In silico model-driven assessment of the effects of single nucleotide polymorphisms (SNPs) on human red blood cell metabolism

Abstract

The completion of the human genome project and the construction of single nucleotide polymorphism (SNP) maps have lead to significant efforts to find SNPs that can be linked to pathophysiology. In silico models of complete biochemical reaction networks relate a cell's individual reactions to the function of the entire network. Sequence variations can in turn be related to kinetic properties of individual enzymes, thus allowing an in silico model-driven assessment of the effects of defined SNPs on overall cellular functions. This process is applied to defined SNPs in two key enzymes of human red blood cell metabolism: glucose-6-phosphate dehydrogenase and pyruvate kinase. The results demonstrate the utility of in silico models in providing insight into differences between red cell function in patients with chronic and nonchronic anemia. In silico models of complex cellular processes are thus likely to aid in defining and understanding key SNPs in human pathophysiology.

Figure 1

A schematic depiction of the genotype-phenotype relationship; from sequence variation (a) to changes in biochemical function (b) to network properties (c) to overall physiological function (d). The upper path signifies a normal pathological situation. The DNA sequence codes a nondeficient enzyme that catalyzes the glucose-6-phosphate dehydrogenase or pyruvate kinase reactions normally resulting in homeostatic red blood cell metabolism with the cell assuming its traditional biconcave shape. The lower path represents a pathological situation, a defect in the DNA sequence that leads to a metabolic enzyme with altered catalytic activity. The result is a loss in the cell's ability to respond to oxidative and/or energy loads, hence an inability to maintain osmotic balance and stability that ultimately results in lysis–the phenotypic expression of the genetic defect.

Figure 2

The location of SNPs in the glucose-6-phosphate dehydrogenase (G6PD) protein. (a) The G6PD variants for which there is full molecular characterization available, are shown with their corresponding parameter values and amino-acid location of the single nucleotide polymorphism (SNP). Note that the SNPs are located in critical regions of the protein. (b) Shows that there is no obvious correlation between the altered numerical values for the two kinetic parameters in either the chronic or nonchronic cases.

Figure 3

The ability of glucose-6-phosphate dehydrogenase (G6PD) variants to respond to oxidative load based on in silico red blood cell model results. The NADPH/NADP is plotted for maximum load (v_ox = max value tolerated) versus no oxidative load (v_ox = 0) for each G6PD variant. The nonchronic cases cluster near the normal case and can sustain a maximum oxidative load of near normal (3 mM/h). The chronic cases can be split into two categories, those that can sustain a near-normal load (chronic high load) and those who are well below normal (chronic low load). The horizontal arrows demarcate two regions: an oxidized state (Log [NADPH/NADP] < 0) where the majority of the cofactor is in the form of NADP and a reduced state (Log [NADPH/NADP] > 0) where the majority of the cofactor is in the form of NADPH. For G6PD enzymopathies, the oxidized state is prevalent, while the reduced state is found in normal individuals (Kirkman et al. 1975). The insert shows no clear correlation between V_max, K_i-NADPH, and the cell's ability to withstand oxidative loads.

Figure 4

The ability of pyruvate kinase (PK) variants to respond to energy load. (a) The numerical values for V_max and K_PEP for the documented PK cases. (b) There is no obvious correlation between the two kinetic parameters amongst the variants studied. (c) As the maximum tolerated ATP load decreases, the ratio of the cells [ATP] at maximum load versus at no load increases resulting in a near linear relationship. (d) The energy charge remains near constant despite the significant drop in the each variant's ability to withstand an ATP load.