An INVITATION A

lysis of the hybrid genomes of brewing ye brewing of lysis genomes hybrid the of AnAlysis of the To attend the defence of my PhD thesis:

hybrid genomes of Analysis of the hybrid genomes of brewing yeasts

on Wednesday, January 6th, 2015 at 15:00 in the Senaatzaal brewing yeAsts of the Aula at TU Delft, Mekelweg 6, Delft

Prior to the defence (14:30) there will be a presentation of the thesis for non-experts

You are also invited to the reception which follows, starting 17:00, in ‘t Keldertje,

A Department of Biotechnology, sts Julianalaan 67, Delft

Irina Bolat [email protected]

Paranymphs:

Barbara Kozak [email protected]

Daniel Solís Escalante [email protected] Irina Bolat Irina Bolat Irina Bolat

Analysis of the hybrid genomes of brewing yeasts

Proefschrift

ter verkrijging van de graad van doctor aan de Technische Universiteit Delft, op gezag van de Rector Magnificus prof. ir. K.Ch.A.M. Luyben, voorzitter van het College voor Promoties, in het openbaar te verdedigen op woensdag, 6 januari 2016 om 15:00

door

Irina BOLAT

Dipl-Ing in Food Science and Engineering, “Dunărea de Jos” University, Galaţi, geboren te Galaţi, Romania The dissertation has been approved by the: Promotor: Prof. dr. J.T. Pronk Copromotor: Dr. ir. J-M. Daran

Composition of the doctoral committee:

Rector Magnificus Chairperson Prof. dr. J.T. Pronk Delft University of Technology, promotor Dr. ir. J-M. Daran Delft University of Technology, copromotor

Independent members: Prof. dr. E.J. Smid Wageningen University Prof. dr. G. Walker University of Abertay Dundee Prof. dr. W.R. Hagen Delft University of Technology Dr. B. Gibson VTT Technical Research Centre, Espoo, Finland Dr. ir. J-M. Geertman Heineken Supply Chain, Zoeterwoude

The research described in this thesis was performed at the Industrial Microbiology Section, Department of Biotechnology, Delft University of Technology, the Netherlands and financed by Heineken Supply Chain via the R&I platform.

Cover design & layout by Alex Wesselink, Persoonlijkproefschrift.nl TABLE OF CONTENTS

Samenvatting/Summary 9

Chapter 1 General introduction 19

Chapter 2 Chromosomal copy number variation in Saccharomyces 53 pastorianus is evidence for extensive genome dynamics in industrial lager brewing strains

Chapter 3 amdSYM, a new dominant recyclable marker cassette for 89 Saccharomyces cerevisiae

Chapter 4 Functional analysis and transcriptional regulation of 115 two orthologs of ARO10, encoding broad-substrate- specificity 2-oxo-acid decarboxylases, in the brewing yeast Saccharomyces pastorianus CBS1483

Chapter 5 Saccharomyces cerevisiae x Saccharomyces eubayanus 141 interspecific hybrid, the best of both worlds and beyond

Curriculum vitae 172

List of Publications 173

Acknowledgments 175

HYMN TO NINKASI

Born of the flowing water, Tenderly cared for by the Ninhursag,(…)

You are the one who handles the dough [and] with a big shovel, Mixing in a pit, the bappir with sweet aromatics, Ninkasi, you are the one who handles the dough [and] with a big shovel, Mixing in a pit, the bappir with [date] - honey,

You are the one who bakes the bappir in the big oven, Puts in order the piles of hulled grains, Ninkasi, you are the one who bakes the bappir in the big oven, Puts in order the piles of hulled grains,

You are the one who waters the malt set on the ground, The noble dogs keep away even the potentates, Ninkasi, you are the one who waters the malt set on the ground, The noble dogs keep away even the potentates,

You are the one who soaks the malt in a jar, The waves rise, the waves fall. Ninkasi, you are the one who soaks the malt in a jar, The waves rise, the waves fall.

You are the one who spreads the cooked mash on large reed mats, Coolness overcomes, Ninkasi, you are the one who spreads the cooked mash on large reed mats, Coolness overcomes,

You are the one who holds with both hands the great sweet wort, Brewing [it] with honey [and] wine Ninkasi, (...)(You the sweet wort to the vessel)

The fermenting vat, which makes a pleasant sound, You place appropriately on a large collector vat. Ninkasi, the fermenting vat, which makes a pleasant sound, You place appropriately on a large collector vat.

When you pour out the filtered beer of the collector vat, It is [like] the onrush of Tigris and Euphrates. Ninkasi, you are the one who pours out the filtered beer of the collector vat, It is [like] the onrush of Tigris and Euphrates.

The oldest recipe for brewing written by Sumerians as a poem dedicated to their beer goddess Ninkasi. Written down in 1800 BC, the hymn is in fact much older. Translation by Miguel Civil. (Joshua J. Mark, 2011. The hymn to Ninkasi, goddess of beer, Ancient Encyclopedia History)

SAMENVATTING SUMMARY

of the PhD thesis “Analysis of the hybrid genomes of brewing yeasts” Samenvatting

SAMENVATTING

Eén van de best bewaarde geheimen van brouwers wordt vertegenwoordigd door de gebruikte gist in het brouwproces, vanwege de enorme impact die deze heeft op het specifieke smaakprofiel van het eindproduct. Dit vindt zijn oorsprong in de genetische aanleg van de gebruikte gist. Gebaseerd op het gedrag van de gistcellen aan het einde van de fermentatiestap in het brouwproces, kunnen er twee hoofdgroepen brouwgist worden onderscheiden: hooggistende Saccharomyces cerevisiae stammen (ale gist) en laaggistende Saccharomyces pastorianus stammen (lager gist). Deze laatste hebben een complexe genetische architectuur vanwege hun hybride genoom, bestaande uit de chromosomen van twee verschillende soorten: S. cerevisiae en S. eubayanus. De S. pastorianus giststammen zijn geclassificeerd in twee groepen, afgeleid van twee afzonderlijke hybridisaties tussen S. cerevisiae en S. eubayanus stammen: de Saaz groep is ontstaan uit de samensmelting van een haploïde S. cerevisiae met een haploïde S. eubayanus; de Frohberg groep vindt zijn oorsprong in de fusie van een diploïde S. cerevisiae met een haploïde S. eubayanus. De giststammen binnen de twee groepen hebben zowel verschillende genoomgroottes als duidelijk verschillende fysiologische eigenschappen. In dit proefschrift worden een aantal studies beschreven die uitgevoerd zijn met ondergistende (lager) giststammen, met CBS1483 uit de Frohberg groep als model.

De oorsprong van lager giststammen wordt beschreven in Hoofdstuk 1. Hoewel al veel langer bekend was dat lager giststammen een hybride genoom hebben, was tot voor kort alleen het S. cerevisiae sub-genoom duidelijk geïdentificeerd. De non-cerevisiae tegenhanger is pas recent ontdekt en aangeduid als S. eubayanus. De complexiteit van het genoom van lager giststammen wordt verder vormgegeven door de brouwomstandigheden. Deze veroorzaken abnormale chromosoom kopieaantallen (aneuploïditeit), inter-chromosomale translocaties, geheel of gedeeltelijk verlies van chromosomen, chromosomale rearrangements met toenemende kopieaantallen en introgressie. Al deze veranderingen spelen een belangrijke rol in de uniekheid van ondergistende stammen. Dit wordt behandeld in Hoofdstuk 2, waar de totale genoomsequentie van zes ondergistende stammen duidelijk laat zien dat de hybride genomen van biergisten meer zijn dan alleen eenvoudige samenvoegingen van twee sub- genomen. De stam CBS1483, die gedurende dit proefschrift als studieobject werd gebruikt, is gesequenced met Illumina HiSeq2500 en 4 gepaarde libraries met verschillende insert groottes. Met deze methode kon de hoogste coverage van een lagergist genoom behaald worden (~270x) die tot nu toe gepubliceerd is. Tevens is het kopieaantal van de chromosomen in CBS1483 bepaald. Hierdoor werd het mogelijk om een genetische kaart samen te stellen, bestaande uit een totaal van 68 chromosomen en 35 verschillende chromosomale structuren. Chromosoom III vertoonde een zeer intrigerende structuur, als een uitzonderlijk chimeer chromosoom, zonder eenvoudige kopieën van één van de twee sub-genomen. De aneuploïditeit van CBS1483 werd benadrukt door de hoge variatie in kopieaantal van de 35 chromosomale structuren, uiteenlopend van 1 tot wel 5 kopieën.

10 Samenvatting

Dergelijke buitengewone kopieaantallen werden bevestigd met qPCR en flow cytometrie. Al met al bleek het genoom van S. pastorianus CBS1483 te bestaan uit 56% van S. cerevisiae, 34% van S. eubayanus en 10% chimeer S. cerevisiae/S. eubayanus chromosomaal DNA, verdeeld over 68 chromosomen. Naast de identificatie van de chromosomale complexiteit werd nog een tweede bijzondere eigenschap vastgesteld in CBS1483 die zich uitte in allel variatie, waarbij kopieën van hetzelfde gen bestonden uit verschillende nucleotide sequenties. Om meer inzicht te krijgen in het genomische landschap van lager giststammen werden nog vijf Frohberg-type gisten gesequenced, resulterend in een complex beeld van de soort S. pastorianus. Niet alleen varieerde het chromosoom kopieaantal significant tussen de verschillende bestudeerde industriële lager stammen (van 49 tot 79), ook werd een duidelijke impact aangetoond op brouweigenschappen als de capaciteit om diacetyl te produceren en de flocculatie capaciteit.

Aneuploïditeit in lager gistcellen geeft overlevingsvoordelen, maar het onderhoud en de expressie van complete extra chromosomen zorgt ook voor een toegenomen energiebehoefte. Dit zou een verklaring kunnen zijn voor de hoge gevoeligheid voor antibiotica die aneuploïde stammen vertonen. Gezien het feit dat dergelijke remmers onderdeel uitmaken van de genetische modificatie strategie die gebruikt wordt bij de selectie van bepaalde eigenschappen, beperkt deze gevoeligheid van lager giststammen het aantal heterologe genen dat als selectiemarker gebruikt kan worden. Tegemoetkomend aan deze beperking, beschrijft Hoofdstuk 3 een nieuwe herbruikbare dominante markercassette amdSYM, succesvol gebruikt in zowel lager als ale giststammen. De amdSYM cassette, gevormd uit de Ashbya gossypii TEF2 promotor en terminator en een codon-geoptimaliseerd aceetamidase gen (Aspergillus nidulans amdS), zorgt ervoor dat de gisten aceetamide als enige stikstofbron kunnen gebruiken. Hergebruik van de amdSYM cassette zonder enige heterologe sequenties achter te laten in het genoom, werd eenvoudig mogelijk gemaakt door te groeien in aanwezigheid van fluoroaceetamide. Met behulp van deze techniek werden de volgende genen gedeleteerd: S. cerevisiae-type HXK1 in de Saccharomyces pastorianus lager giststam CBS1483, S. cerevisiae-type ARO80 in een Scottish Ale stam en S. eubayanus-type ARO80 in de nieuw ontdekte stam CBS12357. Geen van deze stammen bezit de mogelijkheid om aceetamide te gebruiken als enige stikstofbron, wat hen uitstekende kandidaten maakt voor de nieuwe marker. Een ander voordeel van deze marker is dat deze eindeloos hergebruikt kan worden, waardoor meerdere modificaties mogelijk gemaakt worden zonder de extra eiwitbelasting die de overexpressie van verschillende heterologe markers met zich mee zou brengen. De Euroscarf collectie bevat de nieuwe amdSYM cassette.

Er is nog maar weinig bekend over de regulering en de impact van de complexe genoomorganisatie van lager giststammen op de smaakproductie. In het kader hiervan wordt in Hoofdstuk 4 een case studie gepresenteerd over de bijdrage aan de smaakproductie van de S. cerevisiae en S. eubayanus subgenomen van de lager gist CBS1483. De studie concentreerde zich op ARO10, een 2 oxo-acid decarboxylase betrokken bij de productie van hogere alcoholen via de Ehrlich pathway, en de bijbehorende transcriptieregulator

11 Samenvatting

ARO80. Beide genen bevinden zich op chromosoom IV, of een chromosoom met gelijke grootte, met drie S. cerevisiae-type allelen (LgSc) en een S. eubayanus-type allel (LgSeub). De functionele analyse van de twee type allelen die de twee sub-genomen binnen de lager giststam CBS1483 vertegenwoordigen, werd uitgevoerd door beide allelen individueel tot expressie te brengen in een decarboxylase-negatieve laboratoriumstam van S. cerevisiae. Hierbij werden subtiele verschillen aangetoond in de substraatspecificiteit van de S. cerevisiae- achtige en S. eubayanus-achtige isoenzymen van Aro10. Hoewel phenylpyruvaat het voorkeurssubstraat was voor beide, was de activiteit op ketoisovaleraat, een precursor voor isobutanol productie, tweemaal hoger voor SeubAro10. Transcript analyse toonde duidelijke verschillen aan in stikstofbron-afhankelijke regulatie van de twee allelen. Phenylalanine als stikstofbron was de sterkste inductor voor (Lg)ScARO10 in zowel de als controle gebruikte S. cerevisiae stam als de lager giststam CBS1483. Het LgSeubARO10 allel in deze stam werd anders gereguleerd dan het ScARO10 allel, met een hoge basale expressie indien gegroeid op ammonia en leucine als stikstofbron, en matige expressie op phenylalanine. Een andere interessante ontdekking was dat de ratio van de transcriptieniveau’s van LgScARO10 en LgSeubARO10 in phenylalanine-gegroeide cultures 3:1 was, wat overeenstemt met het aantal kopieën dat voor beide allelen bepaald werd. De herbruikbare dominante markercassette amdSYM, beschreven in hoofdstuk 3, werd gebruikt om de enkele kopie van LgSeubARO80 uit CBS1483 te verwijderen. Dit onthulde een trans cross-regulatie tussen de twee type allelen, waarbij LgScerARO80 de regulatierol van zijn tegenhanger overneemt. De studie gaf een duidelijk gebrek in correlatie aan tussen de enzymatische activiteit van ARO10 en de genexpressie, wat er sterk op wijst dat post-transcriptionele regulatie een belangrijk rol speelt in aneuploïde lager giststammen.

Het gecompliceerde genoom van lager stammen vertegenwoordigt een sterk voorbeeld van de kracht van omgevingscondities op de moleculaire mechanismen die verantwoordelijk zijn voor bepaalde brouwkarakteristieken. Om de voordelen voortkomend uit het vroegere hybridisatiefenomeen tussen S. cerevisiae en S. eubayanus naar de S. pastorianus afstammeling te onderschrijven, werd een kunstmatige hybride geproduceerd en bestudeerd in Hoofdstuk 5. Met behulp van mass mating werd een hybride geconstrueerd tussen een haploïde S. cerevisiae stam van de CEN.PK familie en een haploïde stram afgeleid van de S. eubayanus type stam CBS12357. De genomen van de hybride stam en een van de ouders (S. eubayanus) werden gesequenced met behulp van Illumina technologie. Door de temperatuursinvloed tijdens fermentatie te bestuderen, werd duidelijk dat de nieuwe stam voor de meeste temperaturen de groeikarakteristieken had verworven van de best presterende ouder. Voor temperaturen tussen 20 en 30°C vertoonde de nieuwe stam betere prestaties dan de beste ouder. Het vermogen van giststammen om maltose en maltotriose te consumeren, is van het uiterste belang bij het produceren van bier, zowel voor stabiliteit als om economische redenen. In de Sc x Seub intersoort hybride, werd het onvermogen van S. eubayanus om maltotriose te consumeren gecompenseerd door de acquisitie van het S. cerevisiae genoom. De hybride vertoonde duidelijke diauxie bij gebruik van maltose en maltotriose, zoals ook

12 Samenvatting gezien in de S. pastorianus stam CBS1483. Hoewel de huidige data onvoldoende houvast biedt om de onderliggende mechanismen te identificeren die aan de basis liggen van de fysiologische verschillen tussen Saccharomyces soorten bij verschillende temperaturen, zou de verbeterde sequentie van S. eubayanus en de beschikbaarheid van de Sc x Seub hybride in de toekomst kunnen bijdragen een het ontcijferen van de multifactoriële en weinig begrepen moleculaire grondslag van koude tolerantie.

13 Summary

SUMMARY

One of the best guarded secrets of brewers is represented by the brewing yeast employed in beer fermentation, due to its profound impact upon the specific flavor profile of the final product. This is in turn imparted by its genetic make-up. Based on the behavior of the yeast cells at the end of the fermentation step of the brewing process, two main groups of brewing yeast can be distinguished: top fermenting yeast (ale yeast) Saccharomyces cerevisiae and bottom-fermenting yeast (lager yeast) Saccharomyces pastorianus. The latter has a complex genetic architecture due to its hybrid genome, comprising chromosomes from two different species: S. cerevisiae and S. eubayanus. The S. pastorianus brewing yeast strains have been classified in two groupsderived from two distinct hybridization events between S. cerevisiae ale strains and S. eubayanus: the Saaz group resulted from the hybridization of haploid S. cerevisiae with haploid S. eubayanus and the Frohberg group obtained from the fusion of a diploid S. cerevisiae with a haploid S. eubayanus. The yeast strains within the two groups have different genome sizes as well as distinct physiological characteristics. This thesis presents a number of studies performed on bottom-fermenting (lager) brewing yeast strains, using as model the lager yeast CBS1483 from the Frohberg group.

The origin of lager brewing strains as well as their taxonomic classification is presented in Chapter 1. Although lager brewing strains were long known to have a hybrid genome, only the S. cerevisiae sub-genome was clearly identified, while the non-cerevisiae counterpart has only recently (has only been discovered recently) and designated as S. eubayanus. The complexity of the genome of lager brewing strains is further shaped by the brewing conditions. They trigger an unbalanced chromosome copy number (aneuploidy), inter-chromosomal translocations, complete or partial chromosome loss, chromosomal rearrangements with increased gene copy number and introgressions. All these changes play an important role in the unicity of lager brewing strains. This is addressed in Chapter 2 where the whole-genome sequence of six lager brewing strains clearly indicated that the hybrid genomes of brewing yeast strains are more than just simple pairings of two sub-genomes. The strain CBS1483, used as a study case throughout the thesis, was sequenced with Illumina HiSeq2500 and 4 paired libraries with different insert sizes. This method allowed the highest coverage of a lager genome (~270x) published to this day. The copy number of the chromosomes within CBS1483 was also identified, thus allowing the assembly of a genetic map, with a total count of 68 chromosomes and 35 different chromosomal structures. Chromosome III showed such an intriguing structure, standing out as a chimeric chromosome, with no plain copies of either of the subgenomes. The aneuploidy of CBS1483 was underlined by the high variation in copy number of the 35 chromosomal structures, ranging from 1 copy up to 5 copies. Such an exceptional chromosome copy number was also confirmed with qPCR and flow cytometry. Overall, the genome of S. pastorianus CBS1483 was composed of 56% of S. cerevisiae, 34% of S. eubayanus and 10% of chimeric S. cerevisiae/S. eubayanus chromosomal DNA distributed over 68 chromosomes. Alongside the identification of chromosome intricacy, another special feature was distinguished in CBS1483, represented by allelic variation, with copies of the

14 Summary same gene displaying different nucleotide sequences. Further into understanding the genomic landscape of lager brewing strains, five more Frohberg-type yeasts were sequenced, revealing a complex picture of the S. pastorianus species. Not only the chromosome copy number significantly varied among the industrial lager strains studied (49-79), but it clearly impacted their brewing-related traits: the diacetyl production capacity, flocculation capacity.

Aneuploidy in lager yeast cells brings survival advantages but maintaining and expressing entire additional chromosomes also represents an energetic burden. This might also explain the high sensitivity exhibited by aneuploid strains to antibiotics. Considering that such inhibitors are part of the genetic engineering strategy involved in the selection of certain traits, this sensitivity of lager brewing strains restricts the number of heterologous genes that can be used as selectable markers. Addressing this constraint Chapter 3 describes a new recyclable dominant marker cassette amdSYM, successfully used in both lager and ale brewing strains. The amdSYM cassette, formed by the Ashbya gossypii TEF2 promoter and terminator and a codon-optimized acetamidase gene (Aspergillus nidulans amdS), confers the yeasts the ability to use acetamide as sole nitrogen source. The recycling of the amdSYM cassette was easily performed by growth in the presence of fluoroacetamide, without leaving any heterologous sequences in the genome. With this technique the following genes were deleted: S. cerevisiae - HXK1 in the Saccharomyces pastorianus lager brewing strain CBS1483, S. cerevisiae - ARO80 in a Scottish Ale strain and S. eubayanus - ARO80 in newly discovered strain CBS12357. None of these strains have the capability to grow on acetamide as sole nitrogen source, which makes them good candidates for the new marker. Another advantage of this marker is the possibility to be re-used an unlimited number of times, thus enabling multiple modifications without the protein burden that would cause the overexpression of several heterologous markers. The Euroscarf collection hosts the new amdSYM cassette.

The impact of the complex genome organization of lager brewing strains on flavour production and its regulation is poorly understood. In this respect a case study on the contribution of the S. cerevisiae and S. eubayanus subgenomes from lager brewing yeast CBS1483 upon aroma production is presented in Chapter 4. The study focused on ARO10, a 2 oxo-acid decarboxylase involved in production of higher alcohols via the Ehrlich pathway and its transcriptional regulator ARO80. Both genes are localized on chromosome IV, or a chromosome of similar size, with three S. cerevisiae-type alleles (LgSc) and one S. eubayanus-type allele (LgSeub). The functional analysis of the two types of alleles reflecting the two sub-genomes within lager brewing strain CBS1483 was performed by individual expression of each allele in a decarboxylase-negative laboratory strain of S. cerevisiae. Subtle differences were revealed in substrate specificity of the S. cerevisiae-like and S. eubayanus- like isoenzymes of Aro10. While phenylpyruvate was the preferred substrate for both, the activity towards ketoisovalerate, a precursor for isobutanol production, was 2-fold higher for LgSeubAro10. The transcript analysis revealed clear differences in nitrogen-source dependent regulation of the two alleles. Phenylalanine as nitrogen source was the strongest

15 Summary inducer for (Lg)ScARO10 in both S. cerevisiae strain used as control and lager brewing strain CBS1483. The LgSeubARO10 allele in the brewing strain was regulated differently from the LgScARO10 allele, showing a high basal expression when growing in ammonia and leucine as nitrogen source and moderate in phenylalanine. Another interesting discovery was that the ratio of the transcript levels of LgScARO10 and LgSeubARO10 in phenylalanine-grown cultures was 3:1, which is consistent with the number of copies identified for each allele. The recyclable dominant marker cassette amdSYM described in Chapter 3, was used to delete the single copy of LgSeubARO80 from CBS1483. This disclosed a trans cross-regulation between the two types of alleles, with LgScARO80 taking over the regulatory role of its counterpart. The study clearly indicated a lack of correlation between the enzymatic activity of ARO10 and the gene expression, strongly suggesting that post-transcriptional regulation is very active in aneuploid lager brewing strains.

The intricate genome of lager strains represents a strong example of the power of the environmental conditions upon the molecular mechanisms, responsible for specific brewing traits. To endorse the advantages brought by the ancient hybridization phenomenon between S. cerevisiae and S. eubayanus onto the new offspring S. pastorianus, an artificial hybrid was produced and studied in Chapter 5. Using mass mating, a hybrid between a haploid S. cerevisiae strain of the CEN.PK family and a haploid strain derived from the S. eubayanus type strain CBS12357 was constructed. The genomes of the hybrid strain and one of the parents (S. eubayanus) were sequenced using Illumina technology. Studying the temperature response during fermentation, it was clear that the new strain acquired the growth characteristics of the best performing parent for most temperatures, and it outperformed the best parent for temperatures ranging from 20 to 30°C. The ability to consume maltose and maltotriose by the brewing yeast strains is of paramout importance in beer production, both for stability as well as economic reasons. In the Sc x Seub interspecific hybrid, the inability of S. eubayanus to consume maltotriose was compensated by the acquisition of the S. cerevisiae genome. Interestingly, the hybrid showed a pronounced diauxic utilization of maltose and maltotriose, also observed in the S. pastorianus strain CBS1483. While the present data are not sufficient to identify the underlying mechanisms that govern the physiological differences between the Saccharomyces species at different temperatures, the improved S. eubayanus sequence and the availability of the Sc x Seub hybrid should in the future contribute to deciphering the multifactorial and poorly understood molecular basis of cold tolerance.

16 Summary

17

CHAPTER 1 General Introduction General Introduction

A BRIEF HISTORY OF BREWING

Beer brewing from germinated barley, together with wine and bread making, is among the oldest biotechnological achievements of humankind. As such, these ancient biotechnological processes helped pave the way from a nomadic lifestyle to more structured societies. The earliest chemical evidence of beer in the archaeological record was discovered in current Iran, nearby the Zagros Mountains. Calcium oxalate (beerstone) inside pottery vessels dating from 3400 - 3000 BC provide clear evidence for their use in brewing in ancient Mesopotamia (Michel, McGovern, 1992; Michel et al., 1993). As soon as the proto-cuneiform writing was invented, texts about beer production and consumption showed a well-developed knowledge and technology base for the brewing process (Figure 1, Figure 2) (Damerow, 2012). Although translations of the Sumerian administrative and literary texts (most notably the Hymn to the Sumerian goddess of beer, Ninkasi) are biased by the current terminology used for beer brewing, they leave no doubt on the ancient roots of modern beer brewing processes.

Figure 1. Proto-cuneiform text Figure 2. Impression on a lapis lazuli from Mesopotamia (ca. 3000 cylinder seal from Queen Pu-abi’s tomb in B.C.) showing calculations of the Royal Cemetery at Ur (ca. 2600-2500 ingredients for different beers B.C.) Top - couple sharing a pot of beer (Nissen, 1990) using long straws (Damerow, 2012).

In Europe, the oldest evidence of beer production dates from 800 BC, in the form of an earthenware amphora discovered in Northern Bavaria that was shown to contain wheat beer residues. Celtic people that inhabited Bavaria fled to the British Isles after the Roman conquest of Central Europe, transferring the brewing knowledge to that part of Europe. The Celts may therefore be considered the ancestors of both German and English brewing culture (Holliland, 2012). As commercial and domestic brewing expanded, regulations were imposed starting as early as 1156 with a decree in the city of Augsburg (Holliland, 2012) stating that ‘the bad beer should be discarded’ and culminating in the Beer Purity Law. This law (Reinheitsgebot) was imposed in 1516 by the Bavarian Duke Wilhelm IV and stipulated that only barley, hops and water be used in Bavarian beers (Kunze, 1996). Later on, this law, which is regarded as the oldest food regulation, was applied to entire Germany and had a great positive impact on all breweries across Europe. In 1516, the nature and role of a fourth key ingredient of beer, the yeast, was still to be discovered.

20 Chapter 1

The presence of yeast cells in beer was first observed by the Dutch scientist and tradesman Antonie van Leeuwenhoek in 1680, using his elegant self-made microscopes (Figure 3A). The important improvements brought by Giovanni Amici (1820) to the resolution of the microscope’s objectives opened the way to further studies concerning the involvement 1 of yeast in alcoholic fermentation. Remarkable research was performed in 1837 by the French physicist Charles Cagniard-Latour, the algologist Friedrich Kützing and the German physiologist Theodor Schwann. They observed that yeasts are living organisms involved in the transformation of sugar into alcohol (Stewart & Russell, 1986; Barnett, 2003). This statement was later unequivocally proven by Louis Pasteur who showed, in 1860, that fermentation is a consequence of yeast metabolism (Pasteur, 1860).

Until the middle of the 16th century all beers were of the ale type. In ale fermentation, the yeast characteristically converts an extract from malt (germinated barley) at relatively high temperatures (20-25º), followed by a short maturation period. After the prohibition of summer brewing imposed in 1553 in Bavaria by Duke Albrecht V (Holliland, 2012) a new type of beer started to be produced, known today as lager beer. The cold environment under which brewing was now performed resulted in the selection of new yeast strains that were capable of undergoing alcoholic fermentation at lower temperatures (8-15º), with long cold maturation periods ‘lagering’ (Kodama et al., 2005). Nowadays, lager beers dominate the beer market while ales and beers brewed by spontaneous fermentation (e.g. Lambic beer) complement the variety in commercially available beers.

A next crucial step in the brewing industry was represented by the novel technique of producing pure yeast cultures. The method was developed by Emil Christian Hansen (1883) in the Carlsberg Laboratory in Copenhagen (Polaina, 2002). Furthermore, he introduced a fed- batch system for yeast propagation which allowed increased biomass production (Boulton, Quain, 2001). This breakthrough, together with the development of the thermometer (G.D.Fahrenheit-1714, A. Celsius-1742) (Figure 3C), the saccharometer (Richardson-1788) for measuring sugar content of wort (Figure 3B), the steam engine (J.Watt-1765) and the refrigeration machine (C von Linde-1871), led to a rapid growth of the number of large, industrially operated breweries. (Kunze, 1996; van Hamersveld, 1996).

Figure 3. (A)Microscope invented by Antonie van Leeuwenhoek (Breig, 2006), (B)Saccharometer invented by J. Richardson (http://www.olney-antiques.co.uk/product/9/35/saccharometer), (C)Thermometer with mercury invented by G.D. Fahrenheit http://www.bornrich.com/original-thermometer-invented-daniel- gabriel-fahrenheit-fetch-157000-london-auction.html.

21 General Introduction

THE OUTLINE OF THE BREWING PROCESS

Beer production comprises three main stages: a) wort production, b) fermentation and c) post-fermentation processing (Figure 4). a) Wort production takes place in the brewhouse, where malt (germinated barley with 4-5% humidity) is grinded to a suitable size and mixed with water in a process called mashing. During mashing, the temperature is gradually increased (55ºC to 78ºC), with rest phases corresponding to the optimum temperature for the activity of the different enzymes present in malt to degrade its abundance of water-insoluble compounds (starch, glucans, proteins) to soluble, short chains molecules, that can be metabolised by the yeast cells during fermentation. The substances that go into solution are referred to as extract. The enzymatic activity is further improved by adjusting the pH of the mash to 5.5-5.6. To separate the insoluble materials, the mash is filtrated in a vessel with perforated floor, so-called lauter tun, that allows the aqueous solids-free extract, called wort, to be separated from the spent grains. To avoid extract loss and improve the brewhouse yield, the spent grains are washed with hot water. These are rich in sugar, proteins and inorganic material are sold as cattle food. The wort is collected in a wort kettle where it is boiled for 60-90 min. During this step, bitter and aromatic hop varieties are added. The high temperature induces the isomerisation of α-acids from hops, thereby giving beer its bitter taste. During wort boiling, other important processes occur: formation and precipitation of protein-polyphenol complexes, inactivation of all enzymes, wort sterilisation, increase in wort color by formation of melanoidins and oxidation of polyphenols. During boiling, the S-methylmethionine SMM is converted to dimethylsulfide (DMS) with an unpleasant smell and taste, and dimethylsulfoxide (DMSO). During this stage of the process, the evaporation rate of DMS must be high enough to reduce it below its flavour threshold value (50 – 60 µg·l-1 DMS) (Kunze, 1996). At the end of boiling, the wort is transferred to a whirlpool, where large particles settle down in the shape of a cone in the middle of the vessel. This compact mass is called ‘coarse break’ or ‘hot trub’ and contains precipitated proteins, lipids and zinc. The clear wort is rapidly cooled down to 7-8ºC for lager beers and to 15-22ºC for ales (Goldammer, 1999). During wort cooling, protein-polyphenol complexes precipitate and form the ‘cold break’ or ‘cold trub’. The particles of this trub are small and remain in suspension for a long period of time. The extent of trub removal from wort has long been studied in relation with yeast metabolism in the fermentation step. A trub-rich wort causes a significant faster fermentation than clear worts because the trub contains unsaturated long-chain fatty acids and ergosterol, which cannot be synthesized by yeast under the anaerobic conditions of large-scale beer fermentation (Kühbeck et al., 2006). Moreover, zinc, one of the essential ions for yeast metabolism, is loosely bound to the trub particles, thus being easily released in the aqueous solution during fermentation (de Nicola, Walker, 2009). Another positive aspect brought by the presence of trub is related to its particulate characteristics that promote the formation of CO2 bubbles from the medium during fermentation, thus reducing the toxic effect of the aqueous species

22 Chapter 1 on yeast cells. The stirring effect produced by the rising bubbles keeps the yeast cells in suspension for a longer period of time, accelerating the fermentation process (Kühbeck et al., 2007). The downside is that turbid worts tend to have a poorer flavour quality and stability, beer filtration-characteristics and foam stability, although the differences compared with a 1 completely clear wort are often minor. b) Fermentation represents the most important process in beer production and takes place in the fermentation cellar: the cold wort is aerated and transferred to the fermentation vessels where it is inoculated (pitched) with yeast suspension. Yeast cells used for inoculation are collected from a previous fermentation. Their cellular membrane is sterol-depleted due to the anaerobic nature of the fermentation from which they were harvested that impairs sterol formation. After pitching the aerated wort with such cells, yeast survives on its glycogen reserve until the fluidity of the membrane is reinstated (Verbelenet al., 2009).

The reuse of yeast suspension harvested from a previous fermentation to a subsequent one is common practice in beer production and, upon each repitching, a brief aeration phase is required to restore the levels of lipids and sterols in the membrane. The build-up of these essential compounds enables multiplication of the yeast cells during the subsequent anaerobic fermentation phase and the amount of oxygen supplied at the beginning of fermentation determines the extent of yeast growth. As soon as oxygen is depleted, anaerobic fermentation commences, during which the yeast cells metabolise wort nutrients into ethanol, carbon dioxide and a series of secondary metabolites essential for beer flavour and stability (higher alcohols, esters, aldehydes, SO2). During fermentation, the temperature is allowed to rise to 10 - 11ºC for lager beers and to 18 - 25ºC for ales (Goldammer, 1999). Due to this difference in temperature, the process lasts longer (5-7 days) in case of bottom-fermentation (lager) than top-fermentation (ale) (3 days) and also the spectrum of flavour compounds differs considerably for these two beer types. Specific characteristics of ale and lager yeast strains are shown in Table 1. In particular for lager beers, a ‘diacetyl rest’ is applied when the majority of wort carbohydrates is consumed. To this end, the temperature is increased to 13-15 ºC to accelerate degradation of the undesirable flavour compound diacetyl, which has a pronounced buttery aroma. In case of ale beers this step is often omitted, as many brands prefer to keep this compound above its sensory threshold level as part of the aroma trademark of the beers. During this warmer phase of the process, other undesirable flavour compounds characteristic to “young/green” beer are also removed. These include acetaldehyde and sulfur compounds, whose removal further contributes to the maturation of the beer flavour.

23

General Introduction WORT WORT Oxygen Yeast Wort cooling Clari fi cation Fermentation /Maturation Hops

Boiling pitcing re- Yeast Yeast storage tank WORT SWEET BEER Lautering Beer fi ltration Spent grains MASH beer tank Bright Mashing Bottling Water Grist Milling Schematic representation of the steps within brewing process: malt is milled and resulted grist mixed with water (generally 1:3 ratio) Adjuncts: barley, maize Malt Figure 4. Figure 4. From here the sweet wort is collected and sent to This is heated according to a defined diagram followed by filtration in lauter tun. forming a mash. the precipitated proteins are being removed from wort in a whirlpool and After boiling, while the spent grains are sold to farmers. copper for the boiling step, followed by beer filtration and bottling. The fermentation starts and maturation continues in the same vessel, clear wort is aerated and pitched with brewing yeast.

24 Chapter 1

Yeast growth in beer fermentation processes ceases when sterols are depleted. Yeast cells then sink to the bottom of the fermenter (lager) or rise to the surface (ale). The yeast biomass can then be harvested and stored at 4ºC, to be used for re-pitching of new fermentation batches later on. Beer conditioning starts with forced cooling of the beer, to 0ºC for lager 1 beers and to 4ºC for ales, with the purpose of saturating the beer with carbon dioxide and to clarify it. Conditioning ensures colloidal stability of the final product by precipitation and removal of protein-tannin complexes and yeast sediment. In modern breweries, high-capacity cylindroconical tanks are used for both stages of fermentation as well as for conditioning. This ‘uni-tank’ system decreases costs and shortens the process time.

The concept of ‘high-gravity brewing’, in which fermentation is performed with very concentrated worts followed by dilution at a later stage, is becoming increasingly popular because it enables increased volumetric productivity. However, suitable yeast strains must be developed to sustain the stress factors brought by this method. A number of breweries employ continuous fermentation systems in which a group of vessels produce a continuous beer stream. This approach is mainly suitable for breweries that produce only a small number of brands, since the advantage of continuous brewing only applies during longer production runs.

There are two types of continuous-brewing systems: (i) those resembling chemostats (systems in which the medium is continuously added while the liquid culture is continuously removed); (ii) plug flow system (system with an elongated form, continuously fed with a mix of wort and yeast). For low or zero-alcohol beers, immobilised yeast reactors are employed, offering the possibility of a short contact between wort and yeast during primary fermentation, to remove wort flavour. c) Post-fermentation processing comprises beer filtration, pasteurization and bottling during which the temperature is kept below 0°C and the sterile conditions impeccable.

THE ROLE OF YEAST IN BEER FERMENTATION

The efficiency of fermentation depends on the physiological properties of the yeast being inoculated as well as on external factors such as wort composition, wort turbidity, aeration, temperature and inoculum density (‘pitching rate’) (Guido et al., 2004).

The main role of brewing yeast is to ferment wort sugars to ethanol and carbon dioxide. Therefore, the correlation between the brewing yeast sugar substrate range and the wort sugar spectrum is of paramount importance. The composition of a typical all-malt wort is shown in

25 General Introduction

Table 2 and comprises glucose, fructose, sucrose, maltose, maltotriose and dextrins. With the exception of dextrins, most S. pastorianus strains are capable of utilizing the entire range of carbohydrates present in wort. A number of physiological studies indicated that certain lager brewing strains lack the ability of utilizing maltotriose (Duval et al., 2010, Gibson et al., 2013). The consumption of these sugars occurs sequentially (Stewart, 2006). Sucrose is hydrolysed to glucose and fructose by invertase and disappears first from the wort (Meneses et al., 2002), then glucose and fructose and finally maltose and maltotriose. Glucose is the preferred substrate for fermentation and the presence of this sugar in the medium represses expression of plasma membrane transporters for the other sugar molecules (Hammond, 1993; Boulton, Quain, 2001). Therefore, only after more than 60% of the glucose in wort has been fermented, maltose is imported by the yeast cells (D’Amore et al., 1989). The length of the fermentation process is to a large extent determined by the rate of maltose fermentation, which is the most abundant sugar in wort. Unlike the facilitated diffusion of glucose and fructose, uptake of the α-glucosides maltose and maltotriose by yeast cells takes place via proton symport. Several genes have been reported to encode α-glucoside transporters in Saccharomyces yeast strains:

AGT1, MALx1 genes, MPH2, MPH3 (maltose permease homolog) and MTT1 (MTY1). Of the encoded transporters, only Mtt1 has a higher affinity for maltotriose than for maltose (Dietvorst et al., 2005; Salema-Oom et al., 2005), while Agt1 has the highest affinity for maltose. At the beginning of fermentation, glucose represses the transcription of these genes and, moreover, inactivates α-glucoside transporters already present in the membrane (Boulton, Quain, 2001). In lager brewing strains, the AGT1 gene is truncated, with a premature stop codon in the sequence that disturbs its functionality (Vidgren et al., 2005; Nakao et al., 2009).

In lager strains, MALx1 genes therefore play a dominant role in maltose uptake. The α-glucoside transporters also differ with respect to their temperature sensitivity. Activity of Agt1 is strongly temperature dependent, while Mtt1 has a comparatively low temperature dependence (Vidgren et al., 2010). Many breweries employ adjuncts in the form of sucrose or maltose syrup to boost the sugar content of worts. Addition of high concentrations of sucrose may lead to incomplete fermentation of wort, caused by a prolonged glucose repression of maltose and maltotriose transporters (Russell 1993). In areas where barley cultivation is not viable for climatic reasons, other cereals are used for beer production. The use of sorghum, maize and rice requires addition of β-amylase for fermentable sugar release, which is absent in their seeds. These alternative cereals contain high amounts of starch with limited solubilisation during mashing and the endosperm remains intact during malting. This causes the release of β-glucans in the mash and difficulties during filtration (Kunze, 2006; Ogbonna, 2011; Tayloret al., 2013).

26 Chapter 1

Table 1. Morphological and physiological differences between ale and lager brewing yeast strains

Ale yeast Lager yeast Reference History Before 3400 BC After 1500 AD Michel et al., 1993; Holliland, 2012 1 Taxonomic name Saccharomyces Saccharomyces van de Walt, 1970 cerevisiae pastorianus Morphology

Barnett, 1992 Branched chains of cells Single/pairs of cells

CBS 1171 CBS 1538 Fermentation 20-25ºC 8-15ºC Kodama et al., 2005 temperature Optimum 30ºC 25ºC Reed, Nagodawithana, temperature 1991 Max growth 37ºC 34ºC Tornai-Lehoczki et al., temperature 1996 Briggs et al., 2004 Flocculation Weak Strong Walker , 1998 Hydrophobic Hydrophilic Rhymes, Smart, 2000 Low surface charge High surface charge Hammond, 1993 Mannose insensitive NewFlo phenotype Dengis et al., 1995; phenotype (Inhibited by mannose, Boulton, Quain, 2001 sucrose, glucose, maltose (Not inhibited by sugars) but not by galactose) Harvesting Top Bottom Kunze, 1996 Carbohydrate Melibiose not utilized Meliobiose fermented Walker, 1998; Briggs et utilization al., 2004 Poor maltotriose Good maltotriose Stewart, 2006; utilization for most strains utilization (Frohberg Duval et al., 2010 group) Gibson et al., 2013 High temperature Low temperature Vidgren et al., 2010 dependence of maltose dependence of maltose transport (AGT1) transport (MTT1) Fructose uptake via Fructose uptake via Briggs et al., 2004 facilitated transport proton symport Raffinose incompletely Raffinose completely Kunze, 1996 utilized utilized Beer aroma Fruity -pronounced Smooth- balanced

27 General Introduction

Table 2. Sugar composition of an all-malt wort (Stewart, Russel, 1993)

Sugar type Wort composition, % Glucose 10-15 Fructose 1-2 Sucrose 1-2 Maltose 50-60 Maltotriose 15-20 Dextrins 20-30

During fermentation, brewer’s yeast produces a wide range of flavour compounds, many of which are by-products of yeast nitrogen metabolism. The nitrogenous components present in wort are listed in Table 3. Due to the fact that proteolytic activity of brewers’ yeast cells is limited and free ammonia is only present in small amounts in wort (Kunze, 2006), the main source of nitrogen in wort is represented by free amino acids. The amino acids are not all directly incorporated into proteins. After entering the cell, certain amino acids (valine, leucine, isoleucine, methionine, phenylalanine) are transaminated, the amino group being donated to other carbon skeletons while free 2-oxo acids are generated (Fontana, Buiatti, 2008). These oxo-acids are further metabolised, yielding aldehydes and higher alcohols with great impact on beer flavour (Hazelwood et al., 2008). Nitrogen starvation of brewers’ yeast can occur when adjuncts are used in amounts that are not balanced with the available nitrogen in wort. The low amount of proteins characteristic to sorghum, maize and rice and the complete absence of nitrogen in sucrose syrups brings a deficit of α-amino nitrogen with direct impact on yeast cells performance. Under these circumstances yeast cells will synthesise de novo the essential aminoacids. Along these biosynthetic pathways a number of by products are formed that impart off-flavours to beer: vicinal diketones produced via the isoleucine-valine biosynthetic pathway (Krogerus, Gibson, 2013) while high concentrations of SO2 and H2S are registered when a deficit in the sulfur-containing amino acid methionine occurs (Fontana, Buiatti, 2008). Yeast metabolism and its viability depends on the (bio) 2- 3- availability of inorganic nutrients, including Zn, Ca, Na, K, Mg, Cu, Fe, Mn, SO4 , PO4 . Moreover brewing yeasts required the organic growth factors biotin and panthotenic acid (Kunze, 2006). The uptake rate and utilization of these nutrients not only depends on their concentration in the wort, but also on their bio-availability (Aleksander et al., 2009). The availability or toxicity of metal ions for bewers yeast in wort exhibits synergistic effects. For example, low concentrations of manganese, an essential element for yeast, inhibits zinc availability (Helin, Slaughter, 1977).

Table 3. Nitrogen components in wort (Reed, Nagodawithana, 1991; Boulton, Quain, 2001)

Nitrogen compound Wort composition, % Aminoacids 30-40% Polypeptides 30-40% Protein 20% Nucleotides 10%

28 Chapter 1

TAXONOMIC CLASSIFICATION OF BREWER’S YEAST

The ultimate goal of taxonomic studies is to classify organisms on the basis of their evolutionary relations (Price et al., 1978). However, the criteria originally used in yeast 1 taxonomy relied heavily on morphological and physiological characteristics of different strains (van der Walt, 1970), with further attempts on more discriminative approaches like serology (Tsuchiya et al., 1974), structure of cell wall mannans (Ballou et al., 1974), DNA base composition (Yarrow, Nakase, 1975) and proton-magnetic resonance spectra of extracted mannans (Fukazawa et al., 1980). Later, DNA sequence similarity became increasingly important in studying the phylogenetic relationships and taxonomy of yeasts.

Species from the genus Saccharomyces were historically classified in two groups: sensu stricto and sensu lato, with the former exhibiting closely related strains with high fermentative ability, displaying a uniform number and chromosome distribution (karyotype) while the species in sensu lato are more heterogeneous (Figure 5) (Naumov et al., 1996a; Špírek et al., 2003). All species initially cumulated under the sensu stricto group display 16 chromosomes upon karyotyping (Naumov et al., 2000c; Fischer et al., 2000) and chromosomal rearrangements were only been detected among closely related species, most probably due to recombinations between non-homologous regions of different chromosomes that might have been beneficial for the organism (Fischer et al., 2000). The exceptions are S. pastorianus with higher number of chromosomes and S. arboricola at the opposite end, with only 12-13 choromosomes.

Figure 5. Karyotypes corresponding to species within (A) Saccharomyces sensu stricto group (Nguyen et al., 2000) and (B) Saccharomyces sensu lato group (Petersen et al., 1999) The species initially assigned to the sensu lato group, were recently reassigned to new genera (Naumovia, Kazachstania, Lachancea) (Kurtzman, 2003)

29 General Introduction

Kurtzman and Robmett (2003) studied the Saccharomyces complex based on multigene sequence analysis, leading to the assignment of 75 species to 14 clades. To classify yeast strains based on their evolutionary similarities, comparison at the genomic level proves to be the most suitable approach. In the DNA renaturation method (Seidler, Mandel, 1971), the extent of sequence identity in two strains is analysed by simply mixing isolated DNA of the two strains, briefly denaturating the DNA and later allowing to renaturate the two strands, followed by a spectrophotometric reading of the resulted sample. Using this approach, 24 species originally grouped in Saccharomyces sensu stricto (van der Walt, 1970; Yarrow , Nakase, 1975) were reduced to only 4 taxa due to the sequence homology displayed by many of the strains: S. cerevisiae (Sc), S. bayanus, S. pastorianus (S. carlsbergensis) and S. kluyveri (Table 4 ) (Vaughan Martini, Kurtzman, 1985). The S. kluyveri was later shown to be evolutionary distant from Saccharomyces sensu stricto (Vaughan-Martini et al., 1993; Ando et al., 1996). Karyotyping (Sheehan et al., 1991), restriction analysis of mitochondrial DNA (Guillamon et al., 1994), hybridization with probes from Ty1 retrotransposable element family (Naumov et al., 1998) and Ty2 element (Codon et al., 1998), PCR fingerprinting (de Barros Lopes et al., 1998), spore formation between species and their viability (Naumov, 2000a) showed clear polymorphisms between and within the four species. Later this allowed fast identification of species just by using specific PCR primers (de Melo Pereiraet al., 2010; Muir et al., 2011). Saccharomyces bayanus was suggested to comprise two variants, displaying different karyotypes: S. bayanus var uvarum and S. bayanus var bayanus. The former one consists of yeast strains capable of fermenting melibiose and unable to grow above 37ºC (Naumov et al., 2000a, Pulvirenti et al., 2000). Moreover, they also display reduce homology at gene level (Casaregola et al., 2001) with semisterile hybrids (low viability ascospores) (Naumov et al., 2000c).

Another species, well-established in the Saccharomyces genra, is S. paradoxus shown to abundantly occur on plant leaves (Glushakova et al., 2007). While species corresponding to S. bayanus and S. pastorianus were predominantly isolated from winery and brewery environments (with few exceptions for S. bayanus), S. cerevisiae and S. paradoxus are readily isolated from similar natural habitats: broad-leaved trees, soil and insects (Table 4). This difference is mirrored in the mitochondrial DNA (mtDNA) of these strains, with conserved genetic organization in case of S. cerevisiae strains and high polymorphism of mtDNA corresponding to wine and brewers’ yeast strains (Codon et al., 1998; Foury et al., 1998).

Three new species isolated in Brazil and Japan were designated as part of the Saccharomyces genra: S. kudriavzevii (Kaneko, Banno, 1991), S. cariocanus (Morais et al., 1992) and S. mikatae (Yamada et al., 1993). Confirmation of these new strains involved a thorough comparative analysis with the already known species (genetic crosses, molecular karyotyping, 18S rRNA sequence analysis) (Naumov et al., 2000c). S. kudriavzevii was recently discovered in oak bark, in Portugal. Unlike the strains from Japan, the Portugese strains were able to metabolise galactose (Sampaio JP, Gonçalves P, 2008) and showed gene sequences closer to the hybrid wine strains from Europe. Isolates from Taiwan (Naumov et al., 2013), confirmed

30 Chapter 1 the presence of the species in this new location, together with S. arboricola and omnipresent S. cerevsiae. The taxonomic location of S. cariocanus is still controversial, with some authors considering it as an independent species (Naumov et al., 2000c), while others (Liti et al., 2006) considering it still as part of the lineages of S. paradoxus due to the lack of aneuploidy 1 in the viable spores resulted from S. cariocanus X S. paradoxus crosses.

