Systems Biology Approaches to Cancer Energy Metabolism - Systems Biology of Metabolic and Signaling Networks - page 223

Table 9.4 Glycolytic flux and metabolite concentrations obtained with a kinetic model of
glycolysis in HeLa cells accounting for enzyme isoforms expressed under normoxia or hypoxia

Normoxia Hypoxia

Control
model

Model plus
isoform ratios

Control

model a Model plus

isoforms ratios a
Glu in NM 0.7 0.49 NM 0.89 0.81
G6P 1.3 0.78 0.85 1.4 0.82 0.83
F6P 0.5 0.018 0.02 0.5 0.02 0.02
FBP 0.38 0.14 0.16 0.23 0.35 0.38
DHAP 0.93 2.0 2.1 0.54 3.0 3.1
G3P ND 0.08 0.08 NM 0.12 0.12
1,3BPG ND 0.0008 0.001 NM 0.002 0.001
3PG ND 0.006 0.006 NM 0.008 0.008
2PG ND 0.002 0.003 NM 0.003 0.003
PEP 0.32 0.0002 0.0002 NM 0.0002 0.0002
Pyr 8.5 2.5 2.5 4.2 2.6 2.6
ATP 8.7 8.3 8.7 7.9 8.7 8.9
ADP 2.7 2.2 2.0 1.8 1.6 1.9
AMP 0.4 1.3 1.1 NM 0.67 0.96
NADH NM 0.005 0.005 NM 0.005 0.005
NAD + NM 1.34 1.34 NM 1.34 1.34
Glycolytic flux 16 19.6 20.7 21 26.1 26.8
Metabolite concentrations in mM; flux in nmol lactate min 1 (mg cellular protein) 1
NM not measured, ND not detected (

In vivo b In vivo b

Metabolite

1 nmol/15 mg cellular protein). The parameters used were
0.0042 (F2,6BP), 1.7 (citrate), 7.2 (Pi), 33 (lactate) mM, and 5 mM (glucose).
a ATPase k value was 3.6

<

10 3 min 1 . For the hypoxia model simulation, the V max values of
GLUT and PFK-1 were 30 and 33.8 nmol/min

mg cellular protein, respectively
b Values were taken from Mar´n-Hern´ndez et al. ( 2011 )

The modified model accounting for isoform ratios showed similar robust behav-
ior, as the model with no different isoforms (Mar ´ n-Hern ´ ndez et al. 2011 ). The
model behavior did not significantly change after altering kinetic parameters
(decreasing by half or increasing by two times) from most glycolytic steps and
PPP flux. Exceptions were given by changes in GLUT, glycogen degradation, and
GAPDH V m values, which led to variations of 50 % or more in the flux-control
distribution and/or metabolite concentrations. In addition, the FBP or DHAP
concentrations varied between 9 and 100 % in response to changes in ALDO V m
and K m values or TPI K m values.
The model was sensitive to changes in the ATPase kinetics, revealing that this
step is the weakest component in conferring stability to the pathway. The ATPase
reaction in the present kinetic model accounts for the energy demand represented
by a multitude of ATP-consuming cellular processes such as ion homeostasis (Na + /
K + ATPase, Ca 2+ ATPases, H + ATPases, MDR ATPases), biosynthesis of proteins,
nucleic acids, lipids, and polysaccharides, and specialized functions (signal trans-
duction, secretion, proliferation). In addition, the rate equation used for the ATPase
reaction has no affinity terms for substrates and products or K eq , but it is rather a