypothesis, a S. cerevisiae mono-culture brew, a B.
claussenii mono-culture brew, and a Wild Ride co-culture brew were carried out. Fermentations
proceeded for twenty-two days according to Parallel 49’s recipe, and specific gravity, dissolved
oxygen and pH were monitored. Daily samples were taken for metabolite analysis via heated
headspace gas chromatography coupled to a flame ionization detector and a mass spectrometer,
transcriptomic analysis via RNA-seq on an Ion S5 System, and proteomic analysis via Waters
Synapt G2 high definition mass spectrometry (Q-TOF MS) coupled to a nanoAcquity ultra
performance liquid chromatography system. The genomes of both yeast cultures were sequenced
on an Ion S5 System. To date, the day 7 transcriptomic profile of the Wild Ride co-culture has been
sequenced, producing 9,734,188 Q20 reads and >= 1,620,889,603 Q20 bases. The Wild Ride reads
were mapped to the Saccharomyces cerevisiae S288C reference genome at 17.65%, with 76.93%
of the reads mapping to rRNA regions located on chromosome XII. Through metabolic analysis,
33 compounds were resolved and identified by mass spectrometry in the three fermentations; of
those, production kinetics for 10 were monitored using flame ionization detection. From whole
genome sequencing. 6,281,650 Q20 reads and >= 1,887,285,253 Q20 bases were produced for
Saccharomyces, while 5,804,220 Q20 reads and >= 1,655,746,215 Q20 bases were produced for
Brettanomyces. From this, a contaminant was detected in the two yeast cultures that shared 99.98%
average nucleotide identity with the bacterium Cellulosimicrobium cellulans strain NEB113; the
average nucleotide identity between NEB113 and the two contaminants sequenced in the two yeast
cultures was >99.9%. Currently, proteomic analysis of samples is underway at the University of

ii

Regina. In the future, additional metabolites will be identified in the liquid phase of fermentation
cultures, and quantities of each compound in all three brews will be determined. Additional
transcriptome sequencing will be carried out so that multiple sample days from all three brews can
be analyzed, and yeast abundance in the Wild Ride co-culture will be monitored by qPCR. The
genomes of S. cerevisiae and B. claussenii will be sequenced at a greater depth of coverage so that
they may be used as reference genomes for transcriptome mapping. The possibility of C. cellulans
contamination will be eliminated by re-sequencing the genomes using new cultures, and different
rRNA removal techniques will be implemented. The mono- and co-culture fermentations will be
repeated so that all aspects of the experiment can be replicated.

Thesis Supervisor: Professor Jonathan D. Van Hamme

iii

ACKNOWLEDGEMENTS

Thank you to the Natural Sciences and Engineering Research Council of Canada (NSERC) for
funding this research through an Undergraduate Student Research Award (USRA) and an NSERC
Engage Grant.

Thank you to Kelsey Dodds and the Parallel 49 Brewing Team for being active and enthusiastic
collaborators throughout this project.

Thank you to Trent Hammer for helping with the GC-MS work for this project.

Thank you to Dr. Tzu-Chaio Chao for the proteomics method development and analysis.

Thank you to Dr. Eric M. Bottos for playing such an instrumental role in this project. Your wealth
of knowledge is invaluable, and your encouragement and willingness to help throughout the
process is so appreciated.

Finally, thank you to Dr. Jonathan D. Van Hamme for being a motivating and patient mentor and
supportive thesis supervisor. Your enthusiasm for science and dedication to student research
continues to inspire me. I feel extremely grateful to have worked under your guidance for the past
four years, and I look forward to more collaboration in the future.
Thank you for helping me realize my potential in science, and for making me into the researcher I
am today.

iv

Table of Contents
ABSTRACT ............................................................................................................................... ii
ACKNOWLEDGEMENTS ...................................................................................................... iv
1

INTRODUCTION .............................................................................................................. 1

2

MATERIALS AND METHODS ........................................................................................ 6
2.1

Yeast storage, propagation and inoculum preparation ................................................ 6

2.2

Genome sequencing and analysis ............................................................................... 6

2.2.1

Genomic DNA isolation ...................................................................................... 6

2.2.2

Sequencing library preparation ............................................................................ 6

2.2.3

Ion Torrent 5S XL sequencing ............................................................................. 7

2.2.4

Genome assembly and analysis ........................................................................... 7

2.3

Wort preparation ......................................................................................................... 8

2.4

Fermentations .............................................................................................................. 8

2.4.1

Bioreactors ........................................................................................................... 8

2.4.2

Wort inoculation .................................................................................................. 8

2.4.3

Sampling conditions............................................................................................. 9

2.4.4

Specific gravity .................................................................................................... 9

2.5

TRIzol extractions ....................................................................................................... 9

2.6

mRNA isolation and sequencing ................................................................................ 9

2.6.1

3

Transcriptome analysis ...................................................................................... 10

2.7

Protein extraction and analysis ................................................................................. 10

2.8

Gas chromatography ................................................................................................. 10

2.8.1

Sample preparation ............................................................................................ 10

2.8.2

Analytical standards ........................................................................................... 11

2.8.3

Headspace chromatography and detector conditions ......................................... 11

2.8.4

Headspace sampling........................................................................................... 12

RESULTS ......................................................................................................................... 12
3.1

Bioreactor parameters ............................................................................................... 12

3.2

Genome sequencing, assembly and analysis ............................................................ 15

v

4

3.3

Transcriptomic Results ............................................................................................. 21

3.4

Volatile metabolites .................................................................................................. 21

DISCUSSION ................................................................................................................... 28
4.1

Genome sequencing and Cellulosimicrobium contamination................................... 28

4.2

Bioreactor parameters measured ............................................................................... 29

4.3

Transcriptomics and rRNA contamination ............................................................... 30

4.4

Proteomics................................................................................................................. 30

4.5

Metabolomics ............................................................................................................ 31

4.5.1

Identified compounds......................................................................................... 31

4.5.2

4-EP and 4-EG ................................................................................................... 31

4.6

Future Work .............................................................................................................. 32

4.6.1

qPCR .................................................................................................................. 32

4.6.2

Repeat fermentations and different brew types ................................................. 32

5

REFERENCES ................................................................................................................. 33

6

SUPPLEMENTAL INFORMATION .............................................................................. 38
6.1

Modified CONDA Pronadisa Brettanomyces Media ................................................ 38

6.2

Sample commands used for metagenome analysis ................................................... 38

6.3

Mass spectrometry parameters .................................................................................. 39

6.4

GC-FID Ethanol Standards ....................................................................................... 48

List of Figures
Figure 3.1 Oxygen consumption in the S. cerevisiae mono-culture, B. claussenii mono-culture,
and the Wild Ride co-culture brews over the course of the fermentations. ........................... 13
Figure 3.2 Variation in pH in the S. cerevisiae mono-culture, B. claussenii mono-culture, and the
Wild Ride co-culture brews over the course of the fermentations......................................... 14
Figure 3.3 Change in specific gravity in the S. cerevisiae mono-culture, B. claussenii monoculture, and the Wild Ride co-culture brews over the course of the fermentations. .............. 15
Figure 3.4 Section of Saccharomyces (bottom) whole genome alignment to Saccharomyces
cerevisiae S288C reference genome (top) showing regions of high homology. ................... 17

vi

Figure 3.5 Section of Brettanomyces (top) whole genome alignment to Brettanomyces
bruxellensis reference genome (bottom) showing regions of high homology. ...................... 17
Figure 3.6 Average nucleotide identity calculation for Brettanomyces contaminant and
Cellulosimicrobium cellulans strain TH-20. .......................................................................... 19
Figure 3.7 Average nucleotide identity calculation for Brettanomyces contaminant and
Saccharomyces contaminant. ................................................................................................. 19
Figure 3.8 Average nucleotide identity calculation for Brettanomyces contaminant and
Cellulosimicrobium cellulans strain NEB113. ...................................................................... 20
Figure 3.9 Average nucleotide identity calculation for Saccharomyces contaminant and
Cellulosimicrobium cellulans strain NEB113. ...................................................................... 20
Figure 3.10 Peak areas of identified compounds or retention times in the Brettanomyces, Wild
Ride co-culture, and Saccharomyces brews as sampled on day 13 of the fermentations as
measured by GC-FID. n = 3, error bars represent standard deviation, p = 0.05. ................... 24
Figure 3.11 Change in peak area of 1-propanol in the B. claussenii (Bc), Wild Ride (WR), and S.
cerevisiae (Sc) brews. ............................................................................................................ 25
Figure 3.12 Change in peak area of ethyl acetate in the B. claussenii (Bc), Wild Ride (WR), and
S. cerevisiae (Sc) brews. ........................................................................................................ 25
Figure 3.13 Change in peak area of 2-methyl-1-propanol in the B. claussenii (Bc), Wild Ride
(WR), and S. cerevisiae (Sc) brews. ...................................................................................... 26
Figure 3.14 Change in peak area of 3-methyl-1-butanol in the B. claussenii (Bc), Wild Ride
(WR), and S. cerevisiae (Sc) brews. ...................................................................................... 26
Figure 3.15 Change in peak area of ethyl hexanoate in the B. claussenii (Bc), Wild Ride (WR),
and S. cerevisiae (Sc) brews. ................................................................................................. 27
Figure 3.16 Change in peak area of ethyl octanoate in the B. claussenii (Bc), Wild Ride (WR),
and S. cerevisiae (Sc) brews. ................................................................................................. 27
Figure 3.17 Change in peak area of ethyl decanoate in the B. claussenii (Bc), Wild Ride (WR),
and S. cerevisiae (Sc) brews. ................................................................................................. 28
Figure 6.1 Detector response and retention time of various ethanol standards run daily to ensure
consistent response during GC-FID. ...................................................................................... 48

vii

List of Tables
Table 1. Brettanomyces genomes available on NCBI. .................................................................. 17
Table 2. Average nucleotide identities (%) between Cellulosimicrobium cellulans contaminants
from yeast cultures and high quality reference genomes available on NCBI ........................ 18
Table 3. Compounds identified by retention time and spectra via GC-MS in S. cerevisiae, B.
claussenii, and Wild Ride fermentations (* International Union of Pure and Applied
Chemistry, ** Chemical Abstracts Service Registry Number, *** http://www.ymdb.ca/,
**** Alternate source: The Good Scents Company). ............................................................ 22

viii

1

INTRODUCTION
Parallel 49 Brewing Company opened in East Vancouver in 2012 and has become an award-

winning microbrewery central to the growing beer culture in British Columbia. With seven yearround beers, Parallel 49 prides itself on producing >35 seasonal and special release brews,
including four unique barrel aged beers. This level of innovation requires significant investment in
research and development, as well as investment in brewing with a range of yeast species, in monoand co-culture. One of their most successful beers, Wild Ride, is brewed with a co-culture of
Saccharomyces cerevisiae and Brettanomyces claussenii (synonym B. anomalus) using a fairly
complex and laborious process. The two yeast species exhibit different growth characteristics, with
B. claussenii growing slower than S. cerevisiae; they also produce different secondary metabolites
as observed in beer flavour profiles. The two members of this community appear to behave
differently in isolation compared to when they are together, suggesting that interactions between
the two species affect fermentation characteristics.
The use of co-cultures in brewing represents a potentially lucrative market; however, these
fermentations are difficult to control and reproduce. This in turn makes them less desirable to be
brewed in large scale fermentations. Because of this, Parallel 49 has halted production of beer
relying on co-cultures of Saccharomyces and Brettanomyces until they can find simpler production
methods. Understanding the fundamental ecology of yeast co-cultures is a step towards control and
reproducibility, and provides an opportunity to model cooperative and competitive interactions
between two related species. While the lineages of Brettanomyces and Saccharomyces diverged
some 200 million years ago, they share some common traits, such as their ability to thrive in
fermentative environments with tolerance to ethanol stress, osmotic stress and low pH (Steensels
et al. 2015).
Saccharomyces cerevisiae has been used for millennia in brewing, wine making, bread
making, and distilling. Its use has been traced back to 3150 BC, having been detected in pots buried
with King Scorpion I, one of the first kings of Egypt (Landry et al. 2006). The name Saccharomyces
cerevisiae was first introduced in 1838 by Franz Meyen (Barnett 1992), and today it is the standard
unicellular eukaryotic model organism (Sulo et al. 2017), critical in our current understanding of
eukaryotic cellular processes. The first type strain of S. cerevisiae isolated for research purposes
was S. cerevisiae S288C (Landry et al. 2006).
1

