Tag Archives: yeast

Review on the use of hybrid yeasts for brewing

I’m really sorry for not posting in this blog more actively. I haven’t really been homebrewing much the past year, since we moved to a new house that needed renovating. We’ve finally renovated the garage into a brewing space and home bar, so I will hopefully be posting more actively about homebrewing now. I actually even brewed two batches of beer here in the new house last weekend, and I’ll be posting an update of them along with some pictures of the new brewing space in a future post.

Besides homebrewing, I like to write about my own and other’s beer and yeast research on the blog. We have some really interesting yeast projects going on, which I hope to be able to share with you soon. In the meanwhile, we were invited to write a review article on the use of hybrid yeasts in brewing for Applied Microbiology and Biotechnology, and I’m happy to say that the article has been published online now. In it we sum up the research that has been done on the use of artificial or de novo yeast hybrids for brewing applications, and discuss what kind of benefits they have to the process. These include creating strains with improved aroma formation, fermentation rate and stress tolerance. There is also a short section on how to create these hybrids. Feel free to have a look if you are interested, the article is open access!


Here is a link to the article: http://link.springer.com/article/10.1007/s00253-016-8007-5


The natural interspecies Saccharomyces cerevisiae × Saccharomyces eubayanus hybrid yeast is responsible for global lager beer production and is one of the most important industrial microorganisms. Its success in the lager brewing environment is due to a combination of traits not commonly found in pure yeast species, principally low-temperature tolerance, and maltotriose utilization. Parental transgression is typical of hybrid organisms and has been exploited previously for, e.g., the production of wine yeast with beneficial properties. The parental strain S. eubayanus has only been discovered recently and newly created lager yeast strains have not yet been applied industrially. A number of reports attest to the feasibility of this approach and artificially created hybrids are likely to have a significant impact on the future of lager brewing. De novo S. cerevisiae × S. eubayanus hybrids outperform their parent strains in a number of respects, including, but not restricted to, fermentation rate, sugar utilization, stress tolerance, and aroma formation. Hybrid genome function and stability, as well as different techniques for generating hybrids and their relative merits are discussed. Hybridization not only offers the possibility of generating novel non-GM brewing yeast strains with unique properties, but is expected to aid in unraveling the complex evolutionary history of industrial lager yeast.

Effects of ploidy on new lager yeast hybrids

As I’ve mentioned previously in many past posts (e.g. here, here and here), I’m working with and researching the properties of newly created lager yeast hybrids (for my PhD project). In the linked posts, you can read about some of our initial results from the project. These mainly established the technique and showed that de novo lager hybrids can exhibit hybrid vigor over their parent strains. Since then I’ve been looking more closely at how hybrids made from the same parent strains, but with varying ploidy levels (i.e. chromosome numbers), behave in regards to fermentation performance, aroma compound production and stress tolerance. We had some very interesting results, and we saw (at least with our hybrids) that the hybrids with higher ploidy level performed better and produced more aroma-rich beer. In order to try to understand why, we sequenced the hybrids and performed transcriptional analysis on selected genes. We saw that the higher ploidy hybrids had higher copy numbers of several genes related to aroma synthesis, and these were also transcribed at higher levels during fermentation. I held a presentation about this research at the 5th International Young Scientists Symposium on Malting, Brewing and Distilling in Chico about a month ago. You can download the presentation slides below! I’m also very happy to announce that we recently had a manuscript on this work accepted for publication in Applied Microbiology and Biotechnology. ‘Ploidy influences the functional attributes of de novo lager yeast hybrids‘ was just published online, and you can find a link to the publication below as well (it is Open Access!).


Link to the publication: http://link.springer.com/article/10.1007/s00253-016-7588-3

Link to the presentation slides: http://beer.suregork.com/wp-content/uploads/2016/04/Krogerus_YSS2016.pdf


The genomes of hybrid organisms, such as lager yeast (Saccharomyces cerevisiae × Saccharomyces eubayanus), contain orthologous genes, the functionality and effect of which may differ depending on their origin and copy number. How the parental subgenomes in lager yeast contribute to important phenotypic traits such as fermentation performance, aroma production, and stress tolerance remains poorly understood. Here, three de novo lager yeast hybrids with different ploidy levels (allodiploid, allotriploid, and allotetraploid) were generated through hybridization techniques without genetic modification. The hybrids were characterized in fermentations of both high gravity wort (15 °P) and very high gravity wort (25 °P), which were monitored for aroma compound and sugar concentrations. The hybrid strains with higher DNA content performed better during fermentation and produced higher concentrations of flavor-active esters in both worts. The hybrid strains also outperformed both the parent strains. Genome sequencing revealed that several genes related to the formation of flavor-active esters (ATF1, ATF2¸ EHT1, EEB1, and BAT1) were present in higher copy numbers in the higher ploidy hybrid strains. A direct relationship between gene copy number and transcript level was also observed. The measured ester concentrations and transcript levels also suggest that the functionality of the S. cerevisiae– and S. eubayanus-derived gene products differs. The results contribute to our understanding of the complex molecular mechanisms that determine phenotypes in lager yeast hybrids and are expected to facilitate targeted strain development through interspecific hybridization.

Report from the 5th Young Scientists Symposium in Chico (Part 2)

It took a little longer than expected to finish the second part of my summaries of the presentations at the 5th International Young Scientists Symposium on Malting, Brewing and Distilling, but here they are. You can find the first part here. I will hopefully have time to post the final part next week. I will also be posting a summary of my own presentation on new lager yeast hybrids.

  • Screening for the brewing ability of non-Saccharomyces yeasts by Maximilian Michel

Maximilian talked about the use of non-conventional yeasts for beer production and he had screened a range of non-Saccharomyces yeasts for brewing potential. Yeast isolates were first identified with genetic fingerprinting and RT-qPCR, and then sent through an initial screening test, which included growth on various carbon sources (glucose, fructose, sucrose, maltose, maltotriose and melibiose), hop resistance (various concentrations of iso-alpha acids), ethanol tolerance (various concentrations of ethanol) and phenolic off-flavour production. Promising strains were then chosen for 2L fermentations. He had focused especially on Torulaspora delbrueckii (but he had also looked at Schizosaccharomyces pombe, Pichia anomala, Hanseniaspora uvarum, Kluyveromyces lactis and Kluyveromyces marxianus), and out of the ten strains he had fermented with at ‘larger’ scale, only one was able to use maltose (and maltotriose). That strain also produced a fruity and berry-like flavour profile. So there are definitely gems to be found in the vast range of wild yeast that are available in nature.