S. arboricola is a recently described Saccharomyces species, discovered in the bark of the trees in mainland China (Wang, Bai, 2008) and only one strain in Taiwan (Naumov et al., 2013). The genome of this new species was recently sequenced and phylogenetic analysis placed it between S. kudriavzevii and S. bayanus (Liti et al., 2013). The strains analysed within this species showed low growth and sensitivity to high temperatures, but good utilization of mannitol and biotin prototrophy.

The most recent and spectacular addition to the Saccharomyces group is represented by S. eubayanus (Seub). This species was initially discovered in Argentina in the galls of the Nothofagus trees in Patagonia. Its genome showed a 99.56% identity with the non- cerevisiae part of the hybrid genome of S. pastorianus (Libkind et al., 2011) and proved to be a cryotolerant strain (Gibson et al., 2013). Recently (Peris et al., 2014), a new S. eubayanus strain was discovered in North America (Wisconsin) being identified as an intraspecific hybrid of two Patagonian populations. At the same time, 3 lineages of S. eubayanus were identified in Far East Asia: Tibet, west and north-west China (Bing et al., 2014) with the Tibetan lineage considered as the real contributor to the lager brewing strains non-cerevisiae sub-genome due to a higher sequence similarity (99.82%) than the Patagonian strains. Low ascopore viability, was noted for S. eubayanus x S. bayanus var. uvarum hybrids (Naumov et al., 2013).

Lager brewing strains are designated as Saccharomyces pastorianus. This name was first assigned by Reess in 1870 to alcoholic fermentation fungi with “sausage-shaped cells” (Hansen, 1895). In 1904, E.C. Hansen established the brewing yeast type I selected in Carlsberg breweries as being true S. pastorianus. Later studies performed on Hansen’s strain showed that it behaved like S. bayanus in terms of galactose and raffinose fermentation (van der Walt, 1965), seemingly refuting its classification as a separate species. Later tests involving single-chromosome transfer from S. pastorianus strains to kar1Δ S. cerevisiae mutants, followed by sequence homology checks at different loci (Nilsson-Tillgren et al., 1981; Holmberg, 1982; Pedersen, 1986), DNA reassociation tests of S. pastorianus strains with S. cerevisiae and S. bayanus species (Vaughan Martini, Martini, 1987), indicated the presence of both genomes in different proportions, thus reinstating S. pastorianus as a separate species.

In 1908, a new bottom-fermenting strain was selected by EC Hansen assigned as S. carlsbergensis (Hansen, 1908). Due to the high similarity with the previously described S. pastorianus both strains correspond to the same species. According to the International

31 General Introduction

Code of Botanical Nomenclature the first name designated to a strain must be kept (Vaughan Martini, Martini, 1987) thus all lager brewing strains should be addressed as S. pastorianus. Nevertheless, many publications still use the name S. carlsbergensis when referring to these type of yeasts. The hybrid nature of lager strains and their chromosomal rearrangements were confirmed by a wide range of molecular techniques, including southern blotting with S. cerevisiae probes (Tamai et al., 1998), comparative DNA sequence analysis via BLASTN, BLASTX (Cliften et al., 2001), fluorescent amplified fragment length polymorphism (AFLP) on genomic DNA of S. pastorianus, S. cerevisiae and S. bayanus strains (Casaregola et al., 2001; de Barros Lopes et al., 2002).

Based on the existing sequences of S. cerevisiae (Goffeau et al., 1996) and S. bayanus var uvarum (Kellis et al., 2003) a comprehensive study was performed by Rainieri et al. (2006), on 35 strains assigned to S. bayanus/S. pastorianus that resulted in three groups, based on the contributing subgenomes: (a) S. cerevisiae/ S. bayanus lager genome, (b) S. cerevisiae/S. uvarum/S. bayanus lager genome, (c) S. bayanus/ S. uvarum lager genome. Lager brewing strains were found in the first group.

All the species currently recognized in the Saccharomyces genra are presented in Table 4.

The first complete sequence of a lager brewing strain genome was published only in 2009 (Nakao et al.) for Weihenstephan 34/70. Since then, other research groups unravelled the sequences of a number of lager brewing strains (Hewitt et al., 2014; Walther et al., 2014; this thesis). Clear genetic evidence of two subgenomes, one similar to S. cerevisiae and the other to S. bayanus, further confirmed the hybrid nature of the genome of lager yeasts. The mitochondrial DNA was proved to be only from the non-cerevisiae parent, while the nucleic genome showed many translocations, deletions and insertions specific for cultured strains.

Lager brewing strains were distributed in two groups by Dunn & Sherlock (2008) based on their origin and subsequent evolution: Group I comprising strains that arose via hybridization between a haploid S. cerevisiae spore with a haploid S. bayanus spore, followed by massive losses of the cerevisiae subgenome (Saaz type strains), and Group II with strains resulted from the fusion of a diploid S. cerevisiae with haploid S. bayanus (Frohberg type strains). This last group kept important parts of both genomes with only limited losses of genetic information from the S. bayanus part. Unlike the Saaz strains, the Frohberg strains have the ability to utilize maltotriose (Gibson et al., 2013) displaying a better fermentation performance. The lager brewing strain CBS1483 used throughout the research described in this thesis, was included in the second group.

32 Chapter 1

Table 4. Currently recognized species in the Saccharomyces genra

Species Identification Geographic range Yeast strains Reference Natural habitats 1 S. cerevisiae Soil, broad- Europe (Finland, e.g. Naumov et al., 1992 leaved trees, the Netherlands), CBS1171 Naumov, 1996b Drosophila Far East Asia, Sniegowski et al., 2002 North America S. paradoxus Oak exudates, Europe, Far East e.g. Naumov et al., 1992 forest soil, peat, Asia, North CBS5829 Naumov et al., 1996a fruit body of America, Africa, Sniegowski et al., 2002 mushrooms, Hawai Drosophila, fungal galls of some trees S. kudriavzevii Decayed leaf Europe e.g. Kaneko, Banno, 1991 Oak bark (Portugal), Asia CBS12752s Naumov et al., 2000, (Japan, Taiwan, 2013 Malaysia) Scannell et al., 2011 Sampaio, Gonçalves, 2008 S. mikatae Soil, decayed leaf Asia (Japan) e.g. Yamada et al., 1993 CBS8839s Naumov et al., 2000 Scannell et al., 2011 S. arboricola Bark tree Asia (China, e.g. Wang, Bai , 2008 Taiwan) CBS10644s Liti et al., 2013 Naumov et al., 2013 S. cariocanus Drosophila South America e.g. Morais et al., 1992 (Brazil) CBS8841 Naumov et al., 2000 S. bayanus var. Mesophylax Europe e.g. Naumov et al., 1992 uvarum adoperus, CBS7001s Tornai-Lehoczki et al., Drosophila, 1996; mushrooms, Scannell et al., 2011 hornbeam Naumov et al., 2013 exudates S. eubayanus Galls on beach South America e.g. Libkind et al., 2011 trees (Patagonia) CBS12357s Peris et al., 2014 High land China Bing et al., 2014 North America Industrial habitats S. cerevisiae Fermentation Europe, Africa e.g. Naumov et al., 1994 processes (high CBS4054 Naumov, Naumova, 2011 temperature) S. pastorianus Breweries Europe e.g. Vaughan Martini, CBS1513, Martini, 1987 (S. cerevisiae x WS34/70s Nakao et al., 2009 S. eubayanus) S. bayanus var Brewing Europe e.g. Rainieri et al., 2006 bayanus contaminant NBRC1948 Libkind et al., 2011 CBS424 Naumov et al., 2013 (S. uvarum x Wineries Pear/ S. eubayanus x Apple juice S. cerevisiae)

33 General Introduction

The genome of the recently discovered cryotolerant species S. eubayanus showed even higher sequence identity with the non-cerevisiae portion of S. pastorianus than the genome of S. bayanus. The discovery of S. eubayanus and its genome sequence have therefore provided an invaluable, previously missing link in the phylogeny of lager brewing strains. The contribution of S. eubayanus rather than S. bayanus to the complex hybrid genome of S. pastorianus has been extensively confirmed by testing specific primer pairs designed on the sequence ofS. eubayanus against the two species with negative results for S. bayanus (Pengelly, Wheals, 2012) and recently by whole-genome sequencing of a number of S. eubayanus strains from different origins (Bing et al., 2014).

Numerous S. eubayanus strains have been detected in Patagonia with a diverse genetic composition. The S. eubayanus moiety within the S. pastorianus genome diverged from one such population, several thousand years ago (Peris et al. 2014). The genomes of lager strains continue to present intriguing genomic puzzles, not only with respect to their functionality and regulation, but also with respect to the question on how two species that have been discovered so far apart from the place where lager beers were first produced (Bavaria, Germany) might have mated. One attractive hypothesis is that the S. eubayanus species also occur in Europe, but that its niche is still waiting to be discovered. As shown for S. paradoxus (Glushakova et al., 2007), S. eubayanus may not be restricted to sugar rich environments but, for example, be an epiphytic species whose abundance follows seasonal dynamics, thereby necessitating year-long investigations for its successful isolation.

While growth and fermentation performance of S. eubayanus from Patagonia are relatively similar with those of the lager strains from the Saaz group, the ester profile of this new species is comparable with the Frohberg lager strains (Gibson et al., 2013).

Chimeric genomes containing subgenomes from species of the Saccharomyces complex have been frequently found in wine regions and breweries (Peris et al., 2012b) but also in human respiratory tract isolates (de Barros Lopes et al., 2002; Peris et al., 2012a) showing that interspecific hybridization amongSaccharomyces species is more common than previously believed.

CHALLENGES IN ANALYSIS OF BREWING YEAST GENOMES

In-depth knowledge of the genome sequence of brewing strains is a prerequisite for understanding the genetic basis of their phenotype and the complex, multi-layered regulation of metabolism under industrially relevant conditions.

Industrial strains evolved to be polyploid or aneuploid, meaning that they harbour more than two sets of homologous chromosomes with, in the case of aneuploidy, different copy numbers for the individual chromosomes. Polyploidy and aneuploidy can occur via whole- genome duplications or by acquiring/losing single chromosomes. It has been observed that

34 Chapter 1 in diploid cells of S. cerevisiae, adaptive mutations are 1.6 times more frequent than in an isogenic haploid population (Paquin, Adams, 1983). Ploidy thus contributes to genetic flexibility and, thereby, to the possibility to rapidly adapt to new environments (Stewart, 1981; Bond et al., 2004). Lager brewing yeasts are invariably aneuploids, with unique 1 chromosomal rearrangements displayed by different strains (Hansen, Kielland-Brandt, 2003).

For a long time, quantitative analysis of overall ploidy has relied on the use of flow fluorocytometry (Hutter, Eipel, 1979), as well as colorimetric methods (Aigle et al., 1983). In such analyses, aneuploidy presented difficulties due to the impossibility of differentiating the copy number for each individual chromosome. Early methods employed the use of microarrays containing S. cerevisiae DNA, culminating in the competitive comparative genome hybridisation (CCGH) method that allowed the determination of the copy number of cerevisiae genes within the S. pastorianus hybrid relative to a haploid S. cerevisiae reference strain (Bond et al., 2004). This method could, however, only resolve half of the genome because the S. eubayanus part had not yet been identified or sequenced. Currently, the availability of whole genome sequences corresponding to S. cerevisiae, S. bayanus MCYC623 and S. pastorianus Weihenstephan 34/70 allows the design of specific primers for each subgenome within lager brewing strains. This in turn enables accurate analysis of chromosome copy number variation by real-time quantitative PCR method using either a control sample with a known ploidy (D’haene et al., 2010) or a reference gene within the analysed genome with an unambiguously defined copy number (Tadamiet al., 2014). In recent studies performed on a meiotic segregant derived from WS34/70 (Ogata et al., 2011) as well as on different lager brewing strains (Tadami et al., 2014), custom DNA microarrays comprising both Sc- and S. eub-type probes allowed the estimation of chromosomes copy number variation.

An additional challenge in analysing brewing yeast genomes is brought about by their hybrid nature, with chromosomes deriving from two species (Figure 6) with largely syntenic genomes (similar order of genetic loci on chromosomes): S. cerevisiae and S. eubayanus (Libkind et al., 2011).

The sequence of different lager strain (Nakao et al., 2009; Hewitt et al., 2014; Wendland et al., 2014), indicated the presence of more than 31 chromosomes. Three types of translocations were observed involving one or both subgenomes: Sc-Sc, Seub-Seub and Sc-Seub. It is well known that sequence divergence acts as a barrier against homologous recombination (Datta et al., 1997). However, in lager brewing strains, the orthologous genes share a high similarity of 78-88% (Hansen, Kielland-Brandt, 2003) displaying many identical short sequences that allow chromosomal translocations. At the same time, double strand breaks in the telomeric regions increased the chances of chromosomal recombination as well as the presence of transposon-related sequences (entire Ty elements or just terminal repeats) that were previously shown to be correlated with the localization of breakpoints (Dunham et al., 2002). The beneficial traits acquired byS. pastorianus are therefore the result of two factors: interspecies

35 General Introduction hybridization and adaptive evolution under the fermentative environment imposed by beer fermentation, lagering and yeast storage. Different mutational events have contributed to the reshaping of brewers’ yeast genomes, including introgressions (Naumova et al., 2011) or translocations that can result in gene silencing through frameshifts in the coding regions (Usher, Bond, 2009), breakpoints in the protein coding regions producing chimeric genes (Hewitt et al., 2014), complete deletions or partial amplification (Jameset al., 2008) and even whole chromosome duplications (Hughes et al., 2002). Such a genetic response of yeast cells to different selection pressures is indicative for an enormous genomic plasticity and was observed in other hybrid strains such as hybrids of S. cerevisiae and S. uvarum (Dunn et al., 2013) as well as in single S. cerevisiae strains: the diploid ancestor of S288C S. cerevisiae (Esberg et al., 2011), evolved clones from S. cerevisiae diploid strains (Dunham et al., 2002) and S. cerevisiae flor yeasts from sherry wines (Infanteet al., 2003). The activity of transcriptional regulators within the hybrid genome is another factor influenced by the environmental conditions (Tirosh et al., 2009) with major impact on lager yeast behaviour. Aneuploidy in yeast cells brings survival advantages but maintaining and expressing entire additional chromosomes also represents an energetic burden. Evolution studies indicate that chromosomal amplification is in fact a “quick fix” for cells under stressful conditions, with a dynamic gain and loss of this state depending on the sustainability of this solution to different growth conditions (Yona et al., 2012).

Figure 6. Schematic representation of ploidy and recombination of a few of the chromosomes within a lager brewing strain, depicting the complex structure of the genome. The two colours reflect contributions by the two different parents: S. cerevisiae (blue) and S. eubayanus (red).

In many wine, brewing and distilling strains, the genes involved in the fermentation of industrially relevant sugars are mainly located in the sub-telomeric regions (e.g. MAL, HXT, SUC, FLO loci) (Louis et al., 1994; Naumov et al., 2000b; Nakao et al., 2009). This clearly indicates the selection pressure of the environmental conditions upon the nuclear genome of yeasts employed by humans in different processes.

36 Chapter 1

Another characteristic of brewing strains that hinders their genetic analysis and impairs their cross-breeding, is represented by their partial, often almost complete, deficiency in sporulation (Gjermansen, Sigsgaard, 1981; Spencer, Spencer, 1977) despite both, S. cerevisiae and S. eubayanus, parental strains being perfectly capable to sporulate (Naumov et al., 2013) 1 (Figure 7). The viability of the few spores that are formed is less than 1%. It is well known that the mating type of S. cerevisiae is determined by the genetic organization of the MAT locus. Poor sporulation of brewing strains was proven to be determined by the lack of function of the mating type genes featuring a sequence that might have been altered during breeding rounds (Tsuboi, Takahashi, 1988) rather than the sporulation genes or their ploidy, as previously believed (Stewart, 1981; Casey, 1986). Studies of chromosome III where the MAT locus is located in two different lager brewing strains (CMBS33, 6701), indicated a mosaic chromosome (S. cerevisiae like and S. non-cerevisiae like parts) with different copy numbers for each strain (Bond et al., 2004). Successful attempts to increase the sporulation frequency in lager brewing yeasts were obtained by overexpressing the Sb-IME1 (regulator of meiosis) in Weihenstephan 34/70 (Ogata et al., 2011), resulting in diploid segregants.

Over the past few years, DNA sequencing methods rapidly became faster, cheaper and more exhaustive. With the development of next generation sequencing technology a higher coverage of the entire genome is possible within a limited time frame. This opens the way for faster sequencing of new yeast strains with complex genomes. The next generation sequencing technique comprises 5 steps: i) DNA library creation; ii) construction of contigs; iii) scaffold assembly; iv) annotation; and v) uncovering copy number variation. All these steps are depicted in Figure 8.

Figure 7. Asci obtained during directed sporulation of S. eubayanus CBS 12357, one of the parental strains of S. pastorianus (Bolat I, unpublished data)

37 General Introduction

GENETIC IMPROVEMENT OF BREWING YEAST STRAINS

The desire to develop new beer brands, the increase in popularity of very high gravity (VHG) fermentation techniques, the diversification of raw materials employed as well as the increase in efficiency and sustainability desired for the current brewing technologies, together create strong incentives for the development of improved brewing yeast strains.

The sequence of a brewing yeast provides a clear picture of the natural capabilities of the strain. Building upon the existing characteristics and optimizing them, constitutes a cumulative sustained effort from geneticists and physiologists in collaboration with brewers. A number of factors involved in reducing brewing time or improving different characteristics can be handled by addition of enzymes. The mosaic genome of lager brewing strains and their abnormal number of chromosomes impose a different genetic engineering approach than for haploid strains. Strain improvement of lager brewing yeasts have been performed using one of the following methods: cross-hybridization, physical (e.g. UV, X-ray) or chemical (e.g. base analogue, de-aminating agent) mutagenesis and gene cloning.

Early studies employed protoplast fusion between haploid and lager brewing strains (de Figueroa , de Van Broock, 1985; Jones et al., 1985; Hansen et al., 1990) with the desire of obtaining viable spores that carry specific traits of the parents. The differences observed in the fusion products compared with the original spheroplasts hindered a structural approach for selective traits (Stewart, 1981; Reed, Nagodawithana, 1991).

Classical breeding was successfully used with a limited number of spores obtained from a production lager brewing strain (Gjermansen, Sigsgaard, 1981) but its applicability is limited by the lack of sporulation of most strains. Nevertheless, methods have been developed allowing the formation of spores from a number of lager brewing strains by employing lower temperatures (21ºC) and using acetate-based sporulation media from which cells were harvested in stationary phase (Bilinski et al., 1987).

Rare mating is another technique employed to produce new industrial strains which relies on the occasional mating type switching. This allows the formation of mating cells capable of hybridizing with haploid strains of known genotype (Schillberg et al., 1991; Spencer, Spencer, 1996; Hammond, 2003). The brewing characteristics of laboratory hybrids differ with the brewery they are tested in, which indicates the important role the environment plays.

Chemical mutagenesis has been employed to reduce the amount of diacetyl, an undesirabe off-flavor compound in beer, whose butter-scotch aroma displays a very low flavour threshold level. In this respect, partial inactivation of the gene ILV2 involved in production of the diacetyl precursor α-aceto-lactate was obtained by selection of spontaneous mutants resistant to the herbicide sulfometuron methyl (Falco, Dumas, 1985; Yadav et al., 1986; Gjermansen et al., 1988).

38 Chapter 1

1 Figure 8. The steps of next generation sequencing technology required for resolving complex genome sequences and chromosome copy number variation number variation copy genome sequences and chromosome resolving complex required for sequencing technology generation The steps of next Figure 8.

39 General Introduction

Genetic transformation of brewing yeasts was first performed by introducing plasmids carrying the desired gene as well as a marker gene for selection of the transformants. When high copy number plasmids (containing 2µm-fragment) were used (Gjermansen, 1983; Meaden, Tubb, 1985; Perry, Meaden, 1988; Villanueba et al., 1990) a loss of the desired plasmid during several rounds of fermentations was encountered. Also the negative effect on fermentation performance of the overproduction of the gene located on the plasmid decreases their suitability. The best option for a stable genetic inheritance is represented by the integration of the desired DNA into the yeast genome. In this method the plasmid contains the desired DNA (deletion cassette) flanked by sequences homologous to the up-/ down-stream chromosome sequence. Using this technique (Scherer, Davis, 1979) plasmid integration was possible, resulting in stable transformants. The counterselection of the marker gene based on the bacteriophage Cre recombinase (Hoess, Abremski, 1985) was originally only easily achievable in ura3 auxotrophic strains (Sauer, 1987) using the plasmid pBS49 with URA3 as selectable marker and GAL1 promoter.

A comprehensive functional analysis of any gene implies the study of the yeast strain completely devoid of all copies of that gene. This is particularly difficult for lager brewing strains due to their high chromosomal copy number and the genetic divergence of the chromosomes. To produce auxotrophy in such strains, at least two alleles (S.cerevisie-type and S. eubayanus–type) must be deleted. This requires the use of heterologous genes as dominant markers for the deletion cassette construction. The choice for different markers must take into account the high sensitivity exhibited by aneuploid strains to protein synthesis inhibitors (Torres et al., 2007). To this day a number of such markers were successfully used in lager brewing strains (Table 5).

The recombinant-DNA techniques are of paramount importance in laboratory research aimed at understanding the molecular basis for optimal performance in the brewing process. The knowledge obtained from such studies can then be used to design strategies for strain optimization that do not rely on direct gene cloning or targeted gene inactivation. The choice of methods for improvement of brewer’s yeast strains is not only determined by technical possibility. Quite on the contrary, despite clear possibilities for strain improvement by recombinant DNA approaches, lack of consumer acceptance has so far prevented the large-scale market introduction of beers produced with genetically engineered yeast strains. Currently, in the EU states, the food products containing genetically modified microorganisms (GMOs) or ingredients obtained from GMOs must be labelled. In the US this issue is still under debate. People reluctance towards such food is a combination of risks and benefits perception (Costa-Font, 2008) strongly influenced by consumers education and knowledge. Surveys regarding people attitude on genetically modified food consumption showed their acceptance being positively influenced when health benefits were displayed (Lusk, 2003; Magnusson, Koivisto Hursti, 2002). As such, communication strategies are needed that cover public concerns rather than avoid them, displaying the potential benefits for consumers rather than for the producers.

40 Chapter 1

Table 5. Heterologous genes used as markers for genetic modification of lager brewing strains via plasmids or direct integration. The alleles corresponding to each subgenome are designated as Scerevisiae-type and Seubayanus-type respectively.

Dominant Trait Origin of the Targeted Reference 1 marker Acquired marker gene Gene Marker on plasmid (Overexpression) CUP1 Resistance to Cu2+ S. cerevisiae Sc-MET25 Henderson et al., 1985 Omura, Shibano, 1995 SMR1 Resistance to S. cerevisiae Sc-ATF1/ATF2 Verstrepen et al., 2003 the herbicide Seub-ATF2 sulfometuron methyl Marker integrated (Deletion) LSD1 Dextranase activity Lipomyces starkeyi Sc-ILV2 Zhang et al., 2008 KanMx Resistance to G418 E. coli Sc-MET10; Hansen, Kielland- Seub-MET10 Brandt, 1996; Sc-BAT1, Strack, Stahl, 2010 Sc-BAT2 Duong et al., 2011 Sc-ILV6 Murakami et al., 2012 Seub-URA3 CUP1 Resistance to Cu2+ S. cerevisiae Sc-ADH2 Wang et al., 2009 ble Resistance to E. coli Sc-ILV6 Duong et al., 2011 phleomycin AMY Utilization Saccharomycopsis Sc-ILV2 Liu et al., 2004 of starch fibuligera hph Resistance Streptomyces Sc-KEX2 Yamagishi et al., 2010 to hygromycin hydroscopicus Sc-URA3 Murakami et al., 2012 AUR1-C Resistance Aureobasidium Seub-KEX2 Yamagishi et al., 2010 to aureobasidin A pullulans PGKp- Resistance to S. cerevisiae Seub-URA3 Akada et al., 2002; YAP1 cerulenin Iijima, Ogata, 2010; SMR1 Resistance to S. cerevisiae Sc-ILV2 Kusunoki, Ogata, 2012 the herbicide sulfometuron methyl

SCOPE AND OUTLINE OF THE THESIS

Lager brewing strains appeared as a result of natural interspecific hybridization between S. cerevisiae and S. eubayanus. The complex genetic make-up of these hybrid strains delayed their exhaustive study. The first step towards a complete understanding of the molecular processes with direct impact on the physiological behaviour of these type of yeasts was made in 2009 when the first genome sequence of a lager strain (S. pastorianus Weihenstephan 34/70) was unravelled (Nakao et al., 2009). Availability of complete, well-annotated genome sequences is an important step towards understanding how the brewing environment has shaped the genetic constitution of different lager strains involved in the beer making process. In particular, comparison of the genome sequences of

41 General Introduction different lager yeast strains will shed light on the molecular mechanisms responsible for industrially relevant, strain-specific traits.

The goal of the present study was to explore the hybrid genome of a lager yeast using physiological, molecular genetics and genomics techniques. As a model for these studies, the lager brewing strain CBS1483, a typical S. cerevisiae-S. eubayanus hybrid, was used.

Chapter 2 describes the sequencing, assembly and annotation of the genome of S. pastorianus CBS1483 and other five additional lager brewing strains covering the two groups: Saaz and Frohberg. In addition to optimizing strategies for rapid assembly of hybrid genomes based on sequence data from ‘next generation’ sequencing techniques, special attention was paid to accurate analysis of chromosome copy number variation in the hybrid genome and their impact upon the phenotypic traits of the strains.

The presence of high copy number for both types of alleles within the aneuploid genome of lager strains requires the use of recyclable markers. Chapter 3 addresses the cumbersome problem of multiple rounds of gene deletion and marker removal in a lager brewing strain. Gene deletions within such yeast strains with conventional dominant markers such as the commonly used KanMX gene represent a challenge due to their high sensitivity to protein synthesis inhibitors (Torres et al., 2007). In this chapter, a new dominant marker cassette amdSYM is introduced and the necessity of long homologous regions for deletion cassette integration in industrial strains such as CBS1483 is demonstrated.

The production of flavour compounds during the brewing process is closely related to yeast metabolism, but it remains unclear to what extent the two subgenomes of lager brewing strains contribute to the production of several important beer flavour compounds. To address this issue, a detailed study was undertaken in Chapter 4 to investigate the contribution of the S. cerevisiae-like and S. eubayanus-like alleles of the ARO10 gene in S. pastorianus CBS1483 to the production of higher alcohols. The results show how differences in copy number and transcriptional regulation of orthologous genes on the two sub-genomes, as well as differences in substrate specificity of the encoded isoenzymes, complicate systems biology approaches in hybrid brewing yeast strains.

A step further in understanding the impact of the two sub-genomes upon the valuable industrial traits of the lager brewing strains was taken in Chapter 5. In this respect a novel S. cerevisiae X S. eubayanus hybrid was produced and studied versus the individual strains for two features: temperature tolerance and oligosaccharide utilization. Moreover the Patagonian S. eubayanus strain CBS12357 was resequenced in view of a reliable sequence comparison of each individual strain with the new experimental hybrid.

42 Chapter 1

REFERENCES

Aigle M, Erbs D, Moll M, 1983. Determination Breig J, 2006. A close look at eighteenth-century of brewing yeast ploidy by DNA measurement, optics, Colonial WilliamsburgJournal, Autumn 1 J.Inst.Brew., 89: 72-74 2006 - http://www.history.org Akada R, Shimizu Y, Matsushita Y, Kawahata Briggs D, Boulton C, Brookes P, Stevens R, M, Hoshida H, Nishizawa Y, 2002. Use of a 2004. Brewing science and practice. Woodhead YAP1 overexpression cassette conferring specific Publishing Limited, Cambridge, England resistance to cerulenin and cycloheximide Casaregola S, Nguyen HV, Lapathitis G, as an efficient selectable marker in the yeast Kotyk G, Gaillardin C, 2001. Analysis of Saccharomyces cerevisiae, Yeast, 19: 17-28 the constitution of the beer yeast genome by Aleksander P, Piotr A, Tadeusz T, Makarewicz PCR, sequencing and subtelomeric sequence M, 2009. Accumulation and release of metal hybridiation, Int J Syst Evol Microbiol, 51: ions by brewer’s yeast during successive 1607-1618 fermentations, J Inst Brew, 115: 78-83 Casey GP, 1986. Molecular and genetic Ando S, Mikata K, Tahara Y, Yamada Y, 1996. analysis of chromosomes X in Saccharomyces Phylogenetic relationships of species of the genus carlsbergensis, Carlsberg Res. Commun., 51: Saccharomyces Meyen ex Reess deduced from 343-362 partial base sequences of 18S and 26S ribosomal Cliften PF, Hillier LW, Fulton L, Graves T, Miner RNAs (Saccharomycetaceae), Biosci. Biotech. T, Gish WR, Waterston RH, Johnston M, 2001. Biochem., 60: 1070-1075 Surveying Saccharomyces genomes to identify Ballou CE, Lipke PN, Raschke WC, 1974. functional elements by comparative DNA Structure and immunochemistry of the cell sequence analysis, Genome Res, 11: 1175-1186 wall mannans from Saccharomyces chevalieri, Codon AC, Benítez T, Korhola M, 1998. Saccharomyces italicus, Saccharomyces Chromosomal polymorphism and adaptation diastaticus, and Saccharomyces carlbergensis, to specific industrial environments of J.Bacteriol., 117: 461-467 Saccharomyces strains, Appl Microbiol Barnett JA, 1992. The taxonomy of the genus Biotechnol, 49: 154-163 Saccharomyces Meyen ex Reess: a short review Costa-Font M, Gil J, Traill W, 2008. Consumer for non-taxonomists, Yeast, 8: 1-23 acceptance, valuation of and attitudes Barnett JA, 2003. Beginnings of microbiology towards genetically modified food: review and and biochemistry: the contribution of yeast implications for food policy, Food Policy, 33: research, Microbiology, 149: 557-567 99-111 Bilinski C, Russell I, Stewart GG, 1987. Phy- Damerow P, 2012. Sumerian beer: the origins siological requirements for induction of sporulation of brewing technology in ancient Mesopotamia, in lager yeast, J Inst Brew, 93: 216-219 CDLJ 2: 1-20 Bing J, Han P-J, Liu W-Q, Wang Q-M, Bai D’Amore T, Russel I, Stewart GG, 1989. Sugar F-Y, 2014. Evidence for a Far East Asian utilization by yeast during fermentation, J Ind origin of lager beer yeast, Curr Biol, May 19; Microbiol, 4: 315-324 24(10):R380-1. doi: 10.1016/j.cub.2014.04.031. Dengis P, Nélissen L, Rouxhet P, 1995. Bond U, Neal C, Donnelly D, James TC, 2004. Mechanisms of yeast flocculation: comparison of Aneuploidy and copy number breakpoints in the top- and bottom- fermenting strains, AEM, 61: genome of lager yeasts mapped by microarray 718-728 hybridisation, Curr Genet, 45: 360-370

43 General Introduction

de Barros Lopes M, Soden A, Martens AL, Duong C, Strack L, Futschik M, Katou Y, Henschke PA, Langridge P, 1998. Differentiation Nakao Y, Fujimura T, Shirahige K, Kodama Y, and species identification of yeasts using PCR, Nevoigt E, 2011. Identification ofSc -type ILV6 Int J Syst Bacteriol, 48: 279-286 as a target to reduce diacetyl formation in lager de Barros Lopes M, Bellon JR, Shirley NJ, brewers’ yeast, Metab Eng, 13: 638-647 Ganter PF, 2002. Evidence for multiple Duval E, Alves S, Dunn B, Sherlock G, Stambuk interspecific hybridization inSaccharomyces B, 2010. Microarray karyotyping of maltose- sensu stricto species, FEMS Yeast Res, 1: 323- fermenting Saccharomyces yeast with different 331 maltotriose utilization profiles reveals copy de Melo Pereira GV, Ramos CL, Galvão C, Soua number variation in genes involved in maltose Dias E, Schwan RF, 2010. Use of specific PCR and maltotriose utilization, Journal of Applied primers to identify three important industrial Microbiology, 109: 248-259 species of Saccharomyces genus: Saccharomyces Esberg A, Muller L, McCusker J, 2011. Genomic cerevisia, Saccharomyces bayanus and structure of and genome-wide recombination in Saccharomyces pastorianus, Lett Appl Microbiol, the Saccharomyces cerevisiae S288C progenitor 51: 131-137 isolate EM93, PLoS ONE, 6: 1-14 de Nicola R, Walker G, 2009. Yeast Falco S, Dumas K, 1985. Genetic analysis of biotechnology: diversity and applications. mutants of Saccharomyces cerevisiae resistant Interaction between yeasts and zinc. to the herbicide sulfometuron methyl, Genetics, Satyanarayana T, Kunze G (eds.), Springer 109: 21-35 Science+Business Media B.V., 238-255 de Figueroa L, de van Broock M, 1985. D’haene B, Vandesompele J, Hellemans J, 2010. Spheroplast fusion of a brewing yeast strain Accurate and objective copy number profiling with Saccharomyces diastaticus, Appl Microbiol using real-time quantitative PCR, Methods, 50: Biotechnol, 21: 206-209 262-270 Fischer G, James SA, Roberts IN, Oliver SG, Dietvorst J, Londesborough J, Steensma HJ, Louis EJ, 2000. Chromosomal evolution in 2005. Maltotriose utilization in lager yeast Saccharomyces, Nature, 405: 451-454 strains: MTT1 encodes a maltotriose transporter, Fontana M, Buiatti S, 2008. Amino acids in beer, Yeast, 22: 775-788. Beer in health and disease prevention, Victor Dunham MJ, Bedrane H, Ferea T, Adams J, Preedy ed., Elsevier Inc. Brown PO, Rosenzweig F, Botstein D, 2002. Foury F, Roganti T, Lecrenier N, Purnelle Characteristic genome rearrangements in B, 1998. The complete sequence of the exeperimental evolution of Saccharomyces mitochondrial genome of Saccharomyces cerevisiae, Proc Natl Acad Sci USA, 99: 16144-9 cerevisiae, FEBS Lett, 440: 325-331 Dunn B, Sherlock G, 2008. Reconstruction of Fukazawa Y, Schinoda T, Nishikawa A, Nakase the genome origins and evolution of the hybrid T, 1980. Synonymy of Saccharomyces cerevisiae lager yeast Saccharomyces pastorianus, Genome Hansen 1883 and Saccharomyces uvarum Res, 18: 1610-1623 Beijerinck 1898: significance of cell wall antigens Dunn B, Paulish T, Stanbery A, Piotrowski J, in yeast classification. Int J Syst Bacteriol, 30: Koniges G, Kroll E, Louis EJ, Liti G, Sherlock G, 196-205 Rosenzweig F, 2013. Recurrent rearrangement Gibson B R, Storgårds E, Krogerus K, Vidgren V, during adaptive evolution in an interspecific 2013. Comparative physiology and fermentation yeast hybrid suggests a model for rapid performance of Saaz and Frohberg lager yeast introgression, PLoS Genet, 9: 1-16 strains and the parental species Saccharomyces eubayanus, Yeast, 30: 255-266

44 Chapter 1

Gjermansen C, Sigsgaard P, 1981. Construction Hansen J, Kielland-Brandt MC, 2003. Topics of a hybrid brewing strain of Saccharomyces in current genetics: Functional Genetics of carlsbergensis by mating of meiotic segregants, Industrial Yeasts. De Winde J.H, ed.), pp. 143- Carlsberg Res.Commun., 46: 1-11 170. Springer-Verlag, Heidelberg. 1 Gjermansen C, 1983. Mutagenisis and genetic Hazelwood LA, Daran JM, van Maris AJA, transformation of meiotic segregants of lager Pronk JT, Dickinson JR, 2008. The Ehrlich yeast , Carlsberg Res Commun, 48: 557-565 pathway for fusel alcohol production: a cebtury Gjermansen C, Nilsson-Tillgren T, Petersen of research on Saccharomyces cerevisiae J, Kielland-Brandt M, Sigsgaard P, Holmberg metabolism, AEM, 74(8): 2259 - 2266 S, 1988. Towards diacetyl-less brewers’ yeast. Hewitt S, Donaldson IJ, Lovell SC, Delneri Influence ofILV2 and ILV5 mutations, J Basic D, 2014. Sequencing and charaterisation of Microbiol, 28: 175-183 rearrangements in three S. pastorianus strains Glushakova AM, Ivannikova YuV, Naumova reveals the presence of chimeric genes and gives ES, Chernov IYu, Naumov GI, 2007. Massive evidence of breakpoint reuse, PLOS One, 9(3), isolation and identification ofSaccharomyces 1-14 paradoxus yeasts from plant phyllosphere, Hoess R, Abremski K, 1985. Mechanism of Microbiology, 76: 205-210 strand cleavage and exchange in the Cre-lox Goldammer T, 1999. The brewers’ handbook. site-specific recombination system, J Mol Biol, The complete book to brewing beer. KVP 181: 351-362 Publishers Clifton, Virginia, USA. Holliland C, 2012. The Oxford companion to Guillamon JM, Barrio E, Huerta T, Querol A, beer, Ed Oliver G., Oxford University Press, NY, 1994. Rapid characterization of four species 102-108 of the Saccharomyces sensu stricto complex Holmberg S, 1982. Genetic differences between according to mitochondrial DNA patterns, Int J Saccharomyces carlsbergensis and S. cerevisiae Syst Bacteriol, 44: 708-714 II. Restriction endonuclease analysis of genes in Hammond JRM, 1993. The Yeasts, Vol.5, (Ed. chromosome III., Carlsberg Res Commun, 47: Rose AH, Harrison JS), Academic Press, London 233-244 Hammond JRM, 2003. Brewing microbiology, Hughes T, Roberts C, Day H, Jones A, Meyer Ed. F. Priest, I Campbell, Kluwer Academic, New M, Slade D, Burchard J, Dow S, Ward T, Kidd York, USA M, Friend S, Marton M, 2000. Widespread aneuploidy revealed by DNA microarray Hansen EC, 1895. Studies on variations of yeast- expression profiling, Nat Genet, 25: 333-337 cells, Ann Bot, 9: 549-560 Hutter KJ, Eipel HE, 1979. Microbial Hansen M, Röcken W, Emeis C-C, 1990. determinations by flow cytometry,J Gen Construction of yeast strains for the production of Microbiol, 113:369-75 low-carbohydrate beer, J Inst Brew, 96: 125-129 Iijima K, Ogata T, 2010. Construction and Hansen EC, 1908. Recherches sur la physiologie evaluation of self-cloning bottom-fermenting et la morphologie des ferments alcooliques. XIII. yeast with high SSU1 expression, J Appl Nouvelles etudes sur les levures de brasserie a Microbiol, 109: 1906-1913 fermentation basse, CR Trav Lab Carlsberg, 7: 179-217 Infante J, Dombek K, Rebordinos L, Cantoral J, Young E, 2003. Genome-wide amplifications Hansen J, Kielland-Brandt MC, 1996. caused by chromosomal rearrangements play a Inactivation of MET10 in brewer’s yeast major role in the adaptive evolution of natural specifically increases SO formation during beer 2 yeast, Genetics, 165: 1745-1759 production, Nat Biotechnol, 14: 1587-1591

45 General Introduction

James T, Usher J, Campbell S, Bond U, 2008. Liti G, Barton D, Louis E, 2006. Sequence Lager yeasts possess dynamic genomes that diversity, reproductive isolation and species undergo rearrangements and gene amplification concepts in Sacchromyces, Genetics, 174: 839- in response to stress, Curr Genet, 53: 139-152 850 Jones R, Russell I, Stewart G, 1985. The use of Liti G, Nguyen Ba AN, Blythe M, Müller C, catabolite derepression as a means of improving Bergström A, Cubillos F, Dafhnis-Calas F, the fermentation rate of brewing yeast strains, Khoshraftar S, Malla S, Mehta N, Siow C, ASBC Journal, 44: 161-166 Warringer J, Moses A, Louis E, Nieduszynski Kaneko Y, Banno I, 1991. Re-examination of C, 2013. High quality de novo sequencing and Saccharomyces bayanus strains by DNA-DNA assembly of the Saccharomyces arboricolus hybridization and electrophoretic karyotyping, genome, BMC Genomics, 14: 69 IFO Res Comm, 15: 30-41 Liu Z, Zhang G, Liu S, 2004. Constructing Krogerus K, Gibson BR, 2013. Influence of an amylolytic brewing yeast Saccharomyces valine and other amino acids on total diacetyl pastorianus suitable for accelerated brewing, J and 2,3-pentanedione levels during fermentation Biosci Bioeng, 98: 414-419 of brewer’s wort, Appl Microbiol Biotechnol, 97: Louis EJ, Naumova ES, Lee A, Naumov G, 6919-6930 Haber JE, 1994. The chromosome end in Kühbeck F, Back W, Krottenthaler M, yeast: its mosaic nature and influence on 2006. Influence of lauter turbidity on wort recombinational dynamics, Genetics, 136: 789- composition, fermentation performance and beer 802 quality, J. Inst. Brew., 112, 215-221 Lusk JL, 2003. Effects of cheap talk on Kühbeck F, Müller M, Back W, Kurz T, consumer willingness-to-pay for golden rice, Am Krottenthaler M, 2007. Effect of hot trub and J Agric Econ, 85: 840-856 particle addition on fermentation performance Magnusson M, Koivisto Hursti U-K, 2002. of Sacharomyces cerevisiae, Enzyme Microb Consumer attitudes towards genetically modified Tech, 41, 711-720 food, Appetite, 39: 9-24 Kunze W, 1996. Technology Brewing and Meaden P, Tubb R, 1985. A plasmid vector Malting, VLB Berlin system for the genetic manipulation of brewing Kurtzman CP, Robnett CJ, 2003. Phylogenetic strains., Proceedings of the European Brewery relationship among yeasts of the ‘Saccharomyces Convention Congress, Helsinki, 219-226 complex’ determined from multigene sequence Meneses FJ, Henschke PA, Jiranek V, 2002. analyses, FEMS Yeast Res, 3: 417-432 A survey of industrial strains Saccharomyces Kurtzman CP, 2003. Phylogenetic cerevisiae reveals numerous altered patterns of circumscription of Saccharomyces, maltose and sucrose utilization, J Inst Brew, Kluyveromices and other members of the 108: 310-321 Saccharomycetaceae, and the proposal of Michel R, McGovern P, 1992. Chemical evidence the new genera Lachancea, Nakaseomyces, for ancient beer, Nature, 360: 24 Naumovia, Vanderwaltozyma and Michel R, McGovern P, Badler V, 1993. The Zygotorulaspora, FEMS Yeast Res, 4: 233-245 first wine & beer. Chemical detection of ancient Kusunoki K, Ogata T, 2012. Construction of fermented beverages, Anal Chem, 65: 408 – 413 self-cloning bottom-fermenting yeast with low Morais PB, Hagler AN, Rosa CA, Mendonça- vicinal diketone production by the homo- Hagler LC, Klaczko LB, 1992. Yeasts associated integration of ILV5, Yeast, 10: 435-442 with Drosophila in tropical forests of Rio de Janeiro, Brazil, Can J Microbiol, 38: 1150-1155

46 Chapter 1

Muir A, Harrison E, Wheals A, 2011. A Naumov G, James S, Naumova E, Louis multiplex set of species-specific primers for E, Roberts I, 2000c. Three new species in rapid identification of members of the genus the Saccharomyces sensu stricto complex: Saccharomyces, FEMS Yeast Res, 11: 552-563 Saccharomyces cariocanus, Saccharomyces 1 Murakami N, Miyoshi S, Yokoyama R, Hoshida kudriavzevii and Saccharomyces mikatae, Int J H, Akada R, Ogata T, 2012. Construction of a Syst Evol Microbiol, 50: 1931-1942 URA3 deletion strain from the allotetraploid Naumov G, Naumova E, 2011. Genetic bottom-fermenting yeast Saccharomyces identification of African cultured yeasts of the pastorianus, Yeast, 29: 155-165 genus Saccharomyce, Microbiol, 80: 386-390 Nakao Y, Kanamori T, Itoh T, Kodama Y, Naumova E, Naumov G, Michailova Y, Rainieri S, Nakamura N, Shimonaga T, Hattori Martynenko N, Masneuf-Pomarède I, M, Ashikari T, 2009. Genome sequence of the 2011. Genetic diversity study of the yeast lager brewing yeast, an interspecies hybrid, DNA Saccharomyces bayanus var uvarum reveals Res, 16: 115-129 introgresed subtelomeric Saccahromyces Naumov I, Naumova E, Korhola M, 1992. cerevisiae genes, Res Microbiol, 162, 204-213 Genetic identification of naturalSaccharomyces Naumov G, Lee C-F, Naumova E, 2013. sensu stricto yeasts from Finland, Holland and Molecular genetic diversity of the Saccharomyces Slovakia, Antonie van Leeuwenhoek, 61: 237-243 yeasts in Taiwan:Saccharomyces arboricola, Naumov G, Naumova E, Michels C, 1994. Saccharomyces cerevisiae and Saccharomyces Genetic variation of the repeated MAL loci in kudriavzevii, Antonie van Leeuwenhoek, 103: natural populations of Saccharomyces cerevisiae 217-228 and Saccharomyces paradoxus, Genetics, 136: Nguyen H-V, Lepingle A, Gaillardin C, 2000. 803-812 Molecular typing demonstrates homogeneity of Naumov G, Naumova E, Sancho E, 1996a. Saccharomyces uvarum strains and reveals the Genetic reidentification ofSaccharomyces strains existence of hybrids between S. uvarum and S. associated with black knot disease of trees in cerevisiae, including the S. bayanus type strain Ontario and Drosophila species in California, CBS380, System Appl Microbiol, 23: 71-85 Can J Microbiol, 42: 335-339 Nilsson-Tillgren T, Gjemansen C, Kielland- Naumov G, 1996b. Genetic identification of Brandt M, Peterson JL, Holmberg S, 1981. biological species in the Saccharomyces sensu Genetic differences between Saccharomyces stricto complex, J Ind Microbiol, 17: 295-302 carlsbergensis and S. cerevisiae. Analysis of chromosome III by single chromosome transfer, Naumov G, Naumova E, Sniegowski P, 1998. Carlsberg Res Commun, 46: 65-76 Saccharomyces paradoxus and Saccharomyces cerevisiae are associated with exudates of North Ogata T, Shikata-Miyoshi M, Tadami H, American oaks, Can J Microbiol, 44: 1045-1050 Nakazawa N, 2011. Isolation of meiotic segregants from a bottom fermenting yeast, J. Naumov G, 2000a. Saccharomyces bayanus var. Inst. Brew., 117(2): 199-205 uvarum comb. nov., a new variety established by genetic analysis, Microbiology, 69: 338-342 Ogbonna A, 2011. Current developments in malting and brewing trials with sorghum in Naumov G, Naumova E, Masneuf I, Aigle Nigeria: a review, J Inst Brew, 117: 394-400 M, Kondratieva V, Dubourdieu D, 2000b. Natural polyploidization of some cultured Omura F, Shibano Y, 1995. Reduction of yeast Saccharomyces sensu stricto: auto- and hydrogen sulphide production in brewing yeast allotetraploidy, Syst Appl Microbiol, 23: 442- by constitutive expression of MET25 gene, J Am 449 Soc Brew Chem, 53: 58-62

47 General Introduction

Paquin C, Adams J, 1983. Frequency of fixation Rainieri S, Kodama Y, Kaneko Y, Mikata K, of adaptive mutations is higher in evolving Nakao Y, Ashikari T, 2006. Pure and mixed diploid than haploid yeast populations, Nature, genetic lines of Saccharomyces bayanus 302: 495-500 and Saccharomyces pastorianus and their Pasteur L, 1860. Mémoire sur la fermentation contribution to the lager brewing strain genome, alcoolique, Imprimerie de Mallet-Bachelier, FR. Appl Environ Microbiol, 72: 3968-3974 Pedersen M, 1986. DNA sequence Reed G, Nagodawithana T, 1991. Yeast polymorphisms in the genus Saccharomyces Technology, Van Nostrand Reinhold, NY IV. Homeologous chromosomes III of Rhymes M, Smart K, 2000. The relationship Saccharomyces bayanus, S. carlsbergensis and S. between flocculation and cell surface physical uvarum, Carlsber Res Commun, 51: 185-202 properties in a FLO1 ale yeast. Brewing yeast Pengelly R Wheals A, 2013. Raid identification fermentation and performance, Blackwelll of Saccharomyces eubayanus and its hybrids. Science, Oxfoed, England. FEMS Yeast Res, 13: 156-161 Salema-Oom M, Pinto VV, Goncalves Peris D, Belloch C, Lopandić K, Álvarez-Pérez P, Spencer-Martins I, 2005. Maltotriose JM, Querol A, Barrio E, 2012a. The molecular utilization by industrial Saccharomyces strains: characterization of new types of Saccharomyces characterization of a new member of the cerevisiae x S. kudriavzevii hybrid yeasts unveils α-glucoside transporter family, Appl Environ a high genetic diversity, Yeast, 29: 81-91 Microb, 71: 5044-5049 Peris D, Lopes CA, Arias A, Barrio E, 2012b. Sampaio JP, Gonçalves P, 2008. Natural Reconstruction of the evolutionary history population of Saccharomyces kudriavzevii in of Saccharomyces cerevisiae x S. kudriavzevii Portugal are associated with oak bark and are hybrids based on multilocus sequence analysis, sympatric with S. cerevisiae and S. paradoxus, Plos One, 9: 1-14 Appl Environ Microb, 74(7): 2144-2152 Peris D, Sylvester K, Libkind D, Gonçalves P, Sauer B, 1987. Functional expression of the Sampaio JP, Alexander WG,Hittinger CT, 2014. cre-lox site-specific recombination system in the Population structure and reticulate evolution of yeast Saccharomyces cerevisiae, Mol Cell Biol, 7: Saccharomyces eubayanus and its lager-brewing 2087-2096 hybrids, Mol Ecol, 23(8): 2031-45 Scannel D, Zill O, Rokas A, Payen C, Dunham Perry C, Meaden P, 1988. Properties of a M, Eisen M, Rine J, Johnston M, Hittinger C, genetically-engineered dextrin-fermenting strain 2011. The awesome power of yeast evolutionary of brewers’ yeast, J Inst Brew, 94: 64-67 genetics: new genome sequences and strain resources for the Saccharomyces sensu stricto Petersen R, Nilsson-Tillgren T, Piškur J, 1999. genu, G3 (Bethesda), 1: 11-25 Karyotypes of Saccharomyces sensu lato species, Int J Syst Bacteriol, 49: 1925-1931 Scherer S, Davis R, 1979. Replacement of chromosome segments with altered DNA Polaina J, 2002. Brewer’s yeast: genetics sequences constructed in vitro, Proc Natl Acad and biotechnology. Applied Mycology and Sci USA, 76, 4951-4955 Biotechnology. Vol.2 Agriculture and food production. Elsevier Science BV. Schillberg S, Zimmermann M, Emeis C, 1991. Analysis of hybrids obtained by rare-mating Price C, Fuson G, Phaff H, 1978. Genome of Saccharomyces strains, Appl Microbiol comparison in yeast systematics: delimitation Biotechnol, 35: 242-246 of species within the genera Schwanniomyces, Saccharomyces, Debaryomyces, and Pichia, Microbiol Rev, 42:161-193.