Brettanomyces was first isolated in 1904 by Hjelte Claussen, though it was not assigned a
genus name until 1921, when it was isolated from lambic beer (Curtin and Pretorius 2014). In
present day, the genus Brettanomyces represents the anamorphic variant of the yeast, while the
genus Dekkera is known as the teleomorphic variant, as some strains demonstrate ascospore
formation (Steensels et al. 2015). Generally, Brettanomyces is viewed as a contaminant in wineries
(Steensels et al. 2015; Serra Colomer et al. 2019); however, it has been suggested that
Brettanomyces is the second most industrially important yeast, with only S. cerevisiae being of
greater importance (Curtin and Pretorius 2014; Gibson et al. 2017). Brettanomyces anomalus, in
particular, is essential in Belgian lambic beers that are produced via spontaneous fermentation
(Gibson et al. 2017; Serra Colomer et al. 2019), where it becomes dominate five to eight months
into the fermentation (Boulton and Quain 2006). Brettanomyces is able to contribute flavours to
beer that Saccharomyces is unable to (Gibson et al. 2017). When fermented, Brettanomyces is
known to produce undesirable taints including horse sweat, barnyard, medicinal or leathery
flavours (Serra Colomer et al. 2019), which is informally known as “Brett character” or “Brett
flavour” (Curtin and Pretorius 2014; Steensels et al. 2015; Gibson et al. 2017). When utilized
correctly, it can produce exotic flavours including pineapple, mango, grape, and pear (Serra
Colomer et al. 2019).
Though through history they have been chosen by accident, the strains of yeast used in beer
production is considered to be the most essential component of the brewing process (Boulton and
Quain 2006). Brewing yeasts have differing optimum growth temperatures and sugar preferences,
which ultimately lead to different flavour profiles; the transcriptomic profile of a yeast can also
vary wildly based on the wort recipe and fermentation conditions (Boulton and Quain 2006;
Wendland 2014). Gene expression in fermenting yeasts typically show increases during the first
48 hours of fermentation (Wendland 2014), with genes encoding for glutamine, asparagine, and
threonine-specific transporters tending to be upregulated around the onset of fermentation
(Procopio et al. 2014). Genes for amino acid metabolism are up- and down-regulated over the
course of fermentation (Schoondermark-Stolk et al. 2006), and as amino acid concentrations
decrease as fermentation progresses, genes for amino acid-specific transmembrane proteins are
upregulated (Procopio et al. 2014). Some genes, such as those providing a measure of resistance to
increasing ethanol concentrations, have been found to increase over the course of fermentation

2

(Wendland 2014). Global gene expression, however, generally decreases over time until the
fermentative process comes to a halt (Wendland 2014).
The proteomic profiles of beers can give insight to the physiological condition of yeast
(Iimure and Sato 2013). Yeast proteomes change in response to environmental conditions during
fermentation. Temperature stress, nutrient limitation, osmotic stress, ethanol concentration, and pH
changes can all affect yeast proteome profiles (Trabalzini et al. 2003; Kobi et al. 2004). Even mild
changes in environmental conditions can elicit complex proteomic responses (Trabalzini et al.
2003), though the majority of proteomic profile changes have been observed to occur at the onset
of anaerobic conditions as yeast transition to fermentative metabolism (Kobi et al. 2004).
Decreased expression of the acetyl-CoA pathway and of carbohydrate metabolism are observed
during this time, and decreased expression of proteins involved in citric acid cycle activity is noted
(Kobi et al. 2004). Under high glucose concentrations, S. cerevisiae has been shown to experience
dramatic changes in protein profiles, including decreased heat-shock protein production, decreased
production of amino acid metabolism-related proteins, and increased production of carbohydrate
metabolism proteins (Pham et al. 2006). Key enzymes involved in alcohol fermentation, such as
protein Pdc5p, are found to be produced in response to high glucose conditions, likely to cope with
osmotic stress (Pham et al. 2006). In general, protein synthesis has been found to decline under
osmotic stress (Pham and Wright 2008).
Hundreds of flavour compounds work together to produce the final taste of a beer (Pires et
al. 2014). Higher alcohols and esters play a key role in the production of a desirable brew (Pires et
al. 2014) and, of the metabolites that contribute to flavour profile, acetate esters are the major group
(Procopio et al. 2014), acting synergistically with other lower concentration compounds to change
beer flavour (Pires et al. 2014). Higher alcohol metabolites such as isoamyl alcohol, isobutyl
alcohol, and hexan-1-ol have been found to reach their highest concentrations four to five days into
the fermentation (Procopio et al. 2014). They are produced during amino acid synthesis or amino
acid catabolism, particularly branched chain amino acids (BCAAs) including leucine, isoleucine,
and valine (Schoondermark-Stolk et al. 2006; Procopio et al. 2014; Olaniran et al. 2017). Higher
alcohols contribute to beer flavour and aroma in both positive and negative ways, with higher
concentrations leading to undesirable characteristics (Olaniran et al. 2017).

3

By studying genome-wide transcriptional expression, protein production, and metabolite
production, the full potential of brewing yeast can be uncovered (Smart 2007). With this
knowledge, there is potential for altering the characteristics and capabilities of brewing yeast
through targeted mating, improving brewing recipes, and producing products more efficiently.
Flavour production, fermentation time, sugar usage, and flocculation are all areas of yeast
physiology that have practical implications in the brewing industry (Dequin 2001; Wendland
2014). For example, there has been interest in the wine industry to combine particular metabolic
and physiologic characteristics of different yeast species using high-throughput mating techniques
(Gibson et al. 2017; Figueiredo et al. 2017). A good example would be exploiting yeast flocculation
genes to lower clarification costs at the brewery. In terms of flavour profile, increased acetate ester
and sulfur dioxide production, dextrin fermentation, and decreased hydrogen sulfide production
were identified as targets for improvement in brewing yeast (Dequin 2001). The reduction of
acetaldehyde concentration is desirable, as this compound tends to lend unfavourable flavours
including grass and walnuts (Shen et al. 2014). Despite the advancements that could be made in
the brewing industry, consumers often avoid food products that have, in their minds, been tampered
with by science. Public opinion is truly the largest obstacle in the way of the introduction of novel
technologies into the brewing industry (Dequin 2001).
Despite being an ancient practice dating back at least 8000 years (Debowski 2014; Liu et
al. 2018), and the significant body of published research in the field dating back to Pasteur (Boulton
and Quain 2006), the biological subtleties of beer brewing still remain a mystery. With the first
Saccharomyces cerevisiae genome being published in 1996 (Goffeau et al. 1996), and the
subsequent development of high-throughput sequencing technologies for DNA, RNA and protein,
there is a growing body of molecular information available for yeast (Wilkening et al. 2013; PerezTraves et al. 2014; Walther et al. 2014; Zhang et al. 2015) given their importance in beer, wine and
spirit brewing, bioethanol production, and medicine. Only a small number of studies have deployed
high-throughput sequencing for yeast transcriptomics and proteomics (Procopio et al. 2011, 2014;
Nookaew et al. 2012; Gibney et al. 2013; Treu et al. 2014; Xu et al. 2014a, 2014b; Sardu et al.
2014) and there are no descriptions of controlled mixed-yeast fermentations for beer brewing
(Carrau et al. 2015).
At present, Parallel 49 critically needs to understand the genetic makeup of the
Saccharomyces and Brettanomyces strains they use; characterize the metabolites produced by the
4

individual strains in relation to their gene expression profiles; and understand how yeast strains
genetically and metabolically interact during mixed-strain fermentations. I hypothesize that the
“omic” profiles of the two individual species will be altered during the mixed-strain fermentations,
and that interactions between the two organisms will be detected.
In order to test this hypothesis, three fermentations were carried out: S. cerevisiae in monoculture; B. claussenii in mono-culture; and a co-culture of both yeast species as found in the Wild
Ride beer. Fermentations proceeded for twenty-two days, and throughout the process, specific
gravity, dissolved oxygen and pH were monitored. Samples were taken daily for chemical analysis
to characterize metabolomic profiles using gas chromatography coupled to both a flame ionization
detector and a mass spectrometer. Thirty-three metabolites have been identified in all three of the
fermentations, and they are currently being quantified. At present, a draft transcriptomic profile of
the Wild Ride co-culture brew at day 7 of the fermentation has been sequenced and analyzed. The
analysis of proteomes from samples that have been sent to the University of Regina is also
underway. Future work includes further analysis of metabolites in the liquid phase, the sequencing
of the two mono-culture transcriptomes from multiple sample days, and transcriptome sequencing
of additional sample days from the Wild Ride co-culture brew. The genomes of S. cerevisiae and
B. claussenii will be sequenced with greater depth of coverage so that they may be used as reference
genomes for transcriptome mapping, and yeast abundance in the co-culture will be monitored by
quantitative PCR to establish growth profiles of the two species.
This project will provide genomic and metabolomic datasets to Parallel 49 that will
hopefully guide their development of a process for brewing their beer with a more reliable and
cost-effective yeast mono-culture, and to gain better control over their current co-cultured products.
This project also has potential to contribute an understanding of fundamental ecological
interactions between microorganisms at the transcriptomic, proteomic, and metabolomic levels.
This process-based understanding has potential to leverage ecological knowledge for industrial
application, and can be viewed as a simple model from which other research endeavors can be
based.

