Content uploaded by Mathias Hutzler
Author content
All content in this area was uploaded by Mathias Hutzler on May 13, 2020
Content may be subject to copyright.
69 March / April 2019 (Vol. 72) BrewingScience
Authors
https://doi.org/10.23763/BrSc19-06hutzler
M. Hutzler, L. Narziß, D. Stretz, K. Haslbeck, T. Meier-Dörnberg, H. Walter, M. Schäfer, T. Zollo,
F. Jacob and M. Michel
Resurrection of the lager strain
Saccharomyces pastorianus TUM 35
Saccharomyces pastorianus lager yeast strains are some of the most important industrially used microbes
used in fermentations. Lager beer types dominate the market with over 90 % of the market share. Although
some popular and widespread lager strains, such as the most used strain Saccharomyces pastorianus TUM
34/70, are well characterized, little or nothing is known about old and seldom used lager strains from
long-standing strain collections. Only two Saccharomyces pastorianus lager strain subgroups are known to
date, ‘Frohberg’ and ‘Saaz’. Most industrial, modern, high-performance lager strains belong to the ‘Frohberg’
group. In this study our group reactivated a freeze-dried stock of a yeast culture (carrier matrix unknown,
probably dry milk powder) of the historic strain TUM 35. The strain was presumed to have been lost over time.
Fortunately, the freeze-dried stock was found in a forgotten box in a storage room (together with other historic
strains) at the Research Center Weihenstephan for Brewing and Food Quality. TUM 35 grew after two weeks
of applying a tailored reactivation protocol in liquid wort. This paper presents research on the history of the
strain TUM 35. Its journey could be traced back from Freising-Weihenstephan to Nuremberg and to its origin
Coburg in upper Franconia. Its history also revealed why this formerly very successful old yeast strain
disappeared completely in the mid to late 1950s. We also confi rmed the species using specifi c qPCR systems
with marker DNA-regions for S. pastorianus identifi cation. PCR-capillary electrophoresis of the IGS2-314
rDNA fragment showed the close relation to the strain TUM 34/70 but also the subtle differences in the
DNA-fi ngerprint pattern. Phenotypic experiments and beer fermentation trials at volumes of 30 L could prove
that TUM 35 performed like a typical Frohberg-type lager strain. It produced a straight, neutral and soft aroma
profi le in the fi nal beer with a high degree of fermentation. Results of fermentation by-product analysis and
other main beer parameters of the beer produced with TUM 35 lie within the specifi cations and reference
values for Frohberg-type lager beers. In contrast to most other lager beer strains, TUM 35 produced no sulfuric
aromas that could be sensed by the tasting panel in the fi nal beer. This study is a fi rst approach to improve the
understanding of old lager beer yeast strains and also opens up opportunities for breweries to use forgotten
old strains for standard or historic lager beer production.
Descriptors: yeast, Saccharomyces pastorianus, TUM 35, lager beer, fermentation, history, identifi cation, differentiation
Mathias Hutzler, Dominique Stretz, Korbinian Haslbeck, Tim Meier-
Dörnberg, Hubert Walter, Fritz Jacob, Maximilian Michel, TU München,
Research Center Weihenstephan for Brewing and Food Quality, Freising,
Germany; Ludwig Narziß, TU München, Wissenschaftszentrum Wei-
henstephan, Freising, Germany; Mario Schäfer, Tobias Zollo, Bayerische
Staatsbrauerei Weihenstephan, Freising, Germany; corresponding author:
m.hutzler@tum.de
1 Introduction
The most important yeast species for fermentation technology
belong to the Saccharomyces genus [1-4]. Presently the genus
includes eight natural species (S. cerevisiae, S. paradoxus, S.
mikatae, S. kudriavzevii, S. arboricola, S. jurei, S. eubayanus and
S. uvarum), together with two artifi cial and hybrid species exclu-
sively associated with human-made fermentative environments
(S. pastorianus and S. bayanus) [5, 6]. Libkind et al. published
that the yeast strains of the species Saccharomyces pastorianus
used in lager beer production are genetic hybrids of Saccharo-
myces cerevisiae and the Patagonian wild yeast Saccharomyces
eubayanus [7]. Older literature (pre-1987) refers to lager yeasts
as S. carlsbergensis instead of S. pastorianus. Although the two
names refer to the same group of yeasts, two distinct type strains
have to be considered. CBS 1538 is the type strain of S. pasto-
rianus and CBS 1513 is the type strain of S. carlsbergensis. As
explained above, the nomenclatural change concerned the belated
recognition that S. pastorianus was an older name than S. carls-
bergensis. This hybrid species is by far the most relevant brewing
yeast, since lager beer accounts for 94 % of the world market
[8]. However, S. pastorianus has never been isolated outside the
brewing environment; therefore, it appears to be an exclusively
domesticated species.
Although the precise origin of lager yeasts is unknown, it is logical
to assume that their emergence is intimately related with the origin
and development of lager-brewing in the 15th century in Bavaria [9,
March / April 2019 (Vol. 72) 70
BrewingScience
10]. S. pastorianus was described by Hansen in 1904, however, the
hybrid nature of lager yeasts was fi rst hypothesized much later in
the 1980s [11-13, 6]. Later S. euybayanus was also discovered in
North America, Asia and New Zealand [14-16]. Dunn and Sherlock
postulated that at least two hybridization events took place and
that all Saccharomyces pastorianus lager strains consist of at least
two types [17]. Other studies also confi rmed the existence of two
different lager yeast groups [18, 1, 19, 20]. The S. pastorianus
strains were divided into the Saaz and Frohberg groups [18, 6, 20].
Gallone et al. described different hypothetical evolutionary models
that could have led to the lager groups Saaz and Frohberg [21].
The terms ‘Frohberg’ and ‘Saaz’ were coined at the VLB at Berlin
by Delbrück and Lindner at the turn of the 19th century, describing
two pure type strains with a characteristic fermentation behavior.
They had been isolated from bottom-fermenting ‘Stellhefen’ (mixed
brewing yeast populations) obtained from Frohberg’s brewery at
Grimma (Saxony) and the Saaz brewhouse (Bohemia) [10, 22]. The
most famous Saaz-type yeast strains are Carlsberg ‘Unterhefe Nr.