  • Lachancea thermotolerans in primary beer fermentations by Jen House

Jen continued on the topic of using wild yeast in beer fermentations. Her research was on the use of Lachancea thermotolerans, which is an interesting species because of its ability to produce lactic acid. Hence, there is potential to use it in pure culture fermentations for the production of sour beer. Jen had tested three different strains of various origins in wort fermentations, and found that all three were able to use maltose, but not maltotriose. The three strains also produced more lactic acid and glycerol than the S. cerevisiae control. They also seemed to have quite low O2 requirements and were resistant to iso-alpha acids up to at least 60 IBU, which makes them interesting for brewing use. The pH only dropped to around 4.2 in her experimental fermentations, which means that they will only produce a mildly tart beer and may not be suitable for sour beers (as the only microbe). Lachancea yeasts have been isolated from the bark of oak trees, so that may be a good place to start looking in case you are interested in trying to isolate your own!

  • Biodiversity of yeast and lab population isolated from Beninese African Sorghum Beer Starter by Sedjro Emile Tokpohozin

Emile has been looking at the biodiversity of Beninese sorghum beers by isolating yeasts and lactic acid bacteria from starter cultures brought from Benin. These starter cultures aren’t made from pure yeast cultures, rather a small amount of beer from the previous batch is used as a starter culture for the next. Emile had isolated (identification by ITS-PCR and MALDI-TOF-MS) a range of yeasts (e.g. Saccharomyces cerevisiae, Candida krusei, Candida ethanolica and Debaryomyces hansenii) and lactic acid bacteria (Lactobacillus plantarum, Lactobacillus fermentum, Lactobacillus helveticus, Lactobacillus brevis and Lactobacillus paracasei) from a starter culture, and he further screened these for the ability to use various carbon and nitrogen sources, as well as beta-glucosidase ability (in order to break down the cyanogenic compound dhurrin that is found in sorghum). Several possible candidate isolates were identified and these are to be used in some pilot-scale fermentations next. Again shows how much ‘wild’ microbes are out there that are potentially useful in brewing!

  • Invited Speaker: Yeast culture collections by Kyria Boundy-Mills

This talk was a bit different, as Kyria talked about the Phaff yeast culture collection (of which she is the curator of). The Phaff collection is the fourth largest in the world, and contains thousands of yeasts. Many of the deposited yeasts have not been characterized very well, so Kyria talked about the possibility of finding ‘hidden gems’ in the collection. These could have some very interesting properties and phenotypes, relevant not only to the brewing industry, but also e.g. the biofuel industry (oleaginous yeast).

  • Relationships between the speed of fermentation and levels of flavor compounds post-fermentation by Maria Josey

Maria had examined the beer aroma compounds and modelled the fermentation kinetics (using a logistic model) of 10 successive fermentations using serially repitched yeast. The 10 fermentations all behaved quite similarly, with only minor differences in fermentation rate. There also didn’t seem to be any relationship between fermentation rate and number of times the yeast was repitched. This shows that you can easily reuse your yeast for over 10 generations without any significant effects on your fermentation (as long as your hygiene practices are good). Positive linear correlations were found though between the concentrations of several aroma compounds and the maximum fermentation rate (the B parameter in the model). Faster fermentation leads to more isoamyl acetate, isobutyl acetate, ethyl hexanoate and ethyl octanoate, which of course is something that seems logical as these compounds are synthesized from metabolic intermediates.

  • Omics analysis revealed multiple stress responses of lager yeast in the process of autolysis by Jinjing Wang

Jinjing had studied the yeast responses associated with autolysis by performing proteomic and transcriptomic analysis on yeast strains with different tendencies to autolyse. She also presented various methods for the quantification of autolysis, including measuring total protein in beer, the stability of the redox potential and nucleic acid release. Using microarray analysis they had identified a range of genes that were down- and upregulated in yeast strains that showed high tendency for autolysis (e.g. RLM1 and UBC4). To confirm the roles of RLM1 and UBC4 in the autolysis process, these genes were both knocked out and overexpressed in a production strain. Overexpression of RLM1 and knocking out UBC4 led to increased autolysis. However, one must keep in mind that autolysis is a complex process that is influenced by a range of cell functions and genes.

  • Energy state model for bottling plants by Isabel Osterroth

Isabel held the only presentation in the ‘Packaging’ topic, and she talked about an energy state model which she had developed for bottling plants. Sustainability and reducing energy use, combined with the fact that bottling plant models haven’t been made before, was the driving force for creating the model. The model described the energy use of various machines in the bottling plant depending on their operational state (machines use energy even when idle). A model that was able to predict the energy use of all the separate functions in the bottling plant was successfully created, and future work will include the use of the model for optimization purposes.

  • Impact of ascorbic acid additions in mashes by Joe Williams

Joe talked about his research on supplementing ascorbic acid to the mash, and gave a virtual tour of the pilot brewery at UC Davis. The motivation for adding ascorbic acid to the mash was to increase thiol and polyphenol formation and to decrease color development in the wort. The study was very preliminary at the moment, and it will be interesting to see the final results. The pilot brewery at UC Davis was quite impressive, featuring a six-vessel 170L brewhouse and four 20L nano-breweries. I am quite jealous.

  • Optimizing hop aroma in beer dry hopped with cascade utilizing glycosidic enzymes by Kaylyn Kirkpatrick

Kaylyn talked about the use of various glycosidic enzymes in dry hopped beer in order to release glycosidically bound aroma compounds. She had tested a range of commercial Rapidase enzymes and what effect they had on the concentrations of various hop aroma compounds in a beer dry hopped with Cascade. The addition of these enzymes seems to have had quite little effect on linalool concentrations, but the concentration of geranyl acetate seems to have been enhanced with the ‘Rapidase Hoptimase’ enzyme. Their sensory panel also noticed an increase in ‘tropical fruit’-like aroma, which could be attributed to several compounds that weren’t quantified in this experiment. It seems like an interesting idea though; using e.g. Cascade in combination with a glycosidic enzyme to replicate the aroma profile of some of the modern aroma hops (e.g. Citra). Not sure how economical such a solution is though?