48 Chapter 1

Sheehan C, Weiss A, Newsom I, Flint V, O’Donnell Tamai Y, Momma T, Yoshimoto H, Kaneko Y, D, 1991. Brewing yeast identification and 1998. Co-existence of two types of chromosome chromosome analysis using high resolution CHEF in the bottom fermenting yeast, Saccharomyces gel electrophoresis, J. Inst. Brew, 97: 163-167 pastorianus, Yeast, 14: 923-933 1 Seidler R, Mandel M, 1971. Quantitative aspects Taylor J, Dlamini B, Kruger J, 2013. The science of deoxyribonucleic acid renaturation: base of the tropical cereals sorghum, maize and rice composition, state of chromosome replication, in relation to lager beer brewing, J Inst Brew, and polynucleotide homologies. Journal of 119(1-2):1-4 Bacteriology, 106: 608-614 journal abbreviation Tirosh I, Reikhav S, Levy AA, Barkai N, 2009. A Sniegowski P, Dombrowski P, Fingerman E, 2002. yeast hybrid provides insight into the evolution Saccharomyces cerevisiae and Saccharomyces of gene expression regulation, Science, 324: paradoxus coexist in a natural woodland site 659-662 in North America and display different levels Tornai-Lehoczki J, Péter G, Dlauchy D, Deák T, of reproductive isolation from European 1996. Some remarks on “a taxonomic key for conspecifics,FEMS Yeast Res, 1: 299-306 the genus Saccharomyces” (Vaughan martini and Spencer J, Spencer D, 1996. Rare-mating and Martini 1993), Antonie van Leeuwenhoek, 69: cytoduction in Saccharomyces cerevisiae, 229-233 Methods MolBiol, 53: 39-44 Torres E, Sokolsky T, Tucker C, Chan L, Špírek M, Yang J, Groth C, Petersen R, Langkjær Boselli M, Dunham M, Amon A, 2007. Effects R, Naumova E, Sulo P, Naumov G, Piškur J, of aneuploidy on cellular physiology and cell 2003. High-rate evolution of Saccharomyces division in haploid yeast, Science, 317: 916-923 sensu lato chromosomes, FEMS Yeast Res., 3: Tsuboi M, Takahashi T, 1988. Genetic analysis 363-373 of the non-sporulating phenotype of brewer’s Stewart G, 1981. The genetic manipulation of yeasts, Ferment Technol, 66: 605-613 industrial yeast strains, Can J Microbiol, 27: Tsuchiya T, Fukazawa Y, Taguchi M, Nakase T, 973-990 Shinoda T, 1974. Serological aspects on yeast Stewart G, Russell I, 1986. One hundred years of classification,Mycopathol. Mycol. Appl., 53: yeast research and development in the brewing 77-91 industry, J Inst.Brew, 92: 537 – 558 Usher J, Bond U, 2009. Recombination between Stewart G, Russell I, 1993. Fermentation-the homoeologous chromosomes of lager yeasts “black box” of the brewing process, MBAA TQ, leads to loss of function of the hybrid GPH1 30:159-168 gene, Appl Environ Microbiol, 75: 4573-4579 Stewart G, 2006. Studies on the uptake and van der Walt J, 1965. Saccharomyces inusitasus metabolism of wort sugars during brewing sp.n., Antonie van Leeuwenhoek, 31: 277-281 fermentation, MBAA TQ, 43: 265 – 269 van der Walt J, 1970. Criteria and methods used Strack L, Stahl U, 2010. Brewer’s yeast in classification. J Lodder (ed), The Yeasts – a modification: effects both on diacetyl production taxonomic study. North-Holland Publishing Co., and alcohol formation, Brewing Science, 63: 6-9 Amsterdam. Tadami H, Shikata-Miyoshi M, Ogata T, 2014. van Hamersveld E, 1996. Floccculation of Aneuploidy, copy number variation and unique brewers’ yeast – quantification, modelling and chromosomal structures in bottom-fermenting control, PhD Thesis, TU Delft yeast revealed by array-CGH, J. Inst. Brew., 120: 27-37

49 General Introduction

Vaughan Martini A, Kurtzman C, 1985. Walther A, Hesselbart A, Wendland J, Deoxyribonucleic acid relatedness among species 2014. Genome sequence of Saccharomyces of the genus Saccharomyces sensu stricto, Int J carlsbergenesis, the world’s first pure culture Syst Bacteriol, 35: 508-511 lager yeast, G3, 4, 783-793 Vaughan Martini A, Martini A, 1987. Three Wang S, Bai F, 2008. Saccharomyces arboricolus newly delimited species of Saccharomyces sensu sp. nov., a yeast species from tree bark, Int J Syst stricto, Antonie van Leeuwenhoek, 53: 77-84 Evol Microbiol, 58: 510-514 Vaughan Martini A, Martini A, Cardinali Wang Z, Wang J, Liu X, He X, Zhang B, 2009. G, 1993. Electrophoretic karyotyping as a Recombinant industrial brewing yeast strains taxonomic tool in the genus saccharomyces, with ADH2 interruption using self-cloning Antonie van Leeuwenhoek, 63: 145-156 GSH1+CUP1 cassette, FEMS Yeast Res, 9: 574- Verbelen P, Dekoninck T, Saerens S, van Mulders 581 S, Thevelein J, Delvaux F, 2009. Impact of Yadav N, McDevitt R, Benard S, Falco S, 1986. pitching rate on yeast fermentation performance Single aminoacid substitutions in the enzyme and beer flavor,Appl Microb Cell Physiol, 82: acetolactate synthase confer resistance to the 155-167 herbicide sulfometuron methyl, Proc Natl Acad Verstrepen K, van Laere S, Vanderhaegen Sci USA, 83: 4418-4422 B, Derdelinckx G, Dufour J-P, Pretorius Yamada Y, Mikata K, Banno I, 1993. I, Winderickx J, Thevelein J, Delvaux F, Reidentification of 121 strains of the genus 2003. Expression levels of the yeast alcohol Saccahromyces, Bull JFCC, 9: 95-119 acetyltransferase genes ATF1, Lg-ATF1, and Yarrow D, Nakase T, 1975. DNA base ATF2 control the formation of a broad range composition of species of the genus of volatile esters, Appl Environ Microbiol, 69: Saccharomyces. Antonie van Leeuwenhoek, 41: 5228-5237 81-88 Vidgren V, Ruohonen L, Londesborough J, 2005. Yamagishi H, Ohnuki S, Nogami S, Ogata Characterization and functional analysis of the T, Ohya Y, 2010. Role of bottom-fermenting MAL and MPH loci for maltose utilization in brewer’s yeast KEX2 in high temperature some ale and lager yeast strains, Appl Environ resistance and poor proliferation at low Microbiol, 71: 7846-7857 temperatures, J Gen Appl Microbiol, 56: 297- Vidgren V, Multanen J, Ruohonen L, 312 Londesborough J, 2010. The temperature Yona A, Manor Y, Herbst R, Romano G, dependence of maltose transport in ale and lager Mitchell A, Kupiec M, Pilpel Y, Dahan O, strains of brewer’s yeast, FEMS YR, 10: 402-411 2012. Chromosomal duplication is a transient Villanueba K, Goossens E, Masschelein C, 1990. evolutionary solution to tress, PNAS, 109: Subthreshold vicinal diketone levels in lager 21010-15 brewing yeast fermentations by means of ILV5 Zhang Y, Wang Z, He X, Liu N, Zhang B, 2008. gene amplification, J Am Soc Brew Chem, 48: New industrial brewing yeast strains with ILV2 111-114 disruption and LSD1 expression, Int J Food Walker G, 1998. Yeast physiology and Microbiol, 123: 18-24 biotechnology. John Wiley & Sons, NY

50 Chapter 1

1

51

CHAPTER 2 Chromosomal copy number variation in Saccharomyces pastorianus is evidence for extensive genome dynamics in industrial lager brewing strains

Van den Broek M Bolat I Nijkamp J Ramos E Luttik M Koopman F Pronk JT Geertman JM de Ridder D Daran JM

Appl Environ Microbiol 81:6253– 6267. doi:10.1128/AEM.01263-15 Copy number variation in Saccharomyces pastorianus

ABSTRACT

Lager brewing strains of Saccharomyces pastorianus are natural interspecific hybrids originating from spontaneous hybridization of Saccharomyces cerevisiae and Saccharomyces eubayanus. Over the past 500 years, S. pastorianus has been domesticated to become one of the most important industrial microorganisms. Production of lager-type beers requires a set of essential phenotypes, including the ability to ferment maltose and maltotriose at low temperature, production of flavors and aromas and the ability to flocculate. Understanding the molecular basis of complex brewing-related phenotypic traits is a prerequisite for rational strain improvement. While genome sequences have been reported, the variability and dynamics of S. pastorianus genomes has not been investigated in detail. Here, using deep sequencing and chromosome copy number analysis, we show that S. pastorianus strain CBS1483 exhibits extensive aneuploidy. This was confirmed by quantitative PCR and by flow cytometry. As a direct consequence of this aneuploidy, a massive number of sequence variants was identified, leading to at least 1800 additional protein variants in S. pastorianus CBS1483. Analysis of eight additional S. pastorianus strains revealed that the previously defined ‘Group I’ strains showed comparable karyotypes, while ‘Group II’ strains showed large inter-strain karyotypic variability. Comparison of three strains with near-identical genome sequences revealed substantial chromosome copy number variation, which may contribute to strain-specific phenotypic traits. The observed variability of lager yeast genomes demonstrates that systematic linking of genotype to phenotype requires a three-dimensional genome analysis, encompassing physical chromosomal structures, copy number of individual chromosomes or chromosomal regions, and allelic variation of copies of individual genes.

54 Chapter 2

INTRODUCTION

Lager brewing yeasts have been inadvertently selected and evolved over many hundreds of years by brewers to survive under the stressful conditions that they encounter during wort fermentation. These include high osmotic and hydrostatic pressures, low temperatures, high CO and high ethanol concentrations (Gibson et al., 2007). More recently, strain selection has 2 2 focused on specific flavor profiles, which can be used as ID bar-codes for specific beverages and for the yeast strain used to ferment it (Steensels et al., 2014). However, understanding of the genetic and molecular basis for brewing yeast performance is very incomplete. A deeper knowledge and understanding of the genome content and structure of S. pastorianus strains is essential for further, knowledge based strain improvement programs.

Beer brewing yeasts include ale yeasts, which are taxonomically classified as Saccharomyces cerevisiae (Vaughan-Martini et al., 1998) and lager yeasts. The latter were long known as S. carlsbergensis but were recently renamed S. pastorianus (Bond 2009; Rainieri et al., 2006). Over the past decades, it has become generally accepted that S. pastorianus strains are natural, interspecific hybrids of S. cerevisiae and a non-cerevisiae Saccharomyces species (Caesar et al., 2007; Casaregola et al., 2001; Rainieri et al., 2006; Tamai et al., 1998; Vaughan-Martini et al., 1985; Yamagishi et al., 1999). Publication of the first complete genome sequence of S. pastorianus Weihenstephan 34/70 (WS34/70) (Nakao et al., 2009) and the subsequent release of three additional S. pastorianus sequences (Hewitt et al., 2014; Walther et al., 2014) confirmed that S. pastorianus is a hybrid of S. cerevisiae and a species closely related to S. bayanus. This result was further refined by the recent discovery and genome analysis of S. eubayanus, whose genome sequence shows a 99.5% identity with that of the non-cerevisiae part of the S. pastorianus genome (Libkind et al., 2011). Originally isolated from oak trees in Nothofagus (Southern beech) forests and stromata of Cyttaria hariotii (an ascomycetous parasite of Nothofagus spp.) in Patagonia, S. eubayanus has recently also been isolated in North America (Peris et al., 2014) and Asia (Bing et al., 2014). Phylogenetic analysis demonstrated that the Tibetan population of the S. eubayanus lineage was more closely related with S. pastorianus than the type strain CBS12357 (PYCC 6148) from Patagonia (Bing et al., 2014).

The distribution of S. eubayanus is consistent with an earlier hypothesis that lager yeast strains may have originated from multiple, separate locations (Dunn et al., 2008) and, thus, from distinct hybridization events. Indeed, lager strains of S. pastorianus can be divided in two populations based on their DNA content, as estimated by array-based comparative genome hybridization (array-CGH) (Bond et al., 2004; Dunn et al., 2008). The hybrid ancestor of Group I, which includes Saaz-type bottom-fermenting yeast, was proposed be an allo-diploid strain originating from the fusion of a haploid ale S. cerevisiae yeast cell with a haploid S. eubayanus. This initial model was however shown not to be completely correct, as deep sequencing analysis revealed that S. carlsbergensis was basically triploid with a diploid S. eubayanus and haploid S. cerevisiae genome content (Walther et al., 2014). Group

55 Copy number variation in Saccharomyces pastorianus

II strains, which include Frohberg-type bottom-fermenting yeasts, were proposed to have arisen from the fusion of a homozygous diploid S. cerevisiae with a haploid S. eubayanus (Dunn et al., 2008).

The hybrid genomes of brewing yeast strains are not simple juxtapositions of the two sub- genomes. Group I and II strain genomes both harbor numerous specific features, including partial or complete loss of chromosomes, inter-chromosomal translocations between S. cerevisiae or/and eubayanus chromosomes and introgression of sequences from one subgenome into the other. Additionally, chromosomal rearrangements, some resulting in copy-number changes, are likely to have contributed to the series of events that shaped brewing yeast genomes in response to the selective pressures to which they were exposed in man-made brewing environments (Bond 2009; Fischer et al., 2000; Usher et al., 2009).

Availability of full genome sequences for both Group I (S. carlsbergensis CBS1513) (Hewitt et al., 2014; Walther et al., 2014) and Group II (S. pastorianus WS34/70) (Nakao et al., 2009) strains enabled a closer examination of the difference between these two types. Despite their supposedly different origin, Groups I and II strains showed breakpoint reuse in two genes (HSP82 and KEM1) (Hewitt et al., 2014), suggesting that these recurrent rearrangements in independent lineages may have provided advantageous evolutionary innovations in brewing environments (Dunn et al., 2013). While it was established that Group I and II strains harbored different numbers of structurally distinct chromosome types (29 and 35, respectively), recent reports did not unequivocally quantify chromosome counts and aneuploidy. This is illustrated by the fact that two studies identified different chromosome copy numbers and distributions for the same S. pastorianus (syn. carlsbergensis) strain CBS1513 (31 versus 47 chromosomes) (Hewitt et al., 2014; Walther et al., 2014).

The goal of the present study was to quantitatively analyze the diversity of S. pastorianus genomes by i) determining and establishing chromosome copy number in different lager strains of S. pastorianus, ii) analyzing chromosome copy number variation in different brewing strains and iii) examining the relation between variation in chromosome number and strain phenotype. To this end, we sequenced and de novo assembled the genome of S. pastorianus strain CBS1483, which was then used as a model for setting up the methodology for determining chromosome copy number. In a second step, the genome sequence of CBS1483 was compared to five newly sequenced lager strain genomes and to three already published ones. Finally, we compared two brewing-related performance traits (diacetyl production profiles and flocculation characteristics) of three karyotype variants of the strain WS34/70 that exhibited differences in chromosome copy numbers.

56 Chapter 2

MATERIALS AND METHODS

Strains and maintenance The S. pastorianus strains used in this study are listed in Table 1. Stock cultures were grown at 20°C in shake flasks on complex YP (1% yeast extract, 2% peptone) medium supplemented with maltose (20 g·l-1). When stationary phase was reached sterile glycerol was added to 2 30% (vol/vol) and 2 mL aliquots were stored in sterile vials at -80°C.

Table 1. Strains used in this study

Strains Description Reference S. pastorianus CBS1483 Group II (July 1927) (Dunn et al., 2008) S. pastorianus A1 Group II; WS34/70 isolate (Bolat et al., 2008) S. pastorianus A1+B11 Group II; WS34/70 isolate (Bolat et al., 2008) S. pastorianus A2 Group II; WS34/70 isolate (Bolat et al., 2008) S. pastorianus Spy1 Group II; Industrial strain descendant of CBS1483 S. pastorianus Spy2 Group II; Mexican industrial strain S. pastorianus CBS1260 Group II (March 1937) (Hewitt et al., 2014; Walther et al., 2014) S. pastorianus var carlsbergensis Group I (Oct 1947) (Hewitt et al., 2014; CBS1513 Walther et al., 2014) S. pastorianus var monacensis Group I (1908) (Hewitt et al., 2014; CBS1503 Walther et al., 2014) S. cerevisiae MATa SUC2 MAL2-8c (Entian et al., 2007; CEN.PK113-7D Nijkamp et al., 2012) S. cerevisiae CEN.PK112 MATa/MATα Euroscarf S. cerevisiae FRY153 3n

Cultivation and media Weihenstephan 34/70 (Hefezentrum Weihenstphan (TU-Munich), 2005) isolates A1, A1+B11 and A2 and the control strain CEN.PK113-7D were cultured in defined medium -1 -1 -1 (WMM) containing 10 g·l (NH4)2SO4, 6 g·l KH2PO4, 0.25 g·l MgSO4.7H2O. Vitamins, -1 trace elements (with reduced CaCl2, 2H2O concentration to 0.04 mg·l ) and the anaerobic growth factors ergosterol (10 mg·l−1) and Tween 80 (420 mg·l−1) were added (Verduyn et al., 1992). The WMM medium was supplemented with a complex mixture of sugars from corn syrup containing 9.5 g·l-1 maltotriose, 27 g·l-1 maltose, 4.5 g·l-1 glucose, 0.75 g·l-1 fructose as consumable sugars. From stock, the strains were cultivated in 60 ml septa flasks with gas outlet containing 50 ml defined medium. Main cultures were performed in 500 ml septa bottle flasks with gas outlet, containing either 100 ml (for flocculation assay) or 200ml (for vicinal diketone measurement). Cultures were performed in Infors multitron standard bioreactors (Infors HT, Bottmingen, Switzerland) set at 160 RPM with a 50 mm throw at 20 °C.

57 Copy number variation in Saccharomyces pastorianus

Genome sequencing, assembly and analysis Genome sequencing was performed using Illumina HiSeq2000 (Illumina, San Diego, CA) at Baseclear (Leiden, The Netherlands). Genomic DNA of the CBS1483 strain was used to obtain four paired libraries with different insert-sizes. Two 100-cycle paired-end libraries with an insert of 500 and 180 bp, were sequenced. For the latter, the overlapping read pairs were merged into single longer pseudo-reads. Another two 50-cycle mate pair libraries with a 3 and 8 kb insert size were sequenced. The combined libraries comprised more than 100 10+6 reads, representing ~7Gb resulting in ~270x coverage (Table 2).

Table 2. Description of the libraries used in this study

Libraries (Pseudo) Reads 500bp paired-end 3kb mate-pair 8kb mate-pair Number of reads (10+6) 9.26 17.2 40.1 36.3 Read length (bp) 141 100 50 50 Total sequence (Mb) 1305.9 1718.9 2004.6 1816.8 Coverage (-fold) 52.2 68.8 80.2 72.7

The genome assembly was performed on the pseudo-reads library using the GSAssembler 2.6, also known as Newbler (454 Life Sciences, Brandford, CT), using default settings. Further, scaffolding was performed on the assembled contigs using SSPACE, version 2.0 (Boetzer et al., 2011). SSPACE maps the paired 500 bp insert size library reads on the assembled contigs using Bowtie, version 0.12.5 (Langmead et al., 2009) as mapping tool. Based on the paired link information between the different contigs the orientation and distance between consecutive contigs is determined and merged into scaffolds. In an iterative process first the 3- and following the 8-kb mate paired library were mapped to the scaffolds formed in the previous round using SSPACE. During the scaffolding process, gaps were introduced to preserve the distance between two placed contigs. In the next step, Gapfiller, version 1 (Boetzer et al., 2012), was applied and closed 549 out of 812 gaps, solving almost 11000 unknown bases. The scaffolds were annotated by the “MAKER2” pipeline (Holt et al., 2011)

The chromosomal copy number variation was estimated by applying Magnolya (Nijkamp et al., 2012). When there was no prior knowledge of the ploidy the gamma settings was set to ‘‘None”. Magnolya uses a Poisson mixture model to estimate the integer copy number of an assembled contig with a minimum length of 500 bp.

To predict chromosome copy number Magnolya can use two assembler type Newbler (454 Life Sciences) suitable for long reads (>= 100 bp) and ABySS (Simpson et al., 2009) for short reads which both only accept fasta or fastq files in base space format. To be able to process the publically available data of strains CBS1503, CBS1513, CBS1260 (http://www. ebi.ac.uk/ena/data/view/PRJEB4654), the original Solid4 color space data were converted to fastq. The color space data were first mapped to the CBS1483 assembled scaffolds using Bowtie (version 0.12.8) (Langmead et al., 2009). The mapped reads were extracted from the

58 Chapter 2 sequence alignment map and converted into a base space fastq format using SamToFastq. jar from picard-tools (version 1.113) (http://picard.sourceforge.net). In a final step, sequence reads were trimmed and filtered based on quality score using Trimmomatic (Bolger et al., 2014). The converted data suitable for chromosomal copy number with Magnolya using ABySS (version 1.3.7) as assembler with k-mer of 29. 2 Determination of heterozygous positions between alleles was performed using the Genome Analysis Toolkit (GATK) Unified Genotyper (McKenna et al., 2010). The pseudo-reads and paired-end library were mapped to the assembled contigs using BWA (Li et al., 2010). The alignment result file (BAM) was sorted in categories (from n=1 to n=5) based on the contig copy number previously predicted by Magnolya using an in-house script. The five generated ploidy groups were individually analyzed for SNP, by setting the “ploidy” and “max_alternate_alleles” parameters to the corresponding copy number.

Copy number variation analysis PCR analysis To determine the copy number of specific chromosomes or parts of chromosomes in CBS1483, quantitative PCR was employed on genomic DNA, using pUG6 DNA (Guldener et al., 1996) as an internal standard. Genomic DNA of lager brewing strain and the DNA corresponding to plasmid pUG6 previously cut with restriction enzyme HindIII were mixed in a 1:1molar ratio. The primers (Table 3) were designed to exhibited a similar Tm ranged between 59.5 and 61.5 oC. The efficiency of the qPCR assays for the target genes and the control amplicon was calculated using the formula: E = 10(−1/m)-1, where m is the slope of the function derived from the Ct versus log-dilution plot (0.02–200 ng input DNA) of DNA template. Only primers exhibiting an efficiency (E) ranging between 1.92 and 1.94 were used for the quantification reactions (Ramakerset al., 2003). The PCR reactions were performed using SYBR Green PCR kit (Qiagen GmbH) on a Rotor GeneQ instrument (Qiagen GmbH). Each gene quantification was based on a minimum of eight independent replicates. The raw data analysis was performed using the LinRegPCR software (version 11.0) (Ruijter et al., 2009). The relative copy number (CNR) of target gene was calculated using the formula: DCt(target1-kanMX) DCt(target2-kanMX) CNR= (Etarget1) /(Etarget1) .

DNA content determination by flow cytometric analysis Samples of culture broth (equivalent to circa 10+7 cells) were taken from mid-exponential shake-flask cultures on WMM medium and centrifuged (5 min, 4700 x g). The pellet was washed once with cold demi-water (Porro et al., 2003), vortexed briefly, centrifuged again (5 min, 4700 x g) and suspended in 800 µL 70% ethanol while vortexing. After addition of another 800 µL 70% ethanol, fixed cells were stored at 4°C until further staining and analysis. Staining of the cells with SYTOX® Green Nucleic Acid Stain was performed as described in (Haase et al., 2002). Samples were analyzed on a BD Accuri C6 flow cytometer (BD Biosciences, Breda, The Netherlands) equipped with a 488 nm laser. The fluorescence intensity (DNA content) was represented using BD Accuri C6 software (BD Biosciences).

59 Copy number variation in Saccharomyces pastorianus

Analysis Optical density OD660 was measured using a Biochrom Libra S60 (Biochrom, Cambridge, United Kingdom). Sugars and ethanol in culture supernatants were analyzed using HPLC, containing an Aminex HPX-87H ion exchange column (BioRad, Veenendaal, The Netherlands) operating at 60 °C with 5 mM H2SO4 as mobile phase at a flow rate of 0.6 ml·min-1. Vicinal diketones were analyzed using static headspace gas chromatography (GC). 5 ml of culture supernatant was heated to 65 °C for 30 minutes prior to injection using a CTC Combi Pal headspace auto-injector (CTC Analytics AG, Zwingen, Switzerland). Samples were analyzed using an 7890A Agilent GC (Agilent, Amstelveen, The Netherlands) with an electron capture detector on an CP-Sil 8 CB (Agilent, Amstelveen, The Netherlands) (50 m x 530 μm x 1 μm) capillary column. The split ratio was 1:1 with a split flow of 8 ml nitrogen per minute. The injector temperature was set at 120 °C and an oven temperature profile of 35 °C for 3 min followed by an increase of 10 °C·min-1 to 95 °C was used. The ECD detector temperature was set at 150 °C with a make-up flow of 10 ml nitrogen per minute.

Flocculation assay Flocculation ability was analyzed using a modified Helm’s test (Bendiak et al., 1996; Helm et al., 2014). Culture samples were spun down and washed with 2 ml of 50 mM EDTA followed by washing with 30 ml of cold dH2O. After washing, the cells were resuspended in at least 25 ml flocculation buffer (50 mM sodium acetate buffer pH 4,5) to obtain a OD660 of 2.5. Of each strain and a water control, 1.9 ml were distributed in 12 wells of a 2 ml 96 wells deep-well plate (Greiner Bio-one, Alphen a/d Rijn, The Netherlands) of which 6 wells contained 100 μl of dH2O and 6 wells contained 100 μl of a 20% CaCl2 stock solution. After mixing by pipetting, the cells were incubated for 5 minutes followed by 45 minutes vortexing and 5 minutes of sedimentation. From 1 mm below the meniscus 100 µl was drawn and pipetted in a 96 wells microtiter plate (Greiner Bio-one, Alphen a/d Rijn, The Netherlands) containing 100 µl of 50 mM EDTA. OD660 was measured in an Infinite® M200 PRO (Tecan Group Ltd., Männedorf, Switserland) after 10 seconds of vortexing. The flocculation coefficient was calculated on the blank corrected average of the six replicates. Using the following equation; flocculation coefficient =(1-B/A)x100%, where A is the average OD660 of the cells with CaCl2 and B is the average OD660 of the cells with dH2O.

60 Chapter 2

Table 3. Primers used in this study

Primers Sequence 5’→3’ Chromosome copy number CBS1483_Sb1_Contig488_ADE1-Fw GTACTTGGCCCTGGTTCTG CNV SeubADE1 CBS1483_Sb1_Contig488_ADE1-Rv CCTACAAGCTAGGCGAATCC CBS1483_Sc1_Contig530_ADE1-Fw CAACGGCCTTAACGACATAG CNV ScADE1 CBS1483_Sc1_Contig530_ADE1-Rv AGCCAGTTCTGCCACTCTAC CBS1483_Sb3_Contig460_NFS1-Fw CAACGGCCTTAACGACATAG 2 CNV SeubNFS1 CBS1483_Sb3_Contig460_NFS1-Rv TGGGTAAGGATGATGCTTTG CBS1483_Sc3_Contig345_ABP1-Fw GCCCAAAGACACATAATTGC CNV ScABP1 CBS1483_Sc3_Contig345_ABP1-Rv TGATTACGATGCTGCAGAAG CBS1483_Sc7_Contig382_EMP24-Fw CGCCCATAATGTCCTTCTTC CNV ScEMP24 CBS1483_Sc7_Contig382_EMP24-Rv GTCAGCTGGCTACTGGATTG CBS1483_Sc7_Contig260_MEP1-Fw AGCCTCATACCGGGTATGTAG CNV ScMEP1 CBS1483_Sc7_Contig260_MEP1-Rv TTGGTATGGATGGCACTACAG CBS1483_Sb7_Contig350_SHY1-Fw AGCAATTTGTCGCCAGTATC CNV SeubSHY1 CBS1483_Sb7_Contig350_SHY1-Rv AGCTGACCTATGACCCGATAC CBS1483_Sb8_Contig146_DUR3-Fw CACTTGTTTTATGCCATCGCG CNV SeubDUR3 CBS1483_Sb8_Contig146_DUR3-Rv CGACACAAGTAGGGCCAGTTC CBS1483_Sc8_Contig370_DUR3-Fw GATGTTTTAGGCTCACCGGG CNV ScDUR3 CBS1483_Sc8_Contig370_DUR3-Rv TATCAAGGAAAACGGTGCCG CBS1483_Sb8_Contig446_HIS3-Fw TGGCTTCTCTTATGGCAACC CNV SeubHIS3 CBS1483_Sb8_Contig446_HIS3-Rv CATCCTCTAGAGAGAACGCT CBS1483_Sb8_Contig445_MDM20-Fw GTGTTGGATGAACTATTCCC CNV SeubMDM20 CBS1483_Sb8_Contig445_MDM20-Rv GAAGCAGGTTATTGAACTGC CBS1483_Sc8_Contig028_MDM20-Fw ATGCAAAACTGGGGCAATTGC CNV ScMDM20 CBS1483_Sc8_Contig028_MDM20-Rv AACTCTAGCAAGAAGTTGTGC Verification of the sequence assembly S23-CBS1483_Sb4-Sb2_Scf1_ SNQ2-Fw CGTGTACTACCTCTTCCACGTGAGACAGAGTTCTC Verification Scf1 S23-CBS1483_Sb4-Sb2_Scf1_ PDX3-Rv GCCTTAACGTTCGTGTGTGGGTGTGG S24-CBS1483_Sc7-Sb7_Scf2_MPT5-Fw GGATAATTTTGGTAATTATGCGTTACAAACGC Verification Scf2 S24-CBS1483_Sc7-Sb7_Scf2_SUA5-Rv2 GTTGGAGAGAAGTGGATTGAGAGC S24-CBS1483_Sc7-Sb7_Scf2_MPT5-Fw GGATAATTTTGGTAATTATGCGTTACAAACGC Verification Scf2 S24-CBS1483_Sc7-Sb7_Scf2_SUA5-Rv2 GTTGGAGAGAAGTGGATTGAGAGC S25-CBS1483_Sc7-Sb7_Scf2_PXR1-Fw GCTGTTGATCATTTCATAGCGCAAAAGAAC Verification Scf2 S25-CBS1483_Sc7-Sb7_Scf2_IMA1-Rv GTAGATGCCTCTTCCAGAACATTG S30-CBS1483_Sb13-Sc13_Scf12_LIP1-Fw GGACAGAAACTTCAACTTGACCTCGG Verification Scf12 S30-CBS1483_Sb13-Sc13_Scf12_UBP15-Rv CGCTCCAATTCTTTCAAAGGCACAGAC S40-CBS1483_Sc16-Sb16_Scf15_KRE6-Fw CGGTGAGTACGGTGGCTACTTTC Connection Scf15-Scf39 S40-CBS1483_Sc16-Sb16_Scf39_SGV1-Rv CCACTGGGACCACAATCAAGATAC S38-CBS1483_Sc12_Scf19_POM33-Fw CTTATAATGCAATTAAAATGAGGTTGGTCC Connection Scf38-Scf8 S38-CBS1483_Sc12_Scf33_PAU17-Fw CATCTCCAGCGCTCTATCTGC S42-CBS1483_Sb16_Scf39_SKI3-Fw GTCCAGCCTGATAGCGGATAAAC Verification Scf39 S42-CBS1483_Sc16_Scf39_HPA2-Rv CGTGTATTGGTGTACGGATGAGTC S31-CBS1483_Sc10-Sb10_Scf14_POL31-Fw CCGTCATTTCAGTATCTTTTTCCCCTGC Verification Scf14 S31-CBS1483_Sc10-Sb10_Scf14_GPI14-Rv CGGAAATTGACGTTAGCAAGGGTG S34-CBS1483_Sc15-Sc11_Scf21_MCH2-Fw GAGATGAGATTGCTGTGCGTGAAG Verification Scf21 S34-CBS1483_Sc15-Sc11_Scf21_RDR1-Rv CTCATTTAGTCGCGTTAGCACACC S28-CBS1483_Sb2-Sb4_Scf10_YPK3-Fw GTATGATGTTTGTGACAAATGTGGCGAGC Verification Scf10 S28-CBS1483_Sb2-Sb4_Scf10_KCS1-Rv GCCATTATCATGGCTAACCATCAGATGT S32-CBS1483_Sb8-Sb15_Scf16_DIA4-Fw CTCCAAACATCAATTGTCGAGGAACTAGG Verification Scf16 S32-CBS1483_Sb8-Sb15_Scf16_HST3-Rv CCTTCTCTGCATGATGAATCCCAGC S27-CBS1483_Sb15-Sb8_Scf6_AUS1-Fw GCGCATTCTTTTGCTGTGGGTTCAAC Verification Scf6 S27-CBS1483_Sb15-Sb8_Scf6_AUS1-Rv CATTCTTTTGCTGTGGGTTCAACAATG S29-CBS1483_Sc7-Sb7_Scf2_MPT5-Fw2 GTGCTAATCAGGGATAATTTTGGTAATTATGC Connection Scf2-Scf3 S29-CBS1483_Sc7-Sb7_Scf3_SPO74-Rv2 CTTCCATTAGGGATTTTAGAAATCCACTTTTG S36-CBS1483_Sb3-Sc3_Scf27_TAF2-Fw CTTCACGGGCTCGATGACTATAAGG Verification Scf27 S36-CBS1483_Sb3-Sc3_Scf27_RRP43-Rv CAATATTATATCCGTAGACGGCCTTTAATAGTG S47-CBS1483_Sb16_Scf8_YAR1-Rv GTAGAGTTGGACCCGCTTTCTTTG Connection Scf8-Scf38 S47-CBS1483_Sc16_Scf38_IQG1-Fw CGGGAATAGTGTGAACCTTTCTGG S44-CBS1483_Sc7-Sc8_Scf25_ERV29-Fw CAGAGTATGGTCCTTGCGATGATG Connection Scf25-Scf2 S44-CBS1483_Sc7-Sb7_Scf2_IMA1-Rv CCAAGGCCGCCAAGAAGAAGAATAAG S38-CBS1483_Sc12_Scf19_POM33-Fw CTTATAATGCAATTAAAATGAGGTTGGTCC Verification Scf19 S33-CBS1483_Sc12_Scf19_ERP2-Rv CTTCATGCTCGTCAGTCAAAGTTTTTTC S49-CBS1483_Sb10-Sc10_Scf24_GPI14-Fw3 CGTGAACATGATTAAACAGCAAAG Verification Scf24 S49-CBS1483_Sb10-Sc10_Scf24_SAG1-Rv3 GAGATGTTGTCAATATCTGCCG S38-CBS1483_Sc12_Scf33_PAU17-Fw CATCTCCAGCGCTCTATCTGC Verification Scf33 S37-CBS1483_Sc12_Scf33_EFB1-Rv GACATCTAGAGTGACAATGGACTTAGCAG

61 Copy number variation in Saccharomyces pastorianus

RESULTS

Assembly and annotation of Saccharomyces pastorianus CBS1483 To enable and facilitate de novo assembly of the genome sequence of S. pastorianus Group II strain CBS1483, four different DNA libraries were prepared and sequenced using Illumina technology (Table 2). The first two libraries were 100-cycle paired-end libraries with insert sizes of 180 and 500 bp. The overlapping read pairs data of the 180-bp library were merged into single longer “pseudo” reads of c.140 bp. Subsequently the pseudo reads were used to perform the primary assembly which yielded 908 contigs of 500 bp or longer (Table 4). In a second assembly step, the 500-bp insert size library and the two 50-cycle mate pair libraries with 3- and 8-kb insert sizes, respectively, were iteratively used to further assemble the CBS1483 strain genome sequence. Scaffolding using SSPACE (V 2.0) (Boetzer et al., 2011) and Gapfiller(Boetzer et al., 2012) resulted in a reduction from 908 contigs to 59 scaffolds (Table 4). The cumulative size of the 41st longest scaffolds covered 22.1 Mb (which also included 196kb N’s introduced during the gap-filling step). Scaffold 41 (73 kb) was identified as mtDNA. The remaining 17 scaffolds had an average length of 4.7 kb and an aggregate size of 80 kb. Finally, the de novo assembled contigs and scaffolds were aligned to the genome sequences of S. cerevisiae S288C (www.yeastgenome.org/) and to S. eubayanus CBS12357 (Libkind et al., 2011).

Table 4. Assembly statistics of the S. pastorianus CBS1483 strain sequence

Assembly Number of contigs 908 Avg contig size (kbp) 24.2 N50 (kbp) 52.5 Largest contig (kbp) 315.2 Total sequence (Mbp) 21.97 Scaffolding and Gap filling Number of scaffolds (>= 500bp) 59 Avg scaffold size (kbp) 378.8 N50 (kbp) 750.1 Largest scaffold (kbp) 1464.6 Total sequence (Mbp) 22.3 (of which 196391 N’s in 263 gaps) Annotation ORF 10279

Since lager brewing strains have been reported to be highly aneuploid (Bond et al., 2004; Casaregola et al., 2001; Dunn et al., 2008), chromosome copy number is an essential feature of the overall genome architecture. To quantify chromosome copy number, the Magnolya algorithm was used, which applies a Poisson-mixture model for copy number estimation of contigs assembled from sequencing data without mapping reads to a reference

62 Chapter 2 genome (Nijkamp et al., 2012). Subsequently, by combining the de novo assembly and the chromosome copy number analysis, a genome map was established that comprised 35 different chromosomal structures and a total count of 68 chromosomes (Figure 1).

Consistent with three previously sequenced S. pastorianus genomes (Hewitt et al., 2014; Nakao et al., 2009; Walther et al., 2014), a large majority of the translocation events 2 identified in CBS1483 had been already described Table( 5).

Table 5. List of breakpoints identified in theSaccharomyces pastorianus CBS1483 genome.

Chromosome Location Scaffold Primer pair

Break point S. cerevisiae / S. cerevisiae ScXV-ScXI YOR381W/FRE3–YKL220C/ Scaffold21 S50 FRE2 Break points S. eubayanus / S. cerevisiae SeubIII-ScIII BUD5-MATα Scaffold27 S36 SeubVII-ScVII YGL173C/XNR1 Scaffold2 S24 SeubVII-ScVII ZUO1 Scaffold2 S25 SeubXIII-ScXIII YMR304W-YMR302C Scaffold12 S30 SeubXVI-ScXVI YPR159W/KRE6 Scaffold15 → 39 S40 YPR190C-YPR191W Scaffold39 S42 YPL240C/HSP82 Scaffold38 → 8 S47 Break points S. cerevisiae / S. eubayanus ScX-SeubX TDH2–ARS1016 Scaffold14 S31 ScXVI-SeubXVI YPL036C/PMA2 S34 ScXII-SeubXII See Figure S2 Break points found in S. pastorianus and already present in S. eubayanus [12] SeubII-SeubIV YBR030W/RKM3-YDR012W/ Scaffold10 S28 RPL4B SeubVIII-SeubXV YHR014W/SPO13-ARS807 Scaffold16 S32 SeubIV-SeubII YDR011W/SNQ2-YBR031W/ Scaffold1 S23 RPL4A SeubXV-SeubVIII YOR018W/ROD1-YHR015W/ S27 MIP6

63 Copy number variation in Saccharomyces pastorianus

Figure 1. Chromosomal structure and chromosome copy number of the lager brewing strain S. pastorianus CBS1483. The graph represent the ploidy prediction generated with the Magnolya algorithm (Nijkamp et al., 2012) of contigs that were de novo assembled by Newbler (www.454.com) and aligned to the reference S. cerevisiae S288C (Goffeau et al., 1996) and S. eubayanus CBS12357 (Libkind et al., 2011) genomes using NUCMER (MUMmer V3.21) (www.mummer.sourceforge.net). The blue and red blocks represent the contributions of the S. cerevisiae and S. eubayanus subgenomes, respectively, to the S. pastorianus CBS1483 genome organisation. The representation of the genomic organisation of CBS1483 takes into account the individual chromosome ploidy and the 35 chromosomal structures identified in this study.

64 Chapter 2

However, the genome of the strain CBS1483 revealed three specific chromosomal structures that could only be identified by integrating assembled scaffolds and chromosome copy number information. First, in contrast to other sequenced brewing strains (Hewitt et al., 2014; Nakao et al., 2009; Walther et al., 2014), CBS1483 retained neither a full S. cerevisiae-type (Sc) nor a S. eubayanus-type (Seub) CHRIII. Instead, it only harbored a chimeric chromosome consisting of the left part of SeubCHRIII (including the left arm, 2 centromere and the major part of the right arm) and the distal part of the right arm of ScCHRIII. Such a chimeric CHRIII was also found in S. pastorianus WS34/70, but in that strain it occurred together with a full ScCHRIII (Nakao et al., 2009). Second, the CBS1483 strain harbored a new, non-reciprocal introgression on CHRXII: a ca. 6-kb insertion of S. eubayanus DNA into ScCHRXII (Figures 1 and S1). Third, a SeubCHRI intrachromosomal region of 9.03-kb (including BUD14, ADE1, KIN3 and CDC15) (Figure S2) was found to be present at a higher copy number than the rest of the chromosome, which could indicate another introgression of SeubDNA in ScCHRI. However, local reassembly of the reads did not confirm such an introgression, instead, the extra copy was found to be flanked by sequences showing high similarity with Ty-delta sites (Figure S2), which precluded a precise localization of the duplicated region.

Strikingly, the chromosome copy number differed dramatically for the 35 S. pastorianus CBS 1483 chromosomal structures. Copy numbers from one, e.g. for ScCHRVI and VII, to a maximum of five copies for theSc CHRVIII were quantified Table( 6). The chromosomal copy number analysis results were consistent with the assembly results. For instance, two types of chromosome VII were identified in the assembly, a fullSc CHRVII and a chimeric Sc/SeubCHRVII. This was further supported by the Magnolya analysis which predicted a different copy integer for the central part of the ScCHRVII (#1) than for the right and left end of the same chromosome (#4). Moreover, it predicted a copy number of three for the SeubCHRVII. These data established that the S. pastorianus CBS1483 carries one copy of a full S. cerevisiae CHRVII and three copies of the chimeric version (Figure 1).

Overall, the genome of S. pastorianus CBS1483, was composed of 56% of S. cerevisiae, 34% of S. eubayanus and 10% of chimeric S. cerevisiae/S. eubayanus chromosomal DNA distributed over 68 chromosomes. This distribution did not match any of the already formulated assumptions regarding lager yeast genome ploidy, which often refers to these yeasts as allopolyploid or as hybrids formed of a haploid and a diploid genome. Our data showed a much more complex chromosomal distribution picturing a composite genome organization (Figure 1 and Table 6).

65 Copy number variation in Saccharomyces pastorianus

Table 6. Chromosome copy number in nine different S. pastorianus strains. The chromosome copy number was quantified by Magnolya

Group 2 Group 1 Industrial Lg WS34/70 karyotype isolates strains CBS1483 CBS1260 CBS1513 CBS1503 A1+B11

Strain Spy2 Spy1 A1 A2

Total Genome size (Mb) 47.5 50.5 56.0 53.5 39.1 56.9 35.7 33.6 34.1 CHR aSize (kb) Sc1 179 3 2 2 2 2 2 1 1 2 Sc2 780 3 3 3 3 2 3 2 1 1 Sc3 300 0 1.5 1 2 0 2 0 0 0 Sc4 1464 3 2 3 3 2 3 2 1 1 Sc5 572 3 2 3 2 1 3 0 1 0 Sc6 260 1 1 2 2 2 2 1 0 0 Sc7 1050 1 3 2 3 1 2.75 2 1 0 Sc8 504 5 3 3 4 2 3 2 1 1 Sc9 398 3 4 5 4 3 3 0 1 2 Sc10 750 2 3 3 3 2 1 2 1 1 Sc11 639 2 3 3 3 1 2 2 0 1 Sc12 983 2 2 3 2 2 3 1.5 0 0 Sc13 918 2 2 3 3 1 3 1 1 1 Sc14 758 2.25 3 3 3 2 2.25 2 1 0 Sc15 1032 3 3 3 3 3 3 2 1 1 Sc16 914 2 2 3 2 3 3 1 1 0 Seub1 149 2 2 1 2 2 2 2 1 1 Seub2-4 1265 1 2 2 2 1.75 2 1 2 2 Seub3-Sc3 295 4 3 4 3 4 3 2 1 1 Seub4-2 984 1 2 2 2 1 2 1 2 2 Seub5 546 2 2 2 2 2 2 3 2 3 Seub6 253 3 3 2 3 1 3 2 3 3 Seub7-Sc7 1044 3 2 2 1 2 2 1 2 3 Seub8-15 804 1 2 2 2 1 2 1 2 2 Seub9 393 2 1 1 1 1 2 3 2 1 Seub10 650 1 1 3 1 1.25 4 1 2 2 Seub11 652 2 2 2 2 2 3 2.75 3 2.5 Seub12 1036 2 2 2 2 1 2 2 3 3 Seub13 925 2 2 2 2 2 2 2 2 2 Seub14 757 2 2 2 2 2 3 1 2 3 Seub15-8 748 1 1 1 2 1 2 1 2 2 Seub16 910 2 2 2 2 1 2 2 2 3 CHR number (Total) 68.25 70.5 77 75 55 79 49,25 45 46.5

66 Chapter 2

Confirmation of the chromosome copy numbers in CBS1483 by q-PCR and flow cytometry. To check if the extraordinary, high chromosome copy numbers predicted by the Poisson mixture model were correct, these predictions were experimentally tested. Quantitative PCR was used as an independent method to analyze the representation of chromosomal sequences that were predicted to have a different copy number. To verify the model-based predictions of CHRVII structures, PCR primers were designed for three different sites on 2 CHRVII, in S. cerevisiae-type EMP24 and MEP1, and in S. eubayanus type SHY1. These genes were predicted to be present in four, one and three copies respectively (Figure 2). For an unbiased copy number estimation, genomic DNA of CBS1483 was spiked with plasmid DNA (pUG6, Guldener et al., 1996)). The relative, pUG6-normalized abundances of ScEMP24 and ScMEP1 (4.05±0.1) and of SeubSHY1and ScMEP1 (3.2±0.05) (Figure 2) perfectly matched the model-based chromosome copy numbers.

Four additional experimental verifications of predicted copy numbers were performed, targeting sequences on CHRI (S. cerevisiae and S. eubayanus ADE1), CHRIII (S. eubayanus type NFS1 and S. cerevisiae type ABP1), CHRVIII (S. cerevisiae and S. eubayanus DUR3) and CHRXV (S. cerevisiae and S. eubayanus MDM20). The copy number ratio of the normalized Sc and Seub ADE1 alleles (4.3±0.6) was consistent with the model prediction. Similarly, the ratio calculated for SeubNFS1 and ScABP1 (1.3±0.5) was compatible with the presence of a unique, chimeric Sc/SeubCHRIII structure. The ratio between the DUR3 species alleles (5.6±0.8) corroborated the predicted large (5-fold) copy number variation between Sc and SeubCHRVIII. Finally, the ratio measured for the normalized Sc and SeubMDM20 (3.4±0.5) was in agreement with the predicted 3:1 S. cerevisiae : S. eubayanus distribution (Figure 2).

Two previous studies lend further support to the reliability of the model-based copy number estimations for strain CBS1483. Using a similar quantitative PCR approach, CBS 1483 ScARO10 and SeubARO10 alleles were shown to have a copy number ratio of three (Bolat et al., 2013), matching the model-based prediction of ScCHRIV and SeubCHRII-IV (Table 6). Furthermore, the observation that a single transformation was sufficient to completely delete the ScHXK1 allele in S. pastorianus CBS1483 (Solis-Escalante et al., 2013) is consistent with the model-based prediction that this strain contains only a single copy of ScCHRVI.

In a second approach, the total DNA content of the S. pastorianus strain CBS1483 was estimated by comparison with strains of known ploidy (n, CEN.PK113-7D; 2n, CEN. PK122; 3n, FRY153) using flow cytometry.

67 Copy number variation in Saccharomyces pastorianus

Figure 2. Chromosome copy number determination by Real Time Quantitative PCR. A- representation of the chromosomes I, III, VII, VIII and XV and their copy number the blue and red blocks represent the S. cerevisiae and S. eubayanus subgenomes, respectively. Coloured triangles (red or blue) indicate the position of marker genes used in the analysis. B- Quantification of relative copy number ratio of gene pairs located on chromosomes I, III, VII, VIII and XV. The bar graph represent the normalized copy number ratio of a gene pair. Each gene copy number was normalized relative to control DNA spiked in genomic DNA of CBS1483. Data are presented as average and standard deviation of at least eight independent reactions.