1’ (CBS 1513) and ‘Unterhefe Nr. 2’ (CBS 1503) isolated by Emil
Christian Hansen whereas the most famous member of the Frohberg
group is Saccharomyces pastorianus TUM 34/70 [1, 6, 20]. Industrial
strains exhibiting strong fermentation performance belong to the
Frohberg group [18, 1]. Besides distinct genetic compositions, the
two lineages of lager yeasts have different fl avor and fermentation
profi les. Saaz strains grow better at low temperatures (10 °C), have
a relatively poor fermentation performance compared with Froh-
berg strains at 22 °C, do not utilize maltotriose, and produce lower
amounts of esters [18, 20]. Saccharomyces pastorianus lager yeast
strains are bottom-fermenting, while Saccharomyces cerevisiae
brewing yeast strains are top-fermenting. Both differ signifi cantly
from one another with regard to numerous characteristics. The
former are able to ferment at lower temperatures, fl occulate well
during primary fermentation and are harvested from the bottom
of a fermentation tank, e.g. cylindroconical tank (CCT). The latter
cease to function or they ferment very slowly at low temperatures
(6 – 10 °C). They ferment most readily at temperatures between 15
and 25 °C, depending on the strain and type of wort. Cell growth is
more vigorous among top-fermenting brewing yeast strains, and the
cells do not fl occulate as rapidly compared with bottom-fermenting
yeast strains [1, 6]. Therefore, while both groups evidence their
alloploid hybrid nature by combining the cryotolerant phenotype
of S. eubayanus with the good fermentation performance of S.
cerevisiae, detectable differences both in cryotolerance and in
fermentation have been linked to the proportional amount of S.
eubayanus DNA retained in their respective genomes. Moreover,
the different phenotypes offer an explanation for the dominance
of Frohberg strains in modern industrial brewing [6].
A brewing yeast strain should be taxonomically classifi ed by means
of molecular biological methods at the species and strain level. Its
propagation and fermentation performance, as well as its aroma
profi le, should also be characterized [1]. TUM 34/70 is one of the
most abundant lager yeast strains in the brewing industry, and is
a strain with which many other lager strains are compared regard-
ing fermentation performance and the pure fl avor of lager beers
produced with it [1]. The genome of TUM 34/70 was also the fi rst
of the bottom-fermenting strains to be sequenced and published
[23]. A study conducted by Mueller-Auffermann used the charac-
teristics of TUM 34/70 as a reference for developing a method
to rapidly compare the performance of lager yeast strains [24].
Studies that compare a broader set of lager strains phenotypically
and in terms of fermentation and technological characteristics in
brewing are rare. A study by Donhauser et al. also compared a
broad set of lager strains on a 20-L scale pilot fermentation set-
up [25, 26]. This group confi rmed that TUM 34/70 is one of the
best-performing strains in terms of fermentation degree and beer
maturation. The fermentation by-products were within the desired
range and the sensorial impression scored highly. These fi ndings
confi rmed those of Prof. Narziß in 1956, TUM 34 (the ancestor of
TUM 34/70) being the most appropriate lager strain for lager beer
production within a strain set of seven lager strains and one lager
strain yeast mixture [27, 28].
Before S. pastorianus TUM 34/70 started its triumphal march after
the dissertation and publication of Narziß, the yeast strain TUM 35
was one of the most widespread lager yeast strains in Franconia
(Northern Bavaria). The aim of this study was to characterize the
recultivated or ‘resurrected’ lager strain TUM 35. In this study we
gathered historical data on why this formerly widespread indus-
trial lager strain completely disappeared. A lot of the historic facts
related to strain TUM 35 are based on personal correspondence
and records belonging to Prof. Ludwig Narziß, who is one of the
authors of this study. It was not included in the strain set of Narziß’s
studies in the 1960s whereas Wagner used that strain 2002 in a
comparative lager strain study [29]. In 2002, Wagner presented
at the fi rst Weihenstephaner Hefesymposium (yeast symposium)
that TUM 35 was the 16th most appropriate fl occulating lager strain
whereas TUM 34/70 was the most appropriate fl occulating lager
strain with regard to various technological and sensorial evaluation
parameters. In this context TUM 35 was actively used in Wagner’s
studies but it was not actively sold or promoted during that time and
only used for research purposes. Before and after these studies
the lager strain TUM 35 was neither used for industrial applica-
tions nor for scientifi c studies. This paper describes why this strain
was of great importance after the Second World War, that it had
extraordinary aroma profi le properties in the past, and why it dis-
appeared completely for industrial-scale fermentations. Following
Wagner’s study in 2002 the strain was completely forgotten and
was neither present as an active slant agar culture nor as a cryo
stock culture at – 80 °C. The strain was lost. Fortunately, our group
found a freeze-dried stock on milk-powder (composition unknown)
in HPLC fl asks. These were found in a forgotten box in a storage
room also containing other yeast strain stocks at our institute. We
reactivated the yeast from the dried pellet in liquid wort according
to a tailored protocol that is described in the paper. The aim of the
study was to characterize the strain TUM 35 genetically, to evalu-
ate its fermentation characteristics in a 30-L brewing trial and the
quality of the fi nal beer. Additionally, the strain history and route
should be revealed.
2 Materials and methods
2.1 Yeast strains
Two lager yeast strains (Saccharomyces pastorianus TUM 34/70
and TUM 35) and one German wheat beer strain (Saccharomyces
71 March / April 2019 (Vol. 72) BrewingScience
cerevisiae TUM 68) were obtained from the Yeast Center division of
the Research Center Weihenstephan for Brewing and Food Quality
(Table 1) . The origin and the historic importance of the yeast strain
TUM 35 are described in detail in the next chapter ‘reactivation of
a freeze-dried culture’ and in the fi rst chapter of the results section
‘historic information research’.
2.2 Reactivation of a freeze dried yeast stock
An old S. pastorianus TUM 35 conservation stock in an HPLC glass
fl ask (age unknown) was found in 2018 at the Research Center
Weihenstephan in a forgotten box in a storage room containing
approx. 20 different old freeze-dried stock cultures. Historic strains
were stored in a box in the same order and with the same strain
number descriptions as written on the list shown in fi gure 2 (results,
see page 72). Figure 2 only shows the top section of the historic
list. As the order and numbers are identical for all strains we as-
sume that the HPLC vial with the number 35 found in 2018 is really
the historic strain TUM 35. Until then, S. pastorianus TUM 35 had
seemed to be lost. Our group reactivated the freeze-dried culture
of TUM 35 according to the steps shown in fi gure 1.