5

2

MATERIALS AND METHODS

2.1

Yeast storage, propagation and inoculum preparation
Yeast slants of S. cerevisiae (Parallel 49 Brewing, Vancouver, BC) and B. claussenii (White

Labs) were supplied by Parallel 49 Brewing and were stored at 4 °C throughout the experiment.
Both yeasts were streaked on modified CONDA Pronadisa Brettanomyces agar (Appendix section
6.1) and were incubated at room temperature. After three days of growth for S. cerevisiae and 12
days of growth for B. claussenii, duplicate 10-ml tubes of liquid CONDA Pronadisa Brettanomyces
media were inoculated with each yeast strain, and incubated for six days in a rolling tube rack at
27 °C. The liquid cultures were checked for purity by streaking on CONDA Pronadisa
Brettanomyces agar. After pure cultures were visually confirmed, 1 ml aliquots of each culture
were pelleted by centrifugation. Supernatants were removed, and the pellets stored at -80 °C for
future isolation of genomic DNA.
Yeast cultures were prepared for fermentation from pure culture plates by inoculating 10
mL CONDA Pronadisa Brettanomyces liquid cultures in duplicate, which were then incubated at
27 °C in a rolling tube rack. After 5 days of growth, Erlenmeyer flasks of CONDA Pronadisa
Brettanomyces liquid media were inoculated with each yeast strain in duplicate, at 5%
(volume/volume), and incubated on an orbital shaker at 150 rpm at 27 °C for three days.
2.2

Genome sequencing and analysis

2.2.1

Genomic DNA isolation
Genomic DNA (gDNA) was isolated from S. cerevisiae and B. claussenii cell pellets using

a PureLink™ Genomic DNA Mini Kit (Invitrogen, Carlsbad, CA, USA). Prior to the start of the
protocol, 1 ml of a 5 mg/ml lyticase stock in Tris-EDTA (TE) buffer was used to resuspend each
yeast pellet for subsequent incubation for 1 hour at 37 °C. The manufacturer’s Gram Negative
Bacterial Cell Lysate protocol was then followed. Eluted DNA was quantified using a Qubit 2.0
Fluorometer (Invitrogen, Carlsbad, CA, USA).
2.2.2

Sequencing library preparation
gDNA libraries for S. cerevisiae and B. claussenii were prepared following the Ion Xpress

Plus gDNA Fragment Library Preparation User Guide (Life Technologies, Carlsbad, CA, USA).
Briefly, gDNA was fragmented via physical fragmentation using a Covaris M220 Focused6

ultrasonicator (Covaris, Woburn, MA, USA) following the 100 ng and 400-base-read protocols.
The libraries were then end-repaired following the 100 ng protocol using the Ion Xpress Plus
Fragment Library Kit (Life Technologies, Carlsbad, CA, USA), followed by purification using an
Agencourt AMPure XP Kit (Beckman Coulter, Brea, CA, USA). Adapter ligation and nick repair
were carried out using the Ion Xpress Plus Fragment Library Kit (Life Technologies, Carlsbad,
CA, USA) and the Ion Xpress Barcode Adapters Kit (Life Technologies, Carlsbad, CA, USA)
following the 50-100 ng standard procedure for barcoded libraries. Adapter-ligated and nickrepaired libraries were purified using an Agencourt AMPure XP Kit, following the 400-600-baseread protocol. The prepared libraries were size selected using a BluePippin System (Sage Science
Inc, Beverly, MA, USA) and Pippin Prep Kit CDF 2010 (Sage Science Inc, Beverly, MA, USA)
following the 400-base-read library size protocol. Size-selected libraries were then purified using
the Agencourt AMPure XP Kit following the 500-base-read or smaller protocol.
2.2.3

Ion Torrent 5S XL sequencing
Template preparation and loading onto an Ion 530 sequencing chip was done with an Ion

Chef (Life Technologies, Carlsbad, CA, USA) using an Ion 510 & Ion 520 & Ion 530 Kit-Chef
(Life Technologies, Carlsbad, CA, USA). Sequencing was performed on an Ion S5 System (Life
Technologies, Carlsbad, CA, USA); demultiplexing and read trimming was performed in Torrent
Suite 5.10.1.
2.2.4

Genome assembly and analysis
Draft de novo culture metagenomes and genome assemblies for Saccharomyces,

Brettanomyces, and bacterial contaminants were carried out using SPAdes 3.13.1 (Bankevich et al.
2012) in careful mode using kmers 21, 33 and 55. Quast 5.0.2 (Gurevich et al. 2013) and Mauve
(Darling et al. 2004) were used to calculate assembly statistics and evaluate alignment to reference
genomes, respectively. Metagenome binning was carried out using MetaBAT 2 (Kang et al. 2019)
after mapping sequencing reads to draft assemblies using the Burrows-Wheeler Aligner (Li and
Durbin 2009) and indexing sequencing reads with SAMtools (Li et al. 2009). Individual read bins,
two each for Saccharomyces and Brettanomyces, and two for the bacterial contaminant(s) were
reassembled using SPAdes as described above. Annotation of bacterial genomes was carried out
using the RAST server (Aziz et al. 2008), and average nucleotide identities were calculated on the

7

ANI Calculator (http://enve-omics.ce.gatech.edu/ani/). BLASTn was used to compare sequences
to databases to determine the statistical significance of resulting matches (Altschul et al. 1990).
2.3

Wort preparation
Wort was prepared based on a recipe supplied by Parallel 49 Brewing Company, modified

for small batch production. The recipe was tripled to allow for an S. cerevisiae mono-culture brew,
a B. claussenii mono-culture brew, and a Wild Ride co-culture brew; the recipe in triplicate is as
follows: 24 L of water was heated to a strike temperature of 70 °C and mixed with 7.29 kg of
Superior Pilsner ground malt in a lauter tun to initiate the mashing process. The mixture was
maintained between 61 and 65 °C for 60 min. After 60 min, the mash liquor was recirculated over
the grain bed along with 6 L of 78 °C water, and drained into a brew kettle. The mash liquor was
brought to a boil for 90 min: 3 kg of white table sugar was added to the wort after 60 min, 22.5 g
of hop type one was added after 60 min, and 37.5 g of hop type two was added at the conclusion
of the boil. The wort was left for 10 min, then rapidly cooled to 18 °C in preparation for transfer to
three bioreactor units.
2.4

Fermentations

2.4.1

Bioreactors
Fermentations were carried out in 14-L BioFlo 110 Fermentors (New Brunswick Scientific,

Edison, NJ, USA) that were pre-sterilized by autoclaving with 5 L of water. Once cool, the water
was removed and replaced with 8 L of freshly prepared wort. Prior to inoculation, the wort was
aerated with 0.2 µm filter sterilized air until the dissolved oxygen concentration levelled off at a
maximum, as determined using a calibrated in-vessel dissolved oxygen probe. During
fermentations, dissolved oxygen, pH and temperature were monitored and recorded.
2.4.2

Wort inoculation
The wort was divided among the three bioreactor units. Six to seven hundred ml of S.

cerevisiae liquid culture was pelleted for the S. cerevisiae mono-culture brew. Post-centrifugation,
the yeast was resuspended in some wort and used to inoculate the brew. This same procedure was
followed for the B. claussenii mono-culture. For the Wild Ride co-culture, 600-700 ml of S.
cerevisiae and 600-700 ml of B. claussenii were pelleted and added to the Wild Ride brew. The

8

brews were maintained at 21 °C within the bioreactors throughout the course of the twenty-twoday fermentations.
2.4.3

Sampling conditions
The fermentations were sampled every day for the first week, and every three days

subsequently until the end of three weeks. Prior to each sampling time point, bioreactors were
mixed for 3 min at 50 rpm. Thirty ml of each fermentation solution were aseptically withdrawn
into I-Chem Amber Septa Vials (Thomas Scientific, Swedesboro, NJ, USA). Ten ml from each of
the fermentation samples were immediately stored for further chemical analysis at -20 °C.
For RNA, DNA, and protein analysis, samples were immediately stored using TRIzol
Reagent (Invitrogen, Carlsbad, CA, USA). Two 1 ml and two 0.25 ml samples were dispensed into
1.5-ml microcentrifuge tubes. The tubes were centrifuged for 1 min at 10,000 xg at room
temperature. Supernatants were discarded, and 0.75 ml of TRIzol Reagent was added to each tube
and vortexed to mix. The TRIzol-suspended samples were stored at -80 °C.
2.4.4

Specific gravity
Specific gravity was measured during each sampling time throughout the fermentations

using a HansTronik Hand Held Refractometer (HansTronik, Heidelberg, Germany). Remaining
supernatant from the centrifuged 1.5-mL microcentrifuge tubes described in the section above was
used for each measurement.
2.5

TRIzol extractions
RNA, DNA, and protein were extracted for analysis using TRIzol Reagent (Invitrogen,

Carlsbad, CA, USA). The manufacturer’s protocol was followed with a few minor changes: RNA
was eluted in 20 µl of low TE buffer prior to being stored at -80 °C; as well, during protein isolation,
the protocol was stopped prior to 1% sodium dodecyl sulfate (SDS) resuspension to ensure dry
protein pellets were produced, which were subsequently frozen at -20 °C.
2.6

mRNA isolation and sequencing
mRNA enrichment was performed from extracted RNA using a polyA Spin mRNA

Isolation Kit (New England BioLabs, Ipswich, MA, USA), and quantified using an RNA 6000 Pico
Kit on a Bioanalyzer 2100 Instrument (Agilent, Santa Clara, CA, USA). The isolated mRNA was
9

treated with DNase using the TURBO DNA-free Kit (Invitrogen, Carlsbad, CA, USA). RNA-seq
libraries were generated using an Ion Total RNA-Seq v2 Kit (Life Technologies, Carlsbad, CA,
USA) according to the manufacturer’s instructions. Template preparation and loading onto an Ion
530 sequencing chip was done with an Ion Chef (Life Technologies, Carlsbad, CA, USA) using an
Ion 510 & Ion 520 & Ion 530 Kit-Chef (Life Technologies, Carlsbad, CA, USA). Sequencing was
performed on an Ion S5 System (Life Technologies, Carlsbad, CA, USA); demultiplexing and read
trimming was performed in Torrent Suite 5.10.1.
2.6.1

Transcriptome analysis
Transcriptome assembly was performed in Geneious Prime 2019.1.3 set to medium-low

sensitivity with 5 fine tuning iterations. Reads with multiple matches were mapped randomly and
gaps were allowed at 10% maximum per read (other parameters: 25 base minimum overlap, 18
base word length, 20% maximum mismatch per read, 15 base maximum gap size, 80% minimum
overlap identity, 13 base index word length, 4 base maximum ambiguity). Expression levels were
calculated in Geneious Prime 2019.1.3 by comparing transcripts normalized by the median of gene
expression ratios as described in (Dillies et al. 2013). Transcripts per million (TPM) were
calculated according to Wagner et al. (Wagner et al. 2012) as: TPM = (CDS read count * mean
read length * 106) / (CDS length * total transcript count).
2.7

Protein extraction and analysis
Protein pellets were shipped on dry ice to the University of Regina so that the proteomes of

the mono- and co-cultures could be analyzed by Dr. Tzu-Chaio Chao at the Institute of
Environmental Change and Society. Samples are being analyzed on a Waters Synapt G2 HDMS
(Q-TOF MS) coupled to a nanoAcquity UPLC. Dr. T-C Chao is optimizing analytical conditions
for the samples provided, and generating databases for data analysis.
2.8

Gas chromatography

2.8.1

Sample preparation
At each time point, bioreactors were mixed for 3 min at 50 rpm, and triplicate (technical)