  • Investigating sources of variation during dry-hopping by Daniel Vollmer

Daniel talked about methods to reduce the amount of variability between replicates in dry hopping experiments. Daniel had noticed in earlier experiments that there was quite large variation between his replicates during dry hopping experiments at pilot-scale, and thus attempted to locate sources for this variation. One of the key findings was that oxygen pickup has a large (negative) effect on hop aroma intensity, and this seemed to have been one of the largest sources of variation. Other sources was the raw material (i.e. the hops cones), which for future experiments will be ground. Another interesting observation, which I mentioned already in the summary of Tom Shellhammer’s keynote lecture, was that there is huge variability in oil content within the same hop cultivar (e.g. Cascade) from different farms. Also very interesting, as I mentioned, was that there seemed to have been no correlation between oil content and aroma intensity. So there are clearly other factors that affect hop aroma intensity as well.

Report from the 5th Young Scientists Symposium in Chico (Part 1)

I apologize again for the inactivity on the blog. I haven’t been brewing much the last half a year. The wife and I bought a house in the end of last year and we’ve been renovating it since. We finally moved in a couple of weeks ago, and have started settling in. So soon I’ll be able to return to brewing again! Anyways, last week I attended the 5th International Young Scientists Symposium on Malting, Brewing and Distilling, which was arranged at Sierra Nevada’s brewery from April 21-23, 2016 in Chico, California, USA. First of all I want to thank Ken Grossman, Sierra Nevada, Charlie Bamforth and all the other organizers for a fantastic conference (especially Sierra Nevada for their generosity)! The conference featured great scientific and social program, awesome food, a relaxed atmosphere, amazing people and delicious beer! I myself presented some of the recent research we’ve been conducting on lager yeast hybrids at VTT the past year (I’ll post a link to the presentation slides soon!). To sum up, we’ve been looking at how the ploidy of new lager yeasts affect their phenotypical properties. I’ll be writing up a more detailed post on this particular research soon, as we just had a manuscript on this work accepted.


As I mentioned, there were a lot of interesting presentations during the conference! I thought I’d write some short notes / summaries of all the presentations in case you are interested. Since there were a lot of presentations, I’m splitting this post into three parts. Anyways, here is the first third of the summaries:

  • Keynote: How Craft Brewing is Transforming the Way We Think About Hops and Hop Flavor by Tom Shellhammer

Tom opened the conference with an interesting talk on the current situation of hop use in the craft industry and hop research at OSU. Craft brewers are using more and more of the global hop production, which also has shifted from being ‘bitter hop’-dominated to being ‘aroma hop’-dominated. Tom also reminded the audience that 1 IBU is not the same as 1 ppm iso-alpha acid. This is particularly relevant with heavily dry hopped beers, where oxidized alpha acids (which are bitter, but not as bitter as iso-alpha acids) can influence the IBU value. In some commercial (dry hopped) beers that had been analysed at OSU, they had observed very high levels of oxidized alpha acids. Another point that was brought up, was that the perceived bitterness gets saturated at high IBU levels (i.e. very little sensorial difference between a 80 IBU beer and a 100 IBU beer). Tom also showed a very interesting figure (which Daniel Vollmer showed again later in his presentation), showing the relationship between hop oil content in Cascade hops sourced from different farms and the hop aroma intensity in beers brewed with these hops (determined by a sensory panel). What was extremely interesting was that there seemed to be no correlation what so ever. The beer brewed with the Cascade hops with lowest oil content actually seemed to have one of the highest aroma intensities. Furthermore, many of the Cascade hops that had the highest oil contents produced beers with the lowest aroma intensities. This just shows that blindly looking at hop oil contents in hops doesn’t actually tell very much about what kind of hop aroma it will give to the beer. If I remember correctly, Tom also suggested that there was no correlation between linalool or myrcene concentrations and the hop aroma intensity either, meaning that there are other key aroma compounds responsible for hop aroma out there that still need to be identified.

  • Towards the release of a 2-row barley variety for California craft malting and brewing by Joshua Hegarty

Joshua talked about how they have attempted to breed a 2-row barley variety that would be suitable for the ‘harsh’ growing conditions in California. These include an abundance of plant pathogens and dry conditions. They had crossed different parent strains, and selected superior varieties which they had then tested in the field. The new breeding lines had shown good yields and malting quality in the field trials. Using gene mapping they had also found several regions associated with disease tolerance in barley.

  • Impact of barley varieties on malt and beer flavor by Lindsay Barr

Lindsay presented some research on the influence of barley varieties on malt and beer flavour that had been carried out at the New Belgium Brewing. Barley variety seems to have quite a big influence on both wort and beer flavour (at least according to their sensory panel). However, there didn’t seem to be any correlation between the flavours that were observed in the wort and the beer. Beer age seemed to have had a bigger impact on the beer flavour than the barley variety.

  • Selective pressurized liquid extraction of hop oil from hop cones by Katy Orr

Katy talked about some of the hop-related research that had been done at Sierra Nevada Brewing. Her background was in environmental chemistry, where she had used different extraction methods to quantify hydrocarbons from environmental samples. Here, she talked about how they had tested two different extraction methods, selective pressurized liquid extraction and Likens Nickerson distillation, to test the efficiency of their hop torpedo. Both methods seemed to have yielded quite similar results for some of the compounds that were analysed. However, the main points that were brought up were that the extracted amount does not equal the actual contents and subsequently the importance of good internal standards (that behave chemically and physically as similarly as the compound of interest as possible).

  • Pro-oxidative effects on the storage stability of German Perle and Czech Saaz pellet hops by Mark Zunkel

Mark had compared the stability of Perle and Saaz hops exposed to oxygen at room temperature during a 9 month period. The hop storage index (HSI; which measures the loss of alpha and beta acids spectrophotometrically) of Perle remained quite stable for around 4 months, after which there was a more rapid loss of the hop acids. Saaz seemed to have remained slightly more stable than Perle, but also experienced a more rapid loss in the latter half of the experiment. Unsurprisingly, both hop varieties suffered a rapid loss of hop oil in the pro-oxidative environment (50% loss of hop oil in a week). This just shows that aroma hops should be stored cold and without the presence of oxygen!