The sealed HPLC fl ask was opened aseptically and the pellet was
placed in 50 mL sterile liquid wort medium in a 100 mL glass fl ask
with sterile cotton plug. The fl ask was incubated at 27 °C for 14
days under aerobic conditions and was shaken continuously at 80
rpm (Wiseshake, Witeg GmbH, Wertheim, Germany). A loop of the
yeast suspension was streaked on wort slant agar and incubated
aerobically at 27 °C for 5 days. The slant agar was stored at 4 °C
and used as new stock culture for further studies.
2.3 Genetic identifi cation and strain determination
2.3.1 DNA extraction
To isolate the DNA from each investigated yeast isolate, cultures
were taken from wort agar slants using a sterile inoculation loop,
transferred to a 1.5 mL micro centrifuge tube, and mixed with an
aliquot of 200 μL InstaGeneTM Matrix solution (Biorad, Munich,
Germany). Each tube was vortexed for ten seconds and incubated
at 56 °C for 30 minutes, followed by another ten seconds of vortex-
ing and incubation at 96 °C for eight minutes. The incubation steps
occurred in a Thermomix 5436 (Eppendorf, Hamburg, Germany).
After incubation, the tubes were centrifuged at 13,000 × g for two
minutes and then a 100 μL aliquot of the supernatant containing
the DNA was transferred to a new 1.5 mL micro centrifuge tube
[30-32]. The DNA concentration was adjusted to 25 ng/μL after
being measured by a Nanodrop 1000 spectrophotometer (Thermo
Scientifi c, Wilmington, USA).
Table 1 Brewing yeast strains, their industrial application and
their original conservation stock type
Yeast strains
(TUM identifi er)
Industrial
application
Conservation
stock type
Saccharomyces
pastorianus TUM 35
Lager beer produc-
tion, bottom-ferment-
ing fl occulant yeast
strain
Freeze-dried culture
on milk powder in
sealed HPLC glass
fl asks
Saccharomyces
pastorianus TUM 34/70
Lager beer produc-
tion, bottom-ferment-
ing fl occulant yeast
strain
Slant agar
Saccharomyces
cerevisiae TUM 68
German wheat beer
production, top-fer-
menting yeast strain
Slant agar
Fig. 1 Reactivation of freeze-dried culture of Saccharomyces pastorianus TUM 35 in aerobic shaking culture in sterile wort with subse-
quent streak on wort agar
March / April 2019 (Vol. 72) 72
BrewingScience
2.3.2 Quantitative polymerase chain reaction (q-PCR)
RT-PCR (Light Cycler® 480 II. Roche Diagnostics Deutschland
GmbH, Mannheim, Germany) was used to taxonomically classify
the isolates. The applied primer and TaqMan® probe sequences
and RT-PCR procedure followed that of Hutzler and publications of
our group [30, 31, 1, 33, 32, 6]. The RT-PCR systems Sc-GRC3,
Sce, TF-COXII, Sbp, Seub, BF-300, BF-LRE1, Sdia are described
as compatible and were performed with 10 μL 2x Mastermix (Light
Cycler® 480 Probe Master, Roche, Germany), 1.4 μL DNA-free
water, 0.8 μL (400 nM) of each primer (Biomers, Ulm, Germany),
0.4 μL (200 nM) probe (Biomers, Ulm, Germany; MGB probe from
ThermoFisher scientifi c, Applied Biosystems®, USA). The yeast
strains S. cerevisiae TUM 68 and S. pastorianus TUM 34/70 were
used as a positive and negative control according to the RT-PCR
system tested.
2.3.3 DNA fi ngerprinting (PCR-amplicon-capillary electro-
phoresis of the IGS2-314 fragment)
In order to determine if isolates represented different or similar
strains, genetic fi ngerprints were generated using the IGS2-
314 method [30, 31]. The IGS2 is a spacer region within of the
ribosomal cluster. To a partial sequence of the intergenic spacer
2 (IGS2-314) the specifi c primers IGS2-314f (5’-CGGGTAACC-
CAGTTCCTCACT-3’) and IGS2-314r (5’-GTAGCATATATTTCTT-
GTGTGAGAAAGGT-3’) (Biomers GmbH, Ulm, Germany) [34,
30, 31] were used at a concentration of 600 nM as described by
Hutzler 2010 [30, 31].
PCR was performed with 22.5 μL RedTaq Mastermix (2x) (Genaxx-
on, Ulm, Germany) and 2.5 μL template DNA with a total reaction
volume of 25 μL. The Mastermix contained 12.5 μL buffer solution
(RedTaq Mastermix), 7.0 μL DNA-free PCR water and 1.5 μL of
each primer (Biomers, Munich, Germany). Cycling parameters
are described by Hutzler [30, 31]. PCR was performed using
a SensoQuest LabCycler48s (SensoQuest GmbH, Gottingen,
Germany). Amplifi ed fragments were analyzed using a capillary
electrophoresis system (Agilent DNA 1000 kit) following the manu-
facturer’s recommendations (lab on a chip, Bioanalyzer Agilent
2100, Agilent Technologies, Santa Clara, CA, USA).The results of
the PCR-product separation (DNA-Fingerprint) using the capillary
electrophoresis were imaged as a gel image and eletropherograms
using the Bioanalyzer 2100 software.
2.4 Brewing trials
2.4.1 Wort
For the fermentation trial, industrial hot wort (IBU 31) for the beer
type Weihenstephaner Original (a Bavarian Helles Lager beer
type with 21 IBU, used hop variety Hallertauer Perle) was sampled
aseptically from the State Brewery Weihenstephan in 20 L Cornelius
KEGs and 30 L was aseptically transferred to pilot fermentation
tanks. When wort reached pitching temperature after cooling, the
wort was pitched as described below. The original gravity and pH
value of the wort are shown in table 3 (results section, see page 73)
2.4.2 Propagation
In order to propagate yeasts, isolates were inoculated from agar
slants (yeast pure culture) into 2 × 60 mL of sterile wort medium
in 2 × 100 mL Erlenmeyer fl asks and incubated for 72 h at ambi-
ent temperature (20 °C) and pressure, and agitated at 80 rpm
using a WiseShake 207 orbital shaker (Witeg Labortechnik
GmbH, Wertheim, Germany). After incubation, the 2 × 60 mL were
transferred to 2 × 2 L of sterile wort medium in 2 × 5 L glass fl asks
closed with sterile cotton plug and further propagated at the same
conditions for an additional 72 hours. After allowing six hours for
sedimentation, the supernatant was decanted and the sediments of
both 5 L fl asks were combined in a new sterile 5 L fl ask. The yeast
cell concentration was determined using a Thoma cell counting
chamber with a chamber depth of 0.1 mm and an area per square
of 0.00025 m2 (Brand GmbH&Co.KG, Wertheim, Germany). The
volume of the yeast suspension was adjusted to pitch the 30 L of
wort at a starting yeast cell concentration of 15 × 106 cells per mL
in the pitched wort.
