13 Oscillations in Yeast Glycolysis
219
called Association-Induction hypothesis developed by Gilbert Ling [26, 30–32],
which, among other things, predicts that intracellular water is in a polarised state and
that the concentration of free intracellular K
+ is low due to binding to intracellular
proteins.
13.4 A Potential Role of Glycolytic Oscillations
Homeostasis implies that a variable, e.g. a concentration or temperature, is actively
regulated to be maintained nearly constant. The concept was originally proposed
by French physiologist Claude Bernard [8]. In biochemistry homeostasis is synonymous with the term steady state. The idea that biological systems operate at a
steady state has survived to this date. Oscillatory dynamics do not fit very well to
the steady state concept. It is interesting that glycolytic oscillations have been found
in multiple yeast strains and thus seems to have been preserved throughout evolution. Furthermore, many other metabolic and signalling pathways in yeast and other
organisms show oscillatory behaviour [23, 24, 34, 36]. Therefore, it is tempting to
speculate that instead of homeostasis one should consider the normal state of the cell
as a homeodynamic state [35, 71]. Homeodynamics consider the living organism
as operating not far from equilibrium on a low duty cycle and with only some constituent processes being far from equilibrium. Since metabolic oscillations seem to be
so widespread it is natural to ask: what are their functions? So far no definite answer
has been given. One possible explanation for the existence of metabolic oscillations
is that the ability to exhibit oscillatory dynamics constitutes a harmless side-effect of
the complex regulatory mechanisms controlling the activities of glycolytic enzymes
[18]. Another explanation could be that the glycolytic oscillations have evolved as
an inevitable side effect of hard trade-offs between robustness and efficiency [13].
According to these two views oscillations have no function at all. On the other hand,
it has also been proposed that the mechanisms responsible for the oscillations have
evolved to reduce the dissipation of energy [60]; that is, oscillations make metabolic
pathways more energy efficient. Here we shall propose a radically new hypothesis
for the glycolytic oscillations, namely that they reflect a living organism’s quest for
maintaining a low-entropy state [66].
Oscillating (bio)chemical processes are considered to be non-linear dissipative
thermodynamic processes [28, 44, 45]. This is in contrast to many oscillating systems
in classical physics, which are non-dissipative adiabatic processes. Traditionally,
irreversible processes have been described by Onsager’s phenomenological equations
[48, 49], which assume a symmetric linear coupling matrix between fluxes and
forces. Onsager-type equations yield fluxes that can be written as J = L · X where
L is an arbitrary matrix. As demonstrated in a recent study [25] any such matrix
can be separated into a symmetric (L
S ) and an antisymmetric (L
A ) matrix. Since
the antisymmetric matrix does not contribute to dissipation, entropy production is
exclusively governed by the symmetric matrix [48, 49]. This situation is illustrated
219
called Association-Induction hypothesis developed by Gilbert Ling [26, 30–32],
which, among other things, predicts that intracellular water is in a polarised state and
that the concentration of free intracellular K
+ is low due to binding to intracellular
proteins.
13.4 A Potential Role of Glycolytic Oscillations
Homeostasis implies that a variable, e.g. a concentration or temperature, is actively
regulated to be maintained nearly constant. The concept was originally proposed
by French physiologist Claude Bernard [8]. In biochemistry homeostasis is synonymous with the term steady state. The idea that biological systems operate at a
steady state has survived to this date. Oscillatory dynamics do not fit very well to
the steady state concept. It is interesting that glycolytic oscillations have been found
in multiple yeast strains and thus seems to have been preserved throughout evolution. Furthermore, many other metabolic and signalling pathways in yeast and other
organisms show oscillatory behaviour [23, 24, 34, 36]. Therefore, it is tempting to
speculate that instead of homeostasis one should consider the normal state of the cell
as a homeodynamic state [35, 71]. Homeodynamics consider the living organism
as operating not far from equilibrium on a low duty cycle and with only some constituent processes being far from equilibrium. Since metabolic oscillations seem to be
so widespread it is natural to ask: what are their functions? So far no definite answer
has been given. One possible explanation for the existence of metabolic oscillations
is that the ability to exhibit oscillatory dynamics constitutes a harmless side-effect of
the complex regulatory mechanisms controlling the activities of glycolytic enzymes
[18]. Another explanation could be that the glycolytic oscillations have evolved as
an inevitable side effect of hard trade-offs between robustness and efficiency [13].
According to these two views oscillations have no function at all. On the other hand,
it has also been proposed that the mechanisms responsible for the oscillations have
evolved to reduce the dissipation of energy [60]; that is, oscillations make metabolic
pathways more energy efficient. Here we shall propose a radically new hypothesis
for the glycolytic oscillations, namely that they reflect a living organism’s quest for
maintaining a low-entropy state [66].
Oscillating (bio)chemical processes are considered to be non-linear dissipative
thermodynamic processes [28, 44, 45]. This is in contrast to many oscillating systems
in classical physics, which are non-dissipative adiabatic processes. Traditionally,
irreversible processes have been described by Onsager’s phenomenological equations
[48, 49], which assume a symmetric linear coupling matrix between fluxes and
forces. Onsager-type equations yield fluxes that can be written as J = L · X where
L is an arbitrary matrix. As demonstrated in a recent study [25] any such matrix
can be separated into a symmetric (L
S ) and an antisymmetric (L
A ) matrix. Since
the antisymmetric matrix does not contribute to dissipation, entropy production is
exclusively governed by the symmetric matrix [48, 49]. This situation is illustrated