  • The effect of hopping regime, cultivar and yeast ß-glucosidase activity on terpene alcohol levels in beer by Daniel Sharp

Daniel talked about the research he had been doing on the release of hop terpenes into beer from hop glycosides. This is an interesting topic for brewers interested in hop aroma, as aroma-active compounds can potentially be released during fermentation through the hydrolysis of hop-derived glycosides in the beer. He had tested the beta-glucosidase activity of a wide range of brewing yeast strains, and then selected strains with high and low activity. Surprisingly, beta-glucosidase activity didn’t seemed to affect the maximum hydrolysis level that was achieved during fermentation (and this level was much lower than the positive control where purified enzyme was added to wort). It just took a slightly longer time to reach this level with the low activity strain. Daniel didn’t seem to see any correlation between beta-glucosidase activity and the amount of aglycones in the beer. Higher glycoside extraction was achieved with whirlpool and dry hopping compared to kettle hopping. Some varieties that seemed to be high in glycosides were Columbus, Centennial, Simcoe and Summit.

  • Creating a gin utilizing novel Scottish Botanicals: A University-Industry collaboration by Margaux Huismann

Margaux talked about her MSc project, which was carried out as a collaboration between Edinburgh Gin and ICBD. During the project, she and 3 other students had developed a gin featuring Scottish (coastal) botanicals. They went to the Scottish coast to forage for interesting botanicals, and then distilled them in lab scale to develop a recipe. The recipe was then used at larger scale at the distillery to produce a commercial product. One botanical in particular, Bladderwrack, seemed to have given off a strong ‘fishy’ aroma during distillation, and its volatile aroma compounds were analysed in more detail. We later got to try the actual gin, and it was really nice (not at all as salty or ‘fishy’ as I first was expecting). Thanks Margaux!

  • Keynote: Impact of brewing practice on yeast performance by Katherine Smart

Katherine talked about some of the research she has been doing the last 15 years. This research has been focused mainly on repitching, yeast viability, stress tolerance and petite mutants. Most interesting to me was the work on why ‘1st Generation’ yeast (i.e. yeast that have already undergone one fermentation) seem to start fermentation faster than ‘0 Generation’ yeast (i.e. yeast that come straight out of the propagator). One cause, is that G1 yeast bud faster than G0 yeast (i.e. enter the replication cycle faster) and (if I remember correctly) are able to use glucose faster from the wort. G1 yeast also seem to use less FAN from the wort, which I found interesting (less nitrogen demand or more biosynthesis?). They had also used high-throughput screening systems to isolate osmo- and ethanol tolerant strains. A quite interesting remark was that strains are seldom good at both, i.e. an osmotolerant strain is rarely ethanol tolerant as well. One good point that was made regarding these high-throughput systems is that you find what you are looking for. These isolates may have high tolerance, but may otherwise perform badly in wort or produce off-flavours.

  • Cambridge Prize Lecture: The Influence of Yeast Handling on Petite Mutant Formation by Stephen Lawrence

Congratulations to Stephen for winning the Cambridge Prize! Stephen talked about the research he had carried out, which won him the Cambridge Prize. His research was focused on petite (or respiratory-deficient) mutants in brewing, and during his presentation he also talked about various stresses the yeast are subjected to during fermentation. Petite mutants (i.e. cells with damaged mitochondrial DNA) form during fermentation as a result of fermentation stresses, and these can accumulate when yeast is repitched for several generations. These petites perform worse in several regards compared to wild type cells, so their accumulation is not desirable from a brewer’s point of view. Some interesting points that were brought up, were that older cells (i.e. cells with more budding scars) were more susceptible to petite formation and that lower mtDNA copy numbers actually didn’t increase the likelihood of petite formation (e.g. older cells tend to have more mtDNA copy numbers). This seems to suggest that the accumulation of mtDNA damage has a higher impact on petite formation than the copy numbers of mtDNA. This was a very interesting talk, and the topic still seems to be quite poorly understood. It will be interesting to follow the topic in the future.


Parts two and three will be posted during next week!

Creating a new ’super fruity’ yeast strain – The best of Conan and WLP644?

In this blog post I’m briefly going to summarize how I created and improved a hybrid yeast strain between Conan (isolated from a can of Heady Topper in 2013) and WLP644. I’ve used both strains in several homebrews previously, and I’ve really liked the fruity esters they produce during fermentation. While Conan produces a really nice aroma profile, it doesn’t seem to attenuate as well as say WLP001 (most probably as a result of incomplete maltotriose use). This means that it often leaves a slight sweetness in the beer, and is a bit unpredictable. With WLP644 I’m not that sure. I’m fairly sure it uses maltotriose and ferments the beer quite dry, but this observation is only based on a single homebrew I’ve done with it. It seems to grow and ferment slightly slower than most ale strains though. So, I had the idea to generate a hybrid between the strains, in order to create the ultimate ‘super fruity’ yeast strain:

  • Lots and lots of fruity esters
  • Complete maltotriose use for high attenuation
  • Bonuses are fast fermentation and moderate flocculation

Spore-to-spore mating is the traditional method of generating yeast hybrids, but I myself am more of a fan of rare mating. Also, spore-to-spore mating couldn’t really be applied here in this case because of the poor sporulation ability of Conan. To begin, I needed to ‘tag’ the parent strains with selection markers. To do this, I selected spontaneous auxotrophs of Conan (ura-) and WLP644 (lys-) on 5-FOA and α-AA agar, respectively. This means that my Conan mutant strain isn’t able to synthesize its own uracil anymore, while the WLP644 mutant strain isn’t able to synthesize its own lysine anymore. Any hybrid between these two would inherit the functional genes from the other parent, meaning the hybrid is prototrophic again, and can again synthesize its own uracil and lysine. Hybrids can thus be selected on minimal media, which doesn’t contain uracil and lysine.


