Systems Biology Approaches to Cancer Energy Metabolism - Systems Biology of Metabolic and Signaling Networks - page 225

simple irreversible mass action law. Irrespective of this simplification, the majority
of metabolite concentrations and fluxes simulated by the model were in quantitative
agreement with the in vivo concentrations, except for F6P and PEP, which were
27 and 1,600 times lower, respectively. These remaining discrepancies indicated
that further kinetic experimentation-based refinement of the model is still needed,
mainly at the level of PFK-1, ALDO, PYK, and ATPases. In this regard, it could be
useful to experimentally assess the effects of PFK-1, ALDO, and PYK binding to
microtubules on their respective kinetic properties (reviewed by Cassimeris
et al. 2012 ), which may account for the significantly lower modeled Fru6P and
PEP concentrations.
An experimentally validated and robust kinetic model can be used to predict the
control distribution under other metabolic steady states such as normoxia versus
hypoxia and normoglycemia versus hypoglycemia. Under these conditions, not
only a change in the isoform expression may occur but also changes in the enzyme
amount (Table 9.3 ; Fig. 9.2 ) and V max (Mar´n-Hern´ndez et al. 2011 ) happen.
Indeed, higher levels of metabolites and a 33 % increased flux were obtained
under hypoxia compared to normoxia. Moreover, under hypoxia a higher HK flux
control coefficient was determined at the expense of a decrease in the control of the
glycogen degradation; this behavior is similar to what was observed when GLUT3
was modeled alone or in combination with HKI (Tables 9.1 and 9.2 ). Again,
including the isoform ratios in the model did not modify metabolite concentrations,
fluxes, and control coefficients, because GLUT3 expression is lower under hypoxia
(Fig. 9.2 ).
The modeling results described above suggested that the initial assumption
made, which considered that the enzyme activity in cells can be solely attributed
to the predominant isoform expressed, is a convenient simplification that avoids the
use of highly complex rate equations. In this regard, it is worth noting that cells
grown in monolayer cultures maintain a high content of low affinity isoforms
(GLUT1 and HKII) because they are always grown in the presence of excess
(25 mM) glucose, even under hypoxia. Under such hyperglycemic conditions,
cells do not require to express and use high affinity isoforms for glucose transport
and phosphorylation (GLUT3 and HKI). Similarly, changes in GLUT isoform
proportions can be achieved under other stressful conditions. For instance, patho-
logical events such as ischemia, starvation, and mitochondria inhibition increase the
content of GLUT3 in normal cells (Nagamatsu et al. 1994 ; Vannucci et al. 1996 ;
Khayat et al. 1998 ). Same results have been observed in tumor cells, where
hypoglycemia or hypoxia plus hypoglycemia increase the mRNA for GLUT3
(Natsuizaka et al. 2007 ).
In conclusion, the simplest and perhaps fastest way cancer cells have to modu-
late the glycolytic flux under cyclically varying conditions (e.g. hypoxia/normoxia;
hypoglycemia/normoglycemia) is to change the ratio of enzyme isoform
corresponding to the most controlling steps, without profoundly altering the
installed, house-keeping over-expression pattern of all the pathway enzymes and
transporters. In this regard, the kinetic model predicts that when GLUT3 is
increased (Fig. 9.3 ), and hence the GLUT3/GLUT1 ratio, which simulates exposure