10 ml samples were aseptically withdrawn, and transferred to 20-ml amber headspace vials
(Canadian Life Science, Peterborough, ON) containing 20 mg NaCl and 100 µl 12 N HCl (Sigma
Aldrich, ACS grade, Oakville, ON). Vials were sealed with 18-mm magnetic caps with butyl
10

rubber/PTFE lined septa, vortexed to mix, and immediately analyzed using a Varian CP-3800 gas
chromatograph equipped with a flame ionization detector and a CombiPAL (CTC Analytics,
Zwingen, Switzerland) autosampler equipped with heated headspace sample handling capabilities
2.8.2

Analytical standards
Ethanol standards were prepared daily using anhydrous ethyl alcohol (Commercial

Alcohols, Brampton, ON) added to 18 MW water at 0.1, 0.5, 1.0, 2.0, 4.0, 6.0, 8.0 and 10.0 % (v/v)
in headspace sample vials with 20 mg NaCl and 100 µl 12 N HCl at a total volume of 10 ml.
Two custom mixed analytical standards were supplied by SPEXCerti Prep (Metuchen, NJ,
USA). One was prepared in ethanol and contained 1000 µg/ml each of trans-2-nonenal, methyl
sufide, isobutyl acetate, isoamyl acetate, ethyl octanoate, ethyl hexanoate, ethyl butanoate, ethyl
acetate, beta-myrcene, acetic acid, acetaldehyde, 3-methyl-1-butanol, 2-phenethyl acetate, 2methyl-1-butanol, 2,3-butanedione and (R)-(+)-limonene. The second was prepared in distilled
water and contained tert-butyl alcohol, methanol, ethylene glycol, ethanol, allyl alcohol, 2propanol, 2-methyl-1-propanol and 1-butanol. Individual pure compounds were from SigmaAldrich (Oakville, ON).
2.8.3

Headspace chromatography and detector conditions
An Agilent (Santa Clara, California) DB-624 UI chromatography column (30 m, 0.320 mm

diameter, 1.80 µm film thickness, widebore) was used with an 1177 injector (gas set to 10 psi; split
ratio 2) equipped with an Agilent Ultra Inert Liner (Split/Splitless, Gooseneck, 2 mm) set to 220
°C, and a flame ionization detector (FID) set to 250 °C. The FID was equipped with an electronic
flow controller set to 25 ml/min He as makeup gas, with 30 ml/min H2 and 300 ml/min air. All
gasses were ultra-high purity from Praxair (Kamloops, BC) The column oven program was: 40 °C
for 4 min, ramped to 200 °C at 7.5 °C/min, held at 200 °C for 6 min, followed by 5 min at 220 °C.
For mass spectrometry of headspace samples, an Agilent 7890B gas chromatograph
coupled to an Agilent 5977A Mass Selective Detector (MSD) was used. Samples were introduced
using a CombiPAL RSI 85 autosampler in heated headspace mode. An Agilent (Santa Clara,
California) DB-624 UI chromatography column (30 m, 0.250 mm diameter, 1.40 µm film
thickness, widebore) was used, and the column oven program was: 40 °C for 4 min, ramped to 200
°C at 7.5 °C/min, held at 200 °C for 6 min, followed by 5 min at 220 °C. The gas flow rate was 1
11

ml/min (7.05 psi) with an average velocity of 36.298 cm/sec. The injector was used in split mode
at a ratio of 2:1 at 2 ml/min with an Agilent 5190-2295 inlet liner. The injector was kept at 220 °C
at a 7.05 psi, with a total flow of helium of 6 ml/min (septum purge 3 ml/min). The transfer line to
the single quadrupole MSD was maintained at 250 °C, and the MSD was kept in scan mode with a
fixed electron energy of 70.0 eV. The electrospray ionization source was maintained at 230 °C,
and the quadrupole temperature was 150 °C. The start mass for scanning was 22 and the end mass
was 400 with a threshold of 150, and the scan speed was 1562.
2.8.4

Headspace sampling
For both GC-FID and GC-MS, samples, analytical standards and blanks were heated on a

CombiPAL autosampler at 70 °C with alternating 5 sec agitation at 350 rpm followed by 5 sec rests
for a total of 15 min. A 1-ml headspace syringe (Canadian Life Science, Peterborough, ON) was
maintained at 70 °C, flushed twice with sample before each injection, filled at 100 µl/sec, and the
1000 µl sample was injected at 600 µl/sec with pre- and post-injection delays of 0.2 sec. The
syringe was flushed for 240 sec post-injection. On each analysis day, instrument, vial and water
blanks were analyzed prior to samples and standards; vial and water blanks were repeated after
standards, and after every third sample. Standards were analyzed each analysis day for
quantification.

3

RESULTS

3.1

Bioreactor parameters
Oxygen consumption data was collected over the first 28 hours of fermentation (Figure

3.1). A notable lag in oxygen consumption was observed in the Brettanomyces only brew. The pH
of the Brettanomyces mono-culture took longer to drop off, and interestingly, the Saccharomyces
mono-culture levelled off at a higher pH than the Brettanomyces or Wild Ride brews (Figure 3.2).
An appreciable lag in the decline of specific gravity was observed in the Brettanomyces monoculture (Figure 3.3).

12

100
90
S_cerevisiae
B_claussenii
Wild_Ride

Oxygen Saturation (%)

80
70
60
50
40
30
20
10
0
0

4

8

12
16
Time (h)

20

24

28

Figure 3.1 Oxygen consumption in the S. cerevisiae mono-culture, B. claussenii mono-culture,
and the Wild Ride co-culture brews over the course of the fermentations.

13

6.0

S_cerevisiae
B_claussenii
Wild_Ride

pH

5.0

4.0

3.0
0

24

48

72

96

120

144

168

192

Time (h)
Figure 3.2 Variation in pH in the S. cerevisiae mono-culture, B. claussenii mono-culture, and
the Wild Ride co-culture brews over the course of the fermentations.

14

1.075
1.070
S. cerevisiae
B. claussenii
Wild Ride

1.065

Specific Gravity

1.060
1.055
1.050
1.045
1.040
1.035
1.030
0

2

4

6

8

10
Time (d)

12

14

16

18

20

Figure 3.3 Change in specific gravity in the S. cerevisiae mono-culture, B. claussenii monoculture, and the Wild Ride co-culture brews over the course of the fermentations.

3.2

Genome sequencing, assembly and analysis
In order to have reference genome data for future transcriptome and proteome analyses,

cultures of Saccharomyces and Brettanomyces obtained from Parallel 49 were subject to whole
genome sequencing. 6,281,650 Q20 reads and >= 1,887,285,253 Q20 bases were produced from
the genomic sequencing run (S5) for Saccharomyces, and 5,804,220 Q20 reads and >=
1,655,746,215 Q20 bases were produced for Brettanomyces. The mean read lengths for
Saccharomyces and Brettanomyces were 342 bp and 332 bp, respectfully. Following de novo
assembly of the yeast genomes using SPAdes, Quast analysis revealed GC maxima at 38% and
75%, indicating the presence of contaminating DNA despite no obvious contamination on yeast
15

culture plates. Saccharomyces and Brettanomyces are expected to have GC contents of 38.2% and
40%, respectively. Given this, the genome sequencing data from both yeast cultures were treated
as metagenomes in an effort to determine the identity of the contaminant(s).
The two yeast culture metagenome assemblies showing contamination were trimmed to
include only contigs greater than 1000 bases, and the original sequencing reads were then mapped
to those assemblies using a Burrows-Wheeler Aligner prior to metagenome binning in MetaBAT
2. Two read bins were generated from each of the Saccharomyces and Brettanomyces culture
metagenomes, each of which was reassembled using SPAdes for a total of four assemblies. Quast
analysis revealed that bin assemblies from both yeast culture metagenomes had single GC maxima
of 38% and 75%. Casual BLAST analysis against the NCBI nucleotide collection (nr/nt) using
BLASTn returned hits to Saccharomyces and Cellulosimicrobium sp. TH-20 from the
Saccharomyces culture metagenome; and yeast genera such as Pichia, Kuraishia and Ogataea, as
well as Cellulosimicrobium sp. TH-20, from the Brettanomyces culture metagenome.
The Saccharomyces bin contained 130,446 reads that assembled into 2,277 contigs over
1,000 bases, for a total length of 7 Mb at an average coverage depth of 6.9. Mauve alignments to
the S. cerevisiae S288C reference genome indicated high homology between the two cultures,
although with only 7 Mb of the Saccharomyces genome sequenced, additional sequencing efforts
using a fresh pure culture is required (Figure 3.4).
The Brettanomyces bin contained 2,407,899 reads that assembled into 2,346 contigs over
1,000 bases, for a total length of 7.4 Mb at an average depth of coverage of 58. Manual examination
of

a

Mauve

alignment

to

Brettanomyces

bruxellensis

(https://www.ncbi.nlm.nih.gov/nuccore/1440771537) revealed high levels of homology between
the two genomes, although the reference genome is 15.4 Mb (41.6% GC) (Figure 3.5).
Brettanomyces genomes available on NBCI ranged from 10.3235 Mb to 13.1606 Mb (Table 1),
and there was an absence of well annotated references.

16

Figure 3.4 Section of Saccharomyces (bottom) whole genome alignment to Saccharomyces
cerevisiae S288C reference genome (top) showing regions of high homology.

Figure 3.5 Section of Brettanomyces (top) whole genome alignment to Brettanomyces
bruxellensis reference genome (bottom) showing regions of high homology.

Table 1. Brettanomyces genomes available on NCBI.
Brettanomyces genomes Genome Size (Mb)

GC Content (%)

NCBI ID #

available on NCBI
B. bruxellensis

13.1606

40.0

11901

B. naardenesis

11.2831

44.5

46556

B. anomalus

12.6919

39.9

38163

B. custersianus

10.3235

40.15

9126

17

The contaminant bin from the Saccharomyces metagenome contained 5,177,117 reads that
assembled into 152 contigs over 1,000 bases with a single GC maximum at 75%, and a total size
of 4.2 Mb at an average coverage depth of 209. The contaminant genome was annotated on the
RAST server, and close strain set analysis returned Cellulosimicrobium sp. TH-20 as the reference
genome with the highest overall similarity. Average nucleotide identity calculations indicated
greater than 98.36% similarity between the contaminant and TH-20, and 4,963,017 of the original
sequencing reads mapped to the TH-20 genome when subjected to a Burrows-Wheeler alignment
(Table 2). Similarly, the contaminant bin from the Brettanomyces metagenome contained
1,488,741 reads that assembled into 183 contigs over 1,000 bases with a single GC maximum at
75%, and a total size of 4.16 Mb at an average coverage depth of 73. The average nucleotide
identify calculation for the Brettanomyces contaminant and Cellulosimicrobium cellulans strain
TH-20 is demonstrated in Figure 3.6.
The contaminants detected in the two yeast cultures share 99.98% average nucleotide
identity (Table 2, Figure 3.7). The complete genome of Cellulosimicrobium cellulans strain
NEB113 was released on July 24, 2019 on NBCI, and the ANIs between NEB113 and the two
contaminants sequenced here are >99.9% (Table 2, Figure 3.8, Figure 3.9).
Table 2. Average nucleotide identities (%) between Cellulosimicrobium cellulans contaminants
from yeast cultures and high quality reference genomes available on NCBI
Brettanomyces

Saccharomyces

contaminant

contaminant

NEB113

TH-20

Brettanomyces
contaminant
Saccharomyces

99.98

contaminant
NEB113

99.97

99.99

98.34

98.36

CP041694.1
TH-20

98.35

NZ_CP020857.1

18

Figure 3.6 Average nucleotide identity calculation for Brettanomyces contaminant and
Cellulosimicrobium cellulans strain TH-20.

