OK, so one more long post. I hope I haven't hijacked this thread, but I get really intrigued by contradictions in the literature, in whatever field I'm researching. Procrastinating more, and doing a search in Google on "Crabtree effect in breweries", I came across a synopsis by Dr. George Fix himself on the division between aerobic respiration and anaerobic fermentation.
Beer and Loafing in Las Vegas - From the HBD Archive
This seems to be a response to a debate (similar to the one we're having now) on a Home Brew Digest forum from 1992:
(George J Fix)
Subject: Yeast Cycles (George Fix)
Date: 1992-10-09 20:45:43 GMT
There has been a lot of interesting discussion concerning yeast
cycles on HBD, and I can not resist inserting my two cents worth.
I have really enjoyed the insights Pierre Jelenc has shared with
us. He is absolutely correct in asserting that the classic division
between aerobic respiration and anaerobic fermentation greatly
oversimplifies what is actually going on in individual yeast cells.
Nevertheless, the aggregate behavior, where the total collection of
cells is viewed as an enzyme system, does display regular behavior.
In fact, the differential equations of enzyme kinetics are derived
from such models much in the same way thermodynamics is derived from
particle models via ensemble averaging. It is important to stress
that the equations of enzyme kinetics are valid only as a description
of the system as a whole. They do not predict the behavior of individual
yeast cells, since the eccentricies of the latter have been averaged
The kinetic models give the following picture of the fermentation. In
the early stages there is a net consumption of dissolved O2 as well as
a reduction of wort lipids such as oleic and linoleic acids. There is a
net increase of metabolic energy, which can be characterized as an
equivalent amount of ATP (adenosine triphosphate). This is accompanied
by a net increase in the cell density N(t) with time t. The biochemistry
is involved, but the associated mathematical description is simple, since
the kinetic equations are linear in this regime. Solution of these equations
shows an exponential growth in N(t) with time t, which is usually written
in terms of logarithms as follows:
log(N(t)) = C*t,
where C is the growth constant. I feel it is valid to call this regime
the aerobic respiratory growth phase, even through individual cells may deviate
from the aggregate behavior.
As the cell density increases and the dissolved O2 level decreases, the
nonlinear regime is approached. Things get really interesting here from a
mathematical point of view. Ironically, the biochemistry is straightforward.
What happens in the aggregate is carbon splitting of elementary sugars such
as glucose (G) and fructose (F) followed by formation of pyruvic acid and
then acetaldehyde. The final step is the reduction of acetaldehyde to
ethanol by yeast enzymes. Since O2 is not involved and since alcohol is formed,
I feel it is valid to call this phase anaerobic fermentation. It is important
to note that there is a net expendure of metabolic energy during this phase.
This is why the ATP buildup in the aerobic phase is so crucial to obtaining
a complete fermentation. Also the curve for N(t) vs. time flattens out.
There are a number of mechanisms that have been identified for inducing the
transition to the nonlinear regime (i.e., from respiration to fermentation).
One of the most important as far as practical brewing is concerned is the
Crabtree effect. It has been shown that a sufficiently high cell concentration
of G and F sugars will strongly induce anaerobic fermentation. Brewing yeast
take G's and F's directly into the cell. Sucrose (G-F) is broken up outside the
cell, and then the G and F fractions are then transported inside. In contrast,
maltose (G-G) and maltotriose (G-G-G) are taken intact into the cell, only
later to be broken down into G units. In an all grain wort, maltose is the
major fermentable sugar followed by maltotriose, the others being under 10% of
the total. This has important practical implications for respiration, for
the maltose concentrations will not induce the Crabtree effect (at least in
the levels that exit in normal beer wort) until the maltose is broken into G
units. At this point a proper respiratory cycle will have occured, assuming
of course that a sufficient amount of O2 is dissolved at the start of the
Because of this, I am in complete agreement with the general principles put
forward in Micah Millspaw's post on yeast propagation. I am less inspired by
the use of dextrose (which is the same as glucose) and sucrose as a substrate
for propagating yeast. The reason centers on the Crabtree effect. I am not
suggesting his methods will not work. Micah has a wall full of ribbons to prove
the contrary. I am suggesting, however, there may be a better ways to go.
Over 85% of commercial propagation and those done in research labs involved
with brewing strains use dilute wort (SG ~ 1.020). Paul Farnsworth has an
excellent discription of this procedure in his article that appeared in the
yeast issue of Zymurgy. I belong to the minority that propagates with full
strength hopped wort. The reasons for this and a description of the procedure
can be found in my article that appeared in Vol. 6 of BREWERY OPERATIONS
published by Brewers Publications. I also use an O2 feed during propagation
to induce the Pasteur effect, which is the exact opposite of the Crabtree
effect. Here fermentation is repressed in favor of respiratory cell growth.
I was working only with half a voice during the AHA conference in June, and
likely many points I was trying to make did not get across. I hope this is not
the case with the point about the practical value of testing yeast that have
been aerobically propagated. Minor technical errors can lead to major problems,
not only with aerobic bacteria but with mutation as well. Both should be
checked. Interestingly, the Wyeast strain 1056, which is the same as Siebel's
BRY-96, does particularly well with aerobic propagation. In fact, I have found
the much discussed tendency of this strain to mutate (something that has
happened with samples from both Wyeast and Siebel) is closely related to the
lack of a proper respiratory cycle. Thus, brewing procedure is the culprit,
not screw ups in Chicago or Portland.
So he seems to make several points that directly contradict the sources that claim S. Cer. do not respire in normal brewery fermentations, and instead suggests that:
1) yeast do respire in the initial stages
2) this early stage respiration is not repressed by a Crabtree effect b/c maltose is the primary sugar being consumed and not glucose
3) once the Crabtree effect does become relevant, a significant phase of respiration has occurred
4) the Pasteur effect can be utilized for S. Cer. to inhibit fermentation in favor of respiration (i.e. the Crabtree effect is effectively microscopically reversible)
So, again, I can't say with 100% certainty that Fix is right about everything within this description, but it seems like a sound argument to me. I haven't read any of the sources that claim S. Cer. do not undergo respiration (or can't be subjected to the Pasteur effect) in typical beer worts, so I can't speak about them, but...if Fix is right, then those sources are way off base, and if those sources are right, then Fix is certainly way off base (and maybe he is, he's a Ph.D in math, not biology - but you would think someone would have directly called him on it by now).