Fig. 2 Section of historic TUM strain list (in German) that docu-
ments the origin of strain TUM 35 and TUM 34/70
Tab le 2 Qualitative results of the qPCR systems used for the investigated isolates to differentiate Saccharomyces sensu stricto species;
positive (+), negative (–)
qPCR System
Species Yeast isolates /
reference strains Sc-GRC3 Sce TF-COXII Sbp Seub BF-LRE1 BF-300 Sdia
S. cerevisiae TUM 68 + + + – – – – –
S. pastorianus
TUM 35 + + – + + + + –
TUM 34/70 + + – + + + + –
73 March / April 2019 (Vol. 72) BrewingScience
2.4.3 Fermentation
Pilot-scale brewing trials were performed with volumes of 30 L
using stainless steel fermentation tanks (Gresser C., Regensburg,
Germany) with a total volume of 60 L. Brewing trials were evaluated
by pitching the calculated volume of yeast suspension to start the
30 L scale fermentation with 15 × 106 cells per mL. The wort was
aerated to obtain an oxygen concentration of 8 mg/L. Specifi c grav-
ity was measured daily as cells in suspension. Yeast isolates were
added at an inoculation rate of 15 million cells/g of homogeneous
mixed wort. Primary fermentation was pitched at 10 °C and main-
tained at 11 °C. After 13 days cooling down to 1 °C started. Lager
time was 14 days at 1 °C. A temperature of 11 °C was maintained
until day 13 and diacetyl reduction and therefore maturation was
almost complete at day 13. The specifi c gravity and pH of samples
were determined from the fi ltered fermentation samples using a
DMA 35N (Anton-Paar GmbH, Graz, Austria) for specifi c gravity
and a pH3210 (WZW, Wissenschaftlich-Technische Werkstätten
GmbH, Weilheim, Germany) for pH measurement. The samples
were fi ltered using a Whatman® folded fi lter paper with a diameter
of 320 mm (GE Healthcare Europe GmbH, Freiburg, Germany).
2.5 Analytical methods
After 14 days lagering, the fi nished beers were analyzed for
physical and chemical attributes, which included the following
parameters: ethanol, pH, specifi c gravity, degree of attenuation,
free vicinal diketones and the concentration of fermentation by-
products. Ethanol, pH, specifi c gravity, and degree of attenuation
were measured using an Anton Paar DMA 5000 Density Meter
with Alcolyzer Plus measuring module, pH measuring module,
and Xsample 122 sample changer (Anton-Paar GmbH, Ostfi ldern,
Germany). Free vicinal diketones were quantifi ed by a Clarus
500 gas chromatograph (Perkin-Elmer, USA) with a headspace
unit and Elite 5 60 m 1.5DF column using a 2,3-hexanedione
internal standard. The fi nal concentrations of fermentation by-
products (e.g. acetaldehyde, ethyl acetate, n-propanol, i-butanol,
isoamyl acetate, amyl alcohols, diacetyl, 2,3-pentanedione) were
measured according to MEBAK II (3.2.21) methods using a gas
chromatograph with a headspace unit and INNOWAX cross-linked
polyethylene-glycol 60 m × 0.32 mm 0.5 μm column (Perkin-Elmer,
USA). Fatty acids and fatty acid esters were determined by gas
chromatography with a fl ame ionization detector (GC-FID) Clarus
500 (Perkin-Elmer, USA) according to the following temperature
protocol (1 min, 60 °C; 3 min, 22 °C (5 °C/min); 8 min, 240 °C
(20°C/min)), detector temperature 250 °C, injector temperature
200 °C, column: 50 m × 0.32 mm, Phenomenex FFAP, 0.25 μm,
carrier gas helium 5.0 ECD-quality, 20 mL/min, 2 bar.
2.5.1 Phenolic off-fl avor test (POF-test)
TUM yeast culture isolates were taken from wort agar slopes
and spread on a YM-agar plate containing one of the precursor
substances: ferulic acid, cinnamic acid and coumaric acid. After
three days of incubation at 24 °C, the three single agar plates per
yeast isolate were evaluated by sniffi ng (blind, 8 testers that were
trained on the target substances 4-VG, 4-VS, 4-VP) to detect
any of the following aromas: ferulic acid becomes 4-vinylguajacol
(4-VG, clove-like), cinnamic acid becomes 4-vinylstyrene (4-VS,
styrofoam-like) and coumaric acid becomes 4-vinylphenol (4-VP,
medicinal-like). S. cerevisiae LeoBavaricus - TUM 68® and S.
pastorianus Frisinga - TUM 34/70® were used as a positive and a
negative control, respectively [30, 31]. A YM-media was made for
the YM-agar plates by adding distilled water to 3.0 g malt extract,
3.0 g yeast extract, 5.0 g peptone, 11.0 g glucose monohydrate
and 20.0 g agar to 1000 mL and autoclaved. After autoclaving,
an aliquot of the following stock solutions was added to the YM-
media at 45 – 50 °C under sterile conditions. For the stock solution
of coumaric acid, 100 mg of the instant was dissolved in 10 mL of
96 % [v/v] ethanol. The stock solution of ferulic and cinnamic acid
was made by dissolving 1 g in 20 mL of 96 % [v/v] ethanol. 10 mL
coumaric acid, 2 mL ferulic acid or 2 mL cinnamic acid stock solu-
tion was added for 1000 mL YM-media.
2.6 Sensory evaluation
The beer produced by TUM 35 was judged by 10 trained panelists
with long-standing experience in the sensorial evaluation of beer.