Figure 3.7 Average nucleotide identity calculation for Brettanomyces contaminant and
Saccharomyces contaminant.

19

Figure 3.8 Average nucleotide identity calculation for Brettanomyces contaminant and
Cellulosimicrobium cellulans strain NEB113.

Figure 3.9 Average nucleotide identity calculation for Saccharomyces contaminant and
Cellulosimicrobium cellulans strain NEB113.
20

3.3

Transcriptomic Results
9,734,188 Q20 reads and >= 1,620,889,603 Q20 bases were generated for the Wild Ride

day 7 transcriptome. The reads were mapped to a Saccharomyces cerevisiae S288C reference
genome (Saccharomyces Genome Database current release reference genome, dated January 31,
2015). Of the total number of Wild Ride reads, 1,717,871 mapped to the Saccharomyces reference
genome (17.65%). 1,321,632 reads mapped to rRNA regions located on chromosome XII (chr XII),
comprising 76.93% of the total mapped reads. 155,153 reads mapped to non-rRNA regions of chr
XII and 241,086 reads mapped to non-rRNA regions located outside of chr XII, comprising 4.07%
of all reads mapped.
3.4

Volatile metabolites
The fermentation progress and odour of the three brews varied. Heated headspace gas

chromatography, coupled to both flame ionization and mass selective detectors, was deployed to
identify volatile yeast metabolites generated. Under the analytical conditions used, 33 compounds,
including ethanol, were resolved from the baseline and identified by mass spectrometry (Table 3).
Of those, 10 were monitored using flame ionization detection (Figure 3.10); two additional
compounds that eluted (4.651 and 6.410 min) before ethanol, and one that eluted immediately after
(9.450 min) were tracked based on peak area, but were not identified. In general, all of the
compounds were produced without appreciable lag in Saccharomyces brews, and after a 3 to 4 day
lag with only Brettanomyces (Figure 3.11 to Figure 3.17). The compounds eluting at 6.410 and
9.450 minutes increased in the Brettanomyces only brew up until day 4 prior to being consumed;
this production was not observed in brews containing Saccharomyces.

21

Table 3. Compounds identified by retention time and spectra via GC-MS in S. cerevisiae, B.
claussenii, and Wild Ride fermentations (* International Union of Pure and Applied Chemistry, **
Chemical Abstracts Service Registry Number, *** http://www.ymdb.ca/, **** Alternate source:
The Good Scents Company).

Name (NIST search) Amount

Compound (IUPAC* Alternate name
name)
(sample)

Flavour (source:
YMDB or
CAS** # YMDB*** ID alternate****)

ethanol

time_course

ethanol

ethyl alcohol

64-17-5

YMDB00883

1-propanol

time_course

propan-1-ol

n -propanol

71-23-8

alcohol, alcoholic,
YMDB01395 fermented, fusel,
musty, peanut, pungent

ethyl acetate

time_course

ethyl acetate

ethyl ethanoate

141-78-6

YMDB00569

anise, balsam, ethereal,
fruity, green,
pineapple, sweet,
weedy

propanoic acid, ethyl
ester

trace

ethyl proponoate

ethyl propionate

105-37-3

YMDB01331

fruity, grape, juicy,
pineapple, rum, sweet

n-propyl acetate

trace

propyl acetate

propyl ethanoate

109-60-4

YMDB01390

bitter, celery, fruity,
fusel, pear, raspberry

2-methyl-1-propanol

time_course

2-methylpropan-1-ol

isobutanol

78-83-1

YMDB00573

bitter, ether, solvent,
wine

2-methyl, ethyl ester
propanoic acid

split_peak

ethyl isobutyrate

ethyl 2methylpropionate

97-62-1

3-methyl, 1-butanol

time_course

3-methylbutan-1-ol

isoamylol

123-51-3

1-butanol, 2-methyl

split_peak

2-methylbutan-1-ol

active amyl alcohol 137-32-6

YMDB00567 malt

butanoic acid, ethyl
ester

small

ethyl butanoate

ethyl butyrate

YMDB01385

(E)pent-2-en-3-yl
hexanoate

trace

[(E)-pent-2-en-3-yl]
hexanoate

thioacetic acid ether

trace

propanoic acid, 2hydroxy, ethyl ester

small

ethyl 2hydroxypropanoate

ethyl lactate

sweet, fruity, creamy,
2676-33-7 YMDB01429 pineapple, caramellic
brown

butanoic acid, 2methyl, ethyl ester

small

ethyl 2methylbutanoate

2-methylbutanoic
acid ethyl ester

fruity, fresh, berry,
7452-79-1 YMDB01353 grape, pineapple,
mango, cherry

butanoic acid, 3methyl, ethyl ester

small

ethyl 3methylbutanoate

ethyl isovalerate

108-64-5

sweet, fruity, spice,
YMDB01334 metallic, green,
pineapple, apple

1 butanol, 3-methyl
acetate

small

3-methylbutyl acetate isoamyl acetate

123-92-2

YMDB00571

105-54-4

Structure

alcohol, ethereal,
medical, strong, sweet

pungent, etherial,
fruity rum, egg nog
fruity, banana,
YMDB00570 alcoholic, burnt, fusel,
malt, oil, whiskey

OH

apple, banana, cognac,
fruity, juicy, pineapple

thioacetic acid ether

banana, bitter, fruity,
solvent, sweet

22

1-butanol, 2-methyl
acetate

trace

2-methylbutyl acetate active amyl acetate 624-41-9

banana, fruit, juicy,
YMDB00568 overripe fruit, peanut,
sweet

pentanoic acid, ethyl
ester

trace

ethyl pentanoate

ethyl valerate

539-82-2

apple, fruity, green,
YMDB01342 pineapple, sweet,
tropical, yeast

hexanoic acid, ethyl
ester

time_course

ethyl hexanoate

ethyl caproate

123-66-0

YMDB01381

apple peel, banana,
fruity, green,
pineapply, sweet,
waxy

butanedioic acid,
diethyl ester

trace

diethyl butanedioate

diethyl succinate

123-25-1

apple, apricot,
chocolate, cooked,
cranberry, fruity,
YMDB01380
grape, mild, musty,
peach, pear, wine,
ylang

ethyl heptanoate

trace

ethyl heptanoate

heptanoic acid,
ethyl ester

106-30-9

cognac, fruity, melon,
YMDB01474 pineapple, plum, rum,
wine

octanoic acid, ethyl
ester

time_course

ethyl octanoate

caprylate

106-32-1

apricot, banana,
brandy, fat, fruity,
YMDB01389
pear, sweet, waxy,
wine

nanoic acid, ethyl
ester

trace

ethyl nonanoate

ethyl caprylate

123-29-5

fruity, natural, rose,
YMDB01354 rum, tropical, waxy,
wine

ethyl cis-4decenoate

7367-84-2

4-decenoic acid, ethyl
trace
ester

O
O

O
O
O
O

apple, brandy, fruity,
YMDB01391 grape, oily, pear,
sweet, waxy

decanoic acid, ethyl
ester

time_course

ethyl decanoate

ethyl caprinate

110-38-3

octanoic acid, 3methyl butyl ester

trace

3-methylbutyl
octanoate

isoamyl octanoate

coconut, fruity, green,
2035-99-6 YMDB01350 oily, pineapple, soapy,
sweet

dodecanoic acid, ethyl
time_course
ester

ethyl dodecanoate

ethyl laurate

106-33-2

D-limonene

trace

Racemic: DL1-Methyl-4-(prop-1-enLimonene;
2-yl)cyclohex-1-ene
Dipentene

138-86-3
(R/S)

citrus

1,3,8-p-menthatriene

trace

1,3,8-paramenthatriene

18368-951

oily, terpy,
camphoreous, cooling,
thymol, herbal, woody,
pine

ethyl 5methylhexanoate

trace

ethyl 5methylhexanoate

ethyl isoamylacetate

10236-109

pentanoic acid, 2hydroxy-4-methyl,
methyl ester

trace

methyl 2-hydroxy-4methylpentanoate

methyl 2-hydroxy-4- 40348-72methylvalerate
9

ethyl 9-decenoate

trace

2-ethyldec-9-enoic
acid

ethyl dec-9-enoate

67233-91YMDB01414 fruity, fatty
4

ethyl undecanoate

small

ethyl undecanoate

Undecanoic acid,
ethyl ester

627-90-7

YMDB01332

clean, floral, leaf,
soapy, sweet, waxy

sweet, fruity, musty

YMDB16012

coconut, cognac, fatty,
soapy, waxy

23

30000

*

* statistically significant

25000

*

Wild Ride

*

20000

Peak Area

Brettanomyces

Saccharomyces

*

15000
10000

*
*

**

5000

*

*

*

ld

ec

an
o

at
e

oa
te
et
hy

lo

an

et
hy

ex
lh

et
hy

ct
an

oa
te

ol
ta
n

l
an
o

yl
-1
-b
u

op

m
et
h

3-

2-

m
et
h

yl
-1
-p
r

ce
ta
te

l

la

no

et
hy

1-

pr
op
a

45
4
9.

41
6.

4.

65
1

0

Retention Time or Compound

Figure 3.10 Peak areas of identified compounds or retention times in the Brettanomyces, Wild
Ride co-culture, and Saccharomyces brews as sampled on day 13 of the fermentations as measured
by GC-FID. n = 3, error bars represent standard deviation, p = 0.05.

24

7000
6000

Peak Area

5000
4000
3000
2000
1000
0
0

2

4

Bc_11.806

6
8
Time (d)
WR_11.806

10

12

14

Sc_11.806

Figure 3.11 Change in peak area of 1-propanol in the B. claussenii (Bc), Wild Ride (WR), and
S. cerevisiae (Sc) brews.

30000
25000

Peak Area

20000
15000
10000
5000
0
0

2
Bc_13.035

4

6
8
Time (d)
WR_13.035

10

12

14

Sc_13.035

Figure 3.12 Change in peak area of ethyl acetate in the B. claussenii (Bc), Wild Ride (WR), and
S. cerevisiae (Sc) brews.

25

4500
4000
3500
Peak Area

3000
2500
2000
1500
1000
500
0
0

2

4

Bc_14.324

6
8
Time (d)
WR_14.324

10

12

14

Sc_14.324

Figure 3.13 Change in peak area of 2-methyl-1-propanol in the B. claussenii (Bc), Wild Ride
(WR), and S. cerevisiae (Sc) brews.

40000
35000

Peak Area

30000
25000
20000
15000
10000
5000
0
0

2
Bc_18.204

4

6
8
Time (d)
WR_18.204

10

12

14

Sc_18.204

Figure 3.14 Change in peak area of 3-methyl-1-butanol in the B. claussenii (Bc), Wild Ride
(WR), and S. cerevisiae (Sc) brews.