The actual hybridization process is easy, all I have to do is mix cultures of both parent strains in rich media (YP-Maltose in my case), incubate for 3 days, pellet, wash and starve the yeast, and finally spread it out on minimal media agar. Any colonies appearing on the minimal media agar are most likely hybrids. The other option is that the colonies are the parent strains, which have undergone spontaneous mutations to return to being prototrophic again. In order to ensure that we in fact have hybrids, the colonies are first purified by restreaking on fresh minimal media agar, and then transferring and streaking a single colony to YPM agar. DNA is then extracted from a single colony on the YPM plate, and interdelta-PCR is performed on the DNA.

Interdelta-PCR is a technique that is usable for differentiating Saccharomyces cerevisiae strains. Below you can see the gel and the profiles that our hybrids (H1-H4) and the parent strains produce. As we can see, the hybrids contain DNA (i.e. the bands) from both parent strains, so it is confirmed that we have successfully generated a hybrid. This is especially evident in the areas I’ve marked with red asterisks on the DNA Ladder. I won’t go into more details on the mating mechanisms, but the short answer is loss of heterozygosity at the MAT locus (for more details see here). Being a result of rare-mating, this hybrid will most likely be tetraploid (assuming that both parent strains are diploid), and contain approximately the whole genome of both parent strains (as can be seen from the gel below, hybrid H2 has already lost some DNA).


To stabilize the hybrids and also generate some potentially interesting phenotypes, I utilized meiotic recombination and segregation! During ascospore formation in yeast, meiosis and chromosomal crossover takes place, during which genes are shuffled between pairs of chromosomes. First we have to sporulate our hybrid (this can be done on 1% potassium acetate), then treat the forming ascospores with a lytic enzyme (e.g. Zymolyase), and finally dissect the ascospores with a micromanipulator. I choose to do this to hybrids H1 and H2. Both strains sporulated quite poorly and the spores had low viabilities as well (34% for H1 and 8% for H2). I selected the fastest growing colonies for further characterization, and this was four segregants from hybrid H1: H1/A4, H1/C3, H1/C4, H1/H1. The characterization was done in some small-scale (35 ml) wort fermentations. This was done in order to confirm that they still are able to grow and ferment well in wort and that they produce a lot of fruity esters.


tetrad_dissectionThe four segregants that were chosen for further characterization

The small-scale fermentations were carried out in 15 °P wort (specific gravity: 1.060), because I thought this would be representative of a typical IPA wort. 35 ml of wort was added to pre-weighed and airlock-capped plastic tubes (50 ml), after which 10 million cells / ml of wort of each yeast strain was added to start the fermentations. Fermentations were carried out at 18 °C (in retrospect, this might have been a little low), and they were monitored daily through mass loss. The alcohol content of the final beer was measured with an Anton Paar DMA5000M + Alcolyzer. Our HS-GC is in heavy use at the moment in other projects, so I wasn’t able to measure the individual ester concentrations. I did sniff the beers though to get a general idea of whether the hybrids are actually fruitier than the parent strains. The fermentations were done in wort that had only been hopped in the beginning of the boil, to make sure that the majority of the beer aroma was yeast- and not hop-derived.

minifermsThe small-scale fermentation vessels

fermentationFermentation progress

abv_ph ABV% and pH of the beers

As you can see, there was considerable differences in fermentation rate between the strains. Of the parent strains, Conan started fast, but ended slow, while WLP644 was the other way around with a slow start and faster finish. In early stages of fermentation, the H1 hybrid was also doing well, but in the end it was two of the meiotic segregants, H1/A4 and H1/C4, that reached the lowest final gravities after two weeks of fermentation (1.013 and 1.009 respectively). The Conan and WLP644 parent strains reached final gravities of 1.015 and 1.017 (this would probably have dropped slightly with a couple of more days of fermentation). There was considerable variation in pH as well, as the lowest pH values were observed in the beers fermented with H1/A4 (4.39), while the highest pH values were observed in the beers fermented with H2, H1/C4 and H1/H1 (4.64, 4.61 and 4.69). The aromas of the beers were very similar, all featuring a similar fruity tone that reminded me of various stone fruits (peach and apricot). While there were some differences in intensity between the beers, I’m not sure how much I would trust just my nose. In my opinion the strongest fruity aroma was found in the beers fermented with Conan, F1/A4 and F1/H1. The weakest aroma was found in the fastest fermenting beer, i.e. F1/C4. However, as I mentioned the actual ester concentrations would need to be measured to actually draw any conclusions. All aromas were ‘clean’, with no signs of any phenolics (both parents are POF-). The WLP644 beer had a slight sulfuric note to it though. Another thing I observed during these fermentations, was that WLP644 flocculates poorly, while Conan and all the hybrids flocculated quite well.

The next step will be to actually brew and taste some beers brewed with these yeast strains, in order to see if there actually is any real world difference. For this I was planning on brewing up a 25-liter batch of APA wort (OG around 1.050, IBUs around 50, whirlpool hops, but no dry hopping), which I would then split into 5 fermentation vessels. To these I would then pitch: Conan, WLP644, H1, H1/A4 and H1/C4. I’ll return when it is time for the brewday and ultimately the tasting notes.

References and additional reading for those interested:

Krogerus K, Magalhães F, Vidgren V, Gibson B. 2015. New lager yeast strains generated by interspecific hybridization. Journal of Industrial Microbiology & Biotechnology 42: 769-778. DOI: 10.1007/s10295-015-1597-6

Steensels J, Snoek T, Meersman E, Nicolino M, Voordeckers K, Verstrepen K. 2014. Improving industrial yeast strains: exploiting natural and artificial diversity. FEMS Microbiology Reviews 38: 947-995. DOI: 10.1111/1574-6976.12073

Physicochemical analysis of Sahti

Last summer, me and a couple of colleagues visited the 23rd National Championship in Sahti brewing, which was organized in northern Espoo. The purpose of our visit was not to compete or just drink Sahti, but rather we were there to collect samples for a research project: we had decided it was time that a thorough physical and chemical analysis on Sahti was to be performed.

First of all, what is Sahti? Well, Sahti is a traditional farmhouse beer that has been produced and consumed in Finland for centuries. The Sahti beers that I have tried have all been sweet and alcoholic, with strong yeast tones in the flavour. The aroma of isoamyl acetate (banana) has particularly stood out. If you are interested in a more thorough introduction, then you can head over to this blog. For instructions (in Finnish) on how to make your own Sahti at home, you can check out this blog.