The beer was judged according to the DLG-scheme (Deutsche
Landwirtschafts-Gesellschaft). In this scheme, the fl avor, taste,
quality of the bitterness, mouthfeel and carbonization of the beer
are judged and each awarded 1 – 5 points, one being the lowest
(negative), and fi ve being the highest (positive) score. A further
descriptive sensory evaluation of the beer was performed.
3 Results and discussion
3.1 Historic information research
Most of the historic facts in this chapter are based on information
and personal correspondence provided by Prof. Ludwig Narziß
who is also a co-author of the present study. Due to the Second
World War (1939 – 1945) and the resulting shortage of malt and
hops, German breweries were forced to produce beer referred to
as “Dünnbier” (translation: thin beer) up until 1948. A maximum
original gravity of 1.7 °P was allowed for beer production. In these
times, there was a variety of different bottom-fermenting yeast strains
used in Bavarian breweries. Breweries around Nuremberg and
Augsburg were supplied with yeast by the Hasen-Bräu Augsburg
brewery. Shortly before 1955, the yeast strain of the Hasen-Bräu
was deposited by the Reif brewery Nuremberg (Brauhaus Nürnberg)
at the TU München as the strain TUM 34. This yeast strain is an
ancestor of today’s most commonly used bottom-fermenting yeast
strain worldwide – Frisinga - TUM 34/70 (progenies of the strain
in other collections are denominated with the identifi ers 34/70, W
34/70 or Weihenstephan 34/70).
Table 3 Chemical analyses of the beer produced with TUM 35
Parameters Units Wort Final beer
Original gravity °P 12.2 12.2
Final gravity °P potentially 1.3* 1.6
Attenuation % potentially 89.0* 87.4
pH value - 5.45 4.63
*determined with an excess amount of the yeast strain TUM 34/70 at
20 °C
March / April 2019 (Vol. 72) 74
BrewingScience
Shortly after the Second World War however, another yeast strain
called “C” was common in many breweries around Franconia
(northern Bavaria). The name came from the city name Coburg
and the brewery Hofbrauhaus Coburg, where it originally came
from (Fig. 2 & 3). The strain we found marked with the number 35
on the HPLC vial with the freeze-dried culture was in a box with
other freeze-dried cultures. The number order was completely
congruent with the order of the numbers
in the historic list (Fig. 2). This is strong
evidence that it must be TUM 35. It is not
known where it came from before the time
in Coburg. This strain was fi rst deposited at
an early research center for brewing called
the Nürnberger Versuchsstation für Bier-
brauerei der Landesgewerbeanstalt (LGA).
In 1948 the German economy recovered,
as did the breweries and the malt and hops
market. From that point, beer with an original
gravity of 8 °P could be brewed. The beers
produced at that time with the yeast strain C
had a mild, balanced fl avor with fi ne notes
of hops according to Prof. Ludwig Narziß.
However, the yeast strain C was again de-
posited by brew master Carl Hager from the
Brauhaus Nürnberg before 1952 at the TU
München and called strain TUM 35. Figure
2 proves that strain 35 derives from Coburg
and was isolated at Nuremberg. Strain 34/70
also originated from the Brauhaus Nürnberg
(brewery Reif) and Professor Narziß Ludwig
knows from his master brewer colleague Carl
Hager that the strain TUM 34 originally came
from Hasen-Bräu, Augsburg.
In 1952, bad weather conditions forced the barley to ripen pre-
maturely. As a result, malt produced from this barley was low in
amino acids and soluble protein. Furthermore, the fi nal attenu-
ation of wort produced from this malt was analyzed at 76 % and
lower. This exceptional composition of the produced worts had
a strong impact on the bottom-fermenting yeast strains used
everywhere. Once fermenting, the yeasts did not fl occulate well
and stayed in suspension much longer than usual. Even though
varying strains were used in different breweries, they all faced
similar problems related to poorly fl occulating yeasts. The strain
TUM 35 had the worst fl occulation of all the strains and caused
the greatest problems during production. It fermented until the
determined fi nal attenuation but it was very diffi cult to crop enough
yeast biomass for further pitching, even with enforced cooling. In
1953 there was a lot of rainfall, which again led to a bad barley
harvest and the brewer’s yeast strains behaved in the opposite
manner. The yeast cells fl occulated early, which is now known as
PYF (premature yeast fl occulation). Again, TUM 35 was one of
the worst-fermenting yeast strains and was therefore no longer
used by any of the breweries. All the breweries switched to a
new yeast strain called TUM 44, which was much more robust
against changes in raw materials. However, the fl avor was not
comparable to the mild balanced aroma profi le of beer produced
using the strain TUM 35. Since this time, the yeast strain TUM 35,
also just called “C”, was lost and not used in any brewery. The
fl avor issues that a lot of the breweries experienced with TUM 44
meant the start of TUM 34’s success story, which continues today
with the yeast strain Frisinga TUM 34/70. The only time TUM 35
was used again, was in strain comparison studies in 1987 and
2002 by Donhauser et al. and Wagner but no further industrial
application followed [25, 26, 29]. The strain was forgotten again
and no active or cryo culture was deposited at the Research
Center Weihenstephan.
Fig. 3 Origin and traffi c of yeast strains TUM 34 and TUM 35
illustrated in a profi le map of Bavaria (free profi le map of
www.d-maps.com)
Fig. 4 DNA Fingerprint of PCR amplicons of the IGS2-314 rDNA fragments of the lager
brewing yeast strains S. pastorianus TUM 35 and TUM 34/70 (left: Bioanalyzer gel
image; right electropherogram images)
75 March / April 2019 (Vol. 72) BrewingScience
3.2 Genetic analyses
The qPCR analysis results of various published marker DNA-
sequences revealed that the DNA of the strain S. pastorianus
TUM 35 has a typical marker DNA-sequence pattern like the S.
pastorianus strain TUM 34/70, which is like a type-strain for the
Frohberg lager yeast subgroup (Table 2). The marker gene pattern
differs from the pattern of the top-fermenting wheat beer strain S.
cerevisiae TUM 68.
Quantitative PCR analyses confi rmed that the strain TUM 35 be-
longs to the Saccharomyces pastorianus species. A genetic strain
comparison using the capillary electrophoresis of PCR amplicons
of the IGS2-314 rDNA fragments revealed that TUM 34/70 and
TUM 35 are very similar in terms of these highly discriminating
regions. There are some quantitative differences for some specifi c
DNA amplicons and there are bands between 80 and 90 s runtime
that can be used for qualitative differentiation. Further investiga-
tions such as the whole-genome sequencing (WGS) of TUM 35
and subsequent comparison with other lager
strains would reveal the detailed phylogenetic
relation of TUM 35 within the Saccharomyces
pastorianus lager clade. Some researchers
conducted detailed WGS comparisons for
Saccharomyces cerevisiae beer strains.
