Animal models of diabetes mellitus (DM) are extensively used in the exploration of the pathophysiological mechanisms underlying the development of cardiovascular complications of DM, including heart failure, a common cardiovascular complication in up to 10–20% of patients with diabetes. These models are limited in that genetic models commonly used do not generally reflect genetic forms of diabetes in humans and pancreatic toxins used such as streptozotocin may have direct myocardial actions themselves. Recently, as part of the mouse mutagenesis programme at the MRC, a novel human relevant mouse model of diabetes (GENA 348) has been described in which an A to T transversion mutation in the glucokinase gene results in an isoleucine to phenylalanine change at amino acid 366 (Toye et al, Diabetes 2004) resulting in severely impaired glucokinase function and significant hyperglycaemia due to decreased insulin secretion. Similar mutations in the glucokinase gene are known to underlie maturity onset diabetes of the young type 2 (MODY 2), a common monogenic form of DM in humans. We have studied the GENA 348 mouse to determine whether it expresses a cardiac phenotype and so can be used in the future to study the pathophysiological mechanisms underlying the development of heart failure in DM.
15 wild-type (WT) GENA 348 controls and eight homozygote mutant (HO) GENA 348 mice were aged to 6 months and had serial echocardiography performed at 3 months and 6 months. At 3 months, echocardiography revealed no evidence of cardiac hypertrophy or contractile dysfunction in the HO compared with the WT mice. Echocardiography performed at 6 months demonstrated significant cardiac hypertrophy in the HO compared with the WT mice, dPW/dD 23.3 ± 1.0 versus 29.1 ± 1.1 (24.8% increase, p<0.01), left ventricular relative wall thickness 0.45 ± 0.01 versus 0.53 ± 0.02 (17.7% increase, p<0.05) and left ventricular diastolic posterior wall thickness 1.1 ± 0.04 versus 1.28 ± 0.06 (16.3% increase, p<0.01) in WT and HO mice, respectively. The mean ejection fraction was not significantly different in WT and KO mice (55.7 ± 2.0 vs 59.6 ± 2.0, p = NS). Western blot analysis of the myocardial tissue demonstrated increased Akt phosphorylation, a known mediator in the hypertrophy signalling cascade but no changes in the phosphorylation of erk, another hypertrophy pathway. In conclusion, in one of the first novel human relevant mouse models of DM we have demonstrated the development of a cardiac phenotype similar to that seen clinically in patients with DM, namely left ventricular hypertrophy with associated activation of the Akt molecular signalling cascade involved in cardiac hypertrophy. The analogous nature of this model to human DM (MODY-2) has significant implications for its use in diabetic cardiovascular research.