The fluorescence intensity of the peaks corresponding to the control strains original ploidy and to its doubling showed a near-perfect linear correlation (R2 = 0.998, Figure 3A). Based on this correlation the overall ploidy values of S. pastorianus CBS1483, based on n and 2n peaks characteristic of the G1+M phase of the cell cycle were 3.8 and 7.5, respectively (Figure 3B). Assuming a haploid genome size of 12Mb, the size of the CBS1483 genome should then be 46Mb. These estimates were in full agreement with the chromosomal count predicted by the Magnolya algorithm which estimated this size to be 47.5 (Table 6). Collectively, these data confirmed the reliability of the prediction model and therefore enabled us to extend the use of this method to multiple strains.

68 Chapter 2

2

Figure 3. DNA content determination by flow cytometry: A- DNA content measured by flow cytometry analysis of Saccharomyces cerevisiae CEN.PK113-7D (n), CEN.PK122 (2n), FRY153 (3n) and of S. pastorianus CBS1483. The cells were harvested in mid-log phase and stained with SYTOX® Green Nucleic Acid Stain. The histogram represent the distribution of ungated FL1-A fluorescence. B- Determination of the ploidy of S. pastorianus CBS 1483. To establish a correlation between fluorescence intensity and ploidy, the control strains (closed circles) CEN.PK113-7D (n/2n), CENPK122 (2n/4n) and FRY153(3n/6n) were used. Data plotted are fluorescence intensities of the flow-cytometry histogram peaks with the highest cell count. The fluorescence intensity of the CBS1483 histogram peaks (open circles) was compared to a standard linear correlation line based on data of the control strains. Dashed lines indicate the 95% confidence interval of the regression analysis. Fluorescence data indicated the average and mean deviation of duplicate experiments.

69 Copy number variation in Saccharomyces pastorianus

Chromosome copy number variation among S. pastorianus strains The extraordinary differences in the copy numbers of individual chromosomes in strain CBS1483 raised the question to what extent this genetic feature differs among S. pastorianus lager brewing strains. To address this question, five additional S. pastorianus strains were newly sequenced, thereby enabling us to extend our analysis beyond the already sequenced strains S. carlsbergensis (syn S. pastorianus) CBS1513, S. monacencis CBS1503 (Hewitt et al., 2014; Walther et al., 2014) and S. pastorianus Weihenstephan 34/70 (Nakao et al., 2009) and CBS1260 (Hewitt et al., 2014). Two of the additionally sequenced strains (Spy1 and Spy2) are industrial brewing strains, the other three are clonal variants of the widely used non- proprietary brewing lager yeast S. pastorianus WS34/70 (www.hefebank-weihenstephan.de/). Karyotyping of WS34/70 cultures indicated that supposedly pure cultures of this strain were, in fact, composed of several karyotypic variants (Bolat et al., 2008). Three Weihenstephan 34/70 isolates with reported karyotype differences (A1, A1+B11 and A2) (Bolat et al., 2008) were sequenced and analyzed for chromosome copy number variation, bringing the total number of S. pastorianus strains included in the analysis to nine (seven Group II and two Group I strains; Table 6).

Two previous studies on S. pastorianus (carlsbergensis) strain CBS1513 (DBVPG 6033) reported different total chromosome copy numbers of 31 (Hewitt et al., 2014) and 47 (Walther et al., 2014). To validate and benchmark use of the Magnolya algorithm for copy number estimations, we used the S. carlsbergensis CBS1513 sequence data generated by Hewitt and co-workers (Hewitt et al., 2014). This analysis predicted a total number of 45 chromosomes (Table 6). Of 29 structurally different chromosomes identified by Walther and co-authors (2014), 28 showed an identical chromosome copy number in our analysis. The main difference between our copy number assessment and that of Walther et al., (2014) concerned CHRIII, for which only one copy was estimated by Magnolya while Walther and co-workers reported three (2014). The resulting copy number difference (#45 vs #47) (Table 6) may result from the different methodologies used for the chromosome quantification. Alternatively, they may be due to the use of an independent genome sequencing dataset (Hewitt et al., 2014) and thus reflect a chromosome copy number difference between clonal lines derived from the same strain.

A large difference was observed in the total chromosome complement of strains CBS1483 (#68) and the CBS1513 (#45-47). Analysis of seven additional lager brewing strains revealed even more extreme differences. The Group I (‘Saaz’) strains CBS1513 and CBS1503 showed similar chromosome number (45 and 46.5 respectively). Conversely, the Group II (‘Frohberg’) strains revealed extremely different chromosome complements, ranging from 49 for strain CBS1260 (DBVPG 6257) to 79 for the industrial strain Spy1. For the latter strain, this would correspond to a total genome size of 57Mb (Table 6), which is almost five-fold higher than the genome size of haploid S. cerevisiae strains (Goffeau et al., 1996).

70 Chapter 2

The single cell isolates of strain WS34/70, A1, A1+B11 and A2 showed strikingly different chromosome complements of 70, 77 and 75 respectively. These results are consistent with an earlier karyotyping study (Bolat et al., 2008) and indicate that chromosomal copy number can be highly dynamic, even in supposedly pure cultures. Exploring allelic variation in S. pastorianus CBS1483 2 The results presented in the previous paragraphs underline that the assembled and already published genome sequences of S. pastorianus strains do not reflect the full variation of the genomic landscape of lager brewing strains. In addition to copy number variation per se, the multiple alleles carried by sister chromosomes may exhibit sequence variations. The extent of heterozygosity in S. pastorianus CBS1483, taking into account the calculated ploidy of individual chromosomes, was estimated using the GATK Unified Genotyper algorithm (www.broadinstitute.org/gatk). This analysis revealed that 13839 nucleotide positions were covered by more than one base, thus confirming the presence of heterozygous loci. These variations affected nucleotides that were located upstream or downstream of open reading frames (43.5%) as well as nucleotides in protein-encoding sequences (56.5%). Of the 7815 mutations located in open reading frames (ORFs) only 4024, located in 1829 ORFs, resulted in an amino acid change (3872 missenses) or in the introduction of a premature stop codon (152 nonsenses) (Figure 4A). At 27 positions, three different nucleotides were encountered in the sequence data, which could lead to occurrence of three protein variants.

To further investigate allelic variation, three open reading frames (ScOAF1, SeubFLO1 and ScHPF1) were more closely inspected. Three alleles of ScOAF1, a gene encoding a regulator involved in transcriptional control of β-oxidation, were located on the three copies of ScCHRI in S. pastorianus CBS1483. These three alleles exhibited 13 different variable positions, of which one was situated in the promoter region and 12 occurred in the open reading frame. Of the latter 12 single-nucleotide variations (SNV), five led to an amino acid change which could lead to proteins with different properties. However, the available sequence data did not allow the complete reconstruction of the two or three potentially separate alleles (Figure 4B). A more precise reconstruction of discrete alleles was possible for SeubFLO1 and ScHPF1. The four copies of SeubFLO1 which encodes a lectin-like protein involved in flocculation, harbored only two variable positions, of which a single position yielded alteration of the protein sequence. Interestingly, three variants were found for this location, encoding predicted protein sequences with, at position 164, either a threonine, a serine or an alanine. These alleles occurred in a 2:1:1 ratio (Figure 4B). Only a single gene, ScHPF1, which encodes a haze-protective mannoprotein involved in aggregated protein particles size reduction, at 5n ploidy showed a position within the open reading frame with multiple nucleotide coverage. Out of two SNV identified in the ScHPF1 gene, only one resulted in an amino acid change, giving rise to two predicted protein sequences represented by a 4:1 allelic ratio (Figure 4B). These examples demonstrate that the theoretical proteome of the strain CBS1483 cannot be simply derived from the 10279 protein sequences (Tables 4) predicted by genome annotation of the consensus sequence. Instead, the in silico CBS1483 proteome should include at least 1829 additional protein variants.

71 Copy number variation in Saccharomyces pastorianus

Figure 4. Allelic variation in S. pastorianus CBS1483. A- Pie chart representing the distribution of the chromosomal position covered by more than one nucleotide as identified by GATK www.broadinstitute.org/gatk( ). The number between brackets denotes the number of position affected by allelic variation. B- Bar charts represents the nucleotide frequency at heterozygous positions in the genes ScOAF1 (3n), SeubFLO1 (4n) and ScHPF1 (5n). The position coverage (%) was calculated by the ratio of the reads covering the position containing one of the nucleotide over the total number of reads covering the evaluated position. The positions tested for allelic variation were covered by a minimum of 110 sequencing reads. The dashed lines indicate the theoretical distribution for the position coverage for 2:1 distribution for 3n ploidy (ScOAF1), for a 2:1:1 distribution for a 4n ploidy (SeubFLO1) and for 3:1:1 and 2:2:1 for 5n ploidy (ScHPF1).

72 Chapter 2

Linking chromosome copy number variation to phenotype To test whether differences in chromosome copy number correlated with phenotype, we analyzed the three WS34/70 isolates (A1, A1+B11, A2) in more detail. In addition to the chromosomal rearrangements that enabled their identification, this revealed differences to the published Weihenstephan 34/70 sequence (Nakao et al., 2009), indicating that these strains were not entirely isogenic. While the strains A1 and A2 exhibited fewer than 200 2 non- and mis-sense mutation differences, they carried a common set of 361 SNVs relative to the reference WS34/70 sequence. Out of the three isolates, A1+B11 was the closest to the reference sequence, but still exhibited over 500 non- and mis-sense differences relative to the A1 and A2 strains (Figure 5). Notwithstanding these differences, the sequences of these lager yeast genomes were highly similar and therefore provided an excellent baseline to study the impact of chromosome copy number on fermentation performance. The three strains were cultivated in WMM medium, a chemically defined medium that contains a mixture of maltose, maltotriose, glucose and fructose in a 64-23-11-2 ratio. The three strains showed an identical specific growth rates (0.15 -1h ) and sugar consumption profiles at 20o C (Figure 6).

Figure 5. Venn diagram representing the single nucleotide variation found in pairwise comparison between A1 and the reference WS34/70 strain sequence (Nakao et al., 2009) (blue circle), between A2 and WS34/70 (green circle) and A1+B11 and WS34/70 (red circle). A- Non sense mutations. B- Missense mutations.

Diacetyl, a vicinal diketone, is a butter-tasting off-flavour that is produced during fermentation as a by-product of valine metabolism. During lagering, diacetyl concentrations need to be reduced by the yeast cells to concentrations below its sensory threshold. The diacetyl concentration produced by a brewing strain is, therefore, a highly relevant industrially characteristic of S. pastorianus strains. Diacetyl production profiles of the three WS34/70 isolates were bi-phasic, with a production phase during the first 25-30 h of cultivation being followed by a diacetyl consumption phase. Strain A1 showed the highest diacetyl production peak (351 ± 10 μg/l), a value that was 1.7 and 2.9-fold higher than found with strains A2 and A1+B11, respectively (Figure 7). Sequence analysis did not reveal any differences

73 Copy number variation in Saccharomyces pastorianus in the valine biosynthesis genes (ILV2, ILV6, ILV5, ILV3, BAT1 and BAT2) of the three strains. However, the chromosomes harboring these genes did show copy number variation. Strain A1+B11, which presented the lowest diacetyl peak, showed the largest differences. Copy numbers of Sc and Seub CHRXIV, which carry ILV2/YMR108W, encoding the catalytic subunit of α-acetolactate synthase, the enzyme that catalyzes the first step in valine biosynthesis, were identical in the three strains (3:2). However, chromosomes XII and X showed one and two additional copies, respectively in strain A1+B11 relative to strains A1 and A2. These chromosomes carry ILV5/YMR108W and ILV3/YJR016C, which encode the enzymes that catalyze the second and third steps in the valine biosynthesis pathway, respectively (Figure 7).

Figure 6. Sugar consumption and growth profiles of the karyotypic variant A1, A1+B11 and A2 of the S. pastorianus WS34/70 strain. A-Sugar consumption profile of theS. pastorianus A1 strain. The strain was grown in WMM medium supplemented with a complex mixture of sugars from corn syrup containing 9.5 g·l-1 maltotriose (closed squared), 27 g·l-1 maltose (closed circle), 4.5 g·l-1 glucose (open circle), 0.75 g·l-1 fructose (open square) as consumable sugars. The strain was grown in shake flask at a temperature of 20oC for 50 hours. Maltotriose, maltose, glucose and fructose were analysed using HPLC. B- Sugar consumption profile of theS. pastorianus A1+B11 strain. C– Sugar consumption profile of theS. pastorianus A2 strain. D- Growth profile of strains A1(closed square), A1+B11 (open circle), A2 (open square). Optical density OD was measured at a wavelength of 660 nm.

74 Chapter 2

Several strategies to control diacetyl in brewing yeasts are based on increasing the metabolic flux through the pathway fromα -acetolactate to valine. Both the incremental reduction of ILV6 copy number (Duong et al., 2011) and the increase in copy number and overexpression of ILV5 in lager yeast (Gjermansen et al., 1988; Kusunoki et al., 2012; Mithieux et al., 1995) have previously been shown to lead to reduced diacetyl formation. The chromosome copy number analysis of the three variants would then suggest that in strain A1+B11, the 2 extra copies of ILV5 ([+1] CHRXII) and ILV3 ([+2] CHRX), genes encoding enzymes that catalyze reactions downstream of the diacetyl precursor α-acetolactate were sufficient to generate a low-diacetyl phenotype.

The ability to flocculate is another highly relevant trait of brewing strains (Bauer et al., 2010). After 38 h of cultivation, when sugars were already completely consumed (Figure 6), strains A1, A1+B11 and A2 were sampled, washed and analyzed for flocculence in the presence and absence of Ca2+. The three strains exhibited extremely different phenotypes in this assay. The A1 strain expressed a phenotype comparable to the non-flocculent laboratory strain CEN.PK113-7D, while strains A2 and A1+B11 showed clear flocculation phenotype, with A1+B11 being more flocculent than A2 Figure( 8).

The flocculin encoding genes LgFLO, FLO1, FLO5, FLO9 and FLO10 are for a large part responsible of the flocculation characteristics of lager yeasts (Van Mulders et al., 2010; Verstrepen et al., 2006). In a first step, we examined sequence variations in the flocculin genes of the three WS34/70 variant strains. Whereas no variation was observed in the LgFLO, FLO1, FLO5 and FLO10, one difference was detected in the FLO9 gene in the A1 strain that led to the replacement of a histidine residue by a tryptophan at position 215. In a second step, we compared the copy number of the various FLO genes in the three strains. All but one SeubFLO genes (FLO1, FLO5 and FLO10) exhibited a copy number matching the chromosomal copy number in all variants, except for SeubFLO1, which was not identified in the A1 strain. Similarly, the LgFLO gene from the A1 strain also showed a copy number comparable to the CHRI copy number. In contrast, the LgFLO gene in A1+B11 and A2 revealed four and two additional copies, respectively, relative to the copy number of CHRI. Although the mutation found in FLO9 and the absence of SeubFLO1 might affect the flocculation properties of the A1 strain, the largest difference between the three strains lay in the LgFLO gene, whose copy number exhibited a strong positive correlation (R2=0.972) with the flocculence of the WS34/70 isolate strains Figure( 8).

75 Copy number variation in Saccharomyces pastorianus

Figure 7. Correlation between diacetyl production profile and chromosome copy number of valine biosynthesis genes. A- The S. pastorianus strains A1 (closed square), A1+B11 (open circle) and A2 (open square) were grown at 20 oC in WMM medium with a complex sugars mixture from corn syrup containing 9.5 g·l-1 maltotriose, 27 g·l-1 maltose, 4.5 g·l-1 glucose, 0.75 g·l-1 fructose. Diacetyl concentrations were measured using static headspace gas chromatography. Data are presented as average and mean deviation of duplicated independent replicates. B- Valine and diacetyl biosynthetic pathways and chromosome copy number of the genes involved in the metabolic routes in strains A1, A1+B11 and A2. Chromosome copy number was predicted using Magnolya (Nijkamp et al., 2012).

76 Chapter 2

2

Figure 8. Flocculation and FLO-gene copy number in S. pastorianus A1, A1+B11 and A2. A- The copy number of the annotated FLO genes was predicted using Magnolya (Nijkamp et al., 2012). The copy number of the contig harboring the FLO gene was compared to the copy number of the contigs comprising the rest of the chromosome, the asterisk (*) denotes when the gene copy number matches the chromosome value. The sequence analysis of the FLO1 and FLO9 genes did not enable an unequivocal assignment of a Sc or Seub allele type; these genes are therefore labelled “Unknown”. B- Flocculation properties of the A1, A1+B11 and A2 strains. The strains were grown in WMM medium at 20 °C for 38 h. Cells were sampled by centrifugation and flocculence was analyzed with the modified Helm’s test (Bendiak et al., 1996; Helm et al., 2014). The flocculation coefficient was calculated from the blank-corrected average of six replicates. C- Correlation between the flocculation coefficient and the copy number of LgFLO1.

77 Copy number variation in Saccharomyces pastorianus

DISCUSSION

High-grade aneuploidy in S. pastorianus: a model for cancer cell biology? Changes in chromosome number can either result in euploidy, with multiple complete sets of chromosomes (3n, 4n) or in aneuploidy, with different copy numbers for individual chromosome(s) (Storchova 2014). Euploidy includes allopolyploidy, which indicates perfect polyploidy of a hybrid genome that contains subgenomes from different ancestors. In higher eukaryotes, aneuploidy is often related to pathological or developmental anomalies (Potter 1991; Torres et al., 2010).

In S. cerevisiae, disomy of individual chromosomes in haploid strains has been implicated in various phenotypes, including defects in cell cycle progression, increased glucose uptake, and increased sensitivity to conditions interfering with protein synthesis and protein folding (Torres et al., 2007). Different disomic S. cerevisiae mutants also shared a characteristic transcriptional response (Sheltzer et al., 2012). However, aneuploidy is generally well tolerated in Saccharomyces species (Bond et al., 2004) and increased ploidy caused by a whole-genome duplication, ca. 100 My ago, in ancestor of Saccharomyces yeasts is seen as a key factor in shaping their genomes and phenotypes (Wolfe et al., 1997). On a shorter time scale, whole-chromosome and segmental aneuploidies are frequently observed in laboratory evolution experiments, in which increased expression of genes carried by duplicated regions confers a selective advantage (de Kok et al., 2012; Oud et al., 2013; Oud et al., 2012; Selmecki et al., 2006; Selmecki et al., 2009).

This study clearly identifies S. pastorianus yeasts as allo-aneuploids, in which high-grade aneuploidy affects both subgenomes. While polyploidy is a common occurrence in nature, high-grade aneuploidy is comparatively rare. In fact, a high-degree aneuploidy as observed for S. pastorianus has only been documented for tumor cells (Davidson et al., 2000; Grigorova et al., 2004; Weaver et al., 2007). For example, in a MCF7 breast cancer cell line, 17 of 22 non-sexual chromosomes were present at three or four copies (Davidson et al., 2000). Hitherto, haploid S. cerevisiae strains with defined disomies have been used as models to study implications of aneuploidy on human pathogenesis (Torres et al., 2007). The present study indicates that S. pastorianus, with its extensive, dynamic aneuploidy, may present a better model to unravel the molecular mechanisms that underlie the development and progression of aneuploidy in cancer cell development.

Different chromosome complements in S. pastorianus: phenotypic and taxonomic implications The S. pastorianus strains analyzed in this study exhibit a remarkably wide range of genome sizes, with total chromosome numbers ranging from 45 to 79 (Table 6). The significant differences in chromosomal content of three clonal isolates of a single, commonly used strain (S. pastorianus Weihenstephan 34/70) indicated that genome copy number in brewing yeast strains is highly dynamic.

78 Chapter 2

In several higher eukaryotes, copy number variation has been associated with phenotypic traits and species separation (Bickhart et al., 2012; Doan et al., 2012; Hu et al., 2011). This study identified clear correlations between chromosome copy number in S. pastorianus and two brewing-related traits: diacetyl production and flocculence Figures( 7 and 8). Three near-isogenic karyotypic variants of S. pastorianus Weihenstephan 34/70 revealed different diacetyl production profiles, although no DNA-sequence differences were found in genes 2 involved in the biosynthesis of this important off-flavor. Instead, diacetyl production profiles directly correlated with copy numbers of structural genes involved in the valine/diacetyl biosynthetic pathway (Figure 7). The biological relevance of this correlation is supported by previous targeted genetic modification studies, in which increased copy numbers of genes involved in this pathway caused altered diacetyl production profiles(Duong et al., 2011; Gjermansen et al., 1988; Mithieux et al., 1995).

The different chromosome complements of the lager yeasts strains analyzed in this study were not limited to ‘subtle’ quantitative changes in chromosome or gene copy number. A particularly pronounced difference between Group I and Group II strains concerned the complete absence of S. cerevisiae chromosomes III, VI and XII in Group I strains. As a consequence, S. pastorianus Group I and II genomes differ by at least 852 genes. One clear impact of this difference is that Group I strains exclusively rely on S. eubayanus rDNA whereas in Group II the rRNA is also transcribed from S. cerevisiae rDNA. The spectacular differences in gene content among S. pastorianus strains is in marked contrast with the high genome conservation observed for S. cerevisiae (Liti et al., 2006; Liti et al., 2009). Recently, based on the genomes of the reference lager yeast strains Weihenstephan 34/70 (Nakao et al., 2009) and CBS1503 (Walther et al., 2014), a change in classification of lager yeasts was proposed. Group I (Saaz) yeasts, originally classified as S. carlsbergensis, were renamed S. pastorianus and, thereby, assigned to the same species as Group II (Frohberg) strains (Rainieri et al., 2006). The highly distinct genome content of the Weihenstephan 34/70 and CBS1503 strains might support a reinstatement of different species names for Group I and Group II S. pastorianus strains, as recently proposed (Wendland 2014). Such a step would also be consistent with the clear-cut physiological characteristic differences exhibited by Saaz and Frohberg strains(Gibson et al., 2013).

Reconstructing evolution and domestication of S. pastorianus The discovery of S. eubayanus, whose genome exhibits over 99.5% identity with the non-S. cerevisiae-type subgenome of lager yeast strains (Bing et al., 2014; Libkind et al., 2011; Peris et al., 2014), represented a scientific breakthrough in understanding the evolutionary history of lager brewing yeasts. Indeed, it recently enabled a recreation, in the laboratory, of the hybridization event between S. cerevisiae and eubayanus which combined properties of the two parents that are advantageous for brewing. Culture collections harbor a vast number of S. pastorianus strains with different phenotypic characteristics (Gibson et al., 2013; Dunn et al., 2005; Steensels et al., 2014) and, based on this study, very likely large genomic variations. This diversity suggests that the their genome plasticity enabled these strains to evolve to adapt to a wide range of environmental conditions and/or to be selected for different product properties.

79 Copy number variation in Saccharomyces pastorianus

Availability of ‘synthetic’ hybrids of S. cerevisiae and S. eubayanus provides unique opportunities to assess the impact of culture conditions on the genome content and dynamics (Dunn et al., 2013; Piotrowski et al., 2012) and, thereby, reconstruct the evolutionary trajectory of S. pastorianus strains and the role of aneuploidy in their evolution. Similar to a recent study on laboratory evolution of plant isolates of the lactic-acid bacterium Lactococcus lactis in milk (Bachmann et al., 2012), such analyses are likely to increase our understanding of how domestication has shaped the genomes of brewing yeasts, which are among the most important industrial micro-organisms on earth.

CONCLUSION

Three-dimensional genomic diversity in S. pastorianus The present analysis of the genomes of different S. pastorianus strains provides new insights in the genetic changes that, after initial hybridization of S. cerevisiae and S. eubayanus, led to current brewing strains. Extensive genome re-organization, involving changes in the copy numbers of genes and chromosomes, has been followed by the generation of allelic diversity and differential expression of paralogs (Querol et al., 2009). Together, these processes have led to a remarkable genomic diversity of lager brewing strains, with important implications for further genetic analysis of brewing-related traits and for strain optimization. In addition to mutations in the original S. cerevisiae and S. eubayanus subgenomes, large variations in chromosome copy number and allelic variation of the thus multiplied genes were observed.

These three dimensions: primary sequence, copy number variation and allelic variation of gene copies within individual strains, should be taken into account in future studies aimed at understanding the domestication of brewing yeasts and their further optimization for beer fermentation.

DATA ACCESS

The sequence data generated in this study are searchable at NCBI –Entrez (http://www. ncbi.nlm.nih.gov/) under the Bioproject PRJNA266750 (accession: SRP049726). Table 4 described the structure of the searchable data.

ACKNOWLEDGMENTS

We would like to thank Marit Hebly and Pascale Daran-Lapujade for technical assistance and Prof. Frank Rosenzweig (University of Montana, Missoula, MT) for kindly providing us the polyploid standard strain FRY153.

80 Chapter 2

REFERENCES

Bachmann H, Starrenburg MJ, Molenaar D, Bolat I, Walsh MC, Turtoi M, 2008. Isolation Kleerebezem M, van Hylckama Vlieg JE, and characterization of two new lager yeast 2012. Microbial domestication signatures strains from the WS34/70 population. Roum of Lactococcus lactis can be reproduced by Biotechnol Lett 6: 62-73. experimental evolution. Genome Res 22: 115- Bolger AM, Lohse M, Usadel B, 2014. 2 124. Trimmomatic: a flexible trimmer for Illumina Bauer FF, Govender P, Bester MC, 2010. Yeast sequence data. Bioinformatics 30: 2114-2120. flocculation and its biotechnological relevance. Bond U, 2009. Chapter 6: The genomes of lager Appl Microbiol Biotechnol 88: 31-39. yeasts. Adv Appl Microbiol 69: 159-182. Bendiak D, Van der Aar P, Barbero F, Benzing Bond U, Neal C, Donnelly D, James TC, 2004. P, Berndt R, Carrick K, Dull C, Dunn S, Eto Aneuploidy and copy number breakpoints in the M, Gonzalez M, Hayashi N, Lawrence D, genome of lager yeasts mapped by microarray Miller J, Pugh T, Rashel L, Rossmore K, Smart hybridisation. Curr Genet 45: 360-370. K, Sobczak J, Speers A, Casey G, 1996. Yeast Caesar R, Palmfeldt J, Gustafsson JS, Pettersson flocculation by absorbance. J Am Soc Brew E, Hashemi SH, Blomberg A, 2007. Comparative Chem 54: 245-248. proteomics of industrial lager yeast reveals Bickhart DM, Hou Y, Schroeder SG, Alkan C, differential expression of the cerevisiae and non- Cardone MF, Matukumalli LK, Song J, Schnabel cerevisiae parts of their genomes. Proteomics 7: RD, Ventura M, Taylor JF, Garcia JF, Van Tassell 4135-4147. CP, Sonstegard TS, Eichler EE, Liu GE, 2012. Casaregola S, Nguyen HV, Lapathitis G, Copy number variation of individual cattle Kotyk A, Gaillardin C, 2001. Analysis of genomes using next-generation sequencing. the constitution of the beer yeast genome by Genome Res 22: 778-790. PCR, sequencing and subtelomeric sequence Bing J, Han PJ, Liu WQ, Wang QM, Bai FY, hybridization. Int J Syst Evol Microbiol 51: 2014. Evidence for a Far East Asian origin of 1607-1618. lager beer yeast. Curr Biol 24: R380-R381. Davidson JM, Gorringe KL, Chin SF, Orsetti B, Boetzer M, Henkel CV, Jansen HJ, Butler D, Besret C, Courtay-Cahen C, Roberts I, Theillet Pirovano W, 2011. Scaffolding pre-assembled C, Caldas C, Edwards PA, 2000. Molecular contigs using SSPACE. Bioinformatics 27: 578- cytogenetic analysis of breast cancer cell lines. Br 579. J Cancer 83: 1309-1317. Boetzer M, Pirovano W, 2012. Toward almost de Kok S, Nijkamp JF, Oud B, Roque FC, de closed genomes with GapFiller. Genome Biol 13: Ridder D, Daran JM, Pronk JT, van Maris AJ, R56-13. 2012. Laboratory evolution of new lactate Bolat I, Romagnoli G, Zhu F, Pronk JT, transporter genes in a jen1Delta mutant of Daran JM, 2013. Functional analysis and Saccharomyces cerevisiae and their identification transcriptional regulation of two orthologs of as ADY2 alleles by whole-genome resequencing ARO10, encoding broad-substrate-specificity and transcriptome analysis. FEMS Yeast Res 2-oxo-acid decarboxylases, in the brewing yeast 10-1364. Saccharomyces pastorianus CBS1483. FEMS Yeast Res 13: 505-517.

81 Copy number variation in Saccharomyces pastorianus

Doan R, Cohen N, Harrington J, Veazey K, Juras Gjermansen C, Nilsson-Tillgren T, Petersen JG, R, Cothran G, McCue ME, Skow L, and Dindot Kielland-Brandt MC, Sigsgaard P, Holmberg SV, 2012. Identification of copy number variants S, 1988. Towards diacetyl-less brewers’ yeast. in horses. Genome Res 22: 899-907. Influence ofilv2 and ilv5 mutations. J Basic Dunn B, Levine RP, Sherlock G, 2005. Microbiol 28: 175-183. Microarray karyotyping of commercial wine Goffeau A, Barrell BG, Bussey H, Davis RW, yeast strains reveals shared, as well as unique, Dujon B, Feldmann H, Galibert F, Hoheisel genomic signatures. BMC Genomics 6: 53- JD, Jacq C, Johnston M, Louis EJ, Mewes HW, Dunn B, Paulish T, Stanbery A, Piotrowski J, Murakami Y, Philippsen P, Tettelin H, Oliver Koniges G, Kroll E, Louis EJ, Liti G, Sherlock G, SG, 1996. Life with 6000 genes. Science 274: Rosenzweig F, 2013. Recurrent rearrangement 563-567. during adaptive evolution in an interspecific Grigorova M, Staines JM, Ozdag H, Caldas yeast hybrid suggests a model for rapid C, Edwards PA, 2004. Possible causes of introgression. PLoS Genet 9: e1003366- chromosome instability: comparison of Dunn B, Sherlock G, 2008. Reconstruction of chromosomal abnormalities in cancer cell lines the genome origins and evolution of the hybrid with mutations in BRCA1, BRCA2, CHK2 and lager yeast Saccharomyces pastorianus. Genome BUB1. Cytogenet Genome Res 104: 333-340. Res 18: 1610-1623. Guldener U, Heck S, Fielder T, Beinhauer J, Duong CT, Strack L, Futschik M, Katou Y, Hegemann JH, 1996. A new efficient gene Nakao Y, Fujimura T, Shirahige K, Kodama Y, disruption cassette for repeated use in budding Nevoigt E, 2011. Identification ofSc -type ILV6 yeast. Nucleic Acids Res 24: 2519-2524. as a target to reduce diacetyl formation in lager Haase SB, Reed SI, 2002. Improved flow brewers’ yeast. Metab Eng 13: 638-647. cytometric analysis of the budding yeast cell Entian KD, Kötter P, 2007. Yeast genetic strain cycle. Cell Cycle 1: 132-136. and plasmid collections. Meth Microbiol 36: Helm E, Nohr B, Thorne R, 2014. The 629-666. measurement of yeast flocculence and its Fischer G, James SA, Roberts IN, Oliver SG, significance in brewing. Wallerstein Lab Louis EJ, 2000. Chromosomal evolution in Commun 16: 315-325. Saccharomyces. Nature 405: 451-454. Hewitt SK, Donaldson IJ, Lovell SC, Delneri Gibson BR, Lawrence SJ, Leclaire JP, Powell D, 2014. Sequencing and characterisation of CD, Smart KA 2007. Yeast responses to stresses rearrangements in three S. pastorianus strains associated with industrial brewery handling. reveals the presence of chimeric genes and gives FEMS Microbiol Rev 31:535–569. doi:.10.1111/ evidence of breakpoint reuse. PLoS One 9: j.1574-6976.2007.00076.x e92203- Gibson BR, Storgårds E, Krogerus K, Vidgren V, Holt C, Yandell M, 2011. MAKER2: an 2013. Comparative physiology and fermentation annotation pipeline and genome-database performance of Saaz and Frohberg lager yeast management tool for second-generation genome strains and the parental species Saccharomyces projects. BMC Bioinformatics 12:491. doi: eubayanus. Yeast 30: 255-266. 10.1186/1471-2105-12-491.: 491-12.

82 Chapter 2

Hu G, Wang J, Choi J, Jung WH, Liu I, Mithieux SM, Weiss AS, 1995. Tandem Litvintseva AP, Bicanic T, Aurora R, Mitchell integration of multiple ILV5 copies and elevated TG, Perfect JR, Kronstad JW, 2011. Variation transcription in polyploid yeast. Yeast 11: 311- in chromosome copy number influences the 316. virulence of Cryptococcus neoformans and Nakao Y, Kanamori T, Itoh T, Kodama Y, occurs in isolates from AIDS patients. BMC Rainieri S, Nakamura N, Shimonaga T, Hattori 2 Genomics 12: 52- M, Ashikari T, 2009. Genome sequence of the Kusunoki K, Ogata T, 2012. Construction of lager brewing yeast, an interspecies hybrid. DNA self-cloning bottom-fermenting yeast with low Res 16: 115-129. vicinal diketone production by the homo- Nijkamp JF, van den Broek MA, Datema E, de integration of ILV5. Yeast 29: 435-442. Kok S, Bosman L, Luttik MA, Daran-Lapujade Langmead B, Trapnell C, Pop M, Salzberg SL, P, Vongsangnak W, Nielsen J, Heijne WHM, 2009. Ultrafast and memory-efficient alignment Klaassen P, Paddon CJ, Platt D, Kotter P, van of short DNA sequences to the human genome. Ham RC, Reinders MJT, Pronk JT, de Ridder D, Genome Biol 10: R25-10. Daran JM, 2012. De novo sequencing, assembly Li H, Durbin R, 2010. Fast and accurate and analysis of the genome of the laboratory long-read alignment with Burrows-Wheeler strain Saccharomyces cerevisiae CEN.PK113-7D, transform. Bioinformatics 26: 589-595. a model for modern industrial biotechnology. Microb Cell Fact 11:36 Libkind D, Hittinger CT, Valerio E, Goncalves C, Dover J, Johnston M, Goncalves P, Sampaio Nijkamp JF, van den Broek MA, Geertman JM, JP, 2011. Microbe domestication and the Reinders MJ, Daran JM, de Ridder D, 2012. De identification of the wild genetic stock of lager- novo detection of copy number variation by co- brewing yeast. Proc Natl Acad Sci U S A 108: assembly. Bioinformatics 28: 3195-3202. 14539-14544. Oud B, Guadalupe-Medina V, Nijkamp JF, de Liti G, Barton DB, Louis EJ, 2006. Sequence Ridder D, Pronk JT, van Maris AJ, Daran JM, diversity, reproductive isolation and species 2013. Genome duplication and mutations in concepts in Saccharomyces. Genetics 174: 839- ACE2 cause multicellular, fast-sedimenting 850. phenotypes in evolved Saccharomyces cerevisiae. Proc Natl Acad Sci U S A 110: E4223-E4231. Liti G, Carter DM, Moses AM, Warringer J, Parts L, James SA, Davey RP, Roberts IN, Burt A, Oud B, van Maris AJ, Daran JM, Pronk JT, Koufopanou V, Tsai IJ, Bergman CM, Bensasson 2012. Genome-wide analytical approaches for D, O’Kelly MJ, van OA, Barton DB, Bailes E, reverse metabolic engineering of industrially Nguyen AN, Jones M, Quail MA, Goodhead I, relevant phenotypes in yeast. FEMS Yeast Res Sims S, Smith F, Blomberg A, Durbin R, Louis 12: 183-196. EJ, 2009. Population genomics of domestic and Peris D, Sylvester K, Libkind D, Goncalves P, wild yeasts. Nature 458: 337-341. Sampaio JP, Alexander WG, Hittinger CT, 2014. McKenna A, Hanna M, Banks E, Sivachenko A, Population structure and reticulate evolution of Cibulskis K, Kernytsky A, Garimella K, Altshuler Saccharomyces eubayanus and its lager-brewing D, Gabriel S, Daly M, DePristo MA, 2010. hybrids. Mol Ecol 23: 2031-2045. The Genome Analysis Toolkit: a MapReduce Piotrowski JS, Nagarajan S, Kroll E, Stanbery A, framework for analyzing next-generation DNA Chiotti KE, Kruckeberg AL, Dunn B, Sherlock G, sequencing data. Genome Res 20: 1297-1303. Rosenzweig F, 2012. Different selective pressures lead to different genomic outcomes as newly- formed hybrid yeasts evolve. BMC Evol Biol 12: 46-

83 Copy number variation in Saccharomyces pastorianus

Porro D, Brambilla L, Alberghina L, 2003. Simpson JT, Wong K, Jackman SD, Schein Glucose metabolism and cell size in continuous JE, Jones SJ, Birol I, 2009. ABySS: a parallel cultures of Saccharomyces cerevisiae. FEMS assembler for short read sequence data. Genome Microbiol Lett 229: 165-171. Res 19: 1117-1123. Potter H, 1991. Review and hypothesis: Solis-Escalante D, Kuijpers NG, Bongaerts N, Alzheimer disease and Down syndrome-- Bolat I, Bosman L, Pronk JT, Daran JM, Daran- chromosome 21 nondisjunction may underlie Lapujade P, 2013. amdSYM, a new dominant both disorders. Am J Hum Genet 48: 1192- recyclable marker cassette for Saccharomyces 1200. cerevisiae. FEMS Yeast Res 13: 126-139. Querol A, Bond U, 2009. The complex and Steensels J, Meersman E, Snoek T, Saels V, dynamic genomes of industrial yeasts. FEMS Verstrepen KJ, 2014. Large-scale selection and Microbiol Lett 293: 1-10. breeding to generate industrial yeasts with Rainieri S, Kodama Y, Kaneko Y, Mikata K, superior aroma production. Appl Environ Nakao Y, Ashikari T, 2006. Pure and mixed Microbiol 80: 6965-6975. genetic lines of Saccharomyces bayanus Storchova Z, 2014. Ploidy changes and genome and Saccharomyces pastorianus and their stability in yeast. Yeast 31: 421-430. contribution to the lager brewing strain genome. Tamai Y, Momma T, Yoshimoto H, Kaneko Y, Appl Environ Microbiol 72: 3968-3974. 1998. Co-existence of two types of chromosome Ramakers C, Ruijter JM, Deprez RH, Moorman in the bottom fermenting yeast, Saccharomyces AF, 2003. Assumption-free analysis of pastorianus. Yeast 14: 923-933. quantitative real-time polymerase chain reaction Torres EM, Sokolsky T, Tucker CM, Chan LY, (PCR) data. Neurosci Lett 339: 62-66. Boselli M, Dunham MJ, Amon A, 2007. Effects Ruijter JM, Ramakers C, Hoogaars WM, Karlen of aneuploidy on cellular physiology and cell Y, Bakker O, van den Hoff MJ, Moorman AF, division in haploid yeast. Science 317: 916-924. 2009. Amplification efficiency: linking baseline Torres EM, Williams BR, Tang YC, Amon A, and bias in the analysis of quantitative PCR 2010. Thoughts on aneuploidy. Cold Spring data. Nucleic Acids Res 37: e45- Harb Symp Quant Biol 75: 445-451. Selmecki A, Forche A, Berman J, 2006. Usher J, Bond U, 2009. Recombination between Aneuploidy and isochromosome formation in homoeologous chromosomes of lager yeasts drug-resistant Candida albicans. Science 313: leads to loss of function of the hybrid GPH1 367-370. gene. Appl Environ Microbiol 75: 4573-4579. Selmecki AM, Dulmage K, Cowen LE, Anderson Van Mulders SE, Ghequire M, Daenen L, JB, Berman J, 2009. Acquisition of aneuploidy Verbelen PJ, Verstrepen KJ, Delvaux FR, 2010. provides increased fitness during the evolution Flocculation gene variability in industrial of antifungal drug resistance. PLoS Genet 5: brewer’s yeast strains. Appl Microbiol e1000705- Biotechnol 88: 1321-1331. Sheltzer JM, Torres EM, Dunham MJ, Amon Vaughan-Martini A, Kurtzman CP, 1985. A, 2012. Transcriptional consequences of Deoxyribonucleic acid relatedness among species aneuploidy. Proc Natl Acad Sci U S A 109: of the genus Saccharomyces sensu stricto. Int J 12644-12649. Syst Evol Microbiol 35: 508-511.

84 Chapter 2

Vaughan-Martini A, Martini A, 1998. Saccharomyces Myen ex Rees. The yeasts, a taxonomic study, Kurtzman, C. P. and Fell, J. W.), pp. 358-371. Verduyn C, Postma E, Scheffers WA, van Dijken JP, 1992. Effect of benzoic acid on metabolic 2 fluxes in yeasts: a continuous-culture study on the regulation of respiration and alcoholic fermentation. Yeast 8: 501-517. Verstrepen KJ, Klis FM, 2006. Flocculation, adhesion and biofilm formation in yeasts.Mol Microbiol 60: 5-15. Walther A, Hesselbart A, Wendland J, 2014. Genome Sequence of Saccharomyces carlsbergensis, the World’s First Pure Culture Lager Yeast. G3 (Bethesda ) 27: 783-793. Weaver BA, Silk AD, Montagna C, Verdier- Pinard P, Cleveland DW, 2007. Aneuploidy acts both oncogenically and as a tumor suppressor. Cancer Cell 11: 25-36. Wendland J, 2014. Lager yeast comes of age. Eukaryotic Cell 10: 1256-1265. Wolfe KH, Shields DC, 1997. Molecular evidence for an ancient duplication of the entire yeast genome. Nature 387: 708-713. Yamagishi H, Ogata T, 1999. Chromosomal structures of bottom fermenting yeasts. Syst Appl Microbiol 22: 341-353.

85 Copy number variation in Saccharomyces pastorianus

SUPPORTING INFORMATION

Figure S1. Identification of a SeubCHRXII DNA non-reciprocal introgression at ScCHRXII in S. pastorianus CBS1483 strain. A- The graphs represent the ploidy prediction of CBS1483 CHRXII made by Magnolya (Nijkamp et al., 2012). The black box indicates the presence of a contig (Contig00589) aligned on SeubCHRXII showing an higher copy number than the rest of the chromosome (three versus two copies). B- Annotation of the contig00589, the 6.4-kb fragments harboured three genes YLR411W/CTR3, YLR412W/BER1 and YLR413W/INA1 encoding a high affinity copper transporter, a protein involved in microtubule-related processes and a putative protein of unknown function respectively. C- Localization of contig00589. The 8-kb library reads were mapped to contig00589 using BWA(Li et al., 2010). A total of 8762 reads were aligned, the mate paired reads of the mapped sequences were selected and assembled. Three contigs were assembled. The first contigs was assembled from 3810 reads and matched a scaffold composing SeubCHRXII. The two other contigs were assembled from 1119 and 515 reads and showed perfect identity with scaffolds (Scf2 and 3) forming ScCHRXII. The locally assembled contigs matching Scf2 harboured YLR409W and YLR410W whereas the locally assembled contig matching Scf3 harboured ORFs YLR415W and YLR417W. Collectively, these data confirmed the presence in ScCHRXII, the presence of a 6.4 kb S. eubayanus DNA introgression.

86 Chapter 2

2

Figure S2. Characterisation of contig00530 carrying an extra ScADE1 copy in S. pastorianus CBS1483 strain. A- The graphs represent the ploidy prediction of CBS1483 CHRI made by Magnolya (Nijkamp et al., 2012). The end right arm of CHRI showed an extra copy relative to the rest of the chromosome (four versus three). B- Annotation of the contig00530, the 9.0-kb fragment harboured four open reading frames YAR014C/ BUD14, YAR015W/ADE1 , YAR018C/KIN3and YAR019C/CDC15 encoding the bud site selection protein 14, the phosphoribosylaminoimidazole-succinocarboxamide synthase, a serine/threonine- protein kinase and the cell division control protein 15 respectively. C- Localization of contig00530. The 8-kb library reads were mapped to contig00530 using BWA (Li et al., 2010). A total of 12126 reads were aligned, the mate paired reads of the mapped sequences were selected and locally assembled. Two contigs were assembled. The first contig was assembled from 4662 reads and matched a scaffold composing ScCHRI as expected. The second contig was assembled from 2351 reads and showed perfect identity with a scaffold forming Sc-SeubCHRX. Analysis of the latter locally assembled contig revealed extremely high homology with multiple Ty sequences and did not allow unambiguous localization of this extra part of ScCHRI.

87

CHAPTER 3 amdSYM, a new dominant recyclable marker cassette for Saccharomyces cerevisiae.

Solis-Escalante D Kuijpers NGA Bongaerts N Bolat I Bosman L Pronk JT Daran JM Daran-Lapujade P

FEMS Yeast Research, 2013, vol. 13 (1), 126-139 amdSYM, a new dominant recyclable marker cassette for Saccharomyces cerevisiae.

ABSTRACT

Despite the large collection of selectable marker genes available for Saccharomyces cerevisiae, marker availability can still present a hurdle when dozens of genetic manipulations are required. Recyclable markers, counterselectable cassettes that can be removed from the targeted genome after use, are therefore valuable assets in ambitious metabolic engineering programmes. In the present work the new recyclable dominant marker cassette amdSYM, formed by the Ashbya gossypii TEF2 promoter and terminator and a codon-optimized acetamidase gene (Aspergillus nidulans amdS) is presented. The amdSYM cassette confers S. cerevisiae the ability to use acetamide as sole nitrogen source. Direct repeats flanking the amdS gene allow for its efficient recombinative excision. As previously demonstrated in filamentous fungi, loss of the amdS marker cassette from S. cerevisiae can be rapidly selected for by growth in the presence of fluoroacetamide. TheamdSYM cassette can be used in different genetic backgrounds and represents the first counterselectable dominant marker gene cassete for use in S. cerevisiae. Furthermore, using astute cassette design, amdSYM excision can be performed without leaving a scar or heterologous sequences in the targeted genome. The present work therefore demonstrates that amdSYM is a useful addition to the genetic engineering toolbox for Sacharomyces laboratory, wild and industrial strains.

90 Chapter 3

INTRODUCTION

The past decade has been marked by the construction of complex cell factories resulting from dozens of genetic manipulations and leading to remarkable new capabilities. For instance, the industrial and model yeast Saccharomyces cerevisiae has been engineered to produce the anti-malaria drug precursor artemisinic acid (Ro et al., 2006), hydrocortisone (Szczebara et al., 2003) and cineole (Ignea et al., 2011), among others. The ever-increasing demand for cheap and sustainable production of complex molecules combined with its attractiveness 3 as a host for pathway engineering will inevitably intensify the exploitation of S. cerevisiae as cell factory in the future (Hong & Nielsen, 2012). Also in fundamental yeast research, extensive genetic manipulation is necessary, for instance to unravel complex transport systems (Wieczorke et al., 1999; Suzuki et al., 2011) and regulatory networks (Baryshnikova et al., 2010). A common and critical feature for all these genetic manipulations is the requirement of selectable markers that enable the selection of mutants carrying the desired genetic modifications. Despite the relatively large number of selection markers available for S. cerevisiae (Table 1), the construction of multiple successive genetic modifications remains a challenge as the number of genetic manipulations typically equals the number of selection markers introduced in the host. Selection markers can be classified in two main categories, auxotrophic markers, which restore growth of specific mutants and dominant markers, which confer completely new functions to their host. Both types suffer from substantial drawbacks. The use of auxotrophic markers is restricted to auxotrophic strains, i.e. strains carrying mutations in one gene leading to a strict requirement for a specific nutrient (Pronk, 2002). This constrain is augmented for industrial strains that are typically prototrophic and for which the aneu- or polyploidy makes the construction of auxotrophic strains a laborious task (Puig et al., 1998). The expression in a single strain of multiple dominant marker genes, under the control of strong promoters, may result in protein burden and other negative effects on host strain physiology (Gopal et al., 1989; Nacken et al., 1996). Additionally, for industrial strains dedicated to food applications such as the production of nutraceuticals, the lack of heterologous DNA is highly desired.

While protein burden might be avoided by expressing marker genes from inducible promoters (Suzuki et al., 2011), many inducible promoters are notoriously leaky (Agha-Mohammadi et al., 2004) and therefore only partly address the problem. A good alternative resides in the use of recyclable markers. Marker recycling was first shown withURA3 (Alani et al., 1987). Loss of URA3, encoding orotidine-5’-phosphate decarboxylase involved in pyrimidine biosynthesis, is lethal in uracil-free media. Since its discovery, URA3 has become a very popular auxotrophic selection marker. This popularity mainly originates from the ability of URA3 to be counter-selected in the presence of 5-fluoroorotic acid (5-FOA), which is converted to a toxic compound (5-fluoro-UMP) by Ura3p. Indeed, when the URA3 marker is flanked by direct repeats, cultivation on 5-FOA enables the efficient selection of strains in which URA3 has been excised via mitotic recombination. URA3 can therefore be re-used for further modifications (Alani et al., 1987; Langlerouault & Jacobs, 1995). Besides the

91 amdSYM, a new dominant recyclable marker cassette for Saccharomyces cerevisiae. obvious need for auxotrophic strains, another drawback of the most frequently used URA3 cassette is that sequences (or scars) are left after marker removal. Commonly, the bacterial sequence hisG is present as direct repeat. However, repeated use of URA3 cassettes carrying hisG increases the probability of mistargeted integrations (Davidson & Schiestl, 2000). An alternative to hisG is the creation of direct repeats upon integration using sequences already present in the host genome, generating seamless marker removal (Akada et al., 2006). Similar to URA3, two other auxotrophic markers MET15 and LYS2 can be counter-selected in the presence of methyl-mercury (Singh & Sherman, 1974) and alpha-aminoadipate (Chattoo et al., 1979) respectively. Nevertheless, their use is limited by the same complications described for URA3.

A successful attempt to recycle virtually any desired marker was the development of the bacteriophage derived LoxP-Cre recombinase system (Hoess & Abremski, 1985; Sauer, 1987; Guldener et al., 1996; Guldener et al., 2002). Exploiting the site-specific activity of the Cre recombinase, this system is used to efficiently remove markers by flanking them with the targeted LoxP sequence. This system exhibits two major limitations, i) it requires the expression of a plasmid-borne recombinase and thereby necessitates an additional selection marker or extensive screening (Schorsch et al., 2009) and ii) the repeated use of this system causes major chromosomal rearrangements (Delneri et al., 2000), personal communication E. Boles).