Saccharomyces pastorianus strains were not
the focus of their studies [35, 36, 6]. Probably
such WGS genetic comparisons for hybrid
species such as Saccharomyces pastorianus
will follow soon.
3.3 Aroma profi ling and brewing
trials
The applied yeast strain TUM 35 fermented
well at 11 °C, resulting in a fi nal gravity of
1.6 °P (original gravity 12.2 °P) and a fi nal
attenuation of 87.4 % after 9 days (Table 3).
The previously determined fi nal attenuation of 89 % using the yeast
strain TUM 34/70 at 20 °C was almost achieved (MEBAK method),
showing a good fermentation performance. After achieving the
fi nal gravity of 1.6 °P and no change in gravity for three following
days, the beer was cooled down for maturation and stored at 0 °C
for 14 days.
The fermentation kinetics of the main fermentation are shown in
fi gure 5. The fermentation kinetics are comparable to the pilot
fermentation kinetics of other studies applying other lager yeast
strains [25, 26, 33, 24, 29].
3.4 Phenolic off-fl avor
Table 4 shows the results of the POF-tests evaluated by sniffi ng.
As shown in table 4, only the wheat beer strain Saccharomyces
cerevisiae TUM 68 (positive control) is capable of building phenolic
off-fl avors. For strain TUM 68, all three corresponding POF-fl avors
were detected by sniffi ng. The lager yeast strains Saccharomyces
pastorianus TUM 34/70 (negative control) and TUM 35 are POF
negative. These two yeast strains cannot decarboxylate any of the
precursor acids. Therefore the PAD or/and FDC activity might be
inactive or blocked [37, 38].
3.5 Flavor compounds and fermentation by-products
The results of the measurements of the fatty acids, esters, higher
alcohols and fermentation by-products of the beer fermented with
TUM 35 presented in tables 5 and 6 show typical values for a
T able 4 POF results of the investigated yeast strains Saccharomy-
ces pastorianus TUM 35 (sample) and TUM 34/70 (negative
control) and Saccharomyces cerevisiae TUM 68 (positive
control)
Product/Precursor TUM 35 TUM 34/70 TUM 68
4-vinylguajacol/ferulic acid – – +
4-vinylphenol/coumaric acid – – +
4-vinylstyrene/cinnamic acid – – +
Table 5 Fatty acids, fatty acid esters and acetate esters detected in the beer sample fermented with TUM 35
Fatty acids Content [mg/L] Fatty acid esters Content [mg/L] Acetate esters Content [mg/L]
Caproic acid 1.3 Butyric acid ethyl ester 0.07 Isoamyl acetate 1.1
Caprylic acid 2 Decanoic acid ethyl ester 0.1 Ethyl acetate 14.1
Decanoic acid 0.69 Acetic acid 2-phenylethyl ester 0.26
Isovaleric acid 1.2 Acetic acid isobutyl ester 0.03
Caproic acid ethyl ester 0.1
Caprylic acid ethyl ester 0.26
Fig. 5 S. pastorianus TUM 35 fermentation profi le of a lager beer wort with an original
gravity of 12.2 °P at a main fermentation temperature of 11 °C
March / April 2019 (Vol. 72) 76
BrewingScience
12 °P lager beer [39, 40]. As lager beer is known for its clear taste
and fl avor, these analyses show exactly what is expected from a
lager beer. Diacetyl, an indicator of maturation, is well below the
threshold of 0.1 mg/L [41].
In his study in 2002, Wagner determined the following values for
the fi nal beer fermented with strain TUM 35: pH value 4.4, diacetyl
0.06 mg/L, sum of higher alcohols 72 mg/L, sum of acetate esters
13 mg/L, acetaldehyde 13 mg/L [29]. In our study, the pH value,
sum of higher alcohols, and sum of acetate esters are slightly
higher than in Wagner’s study. Diacetyl and acetaldehyde were
slightly lower in this study. The fermentation degree in Wagner’s
study is not comparable due to a different time of sampling. All
values in both studies are within the range of values for lager beers.
The slight differences can be explained by different substrates,
fermentation parameters and pitching rates.
3.6 Sensorial evaluation
The beer was judged by ten trained panelists and a mean of the
ten results was calculated. The results were compared with a mean
value of 480 lager beers, which were also tasted by the tasting
panel over one year (2018). The results shown in table 7 indicate
that the beer fermented with TUM 35 was above average in fl avor,
taste and quality of the bitterness in comparison with the mean
of 480 judged lager beers. The DLG evaluation of the beer made
with TUM 35 in the study of Wagner was 4.1 for smell and taste
in 2002 [29]. This is a difference of 0.3 DLG points. In Wagner’s
study, only 5 out of 41 beers made with different lager strains had
evaluations above 4.2. One possible reason for the shift could be
the differing wort substrate and the differing testing panel.
3.7 Sensory description
The beer was described in smell as pure and delightful, fresh and
yeast-typical. The taste was described as being pleasantly fresh-
yeast-typical, full-bodied with a slight bitterness, well balanced and
soft. No sulfuric aromas could be detected in the smell or taste.
Fresh sulfuric aroma profi les are often typical of beers produced
with other lager strains.
4 Conclusion/Summary
The strain history of Saccharomyces pastorianus TUM 35 could
be traced back via Nuremberg to its origin in the city of Coburg in
the Upper Franconia district. It was a dominant and widespread
strain after World War II in Franconia (northern Bavaria) and lost
its impact as its fermentation performance was very susceptible
to the infl uence of bad malt quality. In most cases it was replaced
by other more robust lager strains and completely disappeared as
an industrial lager strain. In 2018, a freeze-dried stock could be
reactivated applying a tailored cultivation/incubation procedure in
liquid wort. Using specifi c qPCR systems the reactivated lager strain
TUM 35 was confi rmed to belong to the species S. pastorianus.