© Nesster / Flickr

We collected samples from 12 random Sahti at the Championships (all from different parts of the country). We then did some thorough analysis on these samples, including: ABV%, residual extract, IBU, colour, foam stability, sugar profile, organic acid profile, higher alcohol profile, ester profile, phenolic acid profile, 4-vinylguaiacol content, and finally we looked for juniper-derived components with GC/MS. We also analyzed seven commercial beers as references (one Sahti, two Pale Lagers, two Hefeweizen, and two Porters).

There was quite a lot of variation between the samples, but in general the Sahti had quite high ABV%, residual extract and ester concentrations, as was expected. The isoamyl acetate (banana aroma) concentration was really high in some of the Sahti (up to around 14.5 mg/L). Since Sahti is unhopped or lightly hopped, the bitterness values were low. Also, since Sahti is typically uncarbonated or lowly carbonated, the samples had really poor foam stability. All samples also had 4-vinylguaiacol (clove-like aroma, typically found in Hefeweizens and Belgian-style ales) concentrations above the flavour threshold. This is a presumably a result of the use of Finnish baker’s yeast. It can be concluded that Sahti indeed is a unique beer style, with some very interesting properties. I’m not that big of a fan myself, but I can understand the fascination behind the style.

You can download a pre-print version (i.e. it hasn’t been formatted yet) of the publication here.


Sahti, a strong, unhopped farmhouse beer flavoured with juniper is still actively brewed in rural areas in Finland. Presented here is the first comprehensive analysis of the physical and chemical properties of this unique beer style. Twelve sahti samples from the southwest of Finland were analysed and while properties varied, the beers generally had high levels of alcohol (mean = 7.9% ABV) and high residual extract (mean = 9.5 °P). Foam stability was negligible, as is typical for the style, and glycerol concentrations at 3.1 – 4.7 g/l were higher than in reference beers (commercial lager, wheat beer and porter). Both of these features may be attributed to the very high gravity conditions employed in brewing sahti beers. Bitterness levels were relatively low (3 – 13 IBU) due to the absence or moderate use of hops. All samples contained detectable levels of the clove-like compound 4-vinylguaiacol due to the use of baker’s rather than brewer’s yeast for brewing. Concentrations of higher alcohols and esters were high, with many individual aroma compounds being above the normal flavour thresholds. Results have highlighted the uniqueness of this style of beer in comparison to commercially available beers and have contributed to our understanding of the reasons for the particular sensorial properties of this traditional beer style.

How new yeast species are inspiring a revolution in brewing

Note, this is a repost of an article I wrote for VTT’s Industrial Biotechnology Blog.

Lager beers – sometimes crisp & light pilsners, sometimes dark & malty doppelbocks, have a common denominator: They are all produced using the lager yeast Saccharomyces pastorianus, the workhorse of the lager brewing industry. This yeast is known for its tolerance to lower temperatures, and brewers take advantage of this when producing lager beers.

These beers typically have a ‘clean’ flavour profile (i.e. lack of yeast character) you see, and by fermenting the beer at colder temperatures, the yeast produces less flavour-active by-products.


Recent analysis of lager brewing yeast genomes has revealed that the many hundreds of strains used in the brewing industry are, in fact, all closely related – more like multiple variants of the same strain than individual strains. Brewers have essentially been using the same strain to brew lager beers for probably 500 years. This is in stark contrast to the other fermented beverage industries, ale, whiskey, wine, cider and so on, where a rich and diverse collection of individual yeast strains is taken for granted.

Therefore, there is huge potential for introducing diversity into the lager brewing industry by generating new strains of lager yeast.

But before one can create new lager yeast it is important to understand what exactly the lager yeast is…

It has been known for some time that lager yeast is actually a hybrid species – more like a mule than the proverbial workhorse. It was clear that one parent was the well-known ale yeast Saccharomyces cerevisiae. It wasn’t until recently that the other side of the family, Saccharomyces eubayanus, was discovered. This discovery has allowed for the improved characterization of lager yeasts, and also opened up the possibility to create new tailor-made lager yeast strains. This is possible through mating of selected strains from the two parent species. These new strains could, e.g. produce unique flavours or ferment the beer more efficiently.

This is exactly what has been the focus of our ongoing research projects at VTT.


The research team. From left to right: Brian Gibson, Kristoffer Krogerus, Virve Vidgren and Frederico Magalhães in VTT’s pilot brewery.

Screening perfect parents to mate

There are four main challenges in generating new lager yeasts: To select suitable parent strains. To get the parents to mate. To separate the hybrid cells from the parents. And finally, to confirm that they actually are hybrids.

We began by screening a range of ale yeast strains, from both VTT’s Culture Collection and commercial yeast suppliers, for beneficial fermentation properties. Once suitable parent ale yeast strains had been identified, the next step was to try to mate them with a strain of S. eubayanus, the other parent of lager yeast.

Before mating, the parent strains still had to be modified with selection markers, so that any hybrid cells could be isolated from the population. We did this by selecting spontaneous auxotrophic mutants of the parent strains, i.e. cells that weren’t able to grow on media lacking certain amino acids. This meant the hybrid cells could be selected by their ability to grow on media lacking these certain amino acids. Mating was then attempted by simply mixing populations of both parent strains, and letting them grow for a couple of days.

Seub_cells© VTT/Ulla Holopainen

After isolating some potential hybrid cells, their hybrid status was confirmed through various PCR tests, which showed whether DNA from both parent strains was present in them. After confirmation that we had produced our own lager yeast hybrids, we wanted to compare them to the parent strains in an actual wort fermentation.

To our pleasant surprise, all hybrid strains performed better than both parent strains, fermenting faster and reaching higher ethanol contents!

The hybrid strains also inherited beneficial properties from both parent strains, such as strong flocculation, cold tolerance and maltotriose utilization.

These first results suggest that this technique is suitable for producing new lager yeast strains with unique properties. These new strains also have the benefit of being non-GMO, which currently at least remains a necessity for brewers.

We are continuing our attempts to find and create perfect lager yeast hybrids at VTT. Our research will especially pay attention to flavour formation and determining how their genetic composition is reflected in their physiology.