The Aspergillus nidulans amdS gene encoding acetamidase has been successfully used as dominant ‘gain of function’ selection marker in different filamentous fungi and the yeast Kluyveromyces lactis (Kelly & Hynes, 1985; Beri & Turner, 1987; Yamashiro et al., 1992; Swinkels et al., 1997; Selten et al., 2000; van Ooyen et al., 2006; Read et al., 2007; Ganatra et al., 2011). Although Selten et al. (2000) suggested that amdS could be used for selection in S. cerevisiae, this statement was not further supported by experimental evidence. Acetamidase catalyzes the hydroxylation of acetamide to acetate and ammonia, thus conferring the ability to the host cell to use acetamide as sole nitrogen or carbon source (Corrick et al., 1987; Hynes, 1994). Similar to URA3, amdS is a recyclable marker that can be counter-selected by growth on media containing the acetamide homologue fluoroacetamide, which is converted by acetamidase to the toxic compound fluoroacetate (Apirion, 1965; Hynes & Pateman, 1970).

This study evaluates the use of the new dominant marker module amdSYM for S. cerevisiae and demonstrates its efficiency for sequentially introducing multiple gene deletions in this yeast. Availability of this dominant, counterselectable marker cassette to the yeast research community should facilitate rapid introduction of multiple genetic modifications into any laboratory and industrial strains of S. cerevisiae.

92 Chapter 3

3 , 2001) , , 1998) , 1998) , 1998) , , 1985) , , 1998) , , 1990) , , 1987) , , 1986) , , 1979) , , 1987) , et al. , 1992) , et al. et al. et al. , 1997) , 1997) , 1994) , et al. , 1987; Langlerouault & Jacobs, 1995) 1987; Langlerouault & Jacobs, , et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. Reference (Alani (Shuster 1989) (Losberger & Ernst, (Wach (Wach (Brachmann (Zhang (Chattoo (Brachmann (Nakayashiki (Brachmann 1974; (Singh & Sherman, Brachmann (Wach (Gatignol (Drocourt 1983) (Gritz & Davies, (Hadfield (Henderson strains Saccharomyces cerevisiae Saccharomyces Recyclable/Method selection with 5-FOA Yes/negative selection with 5-FOA Yes/negative selection with 5-FOA Yes/negative No/-- No/-- No/-- No/-- selection with alpha- Yes/negative aminoadipate No/-- No/-- No/-- selection with methyl- Yes/negative mercury No/-- No/-- No/-- No/-- No/-- No/-- Mode of action Repairs uracil deficiency Repairs uracil deficiency Repairs uracil deficiency Repairs histidine deficiency Repairs histidine deficiency Repairs leucine deficiency Repairs leucine deficiency Repairs lysine deficiency Repairs tryptophan deficiency Repairs adenine deficiency Repairs adenine deficiency Repairs methionine deficiency Resitance to G418 Resistance to phleomycin Resistance to Zeocin Resistance to hygromycin Resistance to chloramphenicol Resistance to Cu2+ Different selectable markers used in laboratory and industrial Marker gene Marker markers Auxotrophic URA3 KlURA3 CaURA3 HIS3 HIS5 LEU2 KlLEU2 LYS2 TRP1 ADE1 ADE2 MET15 Dominant markers KanMX ble Sh ble hph Cat CUP1 Table 1. 1. Table

93 amdSYM, a new dominant recyclable marker cassette for Saccharomyces cerevisiae. , 1998) , et al. , 1991) , et al. Reference (Van den Berg & Steensma, 1997) den Berg & Steensma, (Van 1997) den Berg & Steensma, (Van 1999) (Lackova & Subik, (Hashida-Okado 1999) (Goldstein & McCusker, (Delpozo 1999) (Goldstein & McCusker, 2004) (Cebollero & Gonzalez, 1996) (Xie & Jimenez, 2004) (Cebollero & Gonzalez, 2004) & McCusker, (Vorachek-Warren Recyclable/Method No/-- No/-- No/-- No/-- No/-- No/-- No/-- No/-- No/-- No/-- No/-- Mode of action Resistance to formaldehyde Resistance to fluoroacetate Multi drug resistance Resistance to aureobasidin Resistance to nourseothricin Resistance to cycloheximide Resistance to bialaphos Resistance to o-Fluoro-DL-phenylalanine Resistance to sulfometuron methyl Increased tolerance to sulfite Resistance to D-Serine Continued Marker gene Marker SFA1 dehH1 PDR3-9 AUR1-C nat CYH2 pat ARO4-OFP SMR1 FZF1-4 DsdA Table 1. 1. Table

94 Chapter 3

MATERIALS AND METHODS

Strains and media Propagation of plasmids was performed in chemically competent Escherichia coli DH5α according to manufacturer instructions (Z-competentTM transformation kit, Zymo research, California, US). All yeast strains used in this study are listed in Table 2. Under non selective conditions yeast was grown in complex medium (YPD) containing 10 g l-1 yeast extract, 20 g l-1 peptone and 20 g l-1 glucose. Synthetic media (SM) containing 3 g l-1 KH PO , 2 4 3 -1 -1 -1 0.5 g l MgSO4.7H2O, 5 g l (NH4)2SO4,1 ml l of a trace element solution as previously described (Verduyn et al., 1992), 1 ml l-1 of a vitamin solution (Verduyn et al., 1992) was -1 used. When amdSYM was used as marker, (NH4)2SO4 was replaced by 0.6 g l acetamide as -1 nitrogen source and 6.6 g l K2SO4.to compensate for sulfate (SM-Ac) Recycled markerless cells were selected on SM containing 2.3 g l-1 fluoroacetamide (SM-Fac). SM, SM-Ac and SM-Fac were supplemented with 20 mg l-1 adenine and 15 mg l-1 L-canavanine sulfate when required. In all experiments, 20 g l-1 of glucose were used as carbon source. The pH in all the media was adjusted to 6.0 with KOH. Solid media were prepared by adding 2% agar to the media described above. amdSYM and plasmid construction A codon-optimized (JCat, http://www.jcat.de/, (Grote et al., 2005)) version of A. nidulans amdS flanked bySal I and XhoI restriction sites and attB1 and attB2 recombination sites (Accession number: JX500098) was synthesized at GeneArt AG (Regensburg, Germany) and cloned into the vector pMA (GeneArt AG) generating the plasmid pUD171. The amdS gene was transferred into the destination plasmid pAG426GPD (Alberti et al., 2007) by LR recombination reaction according to the manufacturer recommendations (Invitrogen, California, USA) yielding the plasmid pUDE158.

The Ashbya gossypii TEF2ter from pUG6 (Guldener et al., 1996; Guldener et al., 2002) was amplified and the restriction sites XhoI and KpnI incorporated with the primers p1FW and p1RV and PhusionTM Hot Start Polymerase (Finnzymes, Vantaa, Finland). Restriction with XhoI and KpnI and ligation of the amplified fragment andthe vector p426TEF (Mumberg et al., 1995) resulted in the replacement of the CYC1ter of p426TEF for (A. g.) TEF2ter. The generated vector was termed p426TEF-TEF(t).

Primers amdSBFW and amdSXRV and PhusionTM Hot Start Polymerase (Finnzymes, Vantaa, Finland) were used to amplify amdS and to incorporate BamHI and XhoI sites using pUD171 as template. After digestion with the corresponding enzymes, the gene was cloned into the vector p426TEF-TEF(t) generating the plasmid pUD184. Using primers amdSBFW and amdSSRV, containing BamHI and SacI restriction sites respectively, and PhusionTM Hot Start Polymerase (Finnzymes), the gene amdS together with A. gossypii TEF2 terminator were amplified from pUD184. In order to incorporateBam HI and SacI restriction sites in pUG6, the complete plasmid backbone was amplified by PCR with the

95 amdSYM, a new dominant recyclable marker cassette for Saccharomyces cerevisiae. primers SpUGFW and pUGBRV and and PhusionTM Hot Start Polymerase (Finnzymes).

Digestion with BamHI and SacI of the modified pUG6 and the amdS-(A.g).TEF2ter and ligation led to the construction of pUG-amdSYM (Figure 1). All primers and their sequences are listed in Table 3. The plasmid pUG-amdSYM is available to the research community at Euroscarf (http://web.uni-frankfurt.de/fb15/mikro/euroscarf/) with the accession number: P30669.

Table 2. Saccharomyces sp. strains used in this study

Strain Genotype Reference (van Dijken et al., 2000; Entian & CEN.PK113-7D MATa MAL2-8c SUC2 Kotter, 2007; Nijkamp et al., 2012) (van Dijken et al., 2000; Entian & CEN.PK113-5D MATa MAL2-8c SUC2 ura3-52 Kotter, 2007) Centraal Bureau voor Schimmel- CBS8066 MATa/α HO/ho cultures, http://www.cbs.knaw.nl, CBS, Den Haag, TheNetherlands YSBN MATa/α ho::Ble/ho::HphMX4 (Canelas et al., 2010) MATα SUC2 gal2 mal mel flo1 flo8-1 S288c (Mortimer & Johnston, 1986) hap1 ho bio1 bio6 Centraal Bureau voor Schimmel- cultures, http://www.cbs.knaw.nl, CBS1483 Saccharomyces pastorianus CBS, Den Haag, TheNetherlands (Dunn & Sherlock, 2008) Gift from Dr. JM Geertman Scottish Ale Saccharomyces cerevisiae (Heineken Supply Chain, Zoeterwoude, NL) Centraal Bureau voor Schimmel- cultures, http://www.cbs.knaw.nl, CBS12357 Saccharomyces eubayanus sp.nov CBS, Den Haag, TheNetherlands (Libkind et al., 2011) MATa MAL2-8c SUC2 ura3-52 IME141 This study pAG426GPD (2m URA3 TDH3pr-CYC1ter) MATa MAL2-8c SUC2 ura3-52 pUDE158 IME142 This study (2m URA3 TDH3pr-amdS-CYC1ter) MATa MAL2-8c SUC2 can1Δ::amdSYM IMX168 This study ADE2 IMX200 MATa MAL2-8c SUC2 can1Δ ADE2 This study MATa MAL2-8c SUC2 can1Δ IMX201 This study ade2Δ::amdSYM IMX206 MATa MAL2-8c SUC2 can1Δ ade2Δ This study MATa MAL2-8c SUC2 hxk1Δ::loxP- IMK468 This study amdSYM-loxP IMK470 CBS1483 Sc-hxk1Δ::loxP-amdSYM-loxP This study Scottish Ale Sc-aro80Δ::loxP-amdSYM- IMK473 This study loxP CBS12357 Seub-aro80Δ::loxP-amdSYM- IMK474 This study loxP

96 Chapter 3

3

Figure 1. pUG-amdSYM map. pUG-amdSYM was constructed by replacing the marker KanMX in pUG6 for a codon optimized version of A. nidulans amdS. The plasmid kept all features present in the pUG family (Guldener et al., 1996; Guldener et al., 2002)

Deletion cassette construction. Deletion cassettes were constructed by PCR using PhusionTM Hot Start Polymerase (Finnzymes) and following manufacturer recommendations. Primers used for repeated gene deletions had a similar design described as follows. Forward primers contain two cores: 1) a 50-55 bp sequence homologous to the region upstream the gene to delete and 2) the sequence 5’- CAGCTGAAGCTTCGTACGC-3’ that binds to the region upstream the A. gossypii TEF2 promoter in pUG-amdSYM. Reverse primers contained three cores: 1) a 50-55 bp sequence homologous to the region downstream the gene to be deleted, 2) a 40 bp sequence homologous to the region upstream the targeted region and 3) the sequence 5’- GCATAGGCCACTAGTGGATCTG-3’ that binds downstream the A. gossypii TEF2 terminator in pUG-amdSYM. CAN1 and ADE2 deletion cassettes were constructed using the primer pairs CdcamdSFW and CdcamdSRV, and AdcamdSFW and AdcamdSRV respectively (Table 3).

97 amdSYM, a new dominant recyclable marker cassette for Saccharomyces cerevisiae.

Table 3. Primers used in this study

Primer Sequence 5’ to 3’ a,b Plasmid Construction amdSBFW GGGGGATCCATGCCACAATCTTGGGAAGAA amdSXRV GGGCTCGAGTTATGGAGTAACAACG amdSSRV GGGGAGCTCCAGTATAGCGACCAGCATTC p1FW CCGCTCGAGCAGTACTGACAATAAAAAGATTC p1RV GCGGTACCCAGTATAGCGACCAGCATTC pUGBRV GGGGGATCCTTGTTTATGTTCGGATGTGATGTG SpUGFW GGGCGAGCTCTCGAGAACCCTTAA Deletion cassette Construction TCAGACTTCTTAACTCCTGTAAAAACAAAAAAAAAAAAAGGCATAGCAATCAGCT- CdcamdSFW GAAGCTTCGTACGC ATGCGAAATGGCGTGGGAATGTGATTAAAGGTAATAAAACGTCATATC- CdcamdSRV TAAGAACTCTGAAATAAACTTTCGATTGACGACAGATTGAAAGCATAGGCCAC- TAGTGGATCTG TACTATAACAATCAAGAAAAACAAGAAAATCGGACAAAACAATCAAGTATCAGCT- AdcamdSFW GAAGCTTCGTACGC TCATTTTATAATTATTTGCTGTACAAGTATATCAATAAACTTATATATTAGGAT- AdcamdSRV GTACTTAGAAGAGAGATCCAACGATTTTACGCACCAGCATAGGCCACTAGTG- GATCTG AAACTCACCCAAACAACTCAATTAGAATACTGAAAAAATAAGATGATGACAAGAG- HdcamdSFW GGTCGAACTCCAGCTGAAGCTTCGTACGC AGGGAGGGAAAAACACATTTATATTTCATTACATTTTTTTCATTAGCCTAAGTCG- HdcamdSRV TAATTGAGTCGCATAGGCCACTAGTGGATCTG ScHdcamdSFW CTTCTTATGCCCCTGAACCC ScHdcamdSRV CTATCCTACGACTTTCTCCCTC AGTTAGTCGTAGGAATATATGATCCACGCATAATAAGGTTACATTAAGCACTGCTT- ScARO80dcamdSFW TATCCAGCTGAAGCTTCGTACGC CTTTGTATTTAAAATCATTTTTACGAATAGTGCGGTTGTCTTGGTTGATGACG- ScARO80dcamdSRV TAATTCTGCATAGGCCACTAGTGGATCTG ScARO80g5’FW CACTACCAAAGCCAAATCAGAC ScARO80g5’RV GATAAAGCAGTGCTTAATGTAACC ScARO80g3’FW AGAATTACGTCATCAACCAAGAC ScARO80g3’RV AGCGTAGCTTGCACTACTAG AGCAAGCTAGTCAAAATATTTGCTCTCCGCATGATATAATTACTTTCAGTATCGTT- SeubARO80dcamdSFW GTCCCCAGCTGAAGCTTCGTACGC TAGTAATCAATCATTTATCTAATTAAACATTCCTTTTCTCTAATTTTATGTGTGAG- SeubARO80dcamdSRV GGGCGCATAGGCCACTAGTGGATCTG SeubARO80g5’FW GACATTGATGACAATGGTAGTG SeubARO80g5’RV GGACAACGATACTGAAAGTAAT SeubARO80g3’FW GCCCCTCACACATAAAATTAGAG SeubARO80g3’RV GTGTGGCTTGTACCACAAGA Deletion/Marker removal confirmation CdcFW CGGGAGCAAGATTGTTGTG CdcRV GGTTGCGAACAGAGTAAACC AdcFW AAAGGACACCTGTAAGCGTTG AdcRV AGCATTTCATGTATAAATTGGTGCG HdcFW CCTTAGGACCGTTGAGAGGAATAG HdcRV GACCGCAAAAAAAACATAAGGG ScARO80dcFW TGATCCCGATACTGGAAATCAA amdSdcRV CGACCAGCATTCACATACGA ScARO80gRV TTCATCCTATCTGAACAGAATAC SbARO80dcFW GTCATGCAGGCTCTTCATTG SbARO80gRV TATCCCTCACGTGAATTTAAACC Sequencing C-FW ATCACTTACTGGCAAGTGCG C-RV ATCAGTTGTGCCTGGAAAAG A-FW AACGCCGTATCGTGATTAAC A-RV GGACACTTATATGTCGAGCAAGA a The sequences underlined indicate the position of the introduced restriction. b The sequences in bold indicate the direct repeats used to excise the marker gene upon counter selection

98 Chapter 3

In order to construct the deletion cassette targeting S. cerevisiae HXK1 (ScHXK1) primers HdcamdSFW and HdcamdSRV and the template pUG-amdSYM were used.

The constructed cassette was used to generate the strain IMK468 derived from CEN.PK113- 7D. Genomic DNA of IMK468 was used as template for primers ScHdcamdSFW and ScHdcamdSRV. The resulting cassette contained 500bp homologous sequences upstream and downstream of the ScHXK1 gene and was used for the deletion of ScHXK1 in the Saccharomyces pastorianus lager brewing strain CBS1483. 3 The construction of the deletion cassette corresponding to ScARO80 allele for the S. cerevisiae Scottish Ale strain was performed according to the two steps fusion protocol (Amberg et al., 1995) with the primers ScARO80dcamdSFW and ScARO80dcamdSRV to amplify amdSYM from pUG-amdSYM and ScARO80g5’FW, ScARO80g5’RV, ScARO80g3’FW and ScARO80g 3’RV to generate the 500bp sequence homologous to upstream and downstream sections of ScARO80, genomic DNA from CBS1483 was used as template. The same approach was taken for the deletion of Saccharomyces eubayanus ARO80 (SbARO80) in CBS12357 using the primer pairs SeubARO80dcamdSFW/SeubARO80dcamdSRV, SeubARO80g5’FW/ SeubARO80g5’RV, SeubARO80g3’FW/SeubARO80g3’RV. All primers and their sequences are listed in Table 3.

Selection and marker recycling Yeast transformations were performed using the lithium-acetate protocol (Gietz & Woods, 2002). Integration of the deletion cassettes into the yeast genome was selected by plating the transformation mix on SM-Ac. Targeted integration was verified by PCR with the primer pairs CdcFW/CdcRV, AdcFW/AcdRV, HdcFW/HdcRV, ScARO80dcFW/ amdSdcRV, ScARO80dcFW/ScARO80gRV, SbARO80dcFW/amdSdcRV, SbARO80dcFW/ SbARO80gRV and, when applicable, by transferring single colonies to SM plates containing 15 mg L-1 L-canavanine or by screening for colony pigmentation. A small fraction of single colony was resuspended in 15 μl of 0.02 N NaOH; 2 μl of this cell suspension was used as template for the PCR reaction that was performed using DreamTaq PCR master mix (Fermentas GmbH, St. Leon-Rot, Germany) following the manufacturer recommendations. Marker removal was achieved by growing cells overnight in liquid YPD and transferring 0.2 ml to 100 ml liquid SM-FAc. Marker-free single colonies were obtained by plating 0.1 ml of culture on SM-Fac solid media and confirmed by PCR with the primer pairs CdcFW/CdcRV or AdcFW/AcdFW. All cultures were incubated at 30 °C. In order to confirm that only endogenous sequences were present after marker removal, long run sequencing (Baseclear, Leiden, The Netherlands) was performed using the primers C-FW and C-RV for the CAN1 locus and A-FW and A-RV for ADE2 locus of the marker-free strain IMX206.

99 amdSYM, a new dominant recyclable marker cassette for Saccharomyces cerevisiae.

RESULTS AND DISCUSSION

Expression of amdS in S. cerevisiae confers the ability to grow on acetamide as sole nitrogen source Although the yeast putative amidase gene AMD2 is similar (57.2% similarity and 33.2% identity) to A. nidulans amdS (Chang & Abelson, 1990), there is no report demonstrating acetamidase activity in wild-type S. cerevisiae or on growth of budding yeast on acetamide as sole nitrogen or carbon source. Expression of amdS in S. cerevisiae is therefore expected to bring a new function in this yeast by enabling its growth on acetamide as sole nitrogen source. Although the A. niger amdS promoter is able to drive the expression of genes in S. cerevisiae, this is only possible under carbon-limited conditions (Bonnefoy et al., 1995). Therefore, a codon-optimized amdS sequence was cloned under the control of the strong, constitutive TDH3 promoter in the plasmid pUDE158.

The plasmid pUDE158, containing amdS, and the empty vector pAG426GPD were transformed into CEN.PK113-5D, generating strains IME142 and IME141, respectively. Expression of the acetamidase gene in S. cerevisiae (strain IME142) conferred growth with acetamide as sole nitrogen source while the control strain (strain IME141) was unable to grow (Figure 2). Additionally, the inability of IME142 to grow on acetamide upon loss of amdS by counter-selection with 5-FOA of pUDE158 confirmed that growth on acetamide was fully amdS-dependent (data not shown).

Plasmids and deletion cassettes construction. The coding sequence of the A. nidulans amdS gene, codon-optimized for expression in S. cerevisiae and flanked by the Ashbya gossypii TEF2 promoter and terminator, was cloned into the vector pUG6 (Guldener et al., 1996; Guldener et al., 2002) by replacing the KanMX gene, resulting in the plasmid pUG-amdSYM (Figure 1). The resulting amdSYM module only contained heterologous sequences, thereby reducing the probability of mistargeted integration (Wach et al., 1994). The pUG-amdSYM plasmid can be easily used as template for deletion cassettes containing the new marker module amdSYM and was used for the construction of all deletion cassettes used in this study.

The deletion cassettes contained three major regions (Figure 3A): (i) a 50-55 bp sequence homologous to the upstream part of the gene to be deleted, including the start codon, and a 50-55 bp sequence homologous to the downstream part of the gene to be deleted, including the stop codon. These regions were used for targeted homologous recombination (Baudin et al., 1993); (ii) the amdSYM marker, and (iii) a 40 bp sequence homologous to the region upstream of the targeted locus to create direct repeats in the host genome upon integration, thereby enabling scarless marker excision (Akada et al., 2006) (Figure 3).

100 Chapter 3

3

Figure 2. Growth of strains from the CEN.PK family on acetamide as sole nitrogen source. Expression of amdS on the multicopy plasmid pUDE158 conferred to S. cerevisiae the ability to grow on acetamide as sole nitrogen source. The strains CEN.PK113-5D (ura3-52), IME141 (ura3-52 pAG426GPD (2μ

URA3 TDH3pr-CYC1ter)) and IME142 (ura3-52 pUDE158 (2 μ URA3 TDH3pr-amdS-CYC1ter)) were grown on YPD, SM and SM-Ac media and incubated at 30°C. The plates were read after three days.

Repeated gene deletions in Saccharomyces cerevisiae using amdSYM To evaluate if the new marker amdSYM was suitable for repeated gene knock-out in S. cerevisiae, it was attempted to sequentially delete two genes in the laboratory strain CEN. PK113-7D and, after marker recycling, to construct a marker-free and scarless double- deletion strain. CAN1 and ADE2 were selected for this proof-of-principle experiment because the phenotype caused by CAN1 or ADE2 deletion can be visually screened, giving a fast preliminary evaluation of targeted integration. CAN1 encodes an L-arginine transporter that can also import the toxic compound L-canavanine. Mutants with disrupted CAN1 are able to grow in media containing L-canavanine (Ahmad & Bussey, 1986). ADE2 codes for the enzyme phosphoribosylaminoimidazol carboxylase, which is involved in the biosynthesis of purine nucleotides. ade2 mutants require an external source of adenine and accumulate precursors of purine nucleotides in the vacuole which give colonies a red colour (Zonneveld & Vanderzanden, 1995) (Figure 3B).

101 amdSYM, a new dominant recyclable marker cassette for Saccharomyces cerevisiae.

Figure 3. Sequential gene deletions methodology. A. Cassette design for targeted gene deletion and seamless marker removal. B. Experimental procedure for the sequential deletion of CAN1 and ADE2 in S. cerevisiae by amdSYM recycling. C. Sequencing results of the CAN1 loci of the marker-free IMX206 (can1Δ ade2Δ).

102 Chapter 3

3

Figure 4. Sequential gene deletions of CAN1 and ADE2 using amdSYM in S. cerevisiae. A. The strains IMX168, IMX200 and IMX201 and the parental strain CEN.PK113-7D were grown on SM-Ac, SM-Ac supplemented with adenine and SM supplemented with adenine and L-canavanine. The plates were incubated at 30°C and were read after three days. B. PCR analysis to confirm correct integration of the gene disruption cassettes and their removal at the CAN1 and ADE2 loci. PCR was carried out on reference CEN.PK113-7D, IMX168, IMX200 and IMX201. All PCR’s were performed with the primer pairs CdcFW/CdcRV and AdcFW/AdcRV for CAN1 and ADE2 loci respectively. In the parental strain, amplification of the CAN1 and ADE2 loci generated fragments of 2151bp (a) and 1950bp (d) for CAN1 and ADE2 respectively. PCR on IMX168 DNA generated a fragment of 2835bp (b) due to the incorporation of amdSYM in the CAN1 locus. A short fragment of 333bp (c) was obtained for IMX200 as a result of amdSYM excision from the CAN1 locus. Similarly, the disruption of ADE2 using amdSYM led to a large PCR product of 2691bp in IMX201 (e) while PCR on the ADE2 locus in the marker free strain IMX206 generated a short fragment of 189bp (f). The products obtained were then subjected to agarose gel electrophoresis.

The potential of amdSYM as dominant marker was tested by transforming a deletion cassette to disrupt CAN1 in CEN.PK113-7D. After transformation, cells were grown on synthetic media agar plates containing acetamide as sole nitrogen source (Figure 4A). Targeted gene deletion was confirmed by the ability of single colonies to grow on synthetic media containing L-canavanine (Figure 4A) and by PCR (Figure 4B). The average transformation efficiency was 14 transformants per microgram of DNA, with 100% of the colonies harboring the correct integration. It is important to note that amdSYM did not yield false positives. Although the transformation efficiency reported here for the deletion ofCAN1 may appear

103 amdSYM, a new dominant recyclable marker cassette for Saccharomyces cerevisiae. low as compared to efficiencies reported for other dominant markers such asKanMX (100 transformants/mg DNA) (Guldener et al., 1996), in our experience the targeted region and the sequence of the deletion cassette affects much more the transformation efficiency than the nature of the marker used. To support this it has recently been shown that nucleosome density is a critical factor for transformation efficiency, thereby demonstrating that the localization of a deletion cassette has a strong impact on the transformation efficiency (Aslankoohi et al., 2012). With deletions cassettes using amdSYM-based selection but targeted to other loci than CAN1 we observed transformation efficiencies similar to those reported for classical selection markers (data not shown).

The resulting strain IMX168 was subsequently used to evaluate the potential of marker excision aided by the direct repeats created upon integration (Figure 3B). Under non- selective conditions, i.e. growth in complex media (YPD), mitotic recombination between the direct repeats flankingamdSYM may excise the marker. To select for these recombinants, strain IMX168 (can1D::amdS) was grown overnight in liquid complex media, transferred to synthetic medium containing fluoroacetamide (SM-Fac), then plated on SM-Fac. PCR analysis of the growing colonies confirmed correct marker removal Figure( 4B); the new strain was named IMX200 (can1D). Due to the absence of CAN1, IMX200 (can1D) was able to grow on media containing L-canavanine. As anticipated, it had lost the ability to grow on media containing acetamide as nitrogen source due to the removal of amdSYM (Figure 4A). These results confirmed thatamdSYM can be successfully used as a selection marker in the prototrophic laboratory strain CEN.PK113-7D and that the marker can be removed to avoid protein burden. An important addition is that the design of the deletion cassette allows for marker removal leaving only endogenous sequences (Figure 3C).

A strong feature of recyclable markers is that a single marker is sufficient to perform multiple sequential manipulations in the same strain. The potential of amdSYM for serial deletion was tested by deleting a second gene in the marker-free strain IMX200 (can1D). After transformation with an amdSYM deletion cassette targeted to ADE2 locus, correct transformants could be easily selected for their ability to grow in the presence of L-canavanine, to use acetamide as nitrogen source and for their auxotrophy for adenine and red pigmentation (Figure 4A). The strain IMX201 (can1D ade2D::amdS) was selected using these criteria and correct integration of the deletion cassette was confirmed by PCR (Figure 4B). This second deletion demonstrated that amdSYM is a powerful selection marker for serial gene deletion. IMX201 was further engineered to generate the marker-free strain IMX206 (can1D ade2D) by scarless removal of amdSYM. This was confirmed by sequencing the CAN1 and ADE2 loci in IMX206. Similar to the parental strain CEN.PK113-7D, strain IMX206 (can1D ade2D) was unable to grow on media containing acetamide as sole nitrogen source but, due to the deletions performed, showed a red pigmentation, was not able to grow in absence of adenine and was able to grow on media containing L-canavanine (Figure 4). This marker- and scar-free strain can be subsequently used for additional deletions or other genetic manipulations.

104 Chapter 3

The module amdSYM can be used in a wide range of laboratory, wild and industrial Saccharomyces strains The first condition to useamdSYM as selectable marker is that the parental strain does not have the capability to use acetamide as sole nitrogen source. To verify if other laboratory strains besides the strains of the CEN.PK lineage could be modified using amdSYM as marker, the ability to grow on media containing acetamide as nitrogen source of three popular laboratory strains, namely CBS8066, YSBN (i.e a prototrophic BY strain (Canelas et al., 2010)) and S288c (Mortimer & Johnston, 1986), was tested. None of the laboratory 3 strains were able to use acetamide as sole nitrogen source (Figure 5).

Figure 5. Growth of S. cerevisiae laboratory strains on acetamide as sole nitrogen source. The laboratory strains CEN.PK113-7D,CBS8066, YSBN, S288c and the modified strain IMX168 (can1Δ::amdSYM) were grown on SM-Ac and SM media and incubated at 30°C. The plates were read after three days. Only the strain harboring the amdSYM module, IMX168, was able to grow when acetamide was used as nitrogen source.

To further expand the range of species in which amdS could be used, two brewing strains, the S. cerevisiae Scottish Ale strain and the S. pastorianus lager brewing strain CBS1483, and the wild yeast S. eubayanus CBS12357 were tested for their ability to grow on acetamide as sole nitrogen source. Similarly to the laboratory strains, none of these Saccharomyces species grew with acetamide as sole nitrogen source (Figure 6A). The absence of endogenous acetamidase activity demonstrated that amdSYM could potentially be used as selectable marker in a wide range of laboratory, wild and industrial Saccharomyces strains.

To confirm the universality ofamdSYM as selectable marker, genes were deleted using amdS in the above-mentioned industrial and wild yeast strains. To compensate for the expected lower efficiency of homologous recombination in these non-cerevisiae species, the deletion

105 amdSYM, a new dominant recyclable marker cassette for Saccharomyces cerevisiae. cassettes were designed with longer homologous flanking regions of at least 500bp . The gene HXK1 was deleted in both the laboratory strain CEN.PK113-7D and the S. pastorianus lager brewing strain CBS1483 using cassettes containing amdSYM. S. pastorianus is a hybrid species that contains two subgenomes, one that resembles S. cerevisiae genome and another one similar to S. eubayanus (Libkind et al., 2011). In this study, only the S. cerevisiae allele (Sc.HXK1) was deleted. The strains generated were named IMK468 (Sc.hxk1Δ::amdSYM) for the CEN.PK mutant and IMK470 (Sc.hxk1Δ::amdSYM) for the S. pastorianus mutant. Additionally, the S. cerevisiae Scottish Ale brewing strain and the recently described wild yeast S. eubayanus CBS12357 were also genetically modified usingamdSYM . Making use of amdSYM as marker cassette the gene SbARO80 was deleted in S. eubayanus resulting in the strain IMK474, and ScARO80 was deleted in the S. cerevisiae Scottish Ale strain, generating the strain IMK473. IMK468, IMK470, IMK473 and IMK474 all demonstrated the integration of amdSYM in their genomes by their ability to growth on acetamide as sole nitrogen source (Figure 6A). PCR analysis confirmed the deletion of the targeted genes (Figure 6B). The growth on acetamide plates was slower for industrial strains as compared to CEN.PK-derived deletion mutants, but increasing the acetamide concentration resulted in faster growth, thereby accelerating the screening process (data not shown). Therefore, the amount of acetamide necessary for amdSYM-based strain construction may be strain- dependent and requires optimization.

106 Chapter 3

3

Figure 6: amdSYM as selectable marker for laboratory, wild and industrial Saccharomyces strains. A. The laboratory strain CEN.PK113-7D, the industrial strains CBS1483, Scottish Ale and the wild S. eubayanus CBS12357 and the generated strains IMK468, IMK470, IMK473 and IMK474 were grown on SM-Ac and SM media and incubated at 30°C. The plates were read after five days. B. PCR analysis to confirm correct integration of the gene disruption cassettes was carried out on CEN.PK113-7D, CBS1483, Scottish Ale, CBS12357, IMK468, IMK470, IMK473 and IMK474. Amplification ofSc.HXK1 locus in CEN.PK113-7D and CBS1483 generated fragments of 1837bp (a and c), bigger fragments, 2836bp (b and d) were obtained in the strains IMK468 and IMK470 due to the integration of amdSYM. Amplification of the loci Sc.ARO80 in Scottich Ale and Sb.ARO80 in CBS12357 generated fragments of 4208bp (e) and 4002bp (g) respectively. Confirmation of the deletion ofSc.ARO80 in IMK473 and Sb.ARO80 in IMK474 by PCR generated fragments of 2992bp (f) and 2895bp (h) respectively. The products obtained were then subjected to agarose gel electrophoresis.

107 amdSYM, a new dominant recyclable marker cassette for Saccharomyces cerevisiae.

CONCLUSIONS

When transformed into yeast, A. nidulans amdS conferred the capability to use acetamide as sole nitrogen source. This gain of function allowed the use of amdS as a new dominant marker in S. cerevisiae. In the present work, a new heterologous module, amdSYM, which encompasses the regulatory regions of A. gossypii TEF2 and the A. nidulans gene amdS, was constructed and has been made available for the research community in the widely used plasmid series as pUG-amdSYM. Not only is amdSYM a new dominant marker, but it is also an additional counter-selectable marker in the yeast genetic toolbox. A strong feature of amdSYM is that, contrary to all other counter-selectable markers available, it does not require a specific genetic background for the strain to be selected.

Furthermore, while most marker recycling methods, such as the LoxP-Cre recombinase system, have the disadvantage of leaving scars after each recycling, in the present study marker removal was scarless (Akada et al., 2006), leaving endogenous sequences after marker excision. In principle, amdSYM can be re-used an unlimited number of times, thus enabling multiple modifications without the protein burden that would cause the overexpression of several heterologous markers.

In conclusion, the new marker module amdSYM is an excellent tool for consecutive genetic modifications in the yeastS. cerevisiae and a good alternative to URA3, MET15 or LYS2 with the additional substantial advantage that it is not limited to specific strain backgrounds or to the cerevisiae species. amdSYM has indeed been successfully used as selection marker to perform deletions in the newly discovered wild species of S. eubayanus and in S. cerevisiae and S. pastorianus brewing strains Thanks to this later success in deleting genes from the aneuploid and hybrid genome of S. pastorianus, amdSYM is expected to considerably contribute to the functional analysis of genes in strains with complex genome architecture. As the first dominant recyclable marker,amdSYM opens the door to fast and easy genetic manipulation in Saccharomyces laboratory, wild and industrial strains.

ACKNOWLEDGEMENTS

This work was supported by the Technology foundation STW (Vidi Grant 10776) and by the Kluyver Centre for Genomics of Industrial Fermentation. JTP and J-MD. were also supported by the “Platform Green Synthetic Biology” programme (http://www.pgsb.nl/). The S. cerevisiae Scottish Ale strain was a kind gift of Dr. JM Geertman (Heineken Supply Chain, Zoeterwoude, NL).

108 Chapter 3

REFERENCES

Agha-Mohammadi S, O’Malley M, Etemad A, Baudin A, Ozierkalogeropoulos O, Denouel Wang Z, Xiao X & Lotze MT, 2004. Second- A, Lacroute F & Cullin C, 1993. A simple and generation tetracycline-regulatable promoter: efficient method for direct gene deletion in repositioned tet operator elements optimize Saccharomyces cerevisiae. Nucleic Acids Res 21: transactivator synergy while shorter minimal 3329-3330. promoter offers tight basal leakiness. J Gene Beri RK & Turner G, 1987. Transformation of Med 6: 817-828. Penicillium chrysogenum using the Aspergillus 3 Ahmad M & Bussey H, 1986. Yeast arginine nidulans amdS gene as a dominant selective permease, nucleotide sequence of the CAN1 marker. Curr Genet 11: 639-641. gene. Curr Genet 10: 587-592. Bonnefoy N, Copsey J, Hynes MJ & Davis MA, Akada R, Kitagawa T, Kaneko S, Toyonaga 1995. Yeast proteins can activate expression D, Ito S, Kakihara Y, Hoshida H, Morimura through regulatory sequences of the amdS gene S, Kondo A & Kida K, 2006. PCR-mediated of Aspergillus nidulans. Mol Gen Genet 246: seamless gene deletion and marker recycling in 223-227. Saccharomyces cerevisiae. Yeast 23: 399-405. Brachmann CB, Davies A, Cost GJ, Caputo E, Alani E, Cao L & Kleckner N, 1987. A method Li JC, Hieter P & Boeke JD, 1998. Designer for gene disruption that allows repeated use of deletion strains derived from Saccharomyces URA3 selection in the construction of multiply cerevisiae S288C: a useful set of strains and disrupted yeast strains. Genetics 116: 541-545. plasmids for PCR-mediated gene disruption and Alberti S, Gitler AD & Lindquist S, 2007. A other applications. Yeast 14: 115-132. suite of Gateway (R) cloning vectors for high- Canelas, A. B., Harrison, N., Fazio, A. throughput genetic analysis in Saccharomyces et al., 2010. Integrated multilaboratory cerevisiae. Yeast 24: 913-919. systems biology reveals differences in protein Amberg DC, Botstein D & Beasley EM, 1995. metabolism between two reference yeast strains. Precise gene disruption in Saccharomyces Nature Communications 1:145 doi:10.1038/ cerevisiae by double fusion polymerase chain- ncomms1150 reaction. Yeast 11: 1275-1280. Cebollero E & Gonzalez R, 2004. Comparison Apirion D, 1965. 2-Way selection of mutants of two alternative dominant selectable markers and revertants in respect of acetate utilization for wine yeast transformation. Appl Environ and resistance to fluoro-acetate inAspergillus Microb 70: 7018-7023. nidulans. Genet Res 6: 317-329. Chang TH & Abelson J, 1990. Identification of Aslankoohi E, Voordeckers K, Sun H, Sanchez- a putative amidase gene in yeast Saccharomyces Rodriguez A, van der Zande E, Marchal K & cerevisiae. Nucleic Acids Res 18: 7180-7180. Verstrepen K, 2012. Nucleosomes affect local Chattoo BB, Sherman F, Azubalis DA, Fjellstedt transformation efficiency.Nucleic Acids Res 40: TA, Mehnert D & Ogur M, 1979. Selection 9506-9512. of lys2 mutants of the yeast Saccharomyces Baryshnikova A, Costanzo M, Dixon S, cerevisiae by the utilization of alpha- Vizeacoumar FJ, Myers CL, Andrews B & aminoadipate. Genetics 93: 51-65. Boone C, 2010. Synthetic genetic array (SGA) Corrick CM, Twomey AP & Hynes MJ, 1987. analysis in Saccharomyces cerevisiae and The nucleotide sequence of the amdS gene Schizosaccharomyces pombe. Method Enzymol of Aspergillus nidulans and the molecular 470: 145-179. characterization of 5’ mutations. Gene 53: 63-71.

109 amdSYM, a new dominant recyclable marker cassette for Saccharomyces cerevisiae.

Davidson JF & Schiestl RH, 2000. Mis-targeting Gopal CV, Broad D & Lloyd D, 1989. of multiple gene disruption constructs containing Bioenergetic consequences of protein hisG. Curr Genet 38: 188-190. overexpression in Saccharomyces cerevisiae. Delneri D, Tomlin GC, Wixon JL, Hutter Appl Environ Microb 30: 160-165. A, Sefton M, Louis EJ & Oliver SG, 2000. Gritz L & Davies J, 1983. Plasmid encoded Exploring redundancy in the yeast genome: Hygromycin-B resistance the sequence of an improved strategy for use of the Cre-LoxP Hygromycin-B phosphotransferase gene system. Gene 252: 127-135. and its expression in Escherichia coli and Delpozo L, Abarca D, Claros MG & Jimenez Saccharomyces cerevisiae. Gene 25: 179-188. A, 1991. Cycloheximide resistance as a yeast Grote A, Hiller K, Scheer M, Munch R, cloning marker. Curr Genet 19: 353-358. Nortemann B, Hempel DC & Jahn D, 2005. Drocourt D, Calmels T, Reynes JP, Baron JCat: a novel tool to adapt codon usage of a M & Tiraby G, 1990. Cassettes of the target gene to its potential expression host. Streptoalloteichus hindustanus ble gene for Nucleic Acids Res 33: W526-W531. transformation of lower and higher eukaryotes Guldener U, Heck S, Fiedler T, Beinhauer J to phleomycin resistance. Nucleic Acids Res 18: & Hegemann JH, 1996. A new efficient gene 4009-4009. disruption cassette for repeated use in budding Dunn B & Sherlock G, 2008. Reconstruction of yeast. Nucleic Acids Res 24: 2519-2524. the genome origins and evolution of the hybrid Guldener U, Heinisch J, Koehler GJ, Voss D lager yeast Saccharomyces pastorianus. Genome & Hegemann JH, 2002. A second set of LoxP Res 18: 1610-1623. marker cassettes for Cre-mediated multiple gene Entian KD & Kotter P, 2007. 25 Yeast genetic knockouts in budding yeast. Nucleic Acids Res strain and plasmid collections. Methods in 30: e23-e23. Microbiology, Vol. 36 (Stansfield I & Stark J, Hadfield C, Cashmore AM & Meacock PA, eds.), pp. 629-666. Academic Press, Amsterdam, 1986. An efficient chloramphenicol resistance The Netherlands. marker for Saccharomyces cerevisiae and Ganatra MB, Vainauskas S, Hong JM et al., Escherichia coli. Gene 45: 149-158. 2011. A set of aspartyl protease deficient Hashida-Okado T, Ogawa A, Kato I & strains for improved expression of heterologous Takesako K, 1998. Transformation system for proteins in Kluyveromyces lactis. FEMS Yeast prototrophic industrial yeasts using the AUR1 Res 11: 168-178. gene as a dominant selection marker. FEBS Lett Gatignol A, Baron M & Tiraby G, 1987. 425: 117-122. Phleomycin resistance encoded by the ble gene Henderson RCA, Cox BS & Tubb R, 1985. The from transposon Tn5 as a dominant selectable transformation of brewing yeasts with a plasmid marker in Saccharomyces cerevisiae. Mol Gen containing the gene for copper resistance. Curr Genet 207: 342-348. Genet 9: 133-138. Gietz RD & Woods RA, 2002. Transformation Hoess RH & Abremski K, 1985. Mechanism of of yeast by lithium acetate/single-stranded strand cleavage and exchange in the Cre-Lox carrier DNA/polyethylene glycol method. site-specific recombination system. J Mol Biol Method Enzymol 350: 87-96. 181: 351-362. Goldstein AL & McCusker JH, 1999. Three Hong K & Nielsen J, 2012. Metabolic new dominant drug resistance cassettes for gene engineering of Saccharomyces cerevisiae: a key disruption in Saccharomyces cerevisiae. Yeast 15: cell factory platform for future biorefineries.Cell 1541-1553. Mol Life Sci 69: 2671-2690.

110 Chapter 3

Hynes MJ, 1994. Regulatory circuits of the Nacken V, Achstetter T & Degryse E, 1996. amdS gene of Aspergillus nidulans. Anton Leeuw Probing the limits of expression levels by varying Int J G 65: 179-182. promoter strength and plasmid copy number in Hynes MJ & Pateman JAJ, 1970. Genetic Saccharomyces cerevisiae. Gene 175: 253-260. analysis of regulation of amidase synthesis in Nakayashiki T, Ebihara K, Bannai H & Aspergillus nidulans. 2. Mutants resistant to Nakamura Y, 2001. Yeast [PSI+] “prions” that fluoroacetamide. Mol Gen Genet 108: 107-116. are crosstransmissible and susceptible beyond a Ignea C, Cvetkovic I, Loupassaki S, Kefalas P, species barrier through a quasi-prion state. Mol 3 Johnson CB, Kampranis SC & Makris AM, Cell 7: 1121-1130. 2011. Improving yeast strains using recyclable Nijkamp JF, van den Broek M, Datema E et integration cassettes, for the production al., 2012. De novo sequencing, assembly and of plant terpenoids. Microb Cell Fact 10: analysis of the genome of the laboratory strain doi:10.1186/1475-2859-10-4- Saccharomyces cerevisiae CEN.PK113-7D, a Kelly JM & Hynes MJ, 1985. Transformation model for modern industrial biotechnology. of Aspergillus niger by the amdS gene of Microb Cell Fact 11: doi:10.1186/1475-2859- Aspergillus nidulans. EMBO J 4: 475-479. 11-36- Lackova D & Subik J, 1999. Use of mutated Pronk JT, 2002. Auxotrophic yeast strains in PDR3 gene as a dominant selectable marker in fundamental and applied research. Appl Environ transformation of prototrophic yeast strains. Microb 68: 2095-2100. Folia Microbiol 44: 171-176. Puig S, Ramon D & Perez-Ortin JE, 1998. Langlerouault F & Jacobs E, 1995. A method Optimized method to obtain stable food-safe for performing precise alterations in the yeast recombinant wine yeast strains. J Agr Food genome using a recyclable selectable marker. Chem 46: 1689-1693. Nucleic Acids Res 23: 3079-3081. Read JD, Colussi PA, Ganatra MB & Taron CH, Libkind D, Hittinger CT, Valerio E, Goncalves 2007. Acetamide selection of Kluyveromyces C, Dover J, Johnston M, Goncalves P & lactis cells transformed with an integrative Sampaio JP, 2011. Microbe domestication and vector leads to high-frequency formation of the identification of the wild genetic stock of multicopy strains. Appl Environ Microb 73: lager-brewing yeast. P Natl Acad Sci USA 108: 5088-5096. 14539-14544. Ro DK, Paradise EM, Ouellet M et al., 2006. Losberger C & Ernst JF, 1989. Sequence and Production of the antimalarial drug precursor transcript analysis of the C. albicans URA3 gene artemisinic acid in engineered yeast. Nature 440: encoding Orotidine-5’-Phosphate Decarboxylase. 940-943. Curr Genet 16: 153-157. Sauer B, 1987. Functional expression of the Cre- Mortimer RK & Johnston JR, 1986. Genealogy LoxP site-specific recombination system in the of principal strains of the yeast genetic stock yeast Saccharomyces cerevisiae. Mol Cell Biol 7: center. Genetics 113: 35-43. 2087-2096. Mumberg D, Muller R & Funk M, 1995. Schorsch C, Kohler T & Boles E, 2009. Yeast vectors for the controlled expression Knockout of the DNA ligase IV homolog gene in of heterologous proteins in different genetic the sphingoid base producing yeast Pichia ciferrii backgrounds. Gene 156: 119-122. significantly increases gene targeting efficiency. Curr Genet 55: 381-389.

111 amdSYM, a new dominant recyclable marker cassette for Saccharomyces cerevisiae.

Selten G, Swinkels B & Van Gorcom R, 2000. Vorachek-Warren MK & McCusker JH, 2004. Selection marker gene free recombinant strains: DsdA (D-serine deaminase): a new heterologous A method for obtaining them and the use of MX cassette for gene disruption and selection in these strains. Pat.no.US6051431. Saccharomyces cerevisiae. Yeast 21: 163-171. Shuster JR, Moyer D & Irvine B, 1987. Sequence Wach A, Brachat A, Alberti-Segui C, Rebischung of the Kluyveromyces lactis URA3 gene. Nucleic C & Philippsen P, 1997. Heterologous HIS3 Acids Res 15: 8573-8573. marker and GFP reporter modules for PCR- Singh A & Sherman F, 1974. Association of targeting in Saccharomyces cerevisiae. Yeast 13: methionine requirement with methyl mercury 1065-1075. resistant mutants of yeast. Nature 247: 227-229. Wach A, Brachat A, Pohlmann R & Philippsen Suzuki Y, St Onge RP, Mani R et al., 2011. P, 1994. New heterologous modules for classical Knocking out multigene redundancies via cycles or PCR-based gene disruptions in Saccharomyces of sexual assortment and fluorescence selection. cerevisiae. Yeast 10: 1793-1808. Nat Methods 8: 159-164. Wieczorke R, Krampe S, Weierstall T, Freidel K, Swinkels B, Selten G, Bakhuis J, Bovenberg R Hollenberg CP & Boles E, 1999. Concurrent & Vollebregt A, 1997. The use of homologous knock-out of at least 20 transporter genes amdS genes as selectable markers. Pat. is required to block uptake of hexoses in no.WO/9706261. Saccharomyces cerevisiae. FEBS Lett 464: 123- 128. Szczebara FM, Chandelier C, Villeret C et al., 2003. Total biosynthesis of hydrocortisone from Xie Q & Jimenez A, 1996. Molecular a simple carbon source in yeast. Nat Biotechnol cloning of a novel allele of SMR1 which 21: 143-149. determines sulfometuron methyl resistance in Saccharomyces cerevisiae. FEMS Microbiol Lett Van den Berg MA & Steensma HY, 1997. 137: 165-168. Expression cassettes for formaldehyde and fluoroacetate resistance, two dominant markers Yamashiro CT, Yarden O & Yanofsky C, 1992. in Saccharomyces cerevisiae. Yeast 13: 551-559. A dominant selectable marker that is meiotically stable in Neurospora crassa - the amdS gene van Dijken JP, Bauer J, Brambilla L et al., 2000. of Aspergillus nidulans. Mol Gen Genet 236: An interlaboratory comparison of physiological 121-124. and genetic properties of four Saccharomyces cerevisiae strains. Enzyme Microb Tech 26: Zhang YP, Chen XJ, Li YY & Fukuhara H, 706-714. 1992. LEU2 gene homolog in Kluyveromyces lactis. Yeast 8: 801-804. van Ooyen AJJ, Dekker P, Huang M, Olsthoorn MMA, Jacobs DI, Colussi PA & Taron CH, Zonneveld BJM & Vanderzanden AL, 1995. 2006. Heterologous protein production in the The red ADE mutants of Kluyveromyces lactis yeast Kluyveromyces lactis. FEMS Yeast Res 6: and their classification by complementation 381-392. with cloned ADE1 or ADE2 genes from Saccharomyces cerevisiae. Yeast 11: 823-827. Verduyn C, Postma E, Scheffers WA & Vandijken JP, 1992. Effect of benzoic acid on metabolic fluxes in yeasts - A continuous culture study on the regulation of respiration and alcoholic fermentation. Yeast 8: 501-517.