A fi ngerprint analysis confi rmed its close relation to the well-char-
acterized strain S. pastorianus TUM 34/70 with subtle differences
observed in the IGS2-314 rDNA fi ngerprint pattern. Phenotypic
trials revealed that TUM 35 is POF (phenolic-off-fl avor) negative
and in 30-L pilot fermentations, the strain performs comparably to
a typical fl occulant Frohberg-type lager strain. The analysis of the
fermentation by-products and main beer parameters were compa-
rable to standard lager beers. The sensory evaluation of the beer
produced with TUM 35 achieved higher scores than the mean of
the sensory evaluation points of 480 lager beers in the database.
The descriptive sensory evaluation confi rmed the high point ranking
as the beer was described as an exceptionally neutral, soft and
harmonic beer comprising a typical lager beer aroma profi le. This
study is a fi rst blueprint for the reactivation and characterization
of ‘old, historic’ lager strains. Due to its origin, the strain TUM 35
is now deposited under the brand name ‘Franconia – TUM 35’
at the Research Center Weihenstephan for Brewing and Food
quality and can be requested for further scientifi c studies or for
industrial applications.
5 References
1. Hutzler, M.; Koob, J.; Riedl, R.; Schneiderbanger, H.; Mueller-Auffer-
mann, K. and Jacob, F.: Yeast identifi cation and characterization In:
Hill, A.: Brewing Microbiology – Managing Microbes, Ensuring Quality
and Valorising Waste, 2015.
2. Kurtzman, C. P.; Fell, J. W. and Boekhout, T.: The yeasts, a taxonomic
study, 5th Edition: Elsevier, 2011.
3. Rainieri, S.; Zambonelli, C. and Kaneko, Y.: Saccharomyces sensu
stricto: systematics, genetic diversity and evolution, J. Biosci. Bioeng.,
96 (2003), no. 1, pp. 1-9.
4. Vaughan-Martini, A. and Martini, A.: Saccharomyces Meyen ex Rees
In: Kurtzman, C. P. and Fell, J. W.: The yeasts, a taxonomic study,
2011.
5. Naseeb, S.; James, S. A.; Alsammar, H.; Michaels, C. J.; Gini, B.;
Nueno-Palop, C.; Bond, C. J.; McGhie, H.; Roberts, I. N. and Delneri,
D.: Saccharomyces jurei sp. nov., isolation and genetic identifi cation of
a novel yeast species from Quercus robur, Int. J. Syst. Evol. Microbiol.,
67 (2017), no. 6, pp. 2046-2052.
6. Sampaio, J. P.; Pontes, A.; Libkind, D. and Hutzler, M.: Taxonomy,
Table 6 Higher alcohols and fermentation by-products detected
in the beer sample fermented with TUM 35
Higher alcohols Content
[mg/L]
Fermentation
by-products
Content
[mg/L]
n-propanol 15.0 Diacetyl 0.05
i-butanol 18.1 2,3-pentanedione 0.01
Amyl alcohol
(2,3-methylbutanol) 82.3 Acetaldehyde 5.3
Phenyl ethanol 27.9 Acetoin 1.2
Table 7 Mean results of the sensory evaluation by DLG-scheme
for the beer produced with TUM 35 and a mean of 480
lager beers
Attributes Mean TUM 35 Mean of 480
lager beers
Smell 4.4 4.2
Taste 4.4 4.2
Quality of bitterness 4.4 4.2
Mouthfeel 4.5 4.5
Carbonization 4.5 4.5
77 March / April 2019 (Vol. 72) BrewingScience
diversity, and typing of brewing yeasts In: Bokulich, N. A. and Bamforth,
C. W.: Brewing Microbiology: Current Research, Omics and Microbial
Ecology, 2017.
7. Libkind, D.; Hittinger, C. T.; Valerio, E.; Goncalves, C.; Dover, J.; John-
ston, M.; Goncalves, P. and Sampaio, J. P.: Microbe domestication and
the identifi cation of the wild genetic stock of lager-brewing yeast, Proc.
Natl. Acad. Sci. USA, 108 (2011), no. 35, pp. 14539-44.
8. Riese, J. C. and Eßlinger, H. M.: World beer market In: Eßlinger, H.
M.: Handbook of Brewing Processes, Technology, Markets, 2009.
9. Meußdoerffer, F.: A Comprehensive History of Beer Brewing In: Eßlinger,
H. M.: Handbook of Brewing Processes, Technology, Markets, 2009.
10. Meußdoerffer, F. and Zarnkow, M.: Das Bier, Eine Geschichte von
Hopfen und Malz, 1st Edition: C. H. Beck, 2016.
11. Martini, A. V. and Kurtzman, C. P.: Deoxyribonucleic acid relatedness
among species of the genus Saccharomyces sensu stricto, International
Journal of Systematic Bacteriology, 35 (1985), pp. 508-511.
12. Martini, A. V. and Martini, A.: Three newly delimited species of Sac-
charomyces sensu stricto, Antonie Van Leeuwenhoek, 53 (1987), no.
2, pp. 77-84.
13. Nilsson-Tillgren, T.; Gjermansen, C.; Kielland-Brandt, M. C.; Petersen,
J. G. L. and Holmberg, S.: Genetic differences between Saccharomyces
carlsbergensis and S. cerevisiae. Analysis of chromosome III by single
chromosome transfer, Carlsberg Res. Commun., 46 (1981), pp. 65–76.
14. Bing, J.; Han, P. J.; Liu, W. Q.; Wang, Q. M. and Bai, F. Y.: Evidence
for a Far East Asian origin of lager beer yeast, Current biology: CB,
24 (2014), no. 10, pp. R380-1.
15. Gayevskiy, V. and Goddard, M. R.: Saccharomyces eubayanus and
Saccharomyces arboricola reside in North Island native New Zealand
forests, Environmental microbiology, 18 (2016), no. 4, pp. 1137-47.
16. Peris, D.; Sylvester, K.; Libkind, D.; Goncalves, P.; Sampaio, J. P.;
Alexander, W. G. and Hittinger, C. T.: Population structure and reticulate
evolution of Saccharomyces eubayanus and its lager-brewing hybrids,
Molecular ecology, 23 (2014), no. 8, pp. 2031-45.
17. Dunn, B. and Sherlock, G.: Reconstruction of the genome origins
and evolution of the hybrid lager yeast Saccharomyces pastorianus,
Genome Research, 18 (2008), no. 10, pp. 1610-1623.
18. Gibson, B. R.; Storgards, E.; Krogerus, K. and Vidgren, V.: Comparative
physiology and fermentation performance of Saaz and Frohberg lager
yeast strains and the parental species Saccharomyces eubayanus,
Yeast, 30 (2013), no. 7, pp. 255-66.
