Diapositiva 1

1
A chemostat approach to analyze the distribution
of metabolic fluxes in wine yeasts during
alcoholic fermentation
Quirós, M.1, Martínez-Moreno, R.1,
Barreiro-Vázquez, A.2, Adelantado, N.2,
Vázquez-Lima, F. 3, Morales, P.1, Albiol, J.2,
Ferrer, P.2 and Gonzalez, R.1
1 Instituto de Ciencias de la Vid y del Vino
(CSIC-UR-CAR), Logroño, Spain. 2 Instituto de
Biotecnología y Biomedicina and 3 Departamento de
Ingeniería Química, Universidad Autónoma de
Barcelona, Barcelona, Spain.
Introduction Saccharomyces cerevisiae is the
most relevant yeast species conducting the
alcoholic fermentation that takes place during
winemaking. This biological process can
significantly vary depending on a group of
variables that include the yeast strain employed
and, among others, the concentration of sugar and
nitrogen present in the must, parameters that are
closely altered by the global warming phenomenon.
Unraveling the physiological behaviour and the
distribution of the metabolic fluxes of S.
cerevisiae under winemaking conditions would
allow the development of a metabolic model that
would help us to predict the physiological
behaviour of yeast under different
circumstances. The main goal of the present work
is to achieve different steady states using a
chemostat approach that mimick the main phases
ocurring during winemaking in order to analyze
the metabolic state of yeast during alcoholic
fermentation.
0.25 h-1 0.25 h-1 0.04 h-1 0.04 h-1
Corresponding batch phase Chemostat steady state Corresponding batch phase Chemostat steady state
Culture time (h) Culture time (h) 8 - 17 5.5 RT (1 RT 3.70h) 20 - 35 5.5 RT (1 RT 25h)
Specific growth rate (h-1) Specific growth rate (h-1) 0.37 0.15 0.27 0.023 0.034 0.04
Specific consumption / production rates (mmols g DCW-1 h-1) Glucose -7.3 -10.2 -8.9 -4.4 -7.2 -4.4
Specific consumption / production rates (mmols g DCW-1 h-1) Fructose -2.2 -7.7 -2.8 -2.2 -3.6 -2.5
Specific consumption / production rates (mmols g DCW-1 h-1) Ethanol 6.2 22.0 17.0 11.2 18.6 12.3
Specific consumption / production rates (mmols g DCW-1 h-1) Glycerol 1.08 3.16 1.68 0.67 1.11 0.78
Specific consumption / production rates (mmols g DCW-1 h-1) Acetic acid 0.07 0.19 0.12 0.09 0.15 0.02
Specific consumption / production rates (mmols g DCW-1 h-1) Succinic acid 0.04 0.26 0.08 0.03 0.05 0.07
Specific consumption / production rates (mmols g DCW-1 h-1) Lactic acid 0.04 0.08 0.09 0.03 0.04 0.02

• Near µmax conditions (Stage 1) 79 of the
  carbohydrate carbon consumed came from glucose
  whereas during Stage 2 fructose supplied 36.
• In both phases, carbon was mainly used in
  enerrgy production (etahbol synthesis) but
  increased from 79 in the exponential phase to
  90 in the transition phase. The flux into the
  TCA cycle was kept very low (near zero) in both
  conditions (Fig. 2).
• The flux directed to the pentose phosphate
  pathway was growth rate dependent (Fig. 2).
• The carbon flux distributions were similar to
  those previously obtained for batch cultures
  using synthetic musts (Varela et al., 2004) and
  for S. cerevisiae lab strains grown anaerobically
  in chemostat cultures (Jouhten et al., 2008)

Material and methods Strains and culture
media S. cerevisiae strain EC1118, an industrial
wine yeast commercialized by Lallemand Inc., was
the strain used in this study. Batch and
continuous anaerobic cultures were performed in a
Dasgip parallel fermentation system equipped with
four SR0400SS bioreactors containing 200 mL of a
synthetic must (pH 3.5) described elsewhere
(Quirós et al., 2010) at 28 C, 100 rpm and 10
L/h of N2 for headspace gassing. This synthetic
must was used as feed for the performance of
continuous cultures mimicking the exponential
growth phase during alcoholic fermentation
(D0.25 h-1) while the same must lacking ammonia
and presenting 75 of the aminoacids of the
original must was used for mimicking the
transition to stationary phase (D0.04-1). Metabo
lic flux analysis (MFA) The stoichiometric model
used for MFA was adapted from Varela et al.
(2004). Data consistency analysis was performed
prior to MFA. Experimental data was found to be
consistent within 95 significance level
according to the test proposed by Wang and
Stephanopoulos (1983).

• Results
• Two different phases occurring along a batch
  fermentation were mimicked using continuous
  cultures
• Stage 1 Exponential growth phase. No limiting
  nutrients (D0.25 h-1)
• Stage 2 Transition to stationary phase.
  Depletion of ammonia (D0.04 h-1)
• Figure 1 shows the CO2 transfer rates along the
  steady states performed at the two dilution rates
  studied. Table 1 shows the specific consumption
  and production rates calculated for this two
  first stages along the batch fermentation and
  those calculated for the steady states mimicking
  these conditions.

Figure 2. Comparison of metabolic flux.
Reconciled data (specific consumption and
production rate) were used in the model pathway.

• Conclusions
• The steady states obtained at both dilution
  rates assayed reflect the physiological state of
  S. cerevisiae EC1118 during the two phases of the
  batch fermentation.
• The medium used for feeding the chemostat
  culture performed at D0.04 h-1 needs to be
  improved as most of the specific production and
  consumption rates obtained for the steady states
  are very close to the range bounds calculated
  for the corresponding batch phase.
• This study proves that it is possible to
  reproduce different phases of a wine fermentation
  by performing chemostat cultures, providing a new
  tool for the future construction of a predictive
  model for the global process.

Figure 1. Carbon dioxide transfer rate (CTR) and
carbon dioxide transfer (VCT) dynamic during a
continuous culture at 0.25 h-1 dilution rate.

• References
• Jones et al. (2005) Climate change 73319-343.
• Jouhten et al. (2008) BMC Systems Biology 960.
• Quirós et al. (2010) International Journal of
  Food Microbiology 1399-14
• Varela et al. (2004) Applied Environmental
  Microbiology 703392-3400.
• Wang NS and Stephanopoulos G. (2003)
  Biotechnology and Bioengineering 252177-2208