Decreased serum glucose and glycosylated hemoglobin levels in patients with Chuvash polycythemia: a role for HIF in glucose metabolism

Abstract

In Chuvash polycythemia, a homozygous 598C>T mutation in the von Hippel-Lindau gene (VHL) leads to an R200W substitution in VHL protein, impaired degradation of α-subunits of hypoxia-inducible factor (HIF)-1 and HIF-2, and augmented hypoxic responses during normoxia. Chronic hypoxia of high altitude is associated with decreased serum glucose and insulin concentrations. Other investigators reported that HIF-1 promotes cellular glucose uptake by increased expression of GLUT1 and increased glycolysis by increased expression of enzymes such as PDK. On the other hand, inactivation of Vhl in murine liver leads to hypoglycemia associated with a HIF-2-related decrease in the expression of the gluconeogenic enzyme genes Pepck, G6pc, and Glut2. We therefore hypothesized that glucose concentrations are decreased in individuals with Chuvash polycythemia. We found that 88 Chuvash VHL ( R200W ) homozygotes had lower random glucose and glycosylated hemoglobin A1c levels than 52 Chuvash subjects with wild-type VHL alleles. Serum metabolomics revealed higher glycerol and citrate levels in the VHL ( R200W ) homozygotes. We expanded these observations in VHL ( R200W ) homozygote mice and found that they had lower fasting glucose values and lower glucose excursions than wild-type control mice but no change in fasting insulin concentrations. Hepatic expression of Glut2 and G6pc, but not Pdk2, was decreased, and skeletal muscle expression of Glut1, Pdk1, and Pdk4 was increased. These results suggest that both decreased hepatic gluconeogenesis and increased skeletal uptake and glycolysis contribute to the decreased glucose concentrations. Further study is needed to determine whether pharmacologically manipulating HIF expression might be beneficial for treatment of diabetic patients.

Figure 1

a. Distribution of random serum glucose concentration in VHL wildtype subjects and VHL^R200W homozygotes. The median and interquartile ranges were 89 (81-106) mg/dL in VHL wildtype subjects compared to 75 (56-91) mg/dL inVHL^R200W homozygotes. b. Distribution of hemoglobin A1c in VHL wildtype subjects and VHL^R200W homozygotes. The median and interquartile ranges were 4.1 (3.3-4.6) % in VHL wildtype subjects compared to 3.2 (1.3-3.6) % in VHL^R200W homozygotes. c. Relationship of hemoglobin A1c level (vertical axis) to random serum glucose concentration. A significant positive correlation was observed: r = 0.53, P <0.0001.

Figure 2

a. Glucose tolerance testing in Chuvash polycythemia (CP) mice and C57BL6 wild type (WT) controls. Mice (4-10 per group, aged 3-8 months and age-matched across groups) had fasting glucose tolerance tests performed. Shown are the mean values ±SEM, with the error bars in most cases hidden within the data points. By panel data analysis, the values for CP were significantly different compared to WT (P <0.0001). b. Despite the lower fasting glucose values, insulin levels did not differ between the Chuvash polycythemia and wildtype mice (Chuvash polycythemia males, 0.54 ± 0.05 ng/ml, wildtype males 0.55 ±0.10 ng/ml, P =0.96; Chuvash polycythemia females 0.34 ± 0.05, wildtype females 0.31 ± 0.04 ng/ml, P = 0.66).