19. Liti, G.; Peruffo, A.; James, S. A.; Roberts, I. N. and Louis, E. J.:
Inferences of evolutionary relationships from a population survey of
LTR-retrotransposons and telomeric-associated sequences in the
Saccharomyces sensu stricto complex, Yeast, 22 (2005), no. 3, pp.
177-92.
20. Walther, A.; Hesselbart, A. and Wendland, J.: Genome sequence of
Saccharomyces carlsbergensis, the world’s fi rst pure culture lager
yeast, G3 (Bethesda), 4 (2014), no. 5, pp. 783-93.
21. Gallone, B.; Mertens, S.; Gordon, J. L.; Maere, S.; Verstrepen, K. J.
and Steensels, J.: Origins, evolution, domestication and diversity of
Saccharomyces beer yeasts, Current opinion in biotechnology, 49
(2017), pp. 148-155.
22. Meussdoerffer, F. G.: A Comprehensive History of Beer Brewing In:
Eßlinger, H. M.: Handbook of Brewing Processes, Technology, Markets,
2009.
23. Nakao, Y.; Kanamori, T.; Itoh, T.; Kodama, Y.; Rainieri, S.; Nakamura,
N.; Shimonaga, T.; Hattori, M. and Ashikari, T.: Genome sequence
of the lager brewing yeast, an interspecies hybrid, DNA research:
an international journal for rapid publication of reports on genes and
genomes, 16 (2009), no. 2, pp. 115-29.
24. Mueller-Auffermann, K.; Caldera, A.; Jacob, F. and Hutzler, M.: Charac-
terization of Different Bottom Fermenting Saccharomyces pastorianus
Brewing Yeast Strains, BrewingScience – Monatsschrift für Brauwis-
senschaft, 68 (2015), no. 3/4, pp. 46-57.
25. Donhauser, S.; Wagner, D. and Gordon, D.: Hefestämme und Bier-
qualität, BRAUWELT, 127 (1987), no. 38, pp. 1654-1664.
26. Donhauser, S.; Wagner, D. and Guggeis, H.: Hefestämme und Bier-
qualität, BRAUWELT, 127 (1987), no. 29, pp. 1273-1280.
27. Narziß, L.: Der Einfl uss der Hefe auf die Eigenschaften des Bieres,
Fakultät für Brauwesen, Technische Universität München, dissertation,
1956.
28. Narziß, L.: Der Einfl uß der Hefe auf die Eigenschaften des Bieres,
Brauwissenschaft, 10 (1957), no. 3-5, pp. 58-66, 82-87, 139-144.
29. Wagner, D.: Einfl uß des Hefestammes auf die Bierqualität, 1.Weihen-
stephaner Hefesymposium, 2002, Freising-Weihenstephan,
30. Hutzler, M.: Entwicklung und Optimierung von Methoden zur Identifi -
zierung und Differenzierung von getränkerelevanten Hefen, Wissen-
schaftszentrum Weihenstephan, TU München, dissertation, 2009.
31. Hutzler, M.: Getränkerelevante Hefen – Identifi zierung und Differen-
zierung, 1st Edition: SVH-Verlag, 2010.
32. Meier-Dörnberg, T.; Michel, M.; Wagner, R. S.; Jacob, F. and Hutzler,
M.: Genetic and Phenotypic Characterization of Different Top-fermenting
Saccharomyces cerevisiae Ale Yeast Isolates, BrewingScience, 70
(2017), no. 1/2, pp. 9-25.
33. Meier-Dörnberg, T.; Hutzler, M.; Michel, M.; Methner, F.-J. and Jacob,
F.: The Importance of a Comparative Characterization of Saccharomy-
ces cerevisiae and Saccharomyces pastorianus Strains for Brewing,
Fermentation, 3 (2017), no. 41, pp. 1-25.
34. Buchl, N. R.; Hutzler, M.; Mietke-Hofmann, H.; Wenning, M. and Scherer,
S.: Differentiation of probiotic and environmental Saccharomyces
cerevisiae strains in animal feed, Journal of applied microbiology, 109
(2010), no. 3, pp. 783-91.
35. Gallone, B.; Steensels, J.; Prahl, T.; Soriaga, L.; Saels, V.; Herrera-
Malaver, B.; Merlevede, A.; Roncoroni, M.; Voordeckers, K.; Miraglia,
L.; Teiling, C.; Steffy, B.; Taylor, M.; Schwartz, A.; Richardson, T.;
White, C.; Baele, G.; Maere, S. and Verstrepen, K. J.: Domestication
and Divergence of Saccharomyces cerevisiae Beer Yeasts, Cell, 166
(2016), no. 6, pp. 1397-1410 e16.
36. Goncalves, M.; Pontes, A.; Almeida, P.; Barbosa, R.; Serra, M.; Libkind,
D.; Hutzler, M.; Goncalves, P. and Sampaio, J. P.: Distinct Domestica-
tion Trajectories in Top-Fermenting Beer Yeasts and Wine Yeasts, Curr.
Biol., 26 (2016), no. 20, pp. 2750-2761.
37. Mukai, N.; Masaki, K.; Fujii, T. and Iefuji, H.: Single nucleotide poly-
morphisms of PAD1 and FDC1 show a positive relationship with ferulic
acid decarboxylation ability among industrial yeasts used in alcoholic
beverage production, J. Biosci. Bioeng., 118 (2014), no. 1, pp. 50-5.
38. Richard, P.; Viljanen, K. and Penttila, M.: Overexpression of PAD1
and FDC1 results in signifi cant cinnamic acid decarboxylase activity
in Saccharomyces cerevisiae, AMB Express, 5 (2015), no. 12, pp. 1-5.
39. Dittrich, H.: Mikrobiologie der Lebensmittel – Getränke, 1st Edition:
Behr’s, 1993.
40. Narziss, L.; Back, W.; Gastl, M. and Zarnkow, M.: Abriss der Bierbrau-
erei, 8th Edition: Wiley-VCH, 2017.
41. Meilgaard, M.-C.: Flavour chemistry of beer. Part 2. Flavour and thresh-
old of 239 aroma volatiles, Technical Quarterly MBAA, 12 (1975), no.
3, pp. 151-168.
Received 1 February 2019, accepted 22 March 2019