112 Chapter 3

3

113

CHAPTER 4 Functional analysis and transcriptional regulation of two orthologs of ARO10, encoding broad- substrate-specificity 2-oxo-acid decarboxylases, in the brewing yeast Saccharomyces pastorianus CBS1483

Bolat I Romagnoli G Zhu F Pronk JT Daran JM

FEMS Yeast Research, 2013, vol. 13(6), 505-517 Functional analysis and transcriptional regulation of ARO10

ABSTRACT

The hybrid genomes of Saccharomyces pastorianus consist of subgenomes similar to those of S. cerevisiae and S. eubayanus and impact of the genome structure on flavour production and its regulation is poorly understood. This study focuses on ARO10, a 2 oxo-acid decarboxylase involved in production of higher alcohols. In S. pastorianus CBS1483, four ARO10 copies were identified, three resembledS. cerevisiae ARO10 and one S. eubayanus ARO10. Substrate specificities of lager strain (Lg)ScAro10 and LgSeubAro10 were compared by individually expressing them in a pdc1Δ-pdc5Δ-pdc6Δ-aro10Δ-thi3Δ S. cerevisiae strain. Both isoenzymes catalysed decarboxylation of the 2-oxo-acids derived from branched-chain, sulphur-containing amino-acids isoleucine, and preferably phenylpyruvate. Expression of both alleles was induced by phenylalanine, however in contrast to the S. cerevisiae strain the two genes were not induced by leucine. Additionally LgSeubARO10 showed higher basal expression levels during growth with ammonia. ARO80, which encodes ARO10 transcriptional activator, is located on CHRIV and counts three Sc-like and one Seub-like copies. Deletion of LgSeubARO80 did not affect LgSeubARO10 phenylalanine induction, revealing ‘trans’ regulation across the subgenomes. ARO10 transcript levels showed a poor correlation with decarboxylase activities. These results provide insights on flavour formation in S. pastorianus and illustrate the complexity of functional characterization in aneuploid strains.

116 Chapter 4

INTRODUCTION

The yeast strains used for industrial fermentation of lager beers are taxonomically classified as Saccharomyces pastorianus (Kodama et al., 2005; Rainieri et al., 2006). It has been known for several decades that S. pastorianus strains are natural, aneuploid hybrids of S. cerevisiae and a non-cerevisiae Saccharomyces species (Vaughan Martini & Kurtzman, 1985). Before the first lager yeast genome was completely sequenced, this conclusion was already substantiated by Southern hybridization (Tamai et al., 1998; Yamagishi & Ogata, 1999); PCR/RFLP analysis and partial sequencing (Casaregola et al., 2001; Rainieri et al., 2006) and comparative proteomics (Caesar et al., 2007). The first complete genome 4 sequence of a lager brewing strain, S. pastorianus Weihenstephan 34/70 (WS34/70) was released in 2009. Assembly and interpretation of the WS34/70 genome sequence was based on strong sequence homology of its genome with those of both S. cerevisiae and S. bayanus var. bayanus (Nakao et al., 2009). The recent discovery and genome analysis of S. eubayanus revealed that this cryotolerant yeast, whose genome sequence shows a 99.5% identity with that of the ‘bayanus’ part of lager strains, contributed the non-cerevisiae part of the genome of lager brewing strains of S. pastorianus (Libkind et al., 2011).

The hybrid genomes of lager brewing strains raise intriguing questions about the relative contribution and regulation of genes in the S. cerevisiae and S. eubayanus subgenomes involved in cellular processes relevant for lager brewing. It is, for example, unclear whether the production of specific flavours and off-flavours is predominantly encoded by one of the subgenomes. Furthermore, it has not been investigated whether functions that are encoded on the two subgenomes are due to ‘in trans’ regulation, e.g, by transcriptional regulation of genes in the S. eubayanus subgenome by transcription factors encoded by the S. cerevisiae subgenome.

In beer fermentation, catabolism of wort amino acids and of amino acids generated in yeast metabolism plays a major role in the flavour generation. Fusel alcohols and their esters, which are important amino acid catabolites in brewing yeasts, are key contributors to beer flavour and aroma (Verstrepen et al., 2003; Lodolo et al., 2008). In Saccharomyces species, the Ehrlich pathway (Ehrlich, 1907; Hazelwood et al., 2008) for fusel alcohol production is used for catabolism of amino acids whose carbon skeletons cannot be converted by central metabolism, such as the aromatic, the branched-chain and sulfur-containing amino acids (Dickinson, 1996; Dickinson et al., 2000; Dickinson et al., 1998; Dickinson et al., 1997; Dickinson et al., 2003; Vuralhan et al., 2003). In this pathway, which has been intensively studied in S. cerevisiae (Hazelwood et al., 2008), amino acids are first transaminated to the corresponding α-ketoacids, which are subsequently decarboxylated by thiamin- pyrophosphate-dependent decarboxylases (Romagnoli et al., 2012). The resulting aldehydes are then reduced to their corresponding alcohols.

117 Functional analysis and transcriptional regulation of ARO10

Decarboxylation of the 2-oxo acids derived from amino acid transamination represents the sole irreversible reaction in the Ehrlich pathway (Dickinson et al., 1997). The S. cerevisiae genome harbors five genes that are all showing strong sequence similarity to TPP-dependent decarboxylases. In addition to the three pyruvate decarboxylase genes (PDC1, 5 and 6), these are ARO10 and THI3. Thi3p has so far not been conclusively shown to have a catalytic activity (Romagnoli et al., 2012), while Aro10p is proposed to be the major decarboxylase involved in phenylalanine and tryptophan catabolism (Dickinson et al., 2003; Romagnoli et al., 2012; Vuralhan et al., 2005; Vuralhan et al., 2003). In S. cerevisiae, expression of ARO10 is controlled by Aro80, a positive transcriptional regulator of genes involved in the catabolism of aromatic amino acids (Iraqui et al., 1999).

In view of the central role of the Ehrlich pathway in flavour and aroma generation, ARO10 represents an interesting model to study the contribution and regulation of the S. cerevisiae and S. eubayanus subgenomes. The goals of the present study were to identify the substrate specificities of the Aro10 isoenzymes encoded by the two subgenomes (LgScAro10 and LgSeubAro10), to investigate whether LgScARO10 and LgSeubARO10 are differentially regulated at the transcriptional level and to investigate cross-subgenome transcriptional regulation of LgSeubARO10. To this end, sequences, copy numbers and location of LgScARO10 and LgSeubARO10 genes in the genome of the lager brewing strain S. pastorianus CBS1483 genome were determined. Subsequently, expression levels of the two ARO10 alleles, substrate specificities and kinetic properties of the encoded enzymes were analysed and regulation of LgSeubARO10 was investigated in a LgSeubaro80 deletion mutant of S. pastorianus CBS1483.

MATERIALS AND METHODS

Yeast strains. Yeast strains used in this study are listed in Table 1. The lager brewing strain S. pastorianus CBS1483 (Dunn & Sherlock, 2008) was purchased from Centraalbureau voor Schimmelcultures (http://www.cbs.knaw.nl/collections/). The prototrophic laboratory S. cerevisiae strains CEN.PK113-7D (Nijkamp et al., 2012) and CEN.PK113-5D (ura3- 52) (Entian & Kötter, 2007) were obtained at Euroscarf (http://web.uni-frankfurt.de/fb15/ mikro/euroscarf/) and are isogenic to S. cerevisiae CEN.PK711-7C (Table 1) (Romagnoli et al., 2012; Vuralhan et al., 2005). All strains were maintained as 1 ml aliquots, at -80°C, with 20% v/v glycerol. Stock cultures were grown on complex medium containing 20 g·l-1 glucose, 10 g·l-1 Bacto yeast extract and 20 g·l-1 Bacto peptone, except for strains harbouring URA3-based plasmids, which were grown on synthetic medium (Verduyn et al., 1992) supplemented with 20g·l-1 glucose .

118 Chapter 4

Table 1. Saccharomyces sp. strains used in this study

Strain Genotype Reference CBS1483 S. pastorianus Bottom-fermenting brewing yeast CBSa CEN.PK113-7D S. cerevisiae MATa MAL2-8c SUC2 Euroscarfb CEN.PK113-5D S. cerevisiae MATa MAL2-8c SUC2 ura3 Euroscarfb CEN.PK711-7C S. cerevisiae MATa MAL2-8c SUC2 ura3 pdc1Δ pdc5Δ (Romagnoli et al., pdc6Δ aro10 Δ thi3Δ 2012; Vuralhan et al., 2005) IMZ247 S. cerevisiae MATa MAL2-8c SUC2 ura3 pdc1Δ pdc5Δ This study pdc6Δ aro10 Δ thi3Δ pUDe105(URA3 4

TDH3pr-LgScARO10-CYC1ter) IMZ250 S. cerevisiae MATa MAL2-8c SUC2 ura3 pdc1Δ pdc5Δ This study pdc6Δ aro10 Δ thi3Δ pUDe108(URA3

TDH3pr-LgSeubARO10-CYC1ter) IMK472 S. pastorianus LgSeubaro80Δ::loxP-amdSYM-loxP This study awww.cbs.knaw.nl/ b http://web.uni-frankfurt.de/fb15/mikro/euroscarf/

Plasmid and strain construction Plasmids used in this study are listed in Table 2. The ScARO10 allele was PCR amplified from genomic DNA of the laboratory strain CEN.PK113-7D and from the lager brewing strain CBS1483 using the primer pair XbaI-5’ScARO10 and 3’ScARO10-XhoI (Table 3) (Vuralhan et al., 2005) designed based on the ARO10 sequence found in the Saccharomyces Genome Database (SGD: http://www.yeastgenome.org/). The PCR fragments were digested with XbaI/XhoI and ligated into the p426GPD vector, previously cut with XhoI/SpeI, resulting in pUDe001 and pUDe105. The same fragment was cloned into pCR®II-Blunt-TOPO® resulting in plasmid pTOPOLgScARO10. The two plasmids pUDe001, pUDe105 were transformed into the quintuple deletion strain CEN.PK711-7C (pdc1,5,6Δ aro10Δ thi3Δ) (Romagnoli et al., 2012; Vuralhan et al., 2005) resulting in IMZ002 and IMZ247. The LgSeubARO10 allele was amplified by PCR from chromosomal DNA of CBS1483 and primers 5’SeubARO10/ 3’SeubARO10 that were designed based on the publicly available sequences of S. pastorianus Weihenstephan 34/70 (Nakao et al., 2009) and S. bayanus MCYC 623 (Cliften et al., 2003). The LgSeubARO10 fragment was directionally cloned into the pENTR™ D-TOPO®vector (Life Technologies Europe BV, Bleiswijk, The Netherlands) resulting in the entry clone pENTRLgSeubARO10. This was recombined with the destination vector pAG426GPD- ccdB via the Gateway LR recombination system (Alberti et al., 2007), yielding the expression vector pUDe108. Transformation of this plasmid into strain CEN.PK711-7C yielded the strain IMZ250. Both LgScARO10 and LgSeubARO10 are flanked byTDH3 promoter and CYC1 terminator. To confirm their sequences in lager brewing strain CBS1483, the two ARO10 alleles cloned in TOPO vectors as pTOPOLgScARO10 and pENTRLgSeubARO10 were sequenced by Sanger sequencing at BaseClear BV (Leiden). Yeast transformations were performed with LiAc/ssDNA/PEG method as presiously described in (Gietz & Woods, 2002).

119 Functional analysis and transcriptional regulation of ARO10

Deletion of the LgSeubARO80 allele in the lager brewing yeast CBS1483 yielded strain IMK472. In the construction of this strain, the amdSYM marker cassette (Solis-Escalante et al., 2013) was used. Plasmid pUG-amdSYM (Solis-Escalante et al., 2013) was used for the construction of the loxP-amdSYM-loxP deletion cassette followed by 500bp elongation of the homologous flanks via double fusion PCR protocol(Amberg, 1995). The primer sets used for this long-flank knockout cassette were SeubARO80dcamdSFW/SeubARO80dcamdSRV together with pUG-amdSYM, SeubARO80g5’FW/SeubARO80g5’RV and SeubARO80g 3’FW/SeubARO80g 3’RV (Table 3) on CBS1483 genomic DNA as template. Correct integration of the deletion cassette was verified with primers SeubARO80cdc 5’FW/ SeubARO80cdc 3’FW.

Table 2. Plasmids used in this study

Plasmid Characteristic Reference pCR®II-Blunt-TOPO® Gateway entry plasmid Life Technologies Europe BV pENTR™D-TOPO® Gateway entry plasmid Life Technologies Europe BV

p426GPD URA3 2µ ORI TDH3pr – CYC1ter (Mumberg et al., 1995)

pAG426GPD URA3 2m ORI TDH3pr-ccdB-CYC1ter Alberti (Alberti et al., 2007) URA3 2µ ORI TDH3 –ScARO10- (Romagnoli et al., 2012; pUDE101 pr CYC1ter Vuralhan et al., 2005) URA3 2µ ORI, TDH3 –LgScARO10- pUDE105 pr This study CYC1ter URA3 2µ ORI TDH3 –LgSeubARO10- pUDE108 pr This study CYC1ter pTOPO LgScARO10 Gateway entry plasmid, LgScARO10 This study pENTRLgSeubARO10 Gateway entry plasmid, LgSeubARO10 This study pUG-amdSYM loxP-amdS-loxP cassette Solis-Escalantes et al., 2013

Growth conditions and media Carbon-limited chemostat cultivation was performed in 2.0 l bioreactors (Applikon, Schiedam, the Netherlands) with a working volume of 1.0 l, at 30°C, pH 5.0. Strains S. pastorianus CBS1483, S. cerevisiae CEN.PK113-7D and IMK472 (LgSeubaro80D) were grown anaerobically, at a dilution rate of 0.1h-1, while the TPP dependent 2-oxo acid decarboxylase null- strains IMZ247 and IMZ250 were cultivated aerobically at a dilution rate of 0.05 h-1. Anaerobic and aerobic conditions were obtained by sparging nitrogen or air, respectively, through the cultures at a flow rate of 0.5 l·min-1. A synthetic medium was -1 -1 -1 used that contained 3 g·l KH2PO4, 0.5 g l MgSO4.7H2O, 1 ml·l trace element solution and 1 ml·l-1 vitamin solution as previously described (Verduyn et al., 1992). As carbon source 25 g·l-1 glucose was added for CBS1483, CEN.PK113-7D, IMK472 and 5.7 g·l-1 ethanol for quintuple deletion strains. Different nitrogen sources were used as indicated, at the following -1 -1 -1 concentrations: 5 g·l (NH4)2SO4, 10.0 g·l L-leucine, 12.5 g·l L-phenylalanine. When using amino acids as the nitrogen source, the synthetic medium was supplemented with 6.6 g·l-1

K2SO4. (Boer et al., 2007). In case of anaerobic cultures, Tween-80 and ergosterol were added

120 Chapter 4 to the medium as growth factors (Verduyn et al., 1990). Steady state was assumed to be reached after at least five volume changes, when glucose concentration, culture dry weight, and carbon dioxide production rate changed by less than 2% during one volume change.

Analytical methods Extracellular metabolites such as ethanol, glycerol, glucose, lactate, phenylethanol and phenylpyruvate were determined in supernatants by HPLC analysis with a Bio-Rad Aminex HPX-87H column, at 60°C, eluted with 5 mM sulfuric acid at a flow rate of 0.6 ml·min-1. Table 3. Oligonucleotide primers used in this study 4 Primer Sequence 5’→ 3’ Plasmid construction XbaI-5’ScARO10 GGTCTAGAATGGCACCTGTTACAATTGAAAAG 3’ScARO10-XhoI GGCTCGAGCTATTTTTTATTTCTTTTAAGTGCCGC 5’SeubARO10 CACCATGGCACCTGTTACGATTGACACG 3’SeubARO10 GCACCTTCTCATTTCTTGTTTCTCTTAAGTGCAGCCG Deletion cassette construction and verification AGCAAGCTAGTCAAAATATTTGCTCTCCGCATGATATAAT- SeubARO80dcamdSFW TACTTTCAGTATCGTTGTCCCCAGCTGAAGCTTCGTACGC TAGTAATCAATCATTTATCTAATTAAACATTCCTTTTCTCTAAT- SeubARO80dcamdSRV TTTATGTGTGAGGGGCGCATAGGCCACTAGTGGATCTG SeubARO80g5’FW GACATTGATGACAATGGTAGTG SeubARO80g5’RV GGACAACGATACTGAAAGTAAT SeubARO80g 3’FW GCCCCTCACACATAAAATTAGAG SeubARO80g 3’RV GTGTGGCTTGTACCACAAGA SeubARO80cdc 5’FW GTCATGCAGGCTCTTCATTG SeubARO80cdc 3’FW TATCCCTCACGTGAATTTAAACC Quantitative PCR – gene expression and copy number ScARO10spFW ATATTCGCCTTGTGGACACTAGGC ScARO10spRV CCATTGGTAGGATCTTCTAACCG SeubARO10spFW GGCGAAGAACAGATATTAGAAGGCG SeubARO10spRV GACTTTTTGTAAGTGGACTTGTGTG ACT1spFW GGCTTCTTTGACTACCTTCCA ACT1spRV AGAAACACTTGTGGTGAACGA AmpFw CCAGTGCTGCAATGATACC AmpRv GCGGATAAAGTTGCAGGAC ScARO10sp2FW ATTATATTCGCCTTGTGGACACTA ScARO80spFW ACGACCACCCTTATTAAGAGAT ScARO80spRV TTAATTTCCTCCGTCAAAGGC SeubARO80spFW ACATCAAGAACTGGTATAGCCC SeubARO80spRV GGATCATCACATGTTGCCTC

121 Functional analysis and transcriptional regulation of ARO10

Lactate, phenylethanol and phenylpyruvate were detected by a Waters 2487 dual-wavelength absorbance detector at 214 nm. Ethanol, glycerol, glucose were detected with a Waters 2410 refractive index detector. Biomass dry weight was determined as described in (Verduyn et al., 1990).

Pulsed-field gel electrophoresis and southern hybridization For the identification and localization of ARO10 alleles within the brewing strain, CHEF electrophoresis and Southern blotting were performed. Chromosomes of CBS1483 and CEN.PK113-7D were separated on 1% Megabase DNA agarose gel in 0.5xTBE buffer (1 mM EDTA, 45 mM Tris-Borate, pH 8.0) during a 44 h run programme on the CHEF-DR3 system (Bio-Rad Laboratories B.V. Veenendaal, The Netherlands). DNA was subsequently transferred onto Hybond-N+ nylon membranes (Amersham Biosciences, Piscataway, NJ) and linked by 2 min exposure to UV light. Probe labelling, Southern blotting and signal detection were performed with Gene Images AlkPhos Kit, CPD Star detection reagent and Hyperfilm ECL (Amersham Biosciences). The DNA probes used for Southern blotting were amplified from CBS 1483 using the same primers as those used for plasmid construction.

In vitro enzyme assays Phenylpyruvate decarboxylase activity was measured in vitro using cell extract prepared from samples taken from steady-state chemostat cultures. Cell extracts were prepared as previously described (Romagnoli et al., 2012). Enzyme activity was measured at 30°C by coupling two reactions: (i) decarboxylation of a 2-oxo-acid to the corresponding aldehyde catalysed by the phenylpyruvate decarboxylase and (ii) aldehyde oxidation coupled with the reduction of NAD+ to NADH by an added excess of yeast aldehyde dehydrogenase. The increase in absorption was measured with spectrophotometer Hitachi U-3010 at 340nm -1 (εNADH=6.3mM ). The assay mixture contained 100 mM potassium-phosphate buffer (pH 7.0), 0.2 mM thiamine pyrophosphate, 5.0 mM magnesium chloride, 15.0 mM pyrazole, 2.0 mM NAD+, 1.75 U·ml-1 aldehyde dehydrogenase (Sigma-Aldrich Chemie B.V.Zwijndrecht, Netherlands) dissolved in 1mM dithiothreitol, in a total volume of 1ml. The reaction was started by the addition of one of the following oxo-acids: 10 mM α-ketoisovalerate, 10 mM α-ketoisocaproate, 10 mM α-ketomethylvalerate, 10 mM 4-methylthio-2-oxobutanoate, 5 mM phenylpyruvate. All assays were performed at two concentrations of cell extract to check that activities were proportional to the amount of extract added. Protein concentrations in cell extracts were determined according to the Lowry method (Lowry et al., 1951) using bovine serum albumin as standard. The kinetic parameters Km and Vmax were estimated from assays performed substrate concentrations ranging from 0 to 5 mM for phenylpyruvate and 0 to 40 mM for the other substrates. These kinetic assays were performed in 96 wells plates, using a 300 µl assay mixture, with a TECAN GENios Pro robot (Tecan, Giessen, the Netherlands). The data retrieved were fitted with GraphPad Prism 4.0 (GraphPad Software, Inc., La Jolla, CA), using a nonlinear regression fit to the Michaelis-Menten equation.

122 Chapter 4

Gene expression – real-time QPCR. The relative expression levels of ARO10 alleles were determined by performing real-time quantitative PCR, using ScACT1 as reference gene. Samples corresponding to 240 mg of cells were taken from chemostat cultures during steady state and plunged directly in liquid nitrogen. Frozen samples were broken into small pieces and distributed into 50 ml tubes. After thawing, a cold centrifugation step followed (5000 g, 4 min, 0 °C) and the pellet was re-suspended in 1800 µl ice-cold AE buffer (50 mM sodium acetate, 10 mM EDTA, pH 5.0), 1800 µl Acid phenol/Chloroform/IAA mix and 180µl 10% SDS. The tubes were then placed in a water bath for 5 min at 65°C and subsequently aliquoted in 8 tubes of 800 µl. Total RNA was extracted from a single tube via the phenol/ chloroform method (Schmitt, 1990). For DNase I treatment of the samples, the RNeasy 4 Mini kit and RNase-Free DNase set (Qiagen Benelux B.V., Venlo The Netherlands) were used and quality control was performed with Agilent RNA 6000 Nano Reagents (Agilent Technologies Netherlands B.V., Amstelveen, The Netherlands.) and the Agilent Bioanalyser 2100 according to manufacturer’s instruction. For cDNA synthesis the M-MLV Reverse Transcriptase protocol with random primers was used (Life Technologies Europe BV). Three dilution of cDNA samples were prepared: 10 – 1.0 – 0.1 ng·µl-1 using a pipetting robot (Qiagen) and Rotor-Gene SYBR Green I PCR kit (Qiagen). Real-time QPCR has been performed in a Qiagen-Rotor Gene Q.

Dedicated primers were designed for specific binding to theS. cerevisiae-like allele (LgScARO10 and ScARO10) (ScARO10spFW/ScARO10spRV) and to the S. eubayanus- like allele (LgSeubARO10 and SeubARO10) (SeubARO10spFW/SeubARO10spRV) (Table 3). Prior use the specificity of the primer pairs was evaluated by performing PCR on genomic DNA of S .cerevisiae CEN.PK113-7D and S. eubayanus CBS12357. The primer pair (ScARO10spFW/ScARO10spRV) formed a PCR product with CEN.PK113- 7D genomic DNA only and the primer pair (SeubARO10spFW/SeubARO10spRV) formed a PCR product with CBS12357 genomic DNA only. Both primer pairs could PCR amplify a band with S. pastorianus CBS1483 genomis DNA. The primer pair ACT1spFW/ ACT1spRV corresponded to the reference gene ScACT1. As perfomed for the ARO10 PCR primers, we showed that ACT1 primer pair was specific to the S. cerevisiae ScACT1 and LgScACT1 alleles only. Each sample was analysed in at least triplicate from at least two different chemostats. The expression of each allele relative to ScACT1 was calculated as Amplification^(ScACT1 take-off – sample take-off). A reaction that is 100% efficient would give an amplification value of 2 for every sample, meaning that the amplicon doubled in every cycle. The amplification efficiencies of the reactions were calculated to be 1.75±0.01, 1.80±0.03, 1.79±0.01, 1.82±0.01, 1.81±0.01 and 1.79±0.02 for PCR performed with the primer pair ScARO10sp2FW/ScARO10spRV, specific forLgScARO10 , SeubARO10spFW/ SeubARO10spRV specific for LgSeubARO10, ScARO80spFW/ ScARO80spRV specific for LgScARO80 and SeubARO80spFW/SeubARO80spRV specific for LgSeubARO80, ACT1spFW/ACT1spRV, specific for ScACT1 and AmpFw/AmpRv specific for the ampicillin-

123 Functional analysis and transcriptional regulation of ARO10 resistance gene respectively. The take-off represents the cycle at which the second derivative is at 20% of the maximum level, indicating the end of the noise and the transition to the exponential phase. The average take-off was calculated for each gene of interest.

To identify the copy number of each of the ARO10 alleles in CBS1483, quantitative PCR was employed on genomic DNA, using pUC19 DNA as an internal standard. Genomic DNA of lager brewing strain and the DNA corresponding to plasmid pUC19 previously cut with restriction enzyme HindIII, were mixed in a 1:1 molar ratio. Five different primer sets were used: AmpFw/AmpRv specific for the ampicillin-resistance gene located on pUC19, ScARO10sp2FW/ScARO10spRV specific for LgScARO10, SeubARO10spFW/ SeubARO10spRV specific forLgSeubARO10 , ScARO80spFW/ ScARO80spRV specific for LgScARO80 and SeubARO80spFW/ SeubARO80spRV specific forLgSeubARO80. The expression of each allele relative to the Amp-resistance gene was calculated with the take-off values, as previously described. All analyses were performed in duplicate, with two different dilutions.

RESULTS

Anaerobic growth of S. pastorianus and S. cerevisiae with different nitrogen sources Growth of the aneuploid lager brewing strain CBS1483 (S. pastorianus) and the laboratory strain CEN.PK113-7D (S. cerevisiae) was studied in anaerobic, glucose-limited chemostat cultures on three different nitrogen sources: ammonium, leucine and phenylalanine. Maximum anaerobic specific growth rates with these nitrogen sources were estimated from the CO2 output of the batch cultures that preceded the continuous-feed phase of the chemostat cultures. With phenylalanine as the sole nitrogen source, estimated growth rates of strains CBS1483 and CEN.PK113-7D were comparable (0.25 ± 0.01 and 0.29 ± 0.06 h-1, respectively) while on leucine, the laboratory strain grew faster than the lager brewing strain (growth rates of 0.31 ± 0.01 and 0.22 ± 0.01 h-1, respectively). Strikingly, when ammonium was used as the nitrogen source, the estimated growth rate of the brewing strain was less than half of that of the laboratory strain (0.16 ± 0.02 and 0.37 ± 0.07 h-1, respectively). This may reflect the fact that amino acids rather than free ammonia are the most abundant nitrogen sources in wort (Reed & Nagodawithana, 2001) and/or that some amino acids are better nitrogen than ammonia as a base-substrate (Jones M. & Pierce, 1964; Kunze et al., 1996).

In anaerobic cultures grown at a dilution rate of 0.10 h-1, biomass yields on glucose of the brewing strain CBS1483 were identical in cultures grown on each of the three nitrogen sources (Table 4). In the laboratory strain CEN.PK113-7D, the biomass yield on glucose in ammonia-grown cultures was slightly higher than in cultures grown with leucine or phenylalanine as the nitrogen source (Table 4). Specific rates of ethanol and CO2 production correlated well with the specific rates of glucose uptake. In anaerobic yeast cultures, glycerol

124 Chapter 4 serves as a redox sink to reoxidize NADH generated in biosynthetic reactions (van Dijken & Scheffers, 1986). Although biomass yields on glucose of the lager brewing strain were lower than those of the laboratory strain, glycerol yields on glucose of CBS1483 were higher than those of the laboratory strain for all three nitrogen sources (Table 4). A higher glycerol yield in anaerobic cultures has also been reported for S. kudriavzevii (Arroyo-Lopez et al., 2010), indicating that glycerol metabolism may be different in Saccharomyces species. Specific rates of lactate production were also higher in the brewing strain.

In S. cerevisiae, production of lactate results from detoxification of methylglyoxal, which is formed from dihydroxyacetone phosphate in a chemical reaction (Aguilera & Prieto, 2001). 4 Since dihydroxyacetone phosphate is also a precursor for glycerol formation, the differences between the brewing and lab strains with respect to glycerol and lactate formation may be connected.

Production of phenylethanol, the major catabolite of phenylalanine conversion via the Ehrlich pathway in anaerobic yeast cultures (Etschmann et al., 2002; Vuralhan et al., 2005), was only detectable in cultures grown with phenylalanine as the nitrogen source. Similarly, iso-amyl alcohol, which is formed from either isoleucine or leucine via an Ehrlich-type conversion (Hazelwood et al., 2008), was only observed at very low rates in cultures grown with ammonium or phenylalanine as nitrogen source, while high rates of formation were observed when leucine was the nitrogen source. In anaerobic glucose-limited cultures of S. cerevisiae, the nitrogen content of the biomass is approximately 7 % (Lange & Heijnen, 2001). At a specific growth rate of 0.10 h-1, this corresponds with a rate of nitrogen assimilation of 0.5 mmol N (g biomass)-1 h-1. The observed specific rates of isoamylalcohol and phenylethanol production by the leucine and phenylalanine - grown cultures, were close to this value (Table 4), indicating that uptake and catabolism of these amino acids was closely regulated to meet the cellular nitrogen demand. A notable exception was the growth of the laboratory strain in anaerobic cultures with phenylalanine as nitrogen source, in which the phenylethanol production was 40% higher (0.7 mmol (g biomass)-1 h-1) than the theoretical nitrogen assimilation rate (0.5 mmol N (g biomass)-1 h-1). This suggested that, in this strain, phenylalanine uptake is partially uncoupled from the cellular nitrogen demand.

125 Functional analysis and transcriptional regulation of ARO10

Table 4. Physiological parameters measured during cultivation of S. pastorianus CBS1483 and S. cerevisiae CEN.PK113-7D in anaerobic, carbon-limited chemostat cultures, pH 5, at 30°C, with a dilution rate of 0.1h-1. (data represent the average and SD of data from at least two independent steady- state chemostat cultivations). Carbon recoveries were in all cultures in the interval 101±2%.

S. pastorianus: CBS1483 S. cerevisiae: CEN.PK113-7D Phenyl- Phenyl- Ammonium Leucine Ammonium Leucine alanine alanine

Ysx (g/g) 0.07 ± 0.0 0.07 ± 0.0 0.07 ± 0.0 0.09 ± 0.0 0.08 ± 0.0 0.08 ± 0.0

qglucose (mmol/g/h) 8.0 ± 0.2 8.9 ± 0.1 8.4 ± 0.2 6.2 ± 0.2 7.0 ± 0.4 7.0 ± 0.2

qethanol (mmol/g/h) 12.2 ± 0.3 14.0 ± 0.0 13.0 ± 0.0 10.4 ± 0.4 11.4 ± 0.5 12.0 ± 0.3

qCO2 (mmol/g/h) 13.8 ± 1.0 14.0 ± 0.9 13.8 ± 0.5 10.7 ± 0.4 12.3 ± 0.7 12.8 ± 0.8

qglycerol (mmol/g/h) 1.4 ± 0.0 1.3 ± 0.1 1.9 ± 0.0 0.7 ± 0.0 1.0 ± 0.1 1.2 ± 0.0

qlactate (mmol/g/h) 0.3 ± 0.1 0.6 ± 0.0 0.1 ± 0.0 0.06 ± 0.0 0.1 ± 0.0 0.05 ± 0.0

qphenylethanol (mmol/g/h) N.D. N.D. 0.5 ± 0.0 N.D. N.D. 0.7 ± 0.0

qisoamyl-alcohol (mmol/g/h) 0.01 ± 0.0 0.40 ± 0.0 0.01 ± 0.0 0.01 ± 0.0 0.20 ± 0.1 0.01 ± 0.0

Sequencing and physical mapping of phenylpyruvate decarboxylase alleles To characterize the phenylpyruvate decarboxylase genes in the hybrid genome of S. pastorianus CBS1483, the S. eubayanus-like and S. cerevisiae-like alleles of ARO10 (LgSeubARO10 and LgScARO10, respectively) were PCR-amplified from genomic DNA. For theLgScARO10 allele the primers were designed based on the published sequence of S. cerevisiae S288C (Goffeau et al., 1996) (http://www.yeastgenome.org/ ) and for the LgSeubARO10 allele the primer sequences were based on the sequence of the brewing strain Weihenstephan 34/70 (Nakao et al., 2009). The resulting fragments were sequenced and revealed that the coding regions of the two ARO10 alleles in CBS1483 were of identical length (1908 bp), with LgScARO10 having the same sequence as the one in WS34/70 strain and LgSeubARO10 displaying 2 point mutations compared to the WS34/70 LgSeubARO10 allele. (Table 5, Figure 1A). In both brewing strains, the sequences of the LgSeubARO10 and LgScARO10 alleles showed a sequence identity of 80 % at the DNA level and 84 % at the protein level.

126 Chapter 4

Table 5. DNA sequence identity of ARO10 alleles DNA sequences (%) among different Saccharomyces strains and species.

S288C CBS7001 CBS 12357 WS34/70 CBS1483 S. cerevisiae S. bayanus S. eubayanus S. pastorianus S. pastorianus Sc Sb Seub LgSc LgSeub LgSc LgSeub ARO10 ARO10 ARO10 ARO10 ARO10 ARO10 ARO10 S288C Sc 100 79 80 99 80 99 80 S. cerevisiae ARO10 CBS7001 Sb - 100 93 80 93 80 93 S. bayanus ARO10 CBS 12357 Seub 4 - - 100 80 99 80 99 S. eubayanus ARO10 LgSc - - - 100 80 100 80 WS34/70 ARO10 S. pastorianus LgSeub - - - - 100 80 99 ARO10 LgSc - - - - 100 80 CBS1483 ARO10 S. pastorianus LgSeub ------100 ARO10

To investigate the physical localization of both ARO10 alleles in CBS1483, chromosomes were separated by contour-clamped homogenous electric field electrophoresis (CHEF), blotted to nylon membranes and hybridized with probes specific forLgSeubARO10 and LgScARO10. The results clearly indicated that both alleles were located on CHR IV (Figure 1B). In order to estimate the copy number of each of the ARO10 alleles in CBS1483, quantitative PCR was employed on genomic DNA, using pUC19 DNA as an internal standard. This experiment indicated that the LgSeubARO10 and LgScARO10 alleles were present in a 1:3 ratio in the genome of the brewing strain (Figure 1C). Given the range of ploidies typically observed in brewing strains (Aigle et al., 1983), this suggested that CBS1483 carries a single copy of LgSeubAR010 and three copies of LgScARO10. Consistent with this interpretation, a single knock-out of the S. eubayanus-like allele of ARO80 which is also located on the S. eubayanus type CHR IV (Libkind et al., 2011) and was also present in a 1:3 ratio to the LgScARO80 allele in CBS1483 (Figure 1C), was sufficient to delete this allele from the genome of CBS1483.

127 Functional analysis and transcriptional regulation of ARO10

Figure 1. Genetic relationship, localization and copy number of ARO10 alleles (A) Phylogenetic tree showing the evolutionary relationship of ScARO10 and SeubARO10 alleles among various Saccharomyces species. (B) Karyotypes (G) and Southern blots (B) corresponding to lager brewing strain CBS1483 (L) and the laboratory strain CEN.PK113-7D (C). Probes for LgScARO10 (B1) and LgSeubARO10 (B2) were prepared with primers listed in Table 3. The black boxes indicate the hybridized chromosome. The S. cerevisiae probe hybridizes on CHR IV of both strains tested (B1), while the S. eubayanus probe hybridizes only on CHR IV corresponding to lager brewing strain (B2). (C) The copy number of LgScARO10 and LgScARO80 alleles relative to LgSeubARO10 and LgSeubARO80 determined via quantitative PCR. Technical duplicates were analysed using the genome of pUC19 plasmid as internal control.

Substrate specificity of phenylpyruvate decarboxylase alleles To characterize the substrate specificity of each ARO10 allele in S. pastorianus CBS1483, LgScARO10 and LgSeubARO10 were individually expressed in a S. cerevisiae pdc1Δ pdc5Δ pdc6Δ aro10Δ thi3Δ strain (CEN.PK711-7C), on a multi-copy vector and under the control of the constitutive TDH3 promoter. Since pyruvate-decarboxylase-negative strains cannot grow on glucose as sole carbon source (Flikweert et al., 1999), these strains were grown in aerobic ethanol-limited chemostat cultures, at a dilution rate of 0.05 h-1 (Romagnoli et al., 2012; Vuralhan et al., 2005). Since phenylalanine has been implicated in post-transcriptional regulation of ARO10 (Vuralhan et al., 2005), it was used as nitrogen source in these cultures. It is worth mentioning that the strain CEN.PK711-7C that bears

128 Chapter 4 the deletions of the five TPP-dependent 2-oxo-acid decarboxylases was unable to grow in ethanol limited chemostat with phenylalanine as sole nitrogen source and supplemented with uracil (Vuralhan et al., 2005). However, this cultivation setup enabled analysis of the substrate specificity and kinetic properties (Km, Vmax) of the decarboxylases encoded by each ARO10 allele. To this end, activity of LgScAro10 and LgSeubAro10 in cell extracts was tested with five different substrates (ketoisovalerate, ketoisocaproate, ketomethylvalerate, keto-methylthio-oxobutanoate and phenylpyruvate) over a wide range of concentrations (0 – 40mM) (Table 6). For three of these substrates ketoisovalerate, ketoisocaproate and ketomethylvalerate, 4 activity of both LgScAro10 and LgSeubAro10 increased linearly over the entire range of concentrations tested, without reaching saturation. In contrast, the activity of both decarboxylases with keto-methylthio-oxobutanoate and phenylpyruvate showed typical

Michaelis-Menten saturation kinetics. For both isoenzymes, the Km for phenylpyruvate was an order of magnitude lower than for keto-methylthio-oxobutanoate, while Vmax values were about 50 % higher for phenylpyruvate. No significant difference was observed for the kinetic parameters of the two isoenzymes for these two substrates (Table 6).

Table 6. Kinetic parameters of ScAro10 and SbAro10 from S. pastorianus CBS1483, measured in cell extracts of two S. cerevisiae strains expressing these decarboxylases in a ‘zero-decarboxylase’ (pdc1 pdc5 pdc6 aro10 thi3) background. Strains were grown in aerobic, ethanol-limited chemostats with phenylalanine as sole nitrogen source (dilution rate 0.05 h-1) The results are expressed as the average ± mean deviation of experiments on two independent cultures, with 4 technical duplicates each.

IMZ247 IMZ250

pdc1 pdc5 pdc6 aro10 thi3 pdc1 pdc5 pdc6 aro10 thi3 Genotype LgScARO10 LgSeubARO10

Vmax Vmax Km Km Substrate µmol·min-1·(mg µmol·min-1·(mg protein-1) mM protein-1) mM

Ketoisovalerate >60 >20a >60 >20a

Ketoisocaproate >60 >20a >60 >20a

Ketomethylvalerate >60 >20a >60 >20a

Keto-methylthio- 61 ± 10 2.7 ± 1.3 82 ± 6 2.4 ± 0.5 oxobutanoate

Phenylpyruvate 112 ± 8 0.2 ± 0.0 130 ± 6 0.1 ± 0.0

To further characterize the LgScARO10 and LgSeubARO10 alleles, enzyme activity in cell extracts was compared for all five 2-oxo acid substrates Table( 7). Saturating substrate concentrations were used for phenylpyruvate (5mM) and keto-methylthio-oxobutanoate (10mM), while for the other substrate a fixed concentration of 10 mM was used. The

129 Functional analysis and transcriptional regulation of ARO10 results indicated that both LgScAro10 and LgSeubAro10 are broad substrate-specificity decarboxylases with phenylpyruvate as preferred substrate (Table 7). While, for most substrates, activities of the two isoenzymes were similar, the activity of LgSeubAro10 with ketoisovalerate was ca. 2-fold higher than that of LgScAro10 (Table 7).

Table 7. Activities of of LgScAro10 and LgSeubAro10 from S. pastorianus CBS1483, measured in cell extracts of two S. cerevisiae strains expressing these decarboxylases in a ‘zero-decarboxylase’ (pdc1 pdc5 pdc6 aro10 thi3) background. Activities were measured at concentration of 10 mM for each substrate, with the exception of phenylpyruvate, for which a concentration of 5 mM was used. Numbers in brackets indicate the activity with each substrate relative to that observed with phenylpyruvate. Strains were grown in aerobic, ethanol-limited chemostats with phenylalanine as sole nitrogen source (dilution rate 0.05 h-1) The results are expressed as the average ± mean deviation of experiments on two independent cultures, with 4 technical duplicates each.

IMZ247 (LgScARO10) IMZ250 (LgSeubARO10)

Substrate Activity (µmol·min-1·(mg protein)-1)

Ketoisovalerate 40 ± 10 (37%) 89 ± 19 (68%) Ketoisocaproate 64 ± 16 (61%) 75 ± 9 (57%) Ketomethylvalerate 60 ± 21 (56%) 91 ± 4 (70%) Keto-methylthio-oxobutanoate 42 ± 12 (39%) 71 ± 3 (54%) Phenylpyruvate 106 ± 35 (100%) 131 ± 9 (100%)

Nitrogen-source dependent transcriptional regulation of ARO10 alleles To investigate nitrogen-source-dependent transcriptional regulation of (Lg)ScARO10 and LgSeubARO10, the lager brewing strain CBS1483 and the laboratory strain CEN.PK113- 7D were grown in anaerobic glucose-limited chemostat cultures, using ammonia, leucine or phenylalanine as nitrogen source. To dissect transcriptional responses of LgScARO10 and LgSeubARO10, specific PCR primers were designed and transcript levels were determined by quantitative PCR, using ScACT1 as a reference.

When ammonia was used as nitrogen source, very low expression levels of LgScARO10 were observed in CBS1483 as well as in CEN.PK113-7D (ScARO10) (Table 8). In the laboratory strain, use of leucine as a nitrogen source led to an over 10-fold induction of ScARO10, while leucine did not induce expression of LgScARO10 in the lager brewing strain. In both strains, an over 50-fold induction of (Lg)ScARO10 was observed with phenylalanine as the nitrogen source. The LgSeubARO10 allele in the brewing strain was regulated differently from the ScARO10 allele (Table 8). In particular, it showed high basal transcript levels in cultures grown with ammonia or leucine as the nitrogen source. In cultures grown with phenylalanine as nitrogen source the transcript level of LgSeubARO10 was only 8-fold higher than in cultures grown on ammonia or leucine. Interestingly, in phenylalanine-grown cultures of CBS1483, the ratio of the transcript levels of LgScARO10 and LgSeubARO10 was 3:1 (Table 8), which is consistent with the number of copies identified for each allele Figure( 1).

130 Chapter 4

Table 8. Transcript levels of ARO10 alleles in anaerobic, carbon-limited chemostat cultures (dilution rate of 0.10 h-1) of three yeast strains: the laboratory strain S. cerevisiae CEN.PK113-7D, the lager brewing strain S. pastorianus CBS1483 and its LgSeubaroΔ knock-out strain IMK472, grown on different nitrogen sources. Values represent the expression of the gene of interest relative to the expression of the reference gene ScACT1 as determined by quantitative PCR. Data presented are average ± mean deviation based on at least two independent chemostat cultures and three technical replicates for each transcript level.

Strain S. cerevisiae S. pastorianus S. pastorianus CEN.PK113-7D CBS1483 IMK472 Relevant genotype ScARO80 LgScARO80 LgScARO80 4 LgSeubARO80 LgSeubaro80Δ

Nitrogen source NH4+ Leu Phe NH4+ Leu Phe NH4+ Phe

ScARO10 0.01± 0.14± 0.60± 0.02± 0.01± 2.2± 0.004± 0.41± 0.00 0.03 0.03 0.00 0.00 0.4 0.00 0.05

SeubARO10 0.15± 0.19± 0.70± 0.01± 0.95± - - - 0.00 0.03 0.07 0.00 0.14

To determine whether transcription of LgSeubARO10 can be regulated by the S. cerevisiae- like isoform of the transcriptional activator Aro80 (Iraqui et al., 1999), the single copy of LgSeubARO80 was deleted in strain CBS1483. In the resulting strain (IMK472, LgSeubaro80D)), the three copies of LgScARO80 were still present.

Quantitative PCR was performed on samples taken from anaerobic, carbon-limited cultures of IMK472, grown with ammonia or phenylalanine as nitrogen source. The results indicate that LgScAro80 enabled full induction of LgSeubARO10 during growth on phenylalanine. However, the high basal expression level of LgSeubARO10 observed in ammonia-grown cultures of CBS1483 was not observed in the LgSeubARO80 deletion strain, suggesting that this basal expression specifically required the presence of LgSeubAro80.

Decarboxylase activities in cell extracts do not match ARO10 transcript levels To investigate to what extent the observed nitrogen-source-dependent transcriptional regulation of the LgScARO10 and LgSeubARO10 orthologs controlled decarboxylase activities in S. pastorianus CBS1483, decarboxylase activities with different substrates were analysed in anaerobic chemostat cultures grown on ammonia, leucine or phenylalanine. The use of these three different nitrogen sources had no effect on the activity of pyruvate decarboxylase in cell extracts (Table 9). Surprisingly, in contrast with the strong nitrogen- source-dependent expression of LgScARO10 and LgSeubARO10 at the transcription level, hardly any significant changes were observed in the decarboxylase activities with aromatic, branched-chain and sulfur-containing 2-oxo acids (Table 9). Only for phenylpyruvate, the activity in cultures grown on phenylalanine as nitrogen source were ca. 2-fold higher than in ammonia-grown cultures. These results clearly indicate that decarboxylase activities

131 Functional analysis and transcriptional regulation of ARO10 in the brewing strain are not primarily controlled by transcriptional regulation of either LgScARO10 or LgSeubARO10, suggesting that post-transcriptional regulation is very active in aneuploid lager brewing strains.

Table 9: Phenylpyruvate decarboxylase activity (nmol·min -1·mg protein -1) in S. pastorianus CBS1483 grown on different nitrogen sources, anaerobically, at a dilution rate of 0.10 h-1. Activities were measured at a concentration of 10 mM for each substrate, with the exception of phenylpyruvate, for which a concentration of 5 mM was used. The results represent the average ± standard deviation of at least three independent chemostats, with at least two technical replicates.

Ammonia Leucine Phenylalanine Substrate Enzymatic activity (µmol·min-1·mg protein-1) ± SD Ketoisovalerate 38 ± 6 17 ± 3 40 ± 5 Ketoisocaproate 11 ± 4 7 ± 1 14 ± 9 Ketomethylvalerate 6 ± 3 4 ± 1 11 ± 8 Keto-methylthio-oxobutanoate 38 ± 6 33 ± 2 37 ± 5 Phenylpyruvate 26 ± 9 39 ± 3 49 ± 1 Pyruvate 713 ± 148 667 ± 106 797 ± 171

DISCUSSION

This study focused on functional analysis of two orthologs of a single gene involved in amino acid catabolism, encoded by the S. cerevisiae-derived and S. eubayanus-derived subgenomes of an aneuploid brewing yeast strain. The results show how differences in copy number and transcriptional regulation of orthologous genes on the two subgenomes, as well as differences in substrate specificity of the encoded isoenzymes, complicate systems biology approaches in hybrid brewing yeast strains.

By individual expression in a decarboxylase-negative laboratory strain of S. cerevisiae, differences were revealed in substrate specificity of the S. cerevisiae-like and S. eubayanus- like isoenzymes of Aro10 in the lager brewing yeast strain S. pastorianus CBS1483. Phenylpyruvate was the preferred substrate for both and, consistent with a high degree of sequence identity, the substrate specificity of LgScAro10 closely resembled that of ScAro10 from the laboratory strain S. cerevisiae CEN.PK113-7D (Romagnoli et al., 2012). For most substrates tested, activity of LgSeubAro10 was similar to that of LgScAro10. However, activity towards ketoisovalerate, a precursor for isobutanol production, was 2-fold higher for LgSeubAro10. Such subtle differences in substrate specificity may well affect flavour balance in beer fermentation. In addition, the higher activity of LgSeubAro10 with isobutanol may be applicable in metabolic engineering strategies for isobutanol production by S. cerevisiae, in which Aro10 from S. cerevisiae has been used to catalyse the key decarboxylation step (Avalos et al., 2013; Brat et al., 2012).

132 Chapter 4

Hitherto, few studies have directly compared substrate specificities of S. cerevisiae-like and S. eubayanus-like isoenzymes in brewing yeasts. This study illustrates that expression and functional analysis of isoenzymes in appropriate S. cerevisiae mutant backgrounds is an efficient way to functionally analyse the subgenomes of brewing yeasts. Recently, a similar approach was used to analyse the kinetics of maltotriose transporters encoded by the hybrid genome of the lager brewing strain Weihenstephan 34/70 after their expression in an S. cerevisiae agt1Δ strain (Cousseau et al., 2013).