26

9000
8000
7000
Peak Area

6000
5000
4000
3000
2000
1000
0
0

2

4

Bc_22.235

6
8
Time (d)
WR_22.235

10

12

14

Sc_22.235

Figure 3.15 Change in peak area of ethyl hexanoate in the B. claussenii (Bc), Wild Ride (WR),
and S. cerevisiae (Sc) brews.

6000
5000

Peak Area

4000
3000
2000
1000
0
0

2
Bc_25.868

4

6
8
Time (d)
WR_25.868

10

12

14

Sc_25.868

Figure 3.16 Change in peak area of ethyl octanoate in the B. claussenii (Bc), Wild Ride (WR),
and S. cerevisiae (Sc) brews.

27

10000

Peak Area

9000
8000
7000
6000
5000
4000
3000
2000
1000
0
0

2
Bc_32.111

4

6
8
Time (d)
WR_32.111

10

12

14

Sc_32.111

Figure 3.17 Change in peak area of ethyl decanoate in the B. claussenii (Bc), Wild Ride (WR),
and S. cerevisiae (Sc) brews.

4

DISCUSSION

4.1

Genome sequencing and Cellulosimicrobium contamination
It was determined that Cellulosimicrobium cellulans strain NEB113 was the contaminating

bacterium in the S. cerevisiae and B. claussenii cultures, dominating the genome sequencing
efforts. C. cellulans is Gram positive, and colonies are generally a yellow-white colour and are
circular and convex (Schumann and Stackebrandt 2015). Cells generally start out as rods and, after
medium components has been consumed, can become more circular (Schumann and Stackebrandt
2015). The organism has a GC content of 72.9-76.5% (Schumann and Stackebrandt 2015), has
been shown to grow on peptone-yeast extract-glucose agar, nutrient agar, and tryptic soy agar at
25-28 °C, and is ampicillin resistant (Schumann and Stackebrandt 2015). C. cellulans rarely causes
human illness, generally only occurring when patients are immunocompromised (Schumann and
Stackebrandt 2015).
Interestingly, no obvious growth of C. cellulans was observed on the plated or liquid
cultures of S. cerevisiae or B. claussenii used for gDNA extraction. C. cellulans is a source for
lyticase and zymolyase, and is often used for commercial production of these yeast cell wall lytic
28

enzymes (Ferrer 2006; Miyajima et al. 2009). There is evidence that trace amounts of C. cellulans
DNA can be found in these commercial lytic enzymes (Ferrer 2006; Miyajima et al. 2009), and its
DNA can be found in some lambic beers, though this may be due to contamination that originates
from the yeast DNA extraction step during testing of these brews (De Roos et al. 2018). Despite
this evidence, the large number of reads belonging to C. cellulans from the genomic sequencing
run suggests that the contamination is a result of something more impactful than trace
contamination from lytic enzymes.
Though it is plausible that C. cellulans contaminated the two yeast samples during growth
in either plated or liquid cultures prior to fermentation, it is thought to be unlikely. It is currently
believed that C. cellulans contaminated the yeast slants while at Parallel 49 Brewing, as the slants
sent from the brewery had been previously used. This begs the question of whether other yeast
slants being used for inoculation at Parallel 49 are contaminated with C. cellulans. It likely is not
impacting the beers produced at Parallel 49, but it is still an area of interest. Parallel 49 has since
sent new yeast slants direct from commercial yeast suppliers so that the project may proceed
without potentially contaminated samples.
In the future, pure yeast cultures will be grown from the newly received samples so that
uncontaminated gDNA may be isolated and sequenced to improve sequencing depth and coverage.
This is also of importance as there are currently no Brettanomyces reference genomes on NCBI
that have been sufficiently annotated. In addition to S5 genome sequencing, performing long read
Oxford Nanopore MinION Sequencing (Oxford Nanopore Technologies, Oxford, UK) is of interest
in order to facilitate genome assemblies.
4.2

Bioreactor parameters measured
Parameters measured automatically by the bioreactors and manually on sampling days show

a pattern of Brettanomyces being a slow growing organism in mono-culture. Interestingly, when
brewed in co-culture with Saccharomyces in the Wild Ride brew, oxygen consumption, sugar
usage, and pH drop all occur quicker and, in a fashion, similar to the Saccharomyces mono-culture.
It should be noted that this observation could be due to Saccharomyces dominance in the brew,
wherein the work being done by Saccharomyces overshadows the contributions of Brettanomyces.
Another possible situation is that Saccharomyces is encouraging Brettanomyces to consume

29

oxygen and sugars at an increased rate, or is producing products that Brettanomyces can utilize in
the co-culture.
4.3

Transcriptomics and rRNA contamination
One area that needs more work is the issue of rRNA contamination in the transcriptome

preparations. It has been suggested that over 95% of all RNA in cells can be rRNA, and when not
adequately removed during transcriptome library preparation, rRNA can dominate sequencing
efforts and result in a great reduction in transcriptome coverage (Stewart et al. 2010; Kukurba and
Montgomery 2015). As 76.93% of all transcriptomic reads were mapped to rRNA regions of the
S288C reference genome, rRNA depletion during mRNA enrichment was poor. In the future, a
different mRNA isolation kit should be tested, such as the Ribominus Transcriptome Isolation Kit
(Yeast and Bacteria) (Invitrogen, Carlsbad, CA, USA). To ensure all rRNA is removed in addition
to the depletion steps during mRNA enrichment, utilizing rRNA removing bioinformatics software
would be of use. SortMeRNA (Kopylova et al. 2012), riboPicker (Schmieder et al. 2012) and
rRNAFilter (Wang et al. 2017) are all bioinformatics tools that could be of use in removing rRNA
reads post-sequencing.
Aside from rRNA read overrepresentation, the lack of well-annotated reference genomes
available makes transcriptome mapping more challenging and less accurate, particularly for
Brettanomyces. Once the S. cerevisiae and B. claussenii genomes are re-sequenced at a greater
depth of coverage, through both Ion Torrent S5 and Oxford Nanopore MinION sequencing, as
discussed in section 4.1, reference genomes will be available for improved mapping.
In order to fully assess the transcriptomic profiles of each brew over the course of the
fermentations, Saccharomyces and Brettanomyces RNA-seq libraries will be prepared and
sequenced. Transcriptome libraries will be prepared using RNA from multiple sample days, with
specific importance being placed on the transcriptomic profiles of the mono- and co-culture brews
during the first half of the fermentations, as that is when the majority of changes in expression
should occur. Multiple transcriptomic sequencing runs will likely be needed in order to obtain full
coverage of each transcriptomic profile.
4.4

Proteomics
Work is currently underway with the help of Dr. T-C Chao at the Institute of Environmental

Change and Society at the University of Regina. Preliminary results suggest that the sampling
30

protocol may need adjustment to allow for greater starting protein amounts, as current samples are
close to the detection limits of the bicinchoninic acid protein assay. Once analytical conditions are
fully optimized and samples are deemed sufficient for analysis, the protein samples will be
analyzed on a Waters Synapt G2 HDMS (Q-TOF MS) coupled to a nanoAcquity UPLC.
4.5

Metabolomics

4.5.1

Identified compounds
Currently, 33 compounds have been identified by GC-MS in the Saccharomyces mono-

culture, the Brettanomyces mono-culture, and the Wild Ride co-culture. The quantities of these
compounds in each brew, however, have yet to be determined. Once the quantity of each compound
in each brew have been identified, the major flavour compounds can be determined. In the future,
liquid injections may be performed using a suitable column for GC-MS; this will potentially allow
for the detection of more flavour compounds, as the current volatile metabolite profiles are similar
for all of the brews.
4.5.2

4-EP and 4-EG
The compounds described in Table 3 are of great interest; however, there are two

compounds that have yet to be studied in these brews which are imperative to the study. 4-ethyl
phenol (4-EP) and 4-ethyl guaiacol (4-EG) are the two compounds that are said to lend the
Brettanomyces character to beer (Joseph et al. 2017; Serra Colomer et al. 2019). These volatile
phenols are produced during fermentation by Brettanomyces from a variety of substrates.
Generally, 4-EP and 4-EG produce flavours such as barnyard, horse sweat, BandAid, rancid,
burning tires, smoky, leathery, and goat-like (Steensels et al. 2015; Joseph et al. 2017; Serra
Colomer et al. 2019). Generally, 4-EG is found in higher concentrations in beer (Steensels et al.
2015). These two important compounds will be identified and quantified in the mono- and coculture brews; their presence and abundance will lend to the greater question of how these brews
differ at the metabolic level.

31

4.6

Future Work

4.6.1

qPCR
Currently, it is unknown whether S. cerevisiae and B. claussenii are contributing equally in

the Wild Ride brew. It is plausible that one species may die off due to harmful products made by
the other, or that one yeast species may thrive off of the products of the other. This could directly
relate to key flavour compounds produced during the brew, as well. Studying these competitive
interactions is necessary to understand the full story of Wild Ride. In order to examine this, qPCR
will be carried out in the near future utilizing the TRIzol-extracted DNA as described in section
2.5. The abundance of each yeast species throughout the course of the fermentation will be
uncovered, and the ecological relationship of these two yeasts will be further elucidated.
4.6.2

Repeat fermentations and different brew types
It will be of great benefit to repeat the fermentation experiment again with the mono-

cultures of S. cerevisiae and B. claussenii, and the Wild Ride co-culture. Much has been learned
since carrying out the first fermentation in terms of sampling protocols and analytical methods.
Taking more samples of greater volume would greatly aid in proteomic analysis. Repeating the
brewing process would also allow for the elimination of C. cellulans contamination; by utilizing
the newly received, unopened yeast slants direct from the yeast provider, the fermentations can be
carried out with the confidence that C. cellulans will not influence the fermentative process.
Performing these fermentations in duplicate should suffice to confirm what has already been
determined from the project, permit changes in protocol to be made, and allow for new results to
be produced.
In the greater scope of the project, it could be valuable to look at other beers produced at
Parallel 49 that are carried out in co-culture. The ecological competitions that exist, the dominating
flavour compounds produced, and the transcriptomic and proteomic profiles of these other brews
will elucidate more about fermentative co-cultures as a whole. Viewing these similarities and
differences will tell an interesting story, and talks are currently underway with the Parallel 49 team
regarding possible directions for this project.

