OBJECTIVE: This study examines the extent to which the contractile deficit of diabetic cardiomyopathy is due to altered Ca(2+) homeostasis. RESEARCH DESIGN AND METHODS: Measurements of isometric force and intracellular calcium ([Ca(2+)](i), using fura-2/AM) were made in left ventricular (LV) trabeculae from rats with streptozotocin-induced diabetes and age-matched siblings. RESULTS: At 1.5 mmol/l [Ca(2+)](o), 37 degrees C, and 5-Hz stimulation frequency, peak stress was depressed in diabetic rats (10 +/- 1 vs. 17 +/- 2 mN/mm(2) in controls; P < 0.05) with a slower time to peak stress (77 +/- 3 vs. 67 +/- 2 ms; P < 0.01) and time to 90% relaxation (76 +/- 7 vs. 56 +/- 3 ms; P < 0.05). No difference was found between groups for either resting or peak Ca(2+), but the Ca(2+) transient was slower in time to peak (39 +/- 2 vs. 34 +/- 1 ms) and decay (time constant, 61 +/- 3 vs. 49 +/- 3 ms). Diabetic rats had a longer LV action potential (APD(50), 98 +/- 5 vs. 62 +/- 5 ms; P < 0.0001). Western blotting showed that diabetic rats had a reduced expression of sarco(endo)plasmic reticulum Ca(2+)-ATPase (SERCA)2a, with no difference in expression of the Na(+)/Ca(2+) exchanger. Immunohistochemistry of LV free wall showed that type I collagen was increased in diabetic rats (diabetic 7.1 +/- 0.1%, control 12.7 +/- 0.1%; P < 0.01), and F-actin content reduced (diabetic 56.9 +/- 0.6%; control 61.7 +/- 0.4%; P < 0.0001) with a disrupted structure. CONCLUSIONS: We find no evidence to support the idea that altered Ca(2+) homeostasis underlies the contractile deficit of diabetic cardiomyopathy. The slower action potential and reduced SERCA2a expression can explain the slower Ca(2+) transient kinetics in diabetic rats but not the contractile deficit. Instead, we suggest that the observed LV remodeling may play a crucial role.