Chemostat-based transcript analysis revealed clear differences in nitrogen-source dependent regulation of LgScARO10 and LgSeubARO10, confirming that differential regulation of 4 orthologs on different subgenomes is a relevant phenomenon in hybrid yeast strains (Tirosh et al., 2009). Deletion, in S. pastorianus CBS1483, of the S. eubayanus-like allele of ARO80, which in S. cerevisiae encodes a key transcriptional activator of ARO10, did not eliminate phenylalanine induction of LgSeubARO10. Apparently, LgScAro80 can also activate LgSeubARO10 compensating the loss of S. eubayanus-type activator. This trans cross- regulation is in line with already postulated model (Tirosh et al., 2009). However, the high basal expression of LgSeubARO10 in the absence of phenylalanine was abolished upon this deletion, suggesting that LgSeubAro80 and LgScAro80 are not completely interchangeable. A low efficiency of gene deletion in S. pastorianus CBS1483 precluded us from constructing the complementary triple deletion in ScARO80. Nevertheless, these results highlight the importance of unravelling transcriptional cross-subgenome regulation in hybrid genomes.

Furthermore, all together our results suggest that in beer fermentation, the S. cerevisiae and S. eubayanus 2-oxo acid decarboxylases might play different roles. Formation of higher alcohols derived from catabolism of wort amino acids would involve preferentially LgScAro10 type since it exhibits a wider transcriptional activation and substrate specificity. In contrast, the formation of higher alcohols derived from de novo synthesized amino acids would rely exclusively on the LgSeubAro10 type which present a constitutive expression (Table 8) and contribute to the decarboxylase activity in the absence of extracellular amino acid (Table 9).

This confirms results from genome-wide expression studies with specifically designed microarrays that can discriminate between the expression of S. cerevisiae and S. eubayanus orthologs (Horinouchi et al., 2010; Minato et al., 2009). Availability of high-quality genome sequences of lager brewing strains (Nakao et al., 2009) combined with high-resolution RNAseq methods for transcriptome analysis (Mader et al., 2011; Nookaew et al., 2012) offers excellent opportunities to further disentangle the regulation of brewing yeast subgenomes under the dynamic conditions of industrial beer fermentation.

Differential kinetic properties and/or regulation of isoenzymes encoded by the two subgenomes can have important implications for targeted strain improvement strategies. This is exemplified by a recent study in which expression of theS. cerevisiae-like allele

133 Functional analysis and transcriptional regulation of ARO10 of ILV6, encoding the regulatory subunit of acetolactate synthase, but not that of the S. (eu)bayanus-like allele, was shown to be correlated with diacetyl formation during wort fermentation by a lager yeast strain. Indeed, deletion of the two copies of ScILV6 sufficed to strongly reduce diacetyl levels (Duong et al., 2011). Knowledge on copy number, regulation and activity of S. cerevisiae- and S. eubayanus-like orthologs can also be relevant in classical strain improvement. For example, when the trait of interest requires a loss-of-function mutation, the chance of success in classical mutagenesis and selection is strongly increased when the relevant biological activity is connected to an ortholog of which only a single copy is present in the hybrid genome.

Analysis of decarboxylase activities in cell extracts of cultures grown with different nitrogen sources revealed a remarkable lack of correlation with transcriptional regulation of the two ARO10 orthologs in S. pastorianus CBS1483. In S. cerevisiae, the three isoenzymes of pyruvate decarboxylase and, in particular, Pdc5 also have a broad substrate specificity (Romagnoli et al., 2012). A better understanding of the contribution of different thiamine- pyrophosphate-dependent decarboxylases to flavour production in brewing strains will therefore also require a systematic analysis of the substrate specificities and regulation of the S. cerevisiae-like and S. eubayanus-like isoenzymes of pyruvate decarboxylase.

ACKNOWLEDGEMENTS

We wish to thank Marijke Luttik, Marcel van den Broek and Erik the Hulster for their technical assistance. We also thank [Heineken Supply Chain (Zoeterwoude, The Netherlands)] for performing HPLC measurement of higher alcohols and more especially Jan-Maarten Gertman (Heineken Supply Chain) for his critical comments and advices. The research group of J.T. P. is part of the Kluyver Centre for Genomics of Industrial Fermentation, which is supported by The Netherlands Genomics Initiative. J.T.P and J-M. D. were also supported by the “Platform Green Synthetic Biology” programme (http://www.pgsb.nl) funded by NGI. This research was supported by BIOFLAVOUR, COST Action FA0907 (www.bioflavour. insa-toulouse.fr).

134 Chapter 4

REFERENCES

Aguilera J & Prieto JA, 2001. The Caesar R, Palmfeldt J, Gustafsson JS, Pettersson Saccharomyces cerevisiae aldose reductase is E, Hashemi SH & Blomberg A, 2007. implied in the metabolism of methylglyoxal in Comparative proteomics of industrial lager yeast response to stress conditions. Curr Genet 39: reveals differential expression of the cerevisiae 273-283. and non-cerevisiae parts of their genomes. Aigle M, Erb D & Moll M, 1983. Determination Proteomics 7: 4135-4147. of brewing yeast ploidy by DNA measurement. J Casaregola S, Nguyen HV, Lapathitis G, Inst Brew 89: 72-74. Kotyk A & Gaillardin C, 2001. Analysis of Alberti S, Gitler AD & Lindquist S, 2007. A suite the constitution of the beer yeast genome by 4 of Gateway cloning vectors for high-throughput PCR, sequencing and subtelomeric sequence genetic analysis in Saccharomyces cerevisiae. hybridization. Int J Syst Evol Microbiol 51: Yeast 24: 913-919. 1607-1618. Amberg DC, Botstein D & Beasley EM, 1995. Cliften P, Sudarsanam P, Desikan A, Fulton L, Precise gene disruption in Saccharomyces Fulton B, Majors J, Waterston R, Cohen BA & cerevisiae by double fusion polymerase chain Johnston M, 2003. Finding functional features reaction. Yeast 11: 1275-1280 in Saccharomyces genomes by phylogenetic footprinting. Science 301: 71-76. Arroyo-Lopez FN, Perez-Torrado R, Querol A & Barrio E, 2010. Modulation of the glycerol and Cousseau FE, Alves SL, Jr., Trichez D & ethanol syntheses in the yeast Saccharomyces Stambuk BU, 2013. Characterization of kudriavzevii differs from that exhibited by maltotriose transporters from the Saccharomyces Saccharomyces cerevisiae and their hybrid. Food eubayanus subgenome of the hybrid Microbiol 27: 628-637. Saccharomyces pastorianus lager brewing yeast strain Weihenstephan 34/70. Lett Appl Avalos JL, Fink GR & Stephanopoulos G, 2013. Microbiol 56: 21-29. Compartmentalization of metabolic pathways in yeast mitochondria improves the production Dickinson JR, 1996. ‘Fusel’ alcohols induce of branched-chain alcohols. Nat Biotechnol 31: hyphal-like extensions and pseudohyphal 335-341 formation in yeasts. Microbiology 142: 1391- 1397. Boer VM, Tai SL, Vuralhan Z, Arifin Y, Walsh MC, Piper MD, de Winde JH, Pronk JT & Dickinson JR, Harrison SJ, Dickinson JA & Daran JM, 2007. Transcriptional responses Hewlins MJ, 2000. An investigation of the of Saccharomyces cerevisiae to preferred and metabolism of isoleucine to active amyl alcohol nonpreferred nitrogen sources in glucose-limited in Saccharomyces cerevisiae. J Biol Chem 275: chemostat cultures. FEMS Yeast Res 7: 604-620. 10937-10942. Brat D, Weber C, Lorenzen W, Bode HB & Dickinson JR, Harrison SJ & Hewlins MJ, 1998. Boles E, 2012. Cytosolic re-localization and An investigation of the metabolism of valine to optimization of valine synthesis and catabolism isobutyl alcohol in Saccharomyces cerevisiae. J enables inseased isobutanol production with Biol Chem 273: 25751-25756. the yeast Saccharomyces cerevisiae. Biotechnol Biofuels 5: 65

135 Functional analysis and transcriptional regulation of ARO10

Dickinson JR, Lanterman MM, Danner DJ, Goffeau A, Barrell BG, Bussey H et al., 1996. Pearson BM, Sanz P, Harrison SJ & Hewlins Life with 6000 genes. Science 274: 546, 563- MJ, 1997. A 13C nuclear magnetic resonance 546, 567. investigation of the metabolism of leucine to Hazelwood LA, Daran JM, van Maris AJ, isoamyl alcohol in Saccharomyces cerevisiae. J Pronk JT & Dickinson JR, 2008. The Ehrlich Biol Chem 272: 26871-26878. pathway for fusel alcohol production: a century Dickinson JR, Salgado LE & Hewlins MJ, 2003. of research on Saccharomyces cerevisiae The catabolism of amino acids to long chain and metabolism. Appl Environ Microbiol 74: 2259- complex alcohols in Saccharomyces cerevisiae. J 2266. Biol Chem 278: 8028-8034. Horinouchi T, Yoshikawa K, Kawaide R, Dunn B & Sherlock G, 2008. Reconstruction of Furusawa C, Nakao Y, Hirasawa T & Shimizu the genome origins and evolution of the hybrid H, 2010. Genome-wide expression analysis of lager yeast Saccharomyces pastorianus. Genome Saccharomyces pastorianus orthologous genes Res 18: 1610-1623. using oligonucleotide microarrays. J Biosci Duong CT, Strack L, Futschik M, Katou Y, Bioeng 110: 602-607. Nakao Y, Fujimura T, Shirahige K, Kodama Y & Iraqui I, Vissers S, Andre B & Urrestarazu A, Nevoigt E, 2011. Identification ofSc -type ILV6 1999. Transcriptional induction by aromatic as a target to reduce diacetyl formation in lager amino acids in Saccharomyces cerevisiae. Mol brewers’ yeast. Metab Eng 13: 638-647. Cell Biol 19: 3360-3371. Ehrlich F, 1907. Über die Bedingungen der Jones M. & Pierce JS, 1964. Absorption of Fuselölbildung und über ihren Zusammenhang amino acids from wort by yeasts. J Inst Brew 70: mit dem Eiweissaufbau der Hefe. Ber Dtsch 307-315. Chem Ges 40: 1027-1047- Kodama Y, Kielland-Brandt MC & Hansen Entian KD & Kötter P, 2007. Yeast genetic J, 2005. Lager brewing yeast. Comparative strain and plasmid collections. Meth Microbiol Genomics: Using Fungi as Models (Sunnerhagen 36: 629-666. P & Pi_scur J, eds), pp. 145–164. Springer- Etschmann MM, Bluemke W, Sell D & Schrader Verlag, Heidelberg. J, 2002. Biotechnological production of Kunze W, Wainwright T & Mieth HO, 1996. 2-phenylethanol. Appl Microbiol Biotechnol 59: Technology Brewing and Malting. VLB Berlin, 1-8. Berlin. Flikweert MT, Kuyper M, van Maris AJ, Kotter Lange HC & Heijnen JJ, 2001. Statistical P, van Dijken JP & Pronk JT, 1999. Steady- reconciliation of the elemental and molecular state and transient-state analysis of growth biomass composition of Saccharomyces and metabolite production in a Saccharomyces cerevisiae. Biotechnol Bioeng 75: 334-344. cerevisiae strain with reduced pyruvate- Libkind D, Hittinger CT, Valerio E, Goncalves C, decarboxylase activity. Biotechnol Bioeng 66: Dover J, Johnston M, Goncalves P & Sampaio 42-50. JP, 2011. Microbe domestication and the Gietz RD & Woods RA, 2002. Transformation identification of the wild genetic stock of lager- of yeast by lithium acetate/single-stranded brewing yeast. Proc Natl Acad Sci U S A 108: carrier DNA/polyethylene glycol method. 14539-14544. Methods Enzymol 350: 87-96.

136 Chapter 4

Lodolo EJ, Kock JLF, Axcell BC & Brooks M, Rainieri S, Kodama Y, Kaneko Y, Mikata 2008. The yeast Saccharomyces cerevisiae – the K, Nakao Y & Ashikari T, 2006. Pure and main character in beer brewing. FEMS Yeast Res mixed genetic lines of Saccharomyces bayanus 8: 1018-1036 and Saccharomyces pastorianus and their Lowry OH, Rosebrough NJ, Farr AL & Randall contribution to the lager brewing strain genome. RJ, 1951. Protein measurement with the Folin Appl Environ Microbiol 72: 3968-3974. phenol reagent. J Biol Chem 193: 265-275. Reed G & Nagodawithana TW, 2001. Yeast Mader U, Nicolas P, Richard H, Bessieres P & Technology. van Nostrand Reinhold, New-York. Aymerich S, 2011. Comprehensive identification Romagnoli G, Luttik MA, Kotter P, Pronk JT and quantification of microbial transcriptomes & Daran JM, 2012. Substrate specificity of by genome-wide unbiased methods. Curr Opin thiamine pyrophosphate-dependent 2-oxo-acid 4 Biotechnol 22: 32-41. decarboxylases in Saccharomyces cerevisiae. Appl Environ Microbiol 78: 7538-7548.

Minato T, Yoshida S, Ishiguro T, Shimada E, Mizutani S, Kobayashi O & Yoshimoto H, 2009. Expression profiling of the bottom fermenting Schmitt ME, Brown TA & Trumpower yeast Saccharomyces pastorianus orthologous BL, 1990. A rapid and simple method for genes using oligonucleotide microarrays. Yeast preparation of RNA from Saccharomyces 26: 147-165. cerevisiae. Nucleic Acids Res 18: 3091-3092 Mumberg D, Müller R & Funk M, 1995. Solis-Escalante D, Kuijpers NG, Bongaerts Yeast vectors for the controlled expression N, Bolat I, Bosman L, Pronk JT, Daran JM of heterologous proteins in different genetic & Daran-Lapujade P, 2013. amdSYM, a backgrounds. Gene 156: 119-122 new dominant recyclable marker cassette for Nakao Y, Kanamori T, Itoh T, Kodama Y, Saccharomyces cerevisiae. FEMS Yeast Res 13: Rainieri S, Nakamura N, Shimonaga T, Hattori 126-139. M & Ashikari T, 2009. Genome sequence of the Tamai Y, Momma T, Yoshimoto H & Kaneko Y, lager brewing yeast, an interspecies hybrid. DNA 1998. Co-existence of two types of chromosome Res 16: 115-129. in the bottom fermenting yeast, Saccharomyces Nijkamp JF, van den Broek MA, Datema E et pastorianus. Yeast 14: 923-933. al., 2012. De novo sequencing, assembly and Tirosh I, Reikhav S, Levy AA & Barkai N, analysis of the genome of the laboratory strain 2009. A yeast hybrid provides insight into the Saccharomyces cerevisiae CEN.PK113-7D, a evolution of gene expression regulation. Science model for modern industrial biotechnology. 324: 659-662. Microb Cell Fact 11: 36- van Dijken JP & Scheffers WA, 1986. Redox Nookaew I, Papini M, Pornputtapong N, balances in the metabolism of sugars by yeasts. Scalcinati G, Fagerberg L, Uhlen M & Nielsen FEMS Microbiol Lett 32: 199-224 J, 2012. A comprehensive comparison of Vaughan Martini A & Kurtzman CP, 1985. RNA-Seq-based transcriptome analysis from Deoxyribonucleic acid relatedness among species reads to differential gene expression and cross- of the genus Saccharomyces sensu stricto. Int J comparison with microarrays: a case study in Syst Bacteriol 35: 508-511. Saccharomyces cerevisiae. Nucleic Acids Res 40: 10084-10097.

137 Functional analysis and transcriptional regulation of ARO10

Verduyn C, Postma E, Scheffers WA & van Vuralhan Z, Luttik MA, Tai SL et al., 2005. Dijken JP, 1990. Energetics of Saccharomyces Physiological characterization of the ARO10- cerevisiae in anaerobic glucose-limited chemostat dependent, broad-substrate-specificity 2-oxo cultures. J Gen Microbiol 136: 405-412. acid decarboxylase activity of Saccharomyces Verduyn C, Postma E, Scheffers WA & Vandijken cerevisiae. Appl Environ Microbiol 71: 3276- JP, 1992. Effect of benzoic acid on metabolic 3284. fluxes in yeasts: a continuous-culture study Vuralhan Z, Morais MA, Tai SL, Piper on the regulation of respiration and alcoholic MD & Pronk JT, 2003. Identification fermentation. Yeast 8: 501-517. and characterization of phenylpyruvate Verstrepen KJ, Derdelinckx G, Dufour JP, decarboxylase genes in Saccharomyces cerevisiae. Winderickx J, Thevelein JM, Pretorius IS & Appl Environ Microbiol 69: 4534-4541. Delvaux F, 2003. Flavor-active esters: adding Yamagishi H & Ogata T, 1999. Chromosomal fruitiness to beer. J Biosci Bioeng 96: 110-118 structures of bottom fermenting yeasts. Syst Appl Microbiol 22: 341-353.

138 Chapter 4

4

139

CHAPTER 5 Saccharomyces cerevisiae x Saccharomyces eubayanus interspecific hybrid, the best of both worlds and beyond

Hebly M Brickwedde A Bolat I Driessen MR de Hulster EA van den Broek M Pronk JT Geertman JM Daran JM Daran-Lapujade P

FEMS Yeast Research, 2015, vol. 15(3) doi: 10.1093/femsyr/fov005 Saccharomyces cerevisiae x Saccharomyces eubayanus hybrid

ABSTRACT

Saccharomyces pastorianus lager brewing yeasts have descended from natural hybrids of Saccharomyces cerevisiae and Saccharomyces eubayanus. Their alloploidy has undoubtedly contributed to successful domestication and industrial exploitation. To understand the early events that have led to the predominance of S. pastorianus as lager brewing yeast, an interspecific hybrid betweenS. cerevisiae and S. eubayanus was experimentally constructed. Alloploidy substantially improved the performance of the S.c. x S.e. hybrid as compared to either parent regarding two cardinal features of brewing yeasts: temperature tolerance and polysaccharide utilization. The hybrid’s S. eubayanus subgenome conferred better growth rates and biomass yields at low temperature, both on glucose and on maltose. Conversely, the ability of the hybrid to consume maltotriose, which was absent in the S. eubayanus CBS12357 type strain, was inherited from its S. cerevisiae parent. The S.c. x S.e. hybrid even out-performed its parents, a phenomenon known as transgression, suggesting that fast growth at low temperature and oligosaccharide utilization may have been key selective advantages of the natural hybrids in brewing environments. To enable sequence comparisons of the parental and hybrid strains, the genome of S. eubayanus strain CBS12357 type strain (Patagonian isolate) was resequenced, resulting in an improved publicly available sequence assembly.

142 Chapter 5

INTRODUCTION

Its relatively simple life cycle and ease of manipulation, combined with the frequent occurrence of natural hybrids, have made the genus Saccharomyces a paradigm for evolution of eukaryotic genomes (Morales & Dujon, 2012). Although the seven species of the Saccharomyces genus (i.e. S. cerevisiae, S. paradoxus, S. mikatae, S. kudriavzevii, S. arboricola, S. uvarum and the newly identified S. eubayanus (Libkind et al., 2011)) show significant divergence at the nucleotide-sequence level, they show hardly any prezygotic barriers and, upon mating, form viable diploids (Morales & Dujon, 2012). This compatibility for sexual reproduction explains the rich, reticulate evolution of the Saccharomyces genus, and more particularly that of industrially relevant species. During reticulate evolution, prolonged local adaption of the parents is unlikely to lead to hybrid offsprings that have a competitive advantage in 5 stable environments. However, domestication may have generated new niches that do offer specific advantages to hybrids(Abbott et al., 2013; Verhoeven et al., 2011).

It is now well established that many natural and commercial wine and cider yeasts are double S. cerevisiae x S. bayanus or S. cerevisiae x S. kudriavzevii hybrids or even triple S. cerevisiae x S. uvarum x S. kudriavzevii hybrids (de Barros et al., 2002; González et al., 2006) (for reviews see (Morales & Dujon, 2012; Sipiczki, 2008)). One of the best-known and most commercially relevant interspecific hybrid, the lager-brewing yeast Saccharomyces pastorianus, is an alloploid hybrid of S. cerevisiae and S. eubayanus (Gibson & Liti, 2014). S. pastorianus is a heterogeneous species, whose alloploidy has undoubtedly contributed to its industrial exploitation. Two major groups of lager-brewing yeast, the Saaz and Frohberg lineages (also called hybrid Group 1 and hybrid Group 2, respectively), have been identified and are proposed to be derived from two distinct hybridization events between S. cerevisiae ale strains and S. eubayanus (Dunn & Sherlock, 2008; Libkind et al., 2011). However recent identification of identical break point reuses (cross-overs resulting in S. cerevisiae-S. eubayanus chimeric chromosomes) in two genes (HSP82 and KEM1) in both Group 1 and 2 strains would suggest that both lineages originate from a single hybridization and therefore would share a common hybrid ancestor (Hewitt et al., 2014). These hybridization events occurred centuries ago and the selection pressure resulting from the brewing environment has triggered extensive genome reorganization, resulting in partial loss of heterozygozity and chromosomal rearrangements (Nakao et al., 2009; Walther et al., 2014). Lager-yeast genomes are therefore complex and many aspects of their evolutionary history remain to be elucidated.

Several “artificial” hybrids have been experimentally constructed betweenS. cerevisiae and various of its Saccharomyces relatives, either to investigate fundamental aspects of speciation and reproductive isolation (S. cerevisiae x S. paradoxus hybrid, (Greig et al., 2002)) or for applied perspectives by creating hybrids with novel and potentially industrially relevant properties (Bellon et al., 2011; Bellon et al., 2013; Bizaj et al., 2012). More recently, using S. cerevisiae x S. uvarum de novo constructed hybrids, Piotrowski and coworkers (2012) demonstrated the value of these interspecific hybrids to unravel the context dependency of

143 Saccharomyces cerevisiae x Saccharomyces eubayanus hybrid genome dynamic changes and to bring insight on the evolutionary forces that have shaped the genome of the modern Saccharomyces hybrids. Heterosis (hybrid vigor) of the hybrid offspring in brewing environment has been ascribed to a beneficial combination between the cold tolerance of S. eubayanus and the strong fermentative metabolism of sugars of S. cerevisiae (Gibson & Liti, 2014). The veracity of this hypothesis can be relatively simply investigated by the construction of S. cerevisiae x S. eubayanus hybrids. Surprisingly while S. pastorianus is one of the most investigated and commercially relevant hybrid, hitherto no artificialS. cerevisiae x S. eubayanus hybrids have been reported in the literature.

In the present study, mass mating was used to construct an hybrid between a haploid S. cerevisiae strain of the CEN.PK family (Entian & Kötter, 2007) and a haploid strain derived from the S. eubayanus type strain CBS12357 (isolated from fruiting bodies of Cyttaria hariotii growing on Nothofagus trees in Patagonia by (Libkind et al., 2011)). To enable a detailed analysis of the S.c. x S.e. hybrid’s genome, its S. eubayanus parent CBS12357 was re-sequenced to obtain a high-quality, well-annotated and publicly available genome sequence. The performance of the de novo constructed hybrid was compared to that of its parents in controlled bio-reactor cultures on defined media. Lager-beer production with S. pastorianus is performed under anaerobiosis at low temperatures (10-15°C) and with complex media (i.e. wort) containing mixtures of mono-, di- and tri-saccharides (Gibson et al., 2007; Lodolo et al., 2008). Specific growth rates and sugar consumption rates of the hybrid and its parents was therefore investigated over a wide temperature range (8 to 35°C) under strict anaerobiosis. Furthermore, the ability of the hybrid and its parents to anaerobically consume maltose and maltotriose was evaluated.

MATERIALS AND METHODS

Yeast strains and maintenance The yeast strains used in this study are shown in Table 1. Stock cultures were made by growing strains in 500-ml shake flasks, containing 100 ml YPD medium at 30°C and shaken at 200 rpm. Biomass concentration was evaluated by OD660 measurement and fully-grown cultures were supplemented with 30% (v/v) glycerol, divided in 1 mL aliquots and stored at -80°C until further use. Media: YPD medium contained 10 g.L-1 Bacto yeast extract, 20 g.L-1 -1 -1 Bacto peptone, and 20 g.L glucose. Synthetic medium (SM) contained 3.0 g.L KH2PO4, -1 -1 -1 -1 5.0 g·L (NH4)2SO4, 0.5 g.L MgSO4 - 7 H2O, 1 ml.L trace element solution, and 1 ml.L vitamin solution (as described in (Verduyn et al., 1990)). The pH of SM was set to 6.0 using 2 M KOH. When glucose was used as sole carbon source, 20 g.L-1 of glucose was added to plates and shake flasks and 25 g.L-1 to bioreactors (SM-Glu). To investigate utilization of wort sugars in shake flasks and bioreactors, a sugar mixture (dried glucose syrup *C plus 01987, Cargill, Haubourdin, France; sugar content (w/w): glucose 2.5%, maltose 28.0%, maltotriose 42.0%, higher saccharides 26.4%) was added to SM to a final total sugar concentration of 20 g.L-1 (SM-Mix). For cultivation on plates, media were supplemented

144 Chapter 5 with 2% agar. To trigger sporulation, diploid yeast strains were incubated in sporulation medium (2% potassium acetate, pH 7.0 (Bahalul et al., 2010)). To select for the presence of the KanMX gene, G418 was added to SM-Glu plates at a final concentration of 200 mg·L- 1. To prevent inactivation of G418 by acidification of the medium, ammonium sulfate was -1 -1 replaced by 2.3 g.L urea and 6.6 g.L K2SO4 in G418-containing media. For bioreactor cultures, SM was supplemented with the anaerobic growth factors ergosterol and Tween 80 (0.01 g·L-1 and 0.42 g·L-1 respectively, (Verduyn et al., 1990)) and with 0.15 g·L-1 of antifoam C (Sigma-Aldrich, Zwijndrecht, The Netherlands).

Table 1. Saccharomyces sp. strains used in this study

Strain Species Relevant genotype Ploidy Reference (Entian & 5 Kötter, 2007; CEN.PK113-7D S. cerevisiae MATa URA3 Haploid Nijkamp et al., 2012a) (Entian & CEN.PK122 S. cerevisiae MATa/MAT URA3/URA3 Diploid α Kötter, 2007) (Lopes et al., CR85 S. kudriavzevii Unknown Diploid 2010) (Gonzalez- IMK439 S. cerevisiae MATα ura3Δ::KanMX Haploid Ramos et al., 2013) (Libkind et al., CBS12357 S. eubayanus Unknown Diploid 2011) CBS* CBS1483 S. pastorianus Unknown Aneuploid CBS* S. cerevisiae x MATa/MAT SeubURA3/ IMS0408 α Diploid This study S. eubayanus Scura3Δ::KanMX *www.cbs.knaw.nl/

Construction of interspecific hybrid Stationary phase cells of S. eubayanus CBS12357 grown on YPD at 30°C were collected, washed and incubated for 64 h at 30°C in sporulation medium. Presence of asci was determined microscopically. 100 μL of this culture was treated with zymolyase (5 U·mL-1) for 15 min, after which spores were washed with sterile demineralized water and resuspended in 20 mL YPD medium. To this spore suspension, 100 μL of a mid-exponential culture of S. cerevisiae IMK439 (ura3::KanMX), grown in YPD at 30°C, was added. After incubation for 4 h at 30°C, interspecific hybrids were selected by streaking on a plate containing SM-Glu without uracil and with G418. This medium should only enable growth if both parental genomes are present. Single colonies were selected and restreaked three times on selective plates containing SM-Glu without uracil and with G418, after which one colony was picked to inoculate a YPD shake flask. This interspecific hybrid strain was stored as described above and named IMS0408. Confirmation of its hybrid nature was performed after three repeated batch cultures at 37°C (a non-permissive temperature for S. eubayanus, see results section) by flow cytometry, whole-genome sequencing, chromosome separation using Contour Clamped Homogeneous Electric Field (CHEF) electrophoresis and PCR.

145 Saccharomyces cerevisiae x Saccharomyces eubayanus hybrid

DNA content determination by flow cytometry Samples of culture broth (equivalent to circa 107 cells) were taken from mid-exponential shake-flask cultures on YPD and centrifuged (5 min, 4700 x g). The pellet was washed once with cold phosphate buffer (NaH2PO4 3.3 mM, Na2HPO4 6.7 mM, NaCl 130 mM, EDTA 0.2 mM) (Porro et al., 2003), vortexed briefly, centrifuged again (5 min, 4700 x g) and suspended in 800 µL 70% ethanol while vortexing. After addition of another 800 µL 70% ethanol, fixed cells were stored at 4°C until further staining and analysis. Staining of cells with SYTOX® Green Nucleic Acid Stain (Invitrogen S7020) was performed as described (Haase & Reed, 2002). Samples were analyzed on a Cell Lab Quanta™ SC MPL flow cytometer equipped with a 488 nm laser (Beckman Coulter, Woerden, The Netherlands). The fluorescence intensity (DNA content) was represented using the free CyFlogic software (version 1.2.1, ©Perttu Terhu & ©CyFlo Ltd).

Chromosome separation using Contour Clamped Homogeneous Electric Field (CHEF) electrophoresis Agarose plugs containing the DNA of different strains were prepared using the CHEF yeast genomic DNA plugs Kit (Bio-Rad, Richmond, CA) following manufacturer’s recommendations and used for CHEF electrophoresis. The plugs were placed in a 1% megabase agarose gel in TBE buffer (5.4 g trizma base, 2.75 g boric acid, 2 mL of 0.5M EDTA pH 8.0 and 1L demineralized water) gel. For chromosome separation the CHEF- DRIII Pulsed Field Electrophoresis System (Bio-Rad, Richmond, USA) was used following the manufacturer’s recommendations. For size markers, CHEF DNA Size Marker #170- 3605 (Bio-Rad) was used.

Polymerase chain reaction (PCR) Genomic DNA was extracted using the YeaStar™ Genomic DNA kit (Zymo Research Corporation, Irvine, CA) following the manufacturer’s recommendations. DNA concentrations were measured on a NanoDrop 2000 spectrophotometer (wavelength 260 nm) (Thermo Scientific, Wilmington, DE). Multiplex PCR was performed with DreamTaq PCR Master Mix (2x) (Thermo Fisher Scientific). Primers specific for S. cerevisiae (Scer F2: GCGCTTTACATTCAGATCCCG AG and Scer R2: TAAGTTGGTTGTCAGCAAGATTG) and S. eubayanus (Seub F3: GTCCCTGTACCAATTTAATATTGCGC and Seub R2: TTTCACATCTCTTAGTCTTTTCCAGACG), as described by (Pengelly & Wheals, 2013), were used at a concentration of 300 nM. Cycling parameters were 94°C for 2 min, then 35 cycles of 94°C for 30 s, 55°C for 30 s, and 72°C for 60 s, followed by a 10 min incubation at 72°C. 10 μl samples from each reaction were analyzed by electrophoresis on a 2% (w/v) agarose gel in 0.5x TBE buffer (45 mM Tris-borate pH 8.0 1 mM EDTA), supplemented with SERVA DNA stain G (SERVA Electrophoresis, Heidelberg, Germany) for 40 min at 120 V.

Shake flask experiments Shake flask cultures were grown in 500 mL flasks with a working volume of 100 mL. Temperature and shaking (200 rpm) were controlled in an Innova® 44 incubator shaker (Eppendorf, Nijmegen, The Netherlands).

146 Chapter 5

Anaerobic batches and sequential batch reactors (SBR) Batch cultivations were performed in 2-liter laboratory bioreactors (Applikon, Schiedam, The Netherlands) with a working volume of 1 liter. The pH was controlled at 5.0 by the automatic addition of 2 M KOH, and the stirrer speed was kept at 800 rpm. To ensure anaerobic conditions, bioreactors and medium vessels were continuously sparged with pure -1 -1 nitrogen (N2) gas at flow rates of 0.7 L.min and circa 7 mL.min , respectively, and equipped with Norprene tubing (Saint-Gobain Performance Plastics, Courbevoie, France) and Viton O-rings (Eriks, Alkmaar, The Netherlands). Active temperature regulation was performed and measured online by connecting a platinum resistance thermometer, placed in a socket in the bioreactor, to an RE304 low-temperature thermostat (Lauda, Lauda-Königshofen, Germany). To evaluate utilization of sugar mixtures by different yeast strains, batch cultivations were performed using SM-Mix at 20°C. Inocula were prepared by growing the 5 individual strains at 20°C in shake flasks containing SM-Mix. Each batch experiment was performed in two independent culture replicates.

Aanerobic SBR cultivation was performed as previously described (Hebly et al., 2014). To control the transition between successive batches, carbon dioxide levels were used to automatically control the emptying and refilling of the reactors. CO2 concentration in the exhaust gas below 0.05% indicated depletion of glucose and triggered the removal of 0.9 liters of culture, leaving ca. 0.1 L of fermentation broth as inoculum for the subsequent batch. For starting a new batch the reactor volume was restored to 1 L by automatic addition of fresh anaerobic medium. Growth was constantly monitored via online CO2 measurements in the off-gas. To evaluate the temperature range for optimal growth, yeast strains were cultivated in anaerobic SBRs at various temperatures on SM-Glu. Reactors were inoculated with precultures grown overnight under aerobic environment in shake flasks containing SM-Glu at 30 °C for S. cerevisiae CEN.PK122, S. pastorianus CBS1483 and S. cerevisiae x S. eubayanus IMS0408 and at 24 °C for S. eubayanus CBS12357. Physiology of the yeast strains was evaluated at temperatures ranging from 8 to 35°C. A whole temperature range was performed during each SBR, and each temperature cycle was maintained for two or three successive batches. To avoid evolution of strains resulting from a continuous selection pressure for high or low temperature tolerance, the sequence of the temperatures was chosen in such a way that the temperature neither increased nor decreased for more than three cycles in a row. To check whether the sequence in which the temperatures were analysed affected the experimental outcome, the first and last cycles were performed at the same temperature. Comparison of the growth phenotype during these two temperature cycles revealed no significant differences in growth rate. The maximum specific growth rate (μmax) was always calculated from the second batch at each temperature and based on continuous CO2 measurements in the offgas. To investigate the impact of low temperature on S. cerevisiae CEN.PK122, S. eubayanus CBS12357, the interspecific hybrid IMS0408 and S. pastorianus, a complete quantitative physiological characterization at 12 and 30°C was performed during the second batch of SBRs.

147 Saccharomyces cerevisiae x Saccharomyces eubayanus hybrid

Analytical methods Gas analysis, microscopy, analyses of biomass dry weight and extracelullar metabolite concentrations were analyzed as described previously (Hebly et al., 2014). CO2 in the off-gas was continuously monitored with an NGA 2000 analyzer (Rosemount Analytical, Orrville, OH).

Calculations and statistical analysis of physiological characteristics

Maximum specific growth rates (μmax), biomass yield on substrate (YX/S in gDW·mmol -1 -1­ glucose ), product yields on substrate (YI/S in mmol·mmol glucose ) and biomass specific -1 product formation yields (YI/X in mmol product·gDW ) were calculated via linear regression on at least five experimental data points. Maximum biomass specific glucose consumption max rates (qS ) were calculated by dividing μmax by YX/S and maximum biomass specific production rates were calculated by multiplying YI/X by μmax, based on the assumption that growth stoichiometries remained constant in exponential-phase cultures. CO2 yields on glucose could not be reliably estimated with SBRs due to small variations in the starting volume of each batch.

The standard error of the mean was calculated following equation 1:

! ! 𝑠𝑠!!",! + 𝑠𝑠!!",! 𝑆𝑆𝑆𝑆!!" = 4 in which the standard deviation () was calculated by use of the Microsoft Excel™ linest function (linear regression) on at least five experimental data points. Statistical significance was evaluated using an independent two-sample t-test, assuming unequal variance and equal or unequal sample size (also known as Welch’s t test) according to equation 2:

𝑌𝑌!"! − 𝑌𝑌!"! 𝑡𝑡 = 𝑆𝑆!!",!" in which ! ! 𝑆𝑆!!",!" = 𝑆𝑆𝑆𝑆!!"! + 𝑆𝑆𝑆𝑆!!"! The degrees of freedom used for significance testing was calculated according to the Welch– Satterthwaite equation (Eq. 3, (Welch, 1947)):

! ! ! !! !! !"! !"! !! ! !! ! ! !! !! 𝑑𝑑𝑑𝑑 ≈ !"! !"! ! ∙!" !! ∙!" with being! the! number! ! of replica’s in (in our case 2) and and the degrees of freedom in respectively condition A and B. Differences with a p-value < 0.05 were considered significant. Equations 1-3 are specific for the average biomass specific substrate consumption and product formation yields (), however the same methodology was applied for comparison of the yields on substrate( ).

148 Chapter 5

Whole genome sequencing and analysis of the S. eubayanus CBS12357 and of the interspecific hybrid IMS0408 Genome sequencing was performed using Illumina HiSeq2000 (Illumina, San Diego, CA) at Baseclear (Leiden, The Netherlands). Genomic DNA of S. eubayanus CBS12357 strain was isolated using a YeaStar Genomic DNA kit (Zymo Research, Irvine, CA) and used to obtain a 50-cycle mate pair library with an 8 kb insert size. For a second library, fragments of ~180-200 bp were sequenced paired-end with a read length of 100 bp. Overlapping read pairs were merged into single longer (pseudo)reads, yielding for both libraries over 2200 Mb total sequence, corresponding to a ca. 185x coverage.

Genome assembly was performed on the (pseudo)reads library using the GSAssembler 2.6, also known as Newbler (454 Life Sciences/Roche), using default settings. Mate pair 5 scaffolding was performed on the assembled contigs using SSPACE (version 2.0, (Boetzer et al., 2011)) with Bowtie (version 0.12.5, (Langmead et al., 2009)) as mapping tool. Based on the paired link information between the different contigs, the orientation and distance between consecutive contigs were determined and the contigs were merged into scaffolds. Gaps were introduced during scaffolding and represented by the character N for every base to preserve the distance between two placed contigs. Gapfiller(Boetzer & Pirovano, 2012) was applied and closed 119 out of 286 gaps, solving 17252 unknown bases. The raw read sequences and the assembled contigs have been deposited under the BioProject PRJNA264003 at NCBI (http://www.ncbi.nlm.nih.gov/bioproject).

The ratio of heterozygosity was calculated as follows: the sequencing reads were first re- mapped to the assembled scaffolds using SAMtools (Li et al., 2009). The binary alignment matrix file (.bam) was subsequently processed using GATK(McKenna et al., 2010). The list of heterozygous positions were filtered to exclude positions with sequencing coverage lower than 20-fold and with variation ratio below 0.2. The heterozygosity ratio was calculated as the density of heterozygous positions over the haploid genome size (11.9 Mb).

The genomic DNA of the hybrid IMS0408 strain was used to obtain a 100-cycle paired end library with an 280-bp insert size using Illumina HiSeq2000 (Illumina, San Diego, CA). The library yielded a total of 5.7 Mb pair sequence, corresponding to ca. 20x coverage. The reads of the hybrid strain IMS0408 were mapped on to the reference genome sequences of S. cerevisiae CEN.PK113-7D (accession number: PRJNA52955) and S. eubayanus CBS12357 using Burrows-Wheeler Aligner algorithm with default settings (Li & Durbin, 2009; Li & Durbin, 2010). The raw sequence data of the S.c. x S.e. hybrid IMS0408 have been deposited under the BioProject PRJNA264003 at NCBI. The Magnolya algorithm was used to analyze copy number variation, using Newbler (454 Life Sciences) for assembly (Nijkamp et al., 2012b).

The sequences of S. pastorianus WS34/70 (PRJNA29791, (Nakao et al., 2009)) and of S. eubayanus CDFM21L.1 (PRJNA254367, (Bing et al., 2014)) were used for the comparative analysis of the MAL genes.

149 Saccharomyces cerevisiae x Saccharomyces eubayanus hybrid

RESULTS

Resequencing of Saccharomyces eubayanus CBS12357 Assembly of the available sequence data of S. eubayanus (Libkind et al., 2011) resulted in a highly fragmented genome. To enable and facilitate analysis of the genome sequence of constructed S.c. x S.e. hybrids, an improved sequence assembly of S. eubayanus was necessary. Two different libraries were prepared and sequenced using Illumina technology. The first sequence data set was a 100 cycle paired-end library with an expected insert size of 180 bp. The overlapping read pairs data were merged into 3,293,656 single longer “pseudo” reads of 143 ± 20.5 bp, which represented a total sequence length of 471 Mb. These were subsequently assembled into 372 contigs with a size of 500 bp or longer, resulting in a total sequence of 11.49 Mb (Table 2).

Table 2. Genome sequencing statistics for libraries, assembly, scaffolding of S. eubayanus CBS12357

Assembly Number of contigs (≥ 500bp) 372 Avg contig size (kbp) 30.9 N50 (kbp) 94.1 Largest contig (kbp) 463.9 Total sequence (Mb) 11.5 Scaffolding and Gapfilling Number of scaffolds (≥ 500bp) 76 Avg scaffold size (kbp) 156.7 N50 (kbp) 730.2 Largest scaffold (Gbp) 1.3 Genome size (Mb) 11.9*

In a second assembly step, the 50-cycle mate pair library (17,521,927 pairs) with a 8-kb insert size, representing 1.75 Gb of sequence information, was used to further structure the S. eubayanus CBS12357 strain genome sequence. Scaffolding enabled a reduction of the number of contigs from 372 to 76 (Table 2), which led to a slightly larger haploid genome size of 11.9 Mb, sequenced with a coverage depth of 185-fold. The obtained scaffolded genome sequence was annotated using the MAKER2 pipeline (Holt & Yandell, 2011) resulting in the annotation of 5238 ORFs.

As the S. eubayanus CBS12357 strain sequenced in this study should be identical to the strain sequenced by Libkind and co-workers, we verified that the original sequencing data (Short Read Archive SRP006155 of SRA030851) could be mapped on the resequenced assembly. 98% of the original reads perfectly mapped (no mismatch, no gap allowed) onto the resequenced genome, thereby confirming strain identity. Relative toS. cerevisiae, S. eubayanus harbours reciprocal translocations between chromosomes VIII and XV, and

150 Chapter 5 between chromosomes II and IV, probably due to ectopic recombinations between duplicated RPL2A and RPL2B genes and TY LTR elements, respectively (Fischer et al., 2000). Alignment of the S. eubayanus scaffolds on the S. cerevisiae chromosome template confirmed these occurrences in the S. eubayanus CBS12357 genome (Figure 1A).

S. eubayanus CBS12357 is polyploid, as shown by its ability to sporulate (data not shown and (Libkind et al., 2011)). Flow cytometry showed a DNA pattern compatible with diploidy (Figure 1B), but was not sensitive enough to rule out aneuploidy for one or a few chromosomes. However, copy number estimation of contigs assembled from sequencing data (Nijkamp et al., 2012b) revealed a constant ploidy for all chromosomes (Figure 1C). Taken together, these data demonstrate that S. eubayanus CBS12357 is a strict diploid.

The genome assembly of S. eubayanus CBS12357 resulted logically in the reconstruction of 5 a haploid genome which however gives only a partial representation of its diploid nature. To get a better overview of the difference between the gene alleles, the heterozygosity ratio was estimated. The analysis revealed 246 heterozygous positions exclusively found in upstream and downstream regions of 121 genes. Therefore this corresponds to a very low heterozygosity ratio of 0.0021%, value very similar to the ratio of 0.0039% reported for the S. eubayanus strain CDFM21L.1 (Bing et al., 2014).

Construction of a S. eubayanus x S. cerevisiae hybrid, IMS0408 Interspecific hybrids betweenS. cerevisiae and S. eubayanus were constructed from haploid parents. While Saccharomyces strains of the CEN.PK family are heterothallic, it is yet unknown whether S. eubayanus CBS12357 is homo- or heterothallic and therefore whether pure lines of haploid mating types a and α can be obtained. To simplify and accelerate the construction of hybrids, a ‘mass mating’ strategy (Antunovics et al., 2005; Hawthorne & Philippsen, 1994; Sato et al., 2002) between S. cerevisiae haploid vegetative cells and S. eubayanus spores was undertaken (Figure 2A). True hybrids were selected based on their uracil prototrophy and G418 resistance, contributed by the S. eubayanus and S. cerevisiae parents, respectively (Figure 2A). While hybridization can lead to genetic and phenotypic variation between offspring from identical parents, the diploid S. eubayanus parent, as shown in the previous section, has a very low heterozygosity ratio, and phenotypic variations between the constructed S.c. x S.e. hybrids are thus not expected to be substantial. Therefore, a single interspecific hybrid between S. cerevisiae and S. eubayanus, named IMS0408, was chosen for further analysis. To ensure that the population obtained was solely composed of the hybrid and was not a mixed population of S. cerevisiae and S. eubayanus haploid strains, IMS0408 was grown in three sequential batches at 37°C which, as will be described below, is a non-permissive temperature for S. eubayanus. Culture purity was verified by multiplex PCR (Figure 2B) and flow cytometry analysis of the DNA content of IMS0408 was consistent with diploidy (Figure 2C). Finally, karyotype analysis by CHEF showed that all chromosomes of the parental strains were present in the hybrid IMS0408 (Figure 2D).

151 Saccharomyces cerevisiae x Saccharomyces eubayanus hybrid

The genome of the hybrid strain IMS0408 was sequenced using Illumina technology as described above for S. eubayanus. A 280-bp insert size library was then sequenced to generate a 100 cycle paired end dataset, representing a total sequence amount of 600 Mb. The paired end data were mapped on the reference genomes of the parental strains S. cerevisiae CEN. PK113-7D and S. eubayanus CBS12357.

Out of a total of 5,623,306 reads, 2,692,867 mapped to the CEN.PK113-7D sequence (47.9%) and 2,523,306 (45.6%) reads mapped to the CBS12357 sequence. These results further confirmed the hybrid nature of the genome of the isolate strain IMS0408. In good agreement with the DNA content, mapping of the sequence of IMS0408 to the reference genomes of the parental strains showed that both genomes were present in the same ratio and therefore that the strain has an allodiploid genome of 24 Mb (Figure 2E).

Temperature tolerance of S. cerevisiae, S. eubayanus and their interspecific hybrid To evaluate the temperature range for growth of the hybrid strain and to compare it to that of its parents, quantitative analysis of the physiology of the three strains was performed in anaerobic bioreactor batch cultures at temperatures ranging from 8 to 35°C. To enable acclimation to each temperature, sequential batch reactors (SBR) were used (Abbott et al., 2009; Cruz et al., 2012). In SBRs successive cycles of batch cultivation are performed by automatically emptying near-stationary phase cultures and refilling them with fresh medium. In each emptying step, a small volume of broth is retained in the bioreactor to serve as inoculum for the subsequent cycle.

The anaerobic SBR experiments demonstrated that S. eubayanus shares with S. cerevisiae and S. pastorianus the ability to grow in minimal chemically defined medium in the complete absence of oxygen (Bolat et al., 2013; Visser et al., 1990). As expected, S. cerevisiae grew at all tested temperatures, with highest (0.41±0.01 h-1) and lowest (0.013±0.003 h-1) specific growth rates at 35°C and 8°C, respectively (Figure 3). As previously described (Libkind et al., 2011), S .eubayanus was more cryotolerant than S. cerevisiae, displaying a significantly (t-test p-value below 0.05) higher growth rate at temperatures below 25°C (Figure 3). In contrast, S. eubayanus was unable to grow at 35°C and showed its highest specific growth rate of of 0.28±0.005 h-1 at a temperature of 30°C, which was 22% lower than the growth rate of S. cerevisiae at that temperature. Not only did the interspecific hybrid IMS0408 acquire the growth characteristics of the best performing parent for most temperatures, but it even outperformed the best parent for temperatures ranging from 20 to 30°C. IMS0408 was able to grow at 35°C, albeit with a 22% lower specific growth rate thanS. cerevisiae (Figure 3).

152 Chapter 5

5

Figure 1. Chromosomal architecture of Saccharomyces eubayanus CBS12357. A-Circos plot representation of the sequencing data reveal that, when compared to S. cerevisiae (blue), S. eubayanus (red) harbours reciprocal translocations between chromosomes VIII and XV, and II and IV. B-DNA content measured by flow cytometic analysis of haploid S. cerevisiae IMK439 (red); diploid S cerevisiae CEN.PK122 (green) and S. eubayanus CBS12357 (blue). C-de novo copy number detection using Magnolya (Nijkamp et al., 2012b). The black bars represent the distribution of read counts per assembled contigs. Read counts have been normalised for visualisation purposes to a contig length of 1000 bp. Copy Numbers (CN) of the contigs were estimated using the poisson mixture model (PMM) algorithm in Magnolya. The PMM was fitted on the contigs and plotted as colored lines. The first peak (red line) represents 98% of the assembly showing a constant ploidy. The remaining 2% encompasses redundant sequences (transposons, paralogous genes) that were artificially merged by the assembler into single contigs, resulting in small peaks with large CN (yellow, blue lines).

153 Saccharomyces cerevisiae x Saccharomyces eubayanus hybrid

Figure 2. Construction and validation of the S. cerevisiae x S. eubayanus interspecific hybrid IMS0408. A-Strategy for the construction of interspecific hybrids betweenS. cerevisiae IMK439 (ura3Δ::KanMX) and S. eubayanus CBS12357 (SeubURA3). B-PCR confirmation of the presence of S. eubayanus and S. cerevisiae marker genes in IMS0408. Lane 1: S. cerevisiae’s specific primers. Lane 2S. eubayanus’ specific primers. Lane 3 multiplex PCR with primers specific for both parents. L: GeneRuler 50 bp DNA Ladder. C-Ploidy assessment of the interspecific hybrid IMS0408. DNA content measured by flow cytometric analysis of Saccharomyces cerevisiae IMK439, haploid (red); S. eubayanus CBS12357, diploid (blue) and the interspecific hybrid IMS0408 (pink). D-Karyotyping of IMS0408 and of its parental strains. Chromosomes (Chr) numbers and sizes in kbp are shown and were obtained using S. cerevisiae YNN295 as ladder. E- Mapping of the genome sequence of IMS0408 to the reference genomes of S. cerevisiae CEN.PK113- 7D and S. eubayanus CBS12357.

154 Chapter 5

5

Figure 3. Temperature tolerance of the interspecific hybrid IMS0408 and its parents S. cerevisiae CEN.

PK122 and S. eubayanus CBS12357. The maximum specific growth rate (μmax) of S. cerevisiae CEN. PK122 (green), S. eubayanus CBS12357 (blue) and their interspecific hybrid IMS0408 (purple) at 8, 15,

20, 25, 30 and 35°C was determined in anaerobic SBRs using SM-Glu. μmax was calculated during the second batch at each temperature from continuous CO2 measurements in the offgas. The data represent the average and standard error of the mean of independent duplicate cultures.