32

5

REFERENCES

Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. 1990. Basic local alignment search tool.
J Mol Biol. 215(3): 403–410. doi:10.1016/S0022-2836(05)80360-2.
Aziz RK, Bartels D, Best AA, DeJongh M, Disz T, Edwards RA, Formsma K, Gerdes S, Glass
EM, Kubal M, et al. 2008. The RAST Server: Rapid Annotations using Subsystems
Technology. BMC Genomics. 9:75. doi:10.1186/1471-2164-9-75.
Bankevich A, Nurk S, Antipov D, Gurevich AA, Dvorkin M, Kulikov AS, Lesin VM, Nikolenko
SI, Pham S, Prjibelski AD, et al. 2012. SPAdes: A New Genome Assembly Algorithm and
its Applications to Single-Cell Sequencing. J Comput Biol. 19(5): 455–477.
doi:10.1089/cmb.2012.0021.
Barnett JA. 1992. The taxonomy of the genus Saccharomyces Meyen ex Reess: A short review for
non-taxonomists. Yeast. 8(1): 1–23. doi:10.1002/yea.320080102.
Boulton C and Quain D. 2006. Brewing Yeast and Fermentation. In 1 edition. Wiley-Blackwell,
Oxford; Malden, MA; Ames, Iowa.
Carrau F, Gaggero C, Aguilar PS. 2015. Yeast diversity and native vigor for flavor phenotypes.
Trends Biotechnol. 33(3): 148-154. doi:10.1016/j.tibtech.2014.12.009.
Curtin CD and Pretorius IS. 2014. Genomic insights into the evolution of industrial yeast species
Brettanomyces bruxellensis. FEMS Yeast Res. 14(7): 997–1005. doi:10.1111/15671364.12198.
Darling ACE, Mau B, Blattner FR, Perna NT. 2004. Mauve: Multiple Alignment of Conserved
Genomic Sequence With Rearrangements. Genome Res. 14(7): 1394–1403.
doi:10.1101/gr.2289704.
De Roos J, Vandamme P, De Vuyst L. 2018. Wort Substrate Consumption and Metabolite
Production During Lambic Beer Fermentation and Maturation Explain the Successive
Growth of Specific Bacterial and Yeast Species. Front Microbiol. 9.
doi:10.3389/fmicb.2018.02763.
Debowski J. 2014. Liquid bread: Beer and brewing in cross-cultural perspective. Anthropol. Forum
24(3): 308–310. doi:10/gfxjx9.
Dequin S. 2001. The potential of genetic engineering for improving brewing, wine-making and
baking yeasts. Appl Microbiol Biotechnol. 56(5–6): 577–588.
Dillies MA, Rau A, Aubert J, Hennequet-Antier C, Jeanmougin M, Servant N, Keime C, Marot G,
Castel D, Estelle J, et al. 2013. A comprehensive evaluation of normalization methods for
Illumina high-throughput RNA sequencing data analysis. Brief Bioinformatics. 14(6): 671–
683. doi:10.1093/bib/bbs046.
33

Ferrer P. 2006. Revisiting the Cellulosimicrobium cellulans yeast-lytic β-1,3-glucanases toolbox:
A review. Microb Cell Fact. 5:10. doi:10.1186/1475-2859-5-10.
Figueiredo BIC, Saraiva MAF, Pimenta PP de S, Testasicca MC de S, Sampaio GMS, Cunha AC
da, Afonso LCC, Queiroz MV de, Castro I de M, Brandão RL. 2017. New Lager Brewery
Strains Obtained by Crossing Techniques Using Cachaça (Brazilian Spirit) Yeasts. Appl
Environ Microbiol. 83(20): e01582-17. doi:10.1128/AEM.01582-17.
Gibney PA, Lu C, Caudy AA, Hess DC, Botstein D. 2013. Yeast metabolic and signaling genes
are required for heat-shock survival and have little overlap with the heat-induced genes.
Proc Natl Acad Sci USA. 110(46): E4393–E4402. doi:10.1073/pnas.1318100110.
Gibson B, Geertman JMA, Hittinger CT, Krogerus K, Libkind D, Louis EJ, Magalhães F, Sampaio
JP. 2017. New yeasts—new brews: modern approaches to brewing yeast design and
development. FEMS Yeast Res. 17(4). doi:10.1093/femsyr/fox038.
Goffeau A, Barrell BG, Bussey H, Davis RW, Dujon B, Feldmann H, Galibert F, Hoheisel JD, Jacq
C, Johnston M, et al. 1996. Life with 6000 Genes. Science. 274(5287): 546–567.
doi:10.1126/science.274.5287.546.
Gurevich A, Saveliev V, Vyahhi N, Tesler G. 2013. QUAST: quality assessment tool for genome
assemblies. Bioinformatics. 29(8): 1072–1075. doi:10.1093/bioinformatics/btt086.
Iimure T and Sato K. 2013. Beer proteomics analysis for beer quality control and malting barley
breeding.
Food
Research
International.
54(1):
1013–1020.
doi:10.1016/j.foodres.2012.11.028.
Joseph CML, Albino E, Bisson LF. 2017. Creation and Use of a Brettanomyces Aroma Wheel.
Catalyst: Discovery into Practice. 1(1): 12–20. doi:10.5344/catalyst.2016.16003.
Kang DD, Li F, Kirton E, Thomas A, Egan R, An H, Wang Z. 2019. MetaBAT 2: an adaptive
binning algorithm for robust and efficient genome reconstruction from metagenome
assemblies. PeerJ. 7: e7359. doi:10.7717/peerj.7359.
Kobi D, Zugmeyer S, Potier S, Jaquet-Gutfreund L. 2004. Two-dimensional protein map of an
“ale”-brewing yeast strain: proteome dynamics during fermentation. FEMS Yeast Res.
5(3): 213–230. doi:10.1016/j.femsyr.2004.07.004.
Kopylova E, Noé L, Touzet H. 2012. SortMeRNA: fast and accurate filtering of ribosomal RNAs
in
metatranscriptomic
data.
Bioinformatics.
28(24):
3211–3217.
doi:10.1093/bioinformatics/bts611.
Kukurba KR and Montgomery SB. 2015. RNA Sequencing and Analysis. Cold Spring Harb Protoc.
2015(11): 951–969. doi:10.1101/pdb.top084970.

34

Landry CR, Townsend JP, Hartl DL, Cavalieri D. 2006. Ecological and evolutionary genomics of
Saccharomyces cerevisiae. Mol Ecol. 15(3): 575–591. doi:10.1111/j.1365294X.2006.02778.x.
Li H and Durbin R. 2009. Fast and accurate short read alignment with Burrows–Wheeler transform.
Bioinformatics. 25(14): 1754–1760. doi:10.1093/bioinformatics/btp324.
Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abecasis G, Durbin R.
2009. The Sequence Alignment/Map format and SAMtools. Bioinformatics. 25(16): 2078–
2079. doi:10.1093/bioinformatics/btp352.
Liu L, Wang J, Rosenberg D, Zhao H, Lengyel G, Nadel D. 2018. Fermented beverage and food
storage in 13,000 y-old stone mortars at Raqefet Cave, Israel: Investigating Natufian ritual
feasting. J Archaeol Sci: Reports. 21: 783–793. doi:10.1016/j.jasrep.2018.08.008.
Miyajima Y, Satoh K, Umeda Y, Makimura K. 2009. Quantitation of Fungal DNA Contamination
in Commercial Zymolyase and Lyticase Used in the Preparation of Fungi. Nippon Ishinkin
Gakkai Zasshi. 50(4): 259–262. doi:10.3314/jjmm.50.259.
Nookaew I, Papini M, Pornputtapong N, Scalcinati G, Fagerberg L, Uhlen M, Nielsen J. 2012. A
comprehensive comparison of RNA-Seq-based transcriptome analysis from reads to
differential gene expression and cross-comparison with microarrays: a case study in
Saccharomyces
cerevisiae.
Nucleic
Acids
Res.
40(20):
10084–10097.
doi:10.1093/nar/gks804.
Olaniran AO, Hiralal L, Mokoena MP, Pillay B. 2017. Flavour-active volatile compounds in beer:
production, regulation and control. J Inst Brew. 123(1): 13–23. doi:10.1002/jib.389.
Perez-Traves L, Lopes CA, Querol A, Barrio E. 2014. On the Complexity of the Saccharomyces
bayanus Taxon: Hybridization and Potential Hybrid Speciation. PLoS One. 9(4): e93729.
doi:10.1371/journal.pone.0093729.
Pham TK, Chong PK, Gan CS, Wright PC. 2006. Proteomic Analysis of Saccharomyces cerevisiae
under High Gravity Fermentation Conditions. J Proteome Res. 5(12): 3411–3419.
doi:10.1021/pr060377p.
Pham TK and Wright PC. 2008. The proteomic response of Saccharomyces cerevisiae in very high
glucose conditions with amino acid supplementation. J Proteome Res. 7(11): 4766–4774.
doi:10.1021/pr800331s.
Pires EJ, Teixeira JA, Branyik T, Vicente AA. 2014. Yeast: the soul of beer’s aroma-a review of
flavour-active esters and higher alcohols produced by the brewing yeast. Appl Microbiol
Biotechnol. 98(5): 1937–1949. doi:10/f5tx98.
Procopio S, Brunner M, Becker T. 2014. Differential transcribed yeast genes involved in flavour
formation and its associated amino acid metabolism during brewery fermentation. Eur Food
Res Technol. 239(3): 421–439. doi:10/gfxjx4.
35

Procopio S, Qian F, Becker T. 2011. Function and regulation of yeast genes involved in higher
alcohol and ester metabolism during beverage fermentation. Eur Food Res Technol. 233(5):
721–729. doi:10.1007/s00217-011-1567-9.
Sardu A, Treu L, Campanaro S. 2014. Transcriptome structure variability in Saccharomyces
cerevisiae strains determined with a newly developed assembly software. BMC Genomics.
15(1): 1045. doi:10.1186/1471-2164-15-1045.
Schmieder R, Lim YW, Edwards R. 2012. Identification and removal of ribosomal RNA sequences
from
metatranscriptomes.
Bioinformatics.
28(3):
433–435.
doi:10.1093/bioinformatics/btr669.
Schoondermark-Stolk SA, Jansen M, Verkleij AJ, Verrips CT, Euverink G-JW, Dijkhuizen L,
Boonstra J. 2006. Genome-wide transcription survey on flavour production in
Saccharomyces cerevisiae. World J Microbiol Biotechnol. 22(12): 1347–1356.
doi:10.1007/s11274-006-9182-9.
Schumann P and Stackebrandt E. 2015. Cellulosimicrobium. In Bergey’s Manual of Systematics
of
Archaea
and
Bacteria.
American
Cancer
Society.
pp.
1–9.
doi:10.1002/9781118960608.gbm00127.
Serra Colomer M, Funch B, Forster J. 2019. The raise of Brettanomyces yeast species for beer
production. Curr Opin Biotechnol. 56: 30–35. doi:10.1016/j.copbio.2018.07.009.
Shen N, Wang J, Liu C, Li Y, Li Q. 2014. Domesticating brewing yeast for decreasing acetaldehyde
production and improving beer flavor stability. Eur Food Res Technol. 238(3): 347–355.
doi:10/f5tjqk.
Smart KA. 2007. Brewing yeast genomes and genome-wide expression and proteome profiling
during fermentation. Yeast. 24(11): 993–1013. doi:10/d8x8rq.
Steensels J, Daenen L, Malcorps P, Derdelinckx G, Verachtert H, Verstrepen KJ. 2015.
Brettanomyces yeasts — From spoilage organisms to valuable contributors to industrial
fermentations. Int J Food Microbiol. 206: 24–38. doi:10.1016/j.ijfoodmicro.2015.04.005.
Stewart FJ, Ottesen EA, DeLong EF. 2010. Development and quantitative analyses of a universal
rRNA-subtraction protocol for microbial metatranscriptomics. The ISME Journal. 4(7):
896–907. doi:10.1038/ismej.2010.18.
Sulo P, Szabóová D, Bielik P, Poláková S, Šoltys K, Jatzová K, Szemes T. 2017. The evolutionary
history of Saccharomyces species inferred from completed mitochondrial genomes and
revision in the ‘yeast mitochondrial genetic code.’ DNA Res. 24(6): 571–583.
doi:10.1093/dnares/dsx026.

