Genetic Analysis of Glucose Tolerance in Inbred Mouse Strains: Evidence for Polygenic Control

Abstract

To determine genetic factors involved in diabetes susceptibility in inbred strains of mice, we initially evaluated differences in fed plasma glucose and insulin concentrations among six strains (AKR/J, C3H/HeJ, C57BL/6J, C57L/J, DBA/2J, and SWR/J). There was considerable variation in fed plasma glucose concentration, with C3H/HeJ mice the most glucose tolerant (174 ± 7 mg/dl) and C57BL/6J mice the least glucose tolerant (252 ± 7 mg/dl, P < .0001 vs. C3H/HeJ mice). Glycosylated hemoglobin of C57BL/6J mice (4.0 ± 0.06%) was also higher than that of C3H/HeJ mice (3.52 ± 0.06%, P < .0001). The fed plasma insulin concentration did not differ between these two strains. Glucose tolerance was further evaluated in overnight-fasted C3H/HeJ and C57BL/6J mice by an intraperitoneal glucose tolerance test (IPGTT). Although fasting plasma glucose did not differ, the most remarkable difference in plasma glucose during IPGTT between C57BL/6J and C3H/HeJ mice was noted at 30 min (489 ± 29 vs. 227 ± 20 mg/dl, P < .001). To determine the number of genes involved in the phenotypic difference in glucose tolerance, C57BL/6J males were crossed with C3H/HeJ females (F1, C3H/HeJ × C57BL/6J), and the F1 hybrid females were backcrossed with C57BL/6J males (backcrossed, F1 × C57BL/6J). Plasma glucose after 30 min on IPGTT was 219 ± 8 (n = 21), 456 ± 18 (n = 23), and 292 ± 13 (n = 23) mg/dl for C3H/HeJ, C57BL/6J, and F1 mice, respectively (P < .001 for all comparisons). The range of glucose levels on IPGTT of the backcrossed mice was 250–500 mg/dl. If a single gene were determining the difference in glucose tolerance in C57BL/6J mice, analysis of the plasma glucose level of individual backcrossed mice would show segregation into two classes, one like the F1 hybrid and the other like the inbred parent. This segregation was not observed; therefore, we conclude that modification of glucose tolerance in these two inbred strains is under polygenic control. Plasma insulin levels after 15 min on IPGTT for C57BL/6J mice were lower than those for C3H/HeJ mice (1.0 ± 0.09 vs. 1.3 ± 0.09 ng/ml, n = 6, P < .001). On the other hand, insulin levels after 30 min on IPGTT for C3H/HeJ (0.9 ± 0.08 ng/ml, n = 21), C57BL/6J (1.05 ± 0.1 ng/ml, n = 23), F1 (1.05 ± 0.1 ng/ml, n = 23), and backcrossed mice (1.0 ± 0.1 ng/ml, n = 23) were not different from each other. A positive correlation between plasma glucose and insulin concentrations was observed in C3H/HeJ (r = .42, P < .05) and F1 (r = .46, P < .05) mice, whereas a negative correlation was seen in C57BL/6J mice (r = −.45, P < .05). This suggests that glucose-stimulated insulin secretion may be significantly impaired in C57BL/6J mice, indicating one of the possible mechanisms for the phenotypic expression of glucose intolerance.