Our work will show, for the first time, that such hybrids can be created and how they can be applied in the brewing industry. The results will appear shortly in the Journal of Industrial Microbiology and Biotechology:

Krogerus, K., Magalhães, F., Vidgren, V. & Gibson, B. (2015) New lager yeast strains generated by interspecific hybridization. Journal of Industrial Microbiology and Biotechnology, in press. DOI:10.1007/s10295-015-1597-6.

Maybe someday also you have an opportunity to enjoy these new tasty lager beers in your local pub. Cheers!

Generating new lager yeast hybrids

For my PhD thesis, I’ve been researching the flavour- and stress-related properties of brewing yeast hybrids. It has been known for some time that lager yeast (Saccharomyces pastorianus) is actually a hybrid species, and that one parent was the well-known ale yeast Saccharomyces cerevisiae. In 2011, the other side of the family, Saccharomyces eubayanus, was discovered. This discovery has allowed for the improved characterization of lager yeasts, and also opened up the possibility to create new tailor-made lager yeast strains. This is possible through mating of selected strains from the two parent species.

graphical abstractThis is exactly what I’ve been doing during the past year, and I’m happy to announce that we recently published our first results (New lager yeast strains generated by interspecific hybridization) in the Journal of Industrial Microbiology and Biotechnology. We mated a strongly flocculent production ale strain (from a brewery in the UK) with S. eubayanus, to produce lager yeast hybrids which performed better than the parent strains, and inherited beneficial properties from both. This will open up the possibility to produce a range of new lager yeast strains, with e.g. interesting flavour production and increased stress tolerance. We already have plenty of new interesting hybrid combinations that I’m looking forward to characterizing. I will post more details in a later post, but in the meanwhile feel free to read the publication if you are interested, it is Open Access!

Link to the publication: http://link.springer.com/article/10.1007/s10295-015-1597-6


The interspecific hybrid Saccharomyces pastorianus is the most commonly used yeast in brewery fermentations worldwide. Here, we generated de novo lager yeast hybrids by mating a domesticated and strongly flocculent Saccharomyces cerevisiae ale strain with the Saccharomyces eubayanus type strain. The hybrids were characterized with respect to the parent strains in a wort fermentation performed at temperatures typical for lager brewing (12 °C). The resulting beers were analysed for sugar and aroma compounds, while the yeasts were tested for their flocculation ability and α-glucoside transport capability. These hybrids inherited beneficial properties from both parent strains (cryotolerance, maltotriose utilization and strong flocculation) and showed apparent hybrid vigour, fermenting faster and producing beer with higher alcohol content (5.6 vs 4.5 % ABV) than the parents. Results suggest that interspecific hybridization is suitable for production of novel non-GM lager yeast strains with unique properties and will help in elucidating the evolutionary history of industrial lager yeast.

Physiology of Finnish Baker’s Yeast (Suomen Hiiva)

While ale yeast and baker’s yeast belong to the same species, Saccharomyces cerevisiae, they have over time been adapted to different functions. Brewing with baker’s yeast is not a very common practice, nor generally recommend amongst brewers, mostly because these strains might not exhibit favorable fermentation characteristics, such as production of desirable flavour compounds, adequate attenuation, and flocculation. Of the recognized beer styles, it is more or less only in Northern European Traditional ales (such as Sahti, Gotlandsdricka and Kvass) that it acceptable to ferment with baker’s yeast. As there is little information published on the physiological characteristics of Finnish baker’s yeast in brewing, I thought I’d post some recent results from my related research.


I performed a series of mini-fermentations (30 ml of wort) on a range of yeast strains earlier this spring for screening purposes, and I included the baker’s yeast from Suomen Hiiva, as well as two common homebrewing strains WLP002 (English Ale) and WLP380 (Hefeweizen IV). These mini-fermentations were performed at two different temperatures, 15 and 20 °C. Fermentation progress was monitored by weight loss, and after fermentation the resulting beer was analyzed for ABV% (Anton Paar Alcolyzer ME), Extract (Anton Paar DMA 5000 M), pH (Anton Paar pH ME), aroma compounds (HS-GC/FID), diacetyl (HS-GC/ECD), and phenolics (HPLC/PAD). A 15 °Plato (OG 1.060) wort was used for all fermentations, and the yeast was pitched at a rate of 2.5 g/L (~ 10 million cells/ml).

What first surprised me was how well the baker’s yeast performed during fermentation. Compared to the other two reference strains presented here, it fermented faster at both temperatures and reached a higher final attenuation. It managed to reach a final attenuation of 85% in just 5 days when fermented at 20 °C. Not bad for an all-malt wort at this strength. So the baker’s yeast can definitely give rise to adequate attenuation. Note, a pure culture of the yeast was used (starting from a similar package as in the picture above), so there was no risk of any lactic acid bacteria contamination, which otherwise is probable when using the yeast directly. The pH values of all beers were quite similar irrespective of yeast strain and temperature, and it seems like the baker’s yeast acidified the wort slightly less than the other two reference strains. In the table below you can find a summary of the Extract, ABV% and pH of the beers. In the figure below you can find a plot of the fermentation progress over time (20 °C squares; 15 °C circles).



I have never actually used baker’s yeast for beer fermentation myself, as I’m not that big of a Sahti fan, but all beers (well mainly Sahtis) I’ve tried have had quite a ester-dominated aroma. Especially 3-methylbutylacetate (isoamyl acetate), with its prominent banana aroma, is very pronounced in, and also an integral part of, Sahti. Hence, it comes as no surprise that, compared to the other two reference strains, the baker’s yeast produced more higher alcohols and esters. It was not only 3-methylbutylacetate that was produced in large amounts (even more than the WLP380 Hefeweizen strain), but also the ethyl esters. Ethyl acetate, with its solvent-like aroma, is typically unwanted at higher concentrations in beer, but the other ethyl esters may contribute a fruity aroma to the beer. In the figure below you can see a summary of the flavour impact (i.e. the concentration of the compound in the beer divided by its flavour threshold. An impact above 1 should affect flavour, while an impact between 0.5 and 1 may affect flavour) of the various higher alcohols, esters, and acetaldehyde in the beers fermented with the different yeast strains (20 °C solid; 15 °C striped). As can be seen from the figure, only 3-methylbutanol (the precursor to the banana ester) of the higher alcohols is close to the flavour threshold. As should come to no surprise, less higher alcohols and esters were produced at a lower fermentation temperature. From these aroma compound results, it is clear that the baker’s yeast will produce a fruity, and maybe even slightly solvent-like, beer. Hence, the yeast would probably be suitable for Hefeweizens and Belgian-style beers, and a slightly lower fermentation temperature is probably recommended.sh_aroma