36

Trabalzini L, Paffetti A, Scaloni A, Talamo F, Ferro E, Coratza G, Bovalini L, Lusini P, Martelli
P, Santucci A. 2003. Proteomic response to physiological fermentation stresses in a wildtype wine strain of Saccharomyces cerevisiae. Biochem J. 370(Pt 1): 35–46.
doi:10.1042/BJ20020140.
Treu L, Toniolo C, Nadai C, Sardu A, Giacomini A, Corich V, Campanaro S. 2014. The impact of
genomic variability on gene expression in environmental Saccharomyces cerevisiae strains.
Environ Microbiol. 16(5): 1378–1397. doi:10.1111/1462-2920.12327.
Wagner GP, Kin K, Lynch VJ. 2012. Measurement of mRNA abundance using RNA-seq data:
RPKM measure is inconsistent among samples. Theory Biosci. 131(4): 281–285.
doi:10.1007/s12064-012-0162-3.
Walther A, Hesselbart A, Wendland J. 2014. Genome Sequence of Saccharomyces carlsbergensis,
the World’s First Pure Culture Lager Yeast. G3-Genes Genomes Genet. 4(5): 783–793.
doi:10.1534/g3.113.010090.
Wang Y, Hu H, Li X. 2017. rRNAFilter: A Fast Approach for Ribosomal RNA Read Removal
Without
a
Reference
Database.
J
Comput
Biol.
24(4):
368–375.
doi:10.1089/cmb.2016.0113.
Wendland J. 2014. Lager Yeast Comes of Age. Eukaryot Cell. 13(10): 1256–1265. doi:10/f6k7nb.
Wilkening S, Tekkedil MM, Lin G, Fritsch ES, Wei W, Gagneur J, Lazinski DW, Camilli A,
Steinmetz LM. 2013. Genotyping 1000 yeast strains by next-generation sequencing. BMC
Genomics. 14(1): 90. doi:10.1186/1471-2164-14-90.
Xu W, Wang J, Li Q. 2014a. Microarray studies on lager brewer’s yeasts reveal cell status in the
process of autolysis. FEMS Yeast Res. 14(5): 714–728. doi:10.1111/1567-1364.12156.
Xu W, Wang J, Li Q. 2014b. Comparative proteome and transcriptome analysis of lager brewer’s
yeast in the autolysis process. FEMS Yeast Res. 14(8): 1273–1285. doi:10.1111/15671364.12223.
Zhang H, Richards KD, Wilson S, Lee SA, Sheehan H, Roncoroni M, Gardner RC. 2015. Genetic
characterization of strains of Saccharomyces uvarum from New Zealand wineries. Food
Microbiol. 46: 92–99. doi:10.1016/j.fm.2014.07.016.

37

6

SUPPLEMENTAL INFORMATION

6.1

Modified CONDA Pronadisa Brettanomyces Media

CONDA Pronadisa Brettanomyces Media – Modified:
Dextrose

10 g/L

Peptone

5 g/L

Malt extract

3 g/L

Yeast extract

3 g/L

Yeast nitrogen base

3 g/L

Thiamine stock: dissolve 200 mg of thiamine into 10 ml of water. Filter Sterilize.
Measure all ingredients into a 2-L Erlenmeyer flask, add 1 L of water, and autoclave. Postautoclave, add 1 ml of the thiamine stock to the media and mix with a stir bar.

6.2

Sample commands used for metagenome analysis
Index yeast culture metagenome assembly:

bwa index Brett_SPAdes_denovo_1000plus.fasta
Align sequencing reads to yeast culture metagenome and calculate read coverage:
bwa mem -t 24 Brett_SPAdes_denovo_1000plus.fasta
Brettanomyces_IonXpress_023.fastq | samtools view -b
| samtools sort - -o Brett_mapped_SPAdes1000.bam
jgi_summarize_bam_contig_depths
Brett_mapped_SPAdes1000.bam --outputDepth
depth_Brett_SPAdes1000.txt
Separate assembly contigs into individual genome bins and rename file:
metabat2 -i Brett_SPAdes_denovo_1000plus.fasta -a
depth_Brett_SPAdes1000.txt --minContig 1500 -o 16Jul2019_Brett_metgen_bins/bin -v
mv bin.2.fa 16Jul2019_Brett49_metabatbin.fasta
Index yeast only assembly bin:
bwa index 16Jul2019_Brett49_metabatbin.fasta
Collect original sequencing reads for yeast only genome bin in FASTQ format:
38

bwa mem -t 24 16Jul2019_Brett49_metabatbin.fasta Brettanomyces_IonXpress_023.fastq >
Xpress_to_Brett_bin.sam
samtools view -b -F 4 Xpress_to_Brett_bin.sam > onlyBrettReads.bam
samtools fastq onlyBrettReads.bam >
onlyBrettReads.fastq
Assemble yeast only sequencing reads and collect quality statistics using Quast:
spades.py -s onlyBrettReads.fastq --iontorrent –
careful -k 21,33,55 -o
16Jul2019_SPAdes_Brett_reads_only
quast.py contigs.fasta
Calculate average depth of coverage for assembled contigs using basic Linux commands:
grep '>' contigs.fasta | sed 's/_/,/g' | datamash -t, mean 6
6.3

Mass spectrometry parameters

INSTRUMENT CONTROL PARAMETERS: Agilent GCMS
---------------------------------------------D:\Methods\JonvanHamme\Headspace 2019\Headspace Pal 1 jun 18.M
Tue Jun 18 19:22:10 2019
Control Information
------- ----------Sample Inlet

: GC

Injection Source

: PAL Sampler

Injection Location: Front
Mass Spectrometer

: Enabled

PAL3 Sampler
Serial Number:

Robot
39

Script Name:

HS-STD-V3.0

Installed Syringe
Name:

8010-0265

Maximum Volume (µL):

2500.00

Minimum Volume (µL):

250.00

Script Parameters
Basic
Tool:

HS 1

GC Cycle Time (min):

45

Sample Volume (mL):

1

Agitator:

Agitator 1

Incubation Time (min):

15

Incubation Time Increment (min):
Heat Agitator:

On

Incubation Temperature (°C):
Heat Syringe:

0

70
On

Syringe Temperature (°C):

70

Pre Injection
Flush Time (s):

20

Sample
Sample Vial Penetration Depth (mm):

15
40

Sample Vial Penetration Speed (mm/s):

50

Sample Aspirate Flow Rate (mL/min):

12

Sample Post Aspirate Delay (s):

1

Injection Signal Mode:

Plunger Up

Inlet Penetration Depth (mm):

45

Inlet Penetration Speed (mm/s):

50

Pre Inject Time Delay (s):

0.5

Injection Flow Rate (mL/min):

10

Post Inject Time Delay (s):

0.5

Flush Time (s):

20

Continuous Flush:

Off

Advanced
Agitator Speed (rpm):

350

Agitator On Time (s):

5

Agitator Off Time (s):

5

No Sample Prep method has been assigned to this method.

GC
GC Summary
Run Time
Post Run Time

37.833 min
0 min

41

Oven
Temperature
Setpoint

On

(Initial)

40 °C

Hold Time

4 min

Post Run

45 °C

Program
#1 Rate

7.5 °C/min

#1 Value

200 °C

#1 Hold Time

6 min

#2 Rate

20 °C/min

#2 Value

230 °C

#2 Hold Time

5 min

Equilibration Time

0 min

Max Temperature

325 °C

Maximum Temperature Override
Slow Fan

Disabled

Disabled

Front SS Inlet He
Mode

Split

Heater

On

Pressure

On 7.0481 psi

220 °C

42

Total Flow

On

Septum Purge Flow

6 mL/min
On

Gas Saver

On

Split Ratio

2 :1

Split Flow

2 mL/min

Liner

3 mL/min

20 After 3 min mL/min

Agilent 5190-2295: 870 μL (Universal, low pressure drop, ultra

i)

Thermal Aux 2 (MSD Transfer Line)
Temperature
Setpoint

On

(Initial)

250 °C

Post Run

0 °C

Column
Column #1
Flow
Setpoint

On

(Initial)

1 mL/min

Post Run

Column lock

0.75 mL/min

Unlocked

In

Front SS Inlet He

Out

MSD

(Initial)

40 °C
43

Pressure

7.0481 psi

Flow

1 mL/min

Average Velocity

36.298 cm/sec

Holdup Time

1.3747 min

Column Outlet Pressure

0 psi

Signals
Signal #1: Test Plot
Description

Test Plot

Details
Save
Data Rate

Off
50 Hz

Dual Injection Assignment

Front Sample

Signal #2: Test Plot
Description

Test Plot

Details
Save
Data Rate
Dual Injection Assignment

Off
50 Hz
Back Sample

Signal #3: Test Plot
Description

Test Plot
44

Details
Save

Off

Data Rate

50 Hz

Dual Injection Assignment

Back Sample

Signal #4: Test Plot
Description

Test Plot

Details
Save

Off

Data Rate

50 Hz

Dual Injection Assignment

Back Sample

MS Information
-- ----------General Information
------- ----------Acquisition Mode

: Scan

Solvent Delay (minutes) : 3.5
Tune file

: D:\MassHunter\GCMS\1\5977\ATUNE.U

EM Setting mode Gain

: 1.000000

Normal or Fast Scanning : Normal Scanning
Trace Ion Detection

: On

Run Time (if MS only) : 650 minutes
45

[Scan Parameters]
Start Time

: 3.5

Low Mass

: 22

High Mass

: 400

Threshold

: 150

A/D Samples:

:4

[MSZones]

MS Source

: 230 C maximum 250 C

MS Quad

: 150 C maximum 200 C

Timed Events
----- -----Number Events= 0
END OF MS ACQUISTION PARAMETERS
TUNE PARAMETERS for SN: US1409J201
--------------------------------Trace Ion Detection is ON.
34.593 : EMISSION
70.007 : ENERGY
28.524 : REPELLER
90.331 : IONFOCUS
17.627 : ENTRANCE_LENS
46

1657.755 : EMVOLTS
1879.9 : Actual EMV
1.00 : GAIN FACTOR
1090.000 : AMUGAIN
124.375 : AMUOFFSET
2.000 : FILAMENT
0.000 : DCPOLARITY
14.443 : ENTLENSOFFSET
0.000 : Ion_Body
0.000 : EXTLENS
-731.000 : MASSGAIN
-33.000 : MASSOFFSET
END OF TUNE PARAMETERS
---------------------END OF INSTRUMENT CONTROL PARAMETERS
-----------------------------------

47

GC-FID Ethanol Standards
3.E+06

9.0

detector response

8.8

Detector Response

retention time
Linear (detector response)

8.5

2.E+06

8.3
8.0
7.8

1.E+06
y = 267898x + 61195
R² = 0.9856
(5 analysis days)

7.5

Retention Time (min)

6.4

7.3

0.E+00

7.0
0

2

4

6
8
Ethanol (% v/v)

10

12

Figure 6.1 Detector response and retention time of various ethanol standards run daily to
ensure consistent response during GC-FID.

48