Quantitative physiology of S. eubayanus, S. cerevisiae, their interspecific hybrid andS. pastorianus grown at high and low temperatures A quantitative physiological characterization was performed at low and high temperature in anaerobic SBRs. For these experiments, the high temperature was set at 30°C, as S. eubayanus cannot grow at 35°C (Figure 3). The low temperature was set to 12°C, which is a relevant temperature for brewing (typical temperature range between 10 and 15°C) at which all strains still grew relatively well (around 0.05 h-1).

Comparison of the parental strains revealed that despite the substantial lower specific growth rate of S. eubayanus as compared to S. cerevisiae at 30°C, the overall physiology of the two strains was very similar (Table 3, SI: Table S1). Indeed, while some differences were observed in the specific production rates and yields of minor fermentation products (i.e. pyruvate, lactate and acetate), the biomass, ethanol and glycerol yields were very similar for the two parents (less than 2% difference between strains). Conversely, cultivation at 12°C revealed marked differences. Not only did S. eubayanus grow twice as fast as S. cerevisiae at this temperature, but it also displayed a 46% higher biomass yield on glucose. While the biomass yield on glucose of S. cerevisiae at 12°C was approximately 50 % lower than at 30 °C, the

155 Saccharomyces cerevisiae x Saccharomyces eubayanus hybrid biomass yield of S. eubayanus was remarkably unaffected by low temperature. The different response to temperature of the two parental strains was also apparent in the production rates and yields of minor fermentation by-products (Table 3, Table S1). In particular, the glycerol yield on glucose which, in S. cerevisiae was similar at 12°C than at 30 °C (t-test p-value above 0.05), was temperature-dependent in S. eubayanus and lower at 12°C than at 30 °C (12% lower, t-test p-value below 0.05).

Correspondence of the physiology of the IMS0408 hybrid to each of its parents was temperature-dependent. At 30°C, where the major difference between the parents resided in specific growth rate, and consequently in overall specific consumption and production rates, IMS0408 grew and metabolized glucose at rates similar to that of its S. cerevisiae parent (Table 3, Table S1). Conversely, at 12°C, IMS0408 displayed growth characteristics (i.e. specific growth rate and biomass yield) that were similar to that of itsS. eubayanus parent. Still, at 12°C, the S. cerevisiae ancestry was visible through the fermentation by-products, whose production yields were either close to those observed in S. cerevisiae cultures or intermediate between the two parents (Table 3).

To investigate how the growth characteristics of the artificial hybrid compared to that of a lager brewing yeast, S. pastorianus CBS1483 was grown under the same conditions as the other three Saccharomyces strains. At 30°C, this S. pastorianus strain grew at a specific rate comparable to that of S. eubayanus and displayed biomass and product yields similar to that of both S. cerevisiae and S. eubayanus. Although S. pastorianus has been described as cryotolerant (Gibson et al., 2013), at 12°C, this S. pastorianus strain grew significantly slower than S. eubayanus and at a rate that was similar to that of S. cerevisiae (Table 3). However, as observed for S. eubayanus and IMS0408, its biomass yield at 12 °C was significantly higher than that ofS. cerevisiae (Table 3). Regarding fermentation by-products, S. pastorianus displayed a slightly different pattern than the other strains (Table 3). As observed in S. eubayanus, the glycerol yield on glucose was temperature-dependent and was substantially lower at 12°C than at 30°C.

The ability of the S. eubayanus x S. cerevisiae hybrid IMS0408 to consume maltotriose is inherited from its S. cerevisiae parent The oligosaccharides maltose and maltotriose are the major carbon sources in wort (Hough et al., 1982). S. cerevisiae, S. eubayanus and S. pastorianus can all utilize maltose as carbon source (Gibson & Liti, 2014; Lagunas, 1993; Libkind et al., 2011). However, while S. cerevisiae and S. pastorianus are known to consume maltotriose, the ability of S. eubayanus to consume this trisaccharide is still a matter of debate (Cousseau et al., 2013; Gibson et al., 2013; Vidgren & Londesborough, 2012; Vidgren et al., 2005). When grown at 20°C in anaerobic batch cultures on mixtures of glucose, maltose and maltotriose all three Saccharomyces species, as well as the interspecific hybrid, consumed maltose (Figure 4). As expected, both S. cerevisiae and S. pastorianus consumed all three sugars. In contrast, S. eubayanus did not utilize the maltotriose present in the culture medium and its growth

156 Chapter 5 ceased upon depletion of glucose and maltose. The remarkable observation that glucose was not strictly preferred over maltose and maltotriose in these cultures is probably related to the precultivation of the inoculum, which was grown on the same sugar mixture. While S. pastorianus consumed all available maltotriose, S. cerevisiae left ca. 0.3 g.L-1 of maltotriose in the medium at the end of the experiment, suggesting a lower affinity for this sugar.S. eubayanus grew significantly faster (p-value below 0.05) on maltose than S. cerevisiae and S. -1 -1 -1 pastorianus (μmax of 0.118 ± 0.004 h , 0.071 ± 0.000 h and 0.085 ± 0.001 h respectively). It is noteworthy that the difference in growth rate between S. cerevisiae and S. eubayanus was substantially more pronounced during growth on maltose than during growth on glucose. Indeed, at 20°C, S. eubayanus grew ca. 20% and 66% faster on glucose and maltose than S. cerevisiae respectively (Figure 3 and Figure 4). 5 Table 3. Physiological characteristics of S. cerevisiae, S. eubayanus, the S.c. x S.e. interspecific hybrid and S. pastorianus grown in SBR in SM-Glu at 12 and 30°C. Average of two independent culture replicates and standard error of the mean are shown. DW: dry weight

S. cerevisiae x S. cerevisiae CEN. S. eubayanus S. eubayanus S. pastorianus PK122 CBS12357 IMS0408 CBS1483

30°C 12 °C 30°C 12 °C 30°C 12 °C 30°C 12 °C

0.329 0.039 0.255 0.079 0.370 0.063 0.217 0.035 m (h-1) max ±0.006 ±0.001 ±0.005 ±0.001 ±0.009 ±0.001 ±0.011 ±0.000 Biomass yield 0.016 0.0104 0.0163 0.0152 0.0171 0.0143 0.0165 0.0148

(gDW·mmol ±0.0003 ±0.0003 ±0.0002 ±0.0001 ±0.0004 ±0.0002 ±0.0009 ±0.0002 glucose-1) Yields on glucose (mol·mol glucose-1) Ethanol 1.40 1.27 1.39 1.41 1.37 1.47 1.50 1.45 yield ±0.01 ±0.03 ±0.01 ±0.01 ±0.01 ±0.01 ±0.01 ±0.02 Glycerol 0.194 0.183 0.198 0.173 0.200 0.207 0.194 0.166 yield ±0.004 ±0.005 ±0.004 ±0.003 ±0.003 ±0.002 ±0.003 ±0.004 Pyruvate 0.0061 0.0021 0.0084 0.0051 0.00767 0.0028 0.0088 0.0010 yield ±0.0001 ±0.0001 ±0.0001 ±0.0001 ±0.0001 ±0.0000 ±0.0000 ±0.0001 Succinate 0.0077 0.0029 0.0068 0.0058 0.0083 0.0048 0.0049 0.0102 yield ±0.0009 ±0.0002 ±0.0001 ±0.0002 ±0.0013 ±0.0001 ±0.0001 ±0.0002 0.0130 0.0069 0.0182 0.0154 0.0152 0.0164 0.0167 0.0068 Lactate yield ±0.0004 ±0.0002 ±0.0004 ±0.0002 ±0.0001 ±0.0002 ±0.0003 ±0.0001 0.0235 0.0515 0.0170 0.0118 0.0202 0.0382 0.0112 Acetate yield NDa ±0.0009 ±0.0014 ±0.001 ±0.0005 ±0.0013 ±0.0003 ±0.0011 a The produced acetate was reconsumed by S. pastorianus before the end of the fermentation, thereby biasing calculation of the acetate yield

157 Saccharomyces cerevisiae x Saccharomyces eubayanus hybrid

In the S.c. x S.e. interspecific hybrid, the inability ofS. eubayanus to consume maltotriose was compensated by the acquisition of the S. cerevisiae genome, as indicated by fast consumption of maltotriose by IMS0408 (Figure 4). However, as observed for its S. cerevisiae parent, IMS0408 did not fully consume maltotriose but left a low concentration at the end of growth. From its S. eubayanus parent, the S.c. x S.e. interspecific hybrid inherited the ability to grow more rapidly on maltose at 20°C

Figure 4. Polysaccharides utilization by the S. cerevisiae x S. eubayanus hybrid IMS0408, S. cerevisiae CEN.PK122, S. eubayanus CBS12357 and the lager brewing strain S. pastorianus CBS1483 in anaerobic batch at 20°C. The top graphs represent the concentration of maltotriose (squares), maltose (circles), and glucose (triangles). The bottom graphs represent the biomass density (squares), the % CO2 in the offgas (continuous line) and the ethanol concentration (circles). A single representative batch cultivation is shown for each strain. Duplicate experiments yielded the same results. Indicated specific growth rates

(μmax) were calculated from duplicate experiments based on OD measurements in early exponential phase while maltose was the major consumed carbon source. The average and standard error of the mean are reported.

As observed for S. eubayanus, IMS0408 grew substantially faster than S. cerevisiae (μmax of 0.124 ± 0.001 h-1 and 0.071 ± 0.000 h-1 respectively), resulting in shorter fermentation times (approximately 10 hours shorter). Interestingly, the S.c. x S.e. hybrid showed a pronounced diauxic utilization of maltose and maltotriose, reflected in the CO2 profile Figure( 4, two peaks), that was not observed in S. cerevisiae, which co-consumed these two sugars. This diauxic sugar utilization was also observed in S. pastorianus. The different final biomass concentrations in cultures of IMS0408 and S. cerevisiae grown on sugar mixtures (Figure 4) were consistent with the higher biomass yield of the hybrid observed at low temperature on SM-Glu.

158 Chapter 5

DISCUSSION

An improved genome sequence of S. eubayanus CBS12357 The original genome sequence data of S. eubayanus consist of a single library of 36-nucleotide reads. These reads have not previously been de novo assembled, but only mapped on reference genome sequences (Libkind et al., 2011). Assembly of this original data set using IDBA (iterative De Bruijn Graph De Novo Assembler [http://i.cs.hku.hk/~alse/hkubrg/ projects/idba/]), an assembler compatible with very short reads, produced 14418 contigs, resulting in a highly fragmented genome. This rendered the interpretation and the subsequent utilization extremely difficult. Since these sequence data were obtained, major improvements in next generation sequencing methodologies have enabled increased sequence read length. Additionally, the combination of a paired-end library with short insert size, which enabled 5 collapsing paired reads into longer continuous sequences, and a mate paired library with long (8-kb) insert sizes enabled a new, de novo assembly of the S. eubayanus genome. This genome presents a vastly greater long-range continuity and fewer gaps than the previously available genome sequence (Table 1) allowing identification of key genes such as maltose transporter genes (discussed below). Despite clear improvements, the number of annotated genes is probably still significantly underestimated because the use of short-read sequences did not allow full assembly of most repeated sequences in the S. eubayanus genome. This notwithstanding, availability of the raw data and the structured, annotated sequence of S. eubayanus CBS12357 should benefit the yeast research community.

Physiological evidence of cold adaptation of S. eubayanus Temperature has been implicated as a key environmental factor in the speciation of the Saccharomyces genus, with the different temperature optima of the seven Saccharomyces species enabling their co-existence in specific ecological niches (Goncalves et al., 2011; Salvado et al., 2011; Sampaio & Gonçalves, 2008). S. cerevisiae, the most ‘thermotolerant’ species of the genus, has maximum and optimum growth temperatures of circa 41°C and 33°C, respectively, in aerobic cultures (Postmus et al., 2011). The present study shows that this temperature range is similar in anaerobic cultures. The scarcely characterized species S. eubayanus was first isolated fromNothofagus trees and their surrounding environment (Libkind et al., 2011). The present study demonstrates that, similar to other Saccharomyces species (Merico et al., 2007), S. eubayanus is capable of anaerobic growth in chemically defined minimal medium. Moreover, its previously reported ability to grow faster thanS. cerevisiae at low temperatures (Libkind et al., 2011) was shown to be oxygen-independent. Remarkably, at low temperature S. eubayanus grew even faster than S. kudriavzevii, previously considered to be the most cold tolerant Saccharomyces species (Goncalves et al., 2011; Salvado et al., 2011). S. eubayanus CBS12357 did not only grow faster than S. cerevisiae at temperatures below 25°C, it also exhibited a higher biomass yield on sugar at low temperature (Table 3). While this observation supports the notion that S. eubayanus has adapted to cold environments, the present study does not allow conclusions on the molecular basis for its higher biomass yield.

159 Saccharomyces cerevisiae x Saccharomyces eubayanus hybrid

Future studies on this characteristic should in particular focus on possible differences in biomass composition and, in particular, membrane composition of these two Saccharomyces species. Differences in membrane composition could result in different rates of ion diffusion across cellular membranes and thereby affect maintenance energy requirements (Verduyn et al., 1991). In this respect, it is interesting to note that cold-adapted S. kudriavzevii was found to have a different lipid composition than S. cerevisiae (López-Malo et al., 2013; Tronchoni et al., 2012). The production of glycerol, known cryoprotectant in S. cerevisiae (Aguilera et al., 2007; Hayashi & Maeda, 2006; Panadero et al., 2006; Tulha et al., 2010, has previously been proposed to contribute to the cold tolerance of S. kudriavzevii (Aguilera et al., 2007; Arroyo-Lopez et al., 2010; Gonzalez et al., 2007; Hayashi & Maeda, 2006; Panadero et al., 2006; Tulha et al., 2010). However, S. eubayanus produced less glycerol at low temperature (Table 3), suggesting that glycerol production does not play an important role in this species. The lower glycerol production and higher biomass yield at low temperature of S. eubayanus as compared to S. cerevisiae were shared by S. pastorianus.

While the present data are not sufficient to identify the underlying mechanisms that govern the physiological differences between the Saccharomyces species at different temperatures, the improved S. eubayanus sequence and the availability of the S.c. x S.e. hybrid should in the future contribute to deciphering the multifactorial and poorly understood molecular basis of cold tolerance.

On the origin of maltotriose assimilation in lager brewing strains MAL11/AGT1 has been proposed to encode the transporter responsible for maltotriose uptake in S. cerevisiae (Han et al., 1995). The identification in lager brewing strains of a frame shift leading to an early stop codon and a non-functional S. cerevisiae allele of MAL11/ AGT1 have led to the conclusion that the ability of S. pastorianus to utilize maltotriose was not inherited from S. cerevisiae but from the S. eubayanus ancestor (Cousseau et al., 2013; Vidgren & Londesborough, 2012; Vidgren et al., 2005). The results reported in the present study are consistent with a recent report (Gibson et al., 2013) contradicting this in-silico based hypothesis. A close inspection of the de novo assembled and annotated S. eubayanus genome sequence identified threeMAL loci, on scaffolds 4, 7 and 12. All three MAL genes exhibited higher similarity to MAL31 than to MAL11/AGT1 (Figure 5), consistent with the absence of functional maltotriose transporters in S. eubayanus CBS12357. While our results suggest that the S. cerevisiae subgenome, and not S. eubayanus, conferred the ability to consume maltotriose to S. pastorianus, several other possibilities have to be considered.

Firstly, recent sequencing of S. pastorianus strains has revealed the genome complexity of lager brewing yeasts in which variation and increase in chromosome copy number, a recurrent feature in these strains, has altered gene copy number. Gene duplication is known to promote allelic variation (Conant & Wolfe, 2008) and may have led to the concurrent expression of proteins with properties different from that of the original alleles inherited from the S. cerevisiae and S. eubayanus ancestors. It is therefore very well possible that S. pastorianus harbours ScAGT1 functional alleles.

160 Chapter 5

Figure 5. Topographic phylogenetic tree of annotated S. cerevisiae and S. eubayanus MALX1 genes calculated by the Neighbour joining method. Assembled S. eubayanus scaffold 4, 7 and 12 were aligned to the S288C reference sequences of MAL31/YBR298C and MAL11/YGR2289C.

Secondly, the high sequence homology between the genomes of S. eubayanus CBS12357 5 and the S. eubayanus subgenome in S. pastorianus is not sufficient to conclude that this Patagonian S. eubayanus isolate is a direct ancestor of current S. pastorianus strains. As recently demonstrated (Bing et al., 2014; Peris et al., 2014), S. eubayanus occurs on several continents (i.e. Asia , North and South America). Biodiversity within the S. eubayanus species can therefore be expected, and, contrary to the Patagonian isolate, the original S. eubayanus parent of S. pastorianus may have carried a different set of MALx1 genes, enabling maltotriose utilization. Indeed, analysis of the highly fragmented genome sequence of the recently identified Tibetan S. eubayanus isolate (Bing et al., 2014) identified six very short contigs (<200bp) that exhibited better similarity with MAL11/AGT1 than with MAL31 (Figure 6). Out of these six contigs, five completely aligned on theMAL11/AGT1 nucleotide sequence. 87% of the 106 last nucleotides of the remaining contig (contig11620, 148 bp) could be perfectly aligned to the MAL11/AGT1 sequence, while the first 42 bases showed poor similarity (16%) with the reference sequence, suggesting that S. eubayanus MAL11/ AGT1 might present structural differences.

Additionally, alignment of these short S. eubayanus contigs to the S. pastorianus Weihenstephan 34/70 contig (WS-14.3) containing the putative SeubAGT1 allele [LBYG13187] showed a near-perfect identity (Figure 6). Out of the six Tibetan S. eubayanus contigs, three (contig11221, contig12974, contig12425) were identical to SeubAGT1 Weihenstephan allele (Figure 6). As proposed by Bing and co-workers, the Tibetan S. eubayanus is more closely related to S. pastorianus’ ancestor than the strain isolated from Patagonia.

The ability of newly isolated American strains and of the Tibetan strains to grow on maltotriose is hitherto unknown (Bing et al., 2014; Peris et al., 2014). Further mining of genomic diversity in the genus Saccharomyces, combined physiological characterization of strains and kinetic characterization of transporters, is required to resolve the origin of maltotriose transport in current brewing strains.

161 Saccharomyces cerevisiae x Saccharomyces eubayanus hybrid

Figure 6. Alignment to AGT1/MAL11 from S. cerevisiae of contigs from the model lager brewing strain Weihenstephan 34/70 (Nakao et al., 2009) and from the Tibetan S. eubayanus strain CDFM21L.1 (Bing et al., 2014). S. cerevisiae AGT1/MAL11: YGR289C, Weihenstephan 34/70: contig WS14.3. A: CDFM21L.1 contig11530. B: CDFM21L.1 contig11221 and contig12974. C: CDFM21L.1 contig11620 and contig12425. D: CDFM21L.1 contig12240.

S. cerevisiae x S. eubayanus hybrid, a rare example of best parent heterosis The hybrid displayed an extendend range of growth environments as compared to its parents. From its S. cerevisiae parent the hybrid inherited the ability to grow at 35°C and to consume maltotriose, while its S. eubayanus subgenome endowed the hybrid with a substantially faster and more efficient growth at low temperature. At extreme temperatures of 8 and 35°C, the hybrid was not able to perform as well as its best parent as shown by the intermediate growth rate between that of the parents. However, growth at various temperatures revealed a striking transgression for temperature ranges between 20 and 30°C at which the hybrid substantially and significantly outperformed the best parent. This best parent heterosis was even more pronounced when oligosaccharides were used as carbon source. Maltose is a dimer of glucose and its utilization requires a transporter and a maltase that will hydrolyze maltose into glucose. It has recently been shown that S. pastorianus harbours different

162 Chapter 5

MAL11/AGT1 genes with different temperature sensitivities (Vidgren et al., 2014). The maltose transporters of the more cryotolerant S. eubayanus may similarly perform better than S. cerevisiae’s transporters at low temperature.

The presented hybrid is just one sample of the genetic and phenotypic landscape. However, in view of the rare incidence of best parent heterosis (Stelkens et al., 2014; Timberlake et al., 2011; Zorgo et al., 2012), it is highly likely that this feature is shared by S.c. x S.e. offspring. A surprizing aspect of the reticulate evolution of S. pastorianus is that S. eubayanus has hitherto not been isolated in Europe, the continent where Lager brewing yeast first appeared and evolved (Gibson & Liti, 2014). While encounters between S. cerevisiae and S. eubayanus may have been scarce due to their geographical distribution, the best parent heterosis of S.c.x S.e. hybrids may explain why they have become a predominant brewing yeast. Indeed, the 5 strong selective advantage of this hybrid in the brewing environment may have been decisive in its colonization of wort and subsequent domestication.

As could be anticipated, the physiology of the interspecific hybrid was not identical to that of the lager brewing yeast, even if the two strains shared some interesting features. It would be particularly interesting to investigate the physiology of a range of hybrids constructed from different parents, and more especially from Ale S. cerevisiae’s strains and Chinese S. eubayanus isolates, and with different ploidies (tri- and tetraploids) that may better reflect the genomes of the hybrid ancestors of the modern lager brewing yeast. A recent study has demonstrated the power of directed evolution of interspecific hybrid to investigate the genomic fate of newly-formed interspecific hybrids (Piotrowski et al., 2012). Similarly directed evolution of S.c. x S.e. hybrids in brewing-like environment should prove pivotal in understanding the evolutionary path and forces that have shaped the Lager brewing yeast genome.

ACKNOWLEDGEMENTS

We thank Niels Kuijpers, Daniel Solis-Escalante, Matthijs Martens and Matthijs Niemeijer for their experimental support and Peter Verheijen for helping with statistical analysis of the data. We thank Dr Feng-Yan Bai for sharing the raw sequencing data of the S. eubayanus strain CDFM21L.1 (BioProject PRJNA254367).

This project was carried out within the research program of the Netherlands Consortium for Systems Biology, which is sponsored by the Netherlands Genomics Initiative. AB is funded by the Seventh Framework Programme of the European Union in the frame of the SP3 people support for training and career development of researchers (Marie Curie), Networks for Initial Training (PITN-GA-2013 ITN-2013-606795) YeastCell.

163 Saccharomyces cerevisiae x Saccharomyces eubayanus hybrid

REFERENCES

Abbott DA, Suir E, Duong GH, de Hulster EAF, Bing J, Han PJ, Liu WQ, Wang QM & Bai FY, Pronk JT & van Maris AJA, 2009. Catalase 2014. Evidence for a Far East Asian origin of overexpression reduces lactic acid-induced lager beer yeast. Curr Biol 24: R380-R381. oxidative stress in Saccharomyces cerevisiae. Bizaj E, Cordente AG, Bellon JR, Raspor Appl Environ Microbiol 75: 2320-2325. P, Curtin CD & Pretorius IS, 2012. A Abbott R, Albach D, Ansell S et al., 2013. breeding strategy to harness flavor diversity Hybridization and speciation. J Evol Biol 26: of Saccharomyces interspecific hybrids and 229-246. minimize hydrogen sulfide production.FEMS Aguilera J, Randez-Gil F & Prieto JA, 2007. Yeast Res 12: 456-465. Cold response in Saccharomyces cerevisiae: new Boetzer M, Henkel CV, Jansen HJ, Butler D & functions for old mechanisms. FEMS Microbiol Pirovano W, 2011. Scaffolding pre-assembled Rev 31: 327-341. contigs using SSPACE. Bioinformatics 27: 578- Antunovics Z, Nguyen HV, Gaillardin C & 579. Sipiczki M, 2005. Gradual genome stabilisation Boetzer M & Pirovano W, 2012. Toward almost by progressive reduction of the Saccharomyces closed genomes with GapFiller. Genome Biol. uvarum genome in an interspecific hybrid with 13: 10.1186/gb-2012-13-6-r56. Saccharomyces cerevisiae. FEMS Yeast Res 5: Bolat I, Romagnoli G, Zhu F, Pronk JT & 1141-1150. Daran JM, 2013. Functional analysis and Arroyo-Lopez FN, Perez-Torrado R, Querol A & transcriptional regulation of two orthologs of Barrio E, 2010. Modulation of the glycerol and ARO10, encoding broad-substrate-specificity ethanol syntheses in the yeast Saccharomyces 2-oxo-acid decarboxylases, in the brewing yeast kudriavzevii differs from that exhibited by Saccharomyces pastorianus CBS1483. FEMS Saccharomyces cerevisiae and their hybrid. Food Yeast Res 13: 505-517. Microbiol 27: 628-637. Conant GC & Wolfe KH, 2008. Turning a Bahalul M, Kaneti G & Kashi Y, 2010. Ether- hobby into a job: how duplicated genes find new zymolyase ascospore isolation procedure: an functions. Nat Rev Genet 9: 938-950. efficient protocol for ascospores isolation in Cousseau FE, Alves SL, Jr., Trichez D & Saccharomyces cerevisiae yeast. Yeast 27: 999- Stambuk BU, 2013. Characterization of 1003. maltotriose transporters from the Saccharomyces Bellon JR, Eglinton JM, Siebert TE, Pollnitz AP, eubayanus subgenome of the hybrid Rose L, de Barros LM & Chambers PJ, 2011. Saccharomyces pastorianus lager brewing Newly generated interspecific wine yeast hybrids yeast strain Weihenstephan 34/70. Lett Appl introduce flavour and aroma diversity to wines. Microbiol 56: 21-29. Appl Microbiol Biotechnol 91: 603-612. Cruz AL, Verbon AJ, Geurink LJ, Verheijen Bellon JR, Schmid F, Capone DL, Dunn BL & PJ, Heijnen JJ & van Gulik WM, 2012. Use of Chambers PJ, 2013. Introducing a new breed of sequential-batch fermentations to characterize wine yeast: interspecific hybridisation between the impact of mild hypothermic temperatures a commercial Saccharomyces cerevisiae wine on the anaerobic stoichiometry and kinetics of yeast and Saccharomyces mikatae. PLoS One 8: Saccharomyces cerevisiae. Biotechnol Bioeng 10.1371/journal.pone.0062053. 109: 1735-1744.

164 Chapter 5

de Barros LM, Bellon JR, Shirley NJ & Ganter Gonzalez-Ramos D, van den Broek M, van PF, 2002. Evidence for multiple interspecific Maris AJ, Pronk JT & Daran JM, 2013. hybridization in Saccharomyces sensu stricto Genome-scale analyses of butanol tolerance in species. FEMS Yeast Res 1: 323-331. Saccharomyces cerevisiae reveal an essential role Dunn B & Sherlock G, 2008. Reconstruction of of protein degradation. Biotechnol biofuels 6: the genome origins and evolution of the hybrid 10.1186/1754-6834-6-48. lager yeast Saccharomyces pastorianus. Genome Greig D, Louis EJ, Borts RH & Travisano Res 18: 1610-1623. M, 2002. Hybrid speciation in experimental Entian KD & Kötter P, 2007. Yeast genetic populations of yeast. Science 298: 1773-1775. strain and plasmid collections. Methods in Haase SB & Reed SI, 2002. Improved flow Microbiology, Vol. 36 (Stansfield I & Stark MJR, cytometric analysis of the budding yeast cell eds.), pp. 629-666. Academic Press, Elsevier, cycle. Cell Cycle 1: 132-136. Amsterdam. Han EK, Cotty F, Sottas C, Jiang H & Michels 5 Fischer G, James SA, Roberts IN, Oliver SG CA, 1995. Characterization of AGT1 encoding & Louis EJ, 2000. Chromosomal evolution in a general alpha-glucoside transporter from Saccharomyces. Nature 405: 451-454. Saccharomyces. Mol Microbiol 17: 1093-1107. Gibson B & Liti G, 2014. Saccharomyces Hawthorne D & Philippsen P, 1994. Genetic pastorianus: genomic insights inspiring and molecular analysis of hybrids in the genus innovation for industry. Yeast epub ahaid of Saccharomyces involving S. cerevisiae, S. uvarum print: 10.1002/yea.3033. and a new species, S. douglasii. Yeast 10: 1285- Gibson BR, Lawrence SJ, Leclaire JP, Powell CD 1296. & Smart KA, 2007. Yeast responses to stresses Hayashi M & Maeda T, 2006. Activation associated with industrial brewery handling. of the HOG pathway upon cold stress in FEMS Microbiol Rev 31: 535-569. Saccharomyces cerevisiae. J Biochem 139: 797- Gibson BR, Storgards E, Krogerus K & 803. Vidgren V, 2013. Comparative physiology and Hebly M, de RD, de Hulster EA, de la Torre fermentation performance of Saaz and Frohberg CP, Pronk JT & Daran-Lapujade P, 2014. lager yeast strains and the parental species Physiological and transcriptional responses of Saccharomyces eubayanus. Yeast 30: 255-266. anaerobic chemostat cultures of Saccharomyces Goncalves P, Valerio E, Correia C, de Almeida cerevisiae subjected to diurnal temperature JM & Sampaio JP, 2011. Evidence for divergent cycles. Appl Environ Microbiol 80: 4433-4449. evolution of growth temperature preference in Hewitt SK, Donaldson IJ, Lovell SC & Delneri sympatric Saccharomyces species. PLoS One 6: D, 2014. Sequencing and characterisation of 10.1371/journal.pone.0020739. rearrangements in three S. pastorianus strains González SS, Barrio E, Gafner J & Querol A, reveals the presence of chimeric genes and gives 2006. Natural hybrids from Saccharomyces evidence of breakpoint reuse. PLoS One 9: cerevisiae, Saccharomyces bayanus and e92203- Saccharomyces kudriavzevii in wine Holt C & Yandell M, 2011. MAKER2: an fermentations. FEMS Yeast Res 6: 1221-1234. annotation pipeline and genome-database Gonzalez SS, Gallo L, Climent MA, Barrio E & management tool for second-generation Querol A, 2007. Enological characterization of genome projects. BMC Bioinformatics 12: natural hybrids from Saccharomyces cerevisiae 10.1186/1471-2105-12-491. and S. kudriavzevii. Int J Food Microbiol 116: 11-18.

165 Saccharomyces cerevisiae x Saccharomyces eubayanus hybrid

Hough JS, Briggs DE, Stevens R & Young TW, McKenna A, Hanna M, Banks E et al., 2010. 1982. Hopped wort and beer. Malting and The Genome Analysis Toolkit: a MapReduce Brewing Science, vol. 2. Chapman and Hall. framework for analyzing next-generation DNA Lagunas R, 1993. Sugar transport in sequencing data. Genome Res 20: 1297-1303. Saccharomyces cerevisiae. FEMS Microbiol Rev Merico A, Sulo P, Piskur J & Compagno C, 10: 229-242. 2007. Fermentative lifestyle in yeasts belonging Langmead B, Trapnell C, Pop M & Salzberg SL, to the Saccharomyces complex. FEBS J 274: 2009. Ultrafast and memory-efficient alignment 976-989. of short DNA sequences to the human genome. Morales L & Dujon B, 2012. Evolutionary Genome Biol. 10: 10.1186/gb-2009-10-3-r25. role of interspecies hybridization and genetic Li H & Durbin R, 2009. Fast and accurate exchanges in yeasts. Microbiol Mol Biol Rev 76: short read alignment with Burrows-Wheeler 721-739. transform. Bioinformatics 25: 1754-1760. Nakao Y, Kanamori T, Itoh T, Kodama Y, Li H & Durbin R, 2010. Fast and accurate Rainieri S, Nakamura N, Shimonaga T, Hattori long-read alignment with Burrows-Wheeler M & Ashikari T, 2009. Genome sequence of the transform. Bioinformatics 26: 589-595. lager brewing yeast, an interspecies hybrid. DNA Res 16: 115-129. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abecasis G & Durbin R, Nijkamp JF, van den Broek M, Datema E et 2009. The Sequence Alignment/Map format and al., 2012a. De novo sequencing, assembly and SAMtools. Bioinformatics 25: 2078-2079. analysis of the genome of the laboratory strain Saccharomyces cerevisiae CEN.PK113-7D, a Libkind D, Hittinger CT, Valério E, Gonçalves model for modern industrial biotechnology. C, Dover J, Johnston P, Gonçalves P & Sampaio Microb.Cell Fact. 11: 10.1186/1475-2859-11- JP, 2011. Microbe domestication and the 36. identification of the wild genetic stock of lager- brewing yeast. Proc Natl Acad Sci USA 108: Nijkamp JF, van den Broek MA, Geertman JM, 14539-14544. Reinders MJ, Daran JM & de RD, 2012b. De novo detection of copy number variation by co- Lodolo EJ, Kock JL, Axcell BC & Brooks M, assembly. Bioinformatics 28: 3195-3202. 2008. The yeast Saccharomyces cerevisiae- the main character in beer brewing. FEMS Yeast Res Panadero J, Pallotti C, Rodriguez-Vargas S, 8: 1018-1036. Randez-Gil F & Prieto JA, 2006. A downshift in temperature activates the high osmolarity Lopes CA, Barrio E & Querol A, 2010. Natural glycerol (HOG) pathway, which determines hybrids of S. cerevisiae x S. kudriavzevii share freeze tolerance in Saccharomyces cerevisiae. J alleles with European wild populations of Biol Chem 281: 4638-4645. Saccharomyces kudriavzevii. FEMS Yeast Res 10: 412-421. Pengelly RJ & Wheals AE, 2013. Rapid identification ofSaccharomyces eubayanus and López-Malo M, Chiva R, Rozes N & Guillamon its hybrids. FEMS Yeast Res 13: 156-161. JM, 2013. Phenotypic analysis of mutant and overexpressing strains of lipid metabolism Peris D, Sylvester K, Libkind D, Gonçalves P, genes in Saccharomyces cerevisiae: implication Sampaio JP, Alexander WG & Hittinger CT, in growth at low temperatures. Int J Food 2014. Population structure and reticulate Microbiol 162: 26-36. evolution of Saccharomyces eubayanus and its lager-brewing hybrids. Mol Ecol 23: 2031-2045.

166 Chapter 5

Piotrowski JS, Nagarajan S, Kroll E, Stanbery A, Tronchoni J, Rozes N, Querol A & Guillamon Chiotti KE, Kruckeberg AL, Dunn B, Sherlock JM, 2012. Lipid composition of wine strains of G & Rosenzweig F, 2012. Different selective Saccharomyces kudriavzevii and Saccharomyces pressures lead to different genomic outcomes as cerevisiae grown at low temperature. Int J Food newly-formed hybrid yeasts evolve. BMC.Evol. Microbiol 155: 191-198. Biol. 12: 10.1186/1471-2148-12-46. Tulha J, Lima A, Lucas C & Ferreira C, 2010. Porro D, Brambilla L & Alberghina L, 2003. Saccharomyces cerevisiae glycerol/H+ symporter Glucose metabolism and cell size in continuous Stl1p is essential for cold/near-freeze and freeze cultures of Saccharomyces cerevisiae. FEMS stress adaptation. A simple recipe with high Microbiol Lett 229: 165-171. biotechnological potential is given. Microb.Cell Postmus J, Tuzun I, Bekker M, Muller WH, Fact. 9: 10.1186/1475-2859-9-82. De Mattos MJ, Brul S & Smits GJ, 2011. Verduyn C, Postma E, Scheffers WA & van Dynamic regulation of mitochondrial respiratory Dijken JP, 1990. Physiology of Saccharomyces 5 chain efficiency inSaccharomyces cerevisiae. cerevisiae in anaerobic glucose-limited chemostat Microbiology 157: 3500-3511. cultures. Microbiol Rev 58: 616-630. Salvado Z, Arroyo-Lopez FN, Guillamon Verduyn C, Stouthamer AH, Scheffers WA & JM, Salazar G, Querol A & Barrio E, 2011. van Dijken JP, 1991. A theoretical evaluation Temperature adaptation markedly determines of growth yields of yeasts. Antonie van evolution within the genus Saccharomyces. Appl Leeuwenhoek 59: 49-63. Environ Microbiol 77: 2292-2302. Verhoeven KJ, Macel M, Wolfe LM & Biere A, Sampaio JP & Gonçalves P, 2008. Natural 2011. Population admixture, biological invasions populations of Saccharomyces kudriavzevii in and the balance between local adaptation and Portugal are associated with oak bark and are inbreeding depression. Proc Biol Sci 278: 2-8. sympatric with S. cerevisiae and S. paradoxus. Vidgren V & Londesborough J, 2012. Appl Environ Microbiol 74: 2144-2152. Characterization of the Saccharomyces bayanus- Sato M, Kishimoto M, Watari J & Takashio type AGT1 transporter of lager yeast. J inst M, 2002. Breeding of brewer’s yeast by Brew 118: 148-151. hybridization between a top-fermenting yeast Vidgren V, Ruohonen L & Londesborough J, Saccharomyces cerevisiae and a cryophilic yeast 2005. Characterization and functional analysis Saccharomyces bayanus. J Biosci Bioeng 93: of the MAL and MPH Loci for maltose 509-511. utilization in some ale and lager yeast strains. Sipiczki M, 2008. Interspecies hybridization and Appl Environ Microbiol 71: 7846-7857. recombination in Saccharomyces wine yeasts. Vidgren V, Viljanen K, Mattinen L, Rautio J & FEMS Yeast Res 8: 996-1007. Londesborough J, 2014. Three Agt1 transporters Stelkens RB, Brockhurst MA, Hurst GD, Miller from brewer’s yeasts exhibit different EL & Greig D, 2014. The effect of hybrid temperature dependencies for maltose transport transgression on environmental tolerance in over the range of brewery temperatures (0-20 experimental yeast crosses. J.Evol.Biol. epub degrees C). FEMS Yeast Res 14: 601-613. ahead of print: 10.1111/jeb.12494. Visser W, Scheffers WA, Batenburg-van der Timberlake WE, Frizzell MA, Richards KD & Vegte WH & van Dijken JP, 1990. Oxygen Gardner RC, 2011. A new yeast genetic resource requirements of yeasts. Appl Environ Microbiol for analysis and breeding. Yeast 28: 63-80. 56: 3785-3792.

167 Saccharomyces cerevisiae x Saccharomyces eubayanus hybrid

Walther A, Hesselbart A & Wendland J, 2014. Genome sequence of Saccharomyces carlsbergensis, the world’s first pure culture lager yeast. G3 (Bethesda ) 4: 783-793. Welch BL, 1947. The generalisation of student’s problems when several different population variances are involved. Biometrika 34: 28-35. Zorgo E, Gjuvsland A, Cubillos FA, Louis EJ, Liti G, Blomberg A, Omholt SW & Warringer J, 2012. Life history shapes trait heredity by accumulation of loss-of-function alleles in yeast. Mol biol Evol 29: 1781-1789.

168 Chapter 5 ±1.0 ±1.5 ±0.13 ±0.016 ±0.015 ±0.0086 ±0.0046 19.5 -13.2 2.51 12 °C 0.035±0.000 0.217 0.147 0.1138 0.0633

a CBS1483 ±0.1 ±0.1 n.d. ±0.02 S. pastorianus S. ±0.001 3.5 ±0.0002 ±0.0007 -2.4 0.40 30°C grown in SBR SM-Glu at 12 0.217±0.011 0.016 0.0024 0.0245 ±0.7 ±1.1 ±0.12 ±0.03 ±0.01 ±0.024 S. pastorianus S. 5 ±0.0075 29.6 -21.6 4.31 12 °C 0.327 0.063±0.001 0.1769 0.439 0.1649

IMS0408 ±0.1 ±0.1 ±0.02 ±0.002 ±0.004 6.4 ±0.0003 ±0.0005 -4.4 0.91 30°C 0.370±0.009 0.072 0.167 S. cerevisiae x S. eubayanus eubayanus cerevisiae x S. S. . interspecific hybrid and . 0.0123 0.0212 S.e x S.c. ±0.3 ±0.4 ±0.09 ±0.009 ±0.018 ±0.0035 ±0.0033 21.7 , the , -15.6 3.09 12 °C 0.079±0.001 0.284 0.265 0.1315 0.1058 S. eubayanus S. CBS12357 ±0.0 ±0.1 , S. eubayanus eubayanus S. ±0.02 ±0.003 ±0.001 ±0.001 7.4 before the end of the fermentation, thereby biasing calculation of the acetate yield. before the end of fermentation, ±0.0005 -5.2 0.90 30°C 0.062 0.255±0.005 0.081 0.0304 0.0265 S. cerevisiae S. S. pastorianus S. ±0.5 ±0.7 ±0.10 ±0.02 ±0.01

) ±0.003 ±0.020 -1 28.3 -20.3 .h 12 °C 3.92 0.26 0.48 -1 0.039±0,001 0.124 0.154 ±0.2 ±0.2 ±0.03 ±0.01 S. cerevisiae S. CEN.PK122 Biomass specific rates of ±0.001 ±0.001 4.8 ±0.0003 -3.8 0.69 0.19 30°C 0.329±0.006 0.026 0.0124 0.0078 ) -1 (h

max Lactate Acetate Ethanol Glucose Glycerol Pyruvate m Succinate Biomass specific rates (mmol·g DW rates (mmol·g Biomass specific The produced acetate was reconsumed by SUPPLEMENTARY INFORMATION SUPPLEMENTARY S1. Supplemental Table of two independent culture replicates and standard error the mean are shown. Average and 30°C. a

169

CHAPTER 6 Curriculum Vitae List of Publications Acknowledgements Curriculum Vitae

CURRICULUM VITAE

Irina Bolat was born on 1st of January 1978, in Galaţi, Romania. Between 1996 and 2001 she studied at the Faculty of Food Science and Engineering, “Dunărea de Jos” University, Galaţi. She specialized in brewing and malting technology, earning the title of diplomat- engineer. In 2003 Irina finished her master thesis on “Food control and expertise” at the same University. Starting 2001 she began a trainee program within the company Brau Union (BBAG), working in the brewery plant in Constanţa, Romania, which became part of the Heineken Group starting 2004. Her career in brewing continued until 2009, gaining cross function experience as Technologist, Quality Assistant Manager, Head of the Fermentation- Filtration Department, and Coordinator of the Total Productive Maintenance programme. At the end of 2005, while working in the Fermentation Department, Irina enrolled in a PhD programme within “Dunărea de Jos” University, Galaţi, in the field of Industrial Engineering, under the supervision of Prof. dr. Traian Hopulele and dr. Maria Turtoi. In 2010 she defended her thesis “Studies on the influence of brewing yeast management upon fermentation and the quality of the final product”. In 2008 Irina had the opportunity to work in the R&D Department of Heineken, Zoeterwoude, the Netherlands, for an 8 months secondment, in Brewing Science – Yeas&Fermentation, under the supervision of Michael Walsh. This wonderful experience gave her the possibility to partially lift the veil on brewing yeast genetics and to come into contact with the renowned Industrial Microbiology Group, led by Prof. Jack T. Pronk from the Faculty of Applied Sciences, Delft University of Technology, the Netherlands. In 2009 Irina moved to the Netherlands and started her PhD adventure within the same group, under the supervision of Prof. dr. Jack Pronk and dr. Jean-Marc Daran. Her research focused on expanding knowledge of lager brewing yeasts’ intricate genome, and the results are presented in this thesis. Currently Irina works as Group Quality Manager at the malting company Boortmalt, where she is responsible with the R&D programs, the technological aspects of the process and the quality management system.

172 List of Publications

LIST OF PUBLICATIONS

Van den Broek M, Bolat I, Nijkamp J, Ramos E, Luttik M, Koopman F, Pronk JT, Geertman JM, de Ridder D, JM. Daran, 2015. Chromosomal copy number variation in Saccharomyces pastorianus evidence for extensive genome dynamics in industrial lager brewing strains. Submitted for publication in Applied and Environmental Microbiology.

Hebly M, Brickwedde A, Bolat I, Driessen MR, de Hulster EA, van den Broek M, Pronk JT, Geertman JM, Daran JM, Daran-Lapujade P, 2015. Saccharomyces cerevisiae x Saccharomyces eubayanus interspecific hybrid, the best of both worlds and beyond.FEMS Yeast Research, 15 DOI: http://dx.doi.org/10.1093/femsyr/fov005.

Bolat I, Romagnoli G, Zhu F, Pronk JT, Daran JM, 2013. Functional analysis and transcriptional regulation of two orthologs of ARO10, encoding broad-substrate-specificity 2-oxo-acid decarboxylases, in the brewing yeast Saccharomyces pastorianus CBS1483. FEMS Yeast Research, 13, 505-517.

Solis-Escalante D, Kuijpers N.G.A, Bongaerts N, Bosman L, Bolat I, Pronk JT, Daran JM, Daran-Lapujade P, 2013. amdSYM, a new dominant recyclable marker cassette for Saccharomyces cerevisiae. FEMS Yeast Research, 13, 126-139.

Bolat I, Turtoi M, Walsh M, 2010. Temperature response of new lager brewing strains isolated from WS34/70. Romanian Journal of Food Science, 1, 26-38.

Bolat I, Turtoi M, Walsh M, 2009. Influence of yeast drying process on different lager brewing strains viability. Journal of Agroalimentary Processes and Technologies, 15 (3), 370 – 377.

Bolat I.C., Walsh M.C., Turtoi M., Hopulele T. 2009. The temperature influence on fermentation performance of new lager brewing strains isolated from WS34/70, The Third International Workshop in Agro-Food Issues: Environment – Nutrition – Health Relantionship in the Frame of Eu Policy, B.EN.A., January 16–17, Galati, Romania.

Bolat I, Walsh M, Turtoi M, 2008. Isolation and characterization of two new lager yeast strains from the WS34/70 population, 2008. Romanian Biotechnological Letters, 13, 62-73.

Bolat I, 2008. The importance of trehalose in brewing yeast survival, Innovative Romanian Food Biotechnology, 2, 1-10.

173 Acknowledgements

ACKNOWLEDGEMENTS

I finally reached the moment I have been waiting for so long and the dearest part to write. The time I spent in Delft, in the Industrial Microbiology Group, was the best of my life for so many reasons. From the high scientific spirit to wonderful colleagues that became friends and the extraordinary supervisors.

I would like to thank my promoter, Jack Pronk, who supported me from the start and gave me the opportunity to discover the beauty of fundamental research. Your continuous enthusiasm makes everyone want to work harder and achieve goals that sometimes seem impossible. I will always be grateful for accepting me in your Group even if I came from a different “front”. My great admiration for my co-promoter, Jean-Marc Daran, who introduced me to the fabulous world of genetics and gave me detailed explanations even to my most naive questions.

A big thank you to Erik and Zita for introducing me to the chemostat world, where as soon as you step in you become addicted. Every time I could use more than my dedicated fermenters, it felt like a small victory...but also a lot of cleaning afterwards. Marijke, you were always very calm and positive, a great support to stressed-out PhD students. Thank you also for organizing that amazing food workshop/mini-wedding party! Marinka, thank you for your help with the pipetting robot and the qPCR. They opened up the door to interesting discoveries presented in this thesis. Marcel, without your enormous effort to put the puzzle of the brewing yeast sequence together, this thesis would not have been possible. Who could have imagined that a yeast can harbour 10 alleles for only 3 genes? Pascale, thank you for the fruitful collaboration that translated into two important chapters of this thesis. Everyone contributed to the great atmosphere in the Group: Apilena, Astrid and Jannie keeping all the supplies in order and always ready to help, Jos and Sjaak always radiating happiness with a song and a smile.

I would like to thank the students I had the pleasure to work with: Feibai, Mathijs, Sietske and Philip. You all had an important contribution not only to this thesis but also to my development as a person. I wish you all have great accomplishments in your careers. Emilio, you had an important input to the second chapter of this book. Even if I was not the easiest person to work with during my last part of my PhD, I really appreciate your help and admire your perfectionism.

A big thank you to all the people that made my years in Delft so enjoyable and kept a great dynamic in the IMB Group: Wouter, Rintze, Leonie, Huong, Eline, Ishtar, Barbara, Ton, Tânia, Stefan, Victor, Bart, Jurgen, Mark, Marit, Gabriele, Niels, Nick, Harmen, Alexey, Frank, Nuria, Dani, Tim, Robert, Jeremiah, Filipa, Ishwar, Thiago, Ruben, Benjamin, Bianca, Wesley, Beth, Pilar, Lizanne, Susan, Reno, Tracey.

174 Acknowledgements

I also made incredible friends to whom I would like to thank for the wonderful moments outside work: Lucie, Joost, Katelijne, Fanni, Andre, Luisa, Hilal, Emrah, Elaheh. No matter the distance I am sure we’ll always stay in contact. Joanna en Frank, bedankt voor de gezelligheid en de interessante discussies over alles om me te helpen mijn Nederlandse taal te verbeteren. Jullie positieve gedachten dat “alles komt goed” met mijn promotie maakten me altijd blij.

A special appreciation to Daniel and Barbara, who accepted to stand by me as paranymphs during my defence. A fairly good reason to see each other again!

This journey might have never happened if it hadn’t been for the support and trust of three people that I had the chance to meet: Willem de Jonge, Michael Walsh and Willem van Waesberghe. Thank you for the good advice and the confidence you had in me! Jan-Maarten, your endless energy and readiness to help in finding ways to speed up the trip to the exit from the labyrinth of genome sequences, were a strong force in discovering so many interesting things that are now in this thesis. Thank you for your total support!

Mama şi Ana, mulţumesc pentru susţinerea continuă şi ridicarea moralului în momentele dificile. A durat ceva timp până să ajung aici, dar mă bucur că pot să împart acest final fericit cu voi.

Eduard, I could not thank you enough for all your support, patience and joy you bring into my life. With you by my side everything is possible!

175

An INVITATION A

lysis of the hybrid genomes of brewing ye brewing of lysis genomes hybrid the of AnAlysis of the To attend the defence of my PhD thesis:

hybrid genomes of Analysis of the hybrid genomes of brewing yeasts

on Wednesday, January 6th, 2015 at 15:00 in the Senaatzaal brewing yeAsts of the Aula at TU Delft, Mekelweg 6, Delft

Prior to the defence (14:30) there will be a presentation of the thesis for non-experts

You are also invited to the reception which follows, starting 17:00, in ‘t Keldertje,

A Department of Biotechnology, sts Julianalaan 67, Delft

Irina Bolat [email protected]

Paranymphs:

Barbara Kozak [email protected]

Daniel Solís Escalante [email protected] Irina Bolat Irina Bolat Irina Bolat