Finally we arrive at two other important aroma compounds in beer, diacetyl (butter-like aroma) and 4-vinylguaiacol (clove-like aroma; 4-VG). Diacetyl is always considered an off-flavour (well in some cases in might be acceptable, but in my opinion it is just a sign of poor fermentation practices), while 4-VG is acceptable (and even required) in some styles (such as Hefeweizens and Belgian-style beers). Diacetyl levels decrease towards the end of fermentation, and are highly dependent on fermentation dynamics, so the concentration that was measured here at the end of fermentation doesn’t say that much. The diacetyl concentration was above the flavour threshold (50 ppb) for all strains at both temperatures, which is not that surprising since the measurements were made 120 hours after pitching the yeast, and the lowest flavour impact was observed in the beer fermented with the baker’s yeast at 20 °C. This result again suggests that baker’s yeast is a good candidate for beer fermentations (and especially more rapid ones). The baker’s yeast is also POF+ (i.e. positive for producing ‘phenolic off-flavours’; the yeast produces a phenylacrylic acid decarboxylase enzyme, that decarboxylates ferulic acid (and other phenolic acids) from the wort into 4-vinylguaiacol (and other phenolic compounds)), since 4-VG was observed in the beer. The baker’s yeast produced slightly less 4-VG than WLP380 (the table below displays the percentage of ferulic acid converted into 4-vinylguaiacol; 77% is the theoretical maximum), but it still produced concentrations above the flavour threshold in an all-barley wort (worts made from wheat malts contain more ferulic acid). This slightly limits the beer styles that the baker’s yeast could potentially be used for, but again it could be used for Hefeweizens and Belgian-style beers.


To conclude, it is evident that the Finnish baker’s yeast is perfectly usable for beer fermentations, and it comes to no surprise that it has successfully been used for traditional beer fermentation in the form of Sahti. Of the two reference strains, the baker’s yeast was closer to WLP380 (Hefeweizen IV), and it can be described as a faster-fermenting and fruitier version of it. As a side-note, it can be mentioned that the baker’s yeast flocculated very poorly, which is also similar to the behavior of WLP380. I would not recommend the use of the baker’s yeast for beer styles calling for a clean yeast aroma and no spicy phenolics, but it would be perfectly suitable for a Hefeweizen, Belgian Blond and why not even Belgian Strong Ale (Sahtis are brewed strong, so the baker’s yeast should be quite tolerant to alcohol levels reaching up to 10% ABV). It fermented quite fast at both 20 °C and 15 °C, and I would recommend a lower temperature if you are after a cleaner finish. As mentioned previously, the yeast packages found in the supermarket are all almost certainly contaminated with lactic acid bacteria, so the use of a pure culture is definitely recommended. Good luck with the brewing!

Feel free to contact me if you have any questions!

Reference for flavour thresholds:

  • Meilgaard, M., Prediction of Flavor Differences between Beers from Their Chemical Composition. Journal of Agricultural and Food Chemistry 30 (1982) 1009-1017.

Article on the differences in diacetyl formation between lager yeast strains and the role of the ILV6 gene

I’ve been researching diacetyl formation and removal by brewing yeasts the past year(s), and we recently did some screening of various lager yeast (Saccharomyces pastorianus) strains for diacetyl production and subsequent transcriptional analysis of key genes in the isoleucine-valine catabolic pathway (diacetyl is formed as an indirect by-product of valine biosynthesis in yeast; more information available here). We did this in order to identify what genes contribute most to the differences in wort total diacetyl concentration we observed between the different strains. We observed that particularly the ILV6 gene, encoding for a regulatory subunit of the acetohydroxy acid synthase enzyme (responsible for converting pyruvate into the diacetyl precursor α-acetolactate), showed greater expression early during fermentation in a strain producing more diacetyl. To confirm the role of this ILV6 gene in controlling α-acetolactate/diacetyl formation, we overexpressed both the S. cerevisiae- and S. eubayanus-form of ILV6 in our standard production lager strain A15, and found that overexpression of either form of ILV6 resulted in an identical two-fold increase in wort total diacetyl concentration relative to a control. These results suggest that both forms of the gene influence diacetyl formation, and different levels of transcription (this could be a result of a gene-dosage effect, since different lager yeast hybrids have inherited different amounts of the parental genomes; for a recent review on the subject click here) may contribute to differences in diacetyl production in various lager yeast strains.

Link to the publication: http://onlinelibrary.wiley.com/doi/10.1002/yea.3026/abstract

E-mail me if you are interested in the article, but don’t have access to the full-text.


A screen of 14 S. pastorianus lager-brewing strains showed as much as a nine-fold difference in wort total diacetyl concentration at equivalent stages of fermentation of 15°Plato brewer’s wort. Two strains (A153 and W34), with relatively low and high diacetyl production, respectively, but which did not otherwise differ in fermentation performance, growth or flavour production, were selected for further investigation. Transcriptional analysis of key genes involved in valine biosynthesis showed differences between the two strains that were consistent with the differences in wort diacetyl concentration. In particular, the ILV6 gene, encoding a regulatory subunit of acetohydroxy acid synthase, showed early transcription (only 6 h after inoculation) and up to five-fold greater expression in W34 compared to A153. This earlier transcription was observed for both orthologues of ILV6 in the S. pastorianus hybrid (S. cerevisiae × S. eubayanus), although the S. cerevisiae form of ILV6 in W34 also showed a consistently higher transcript level throughout fermentation relative to the same gene in A153. Overexpression of either form of ILV6 (by placing it under the control of the PGK1 promoter) resulted in an identical two-fold increase in wort total diacetyl concentration relative to a control. The results confirm the role of the Ilv6 subunit in controlling α-acetolactate/diacetyl concentration and indicate no functional divergence between the two forms of Ilv6. The greater contribution of the S. cerevisiae ILV6 to acetolactate production in natural brewing yeast hybrids appears rather to be due to higher levels of transcription relative to the S. eubayanus form.