Rat animal model for heart failure with preserved ejection fraction (HEpEF): A matter of diet

Background. A strong association exists between obesity and heart failure with preserved ejection fraction (HFpEF). Therefore, diet-induced obesity and L-NAME (LN) induced hypertension might be a valuable model for HFpEF. However, the optimal dietary composition and feeding duration required to induce HFpEF are not well established. A high fat, high fructose diet might be effective to induce obesity and an HFpEF phenotype due to its well-known high pathogenicity. Here, we compare different diets in combination with L-NAME for the induction of HFpEF in male Wistar rats.

Methods and results Since excitation-contraction (EC) coupling in mice is quite different to humans, we used male Wistar rats, where EC coupling is more similar to humans. The control group was fed a standard chow alone (Con), and three groups received L-NAME in the drinking water (0.25mg/ml) plus either standard chow (L-NAME), a high fructose but low fat diet (LFD/L-NAME) or a high fructose/high fat diet (HFD/L-NAME). Rats on HFD could choose after 8 weeks between HFD and LFD. After 16 weeks, cardiac ventricular myocytes (at least n=33 from n=3 animals per group) and mitochondria (from n=4-8 hearts per group) were isolated. Only in the HFD/L-NAME group, a substantial increase in plasma BNP was observed, indicating the development of heart failure. Sarcomere length, cytosolic Ca²⁺ (Indo1, AM) and mitochondrial redox state (autofluorescence of NAD(P)H and FAD), membrane potential (TMRM), and ROS (DCF) in myocytes using an automatic Ionoptix fluorescence setup was measured. Pacing at 0.3 Hz, followed by β-adrenergic stimulation and increasing stimulation rate at 3 Hz for 3 minutes was used to subject cardiac myocytes to a physiological stress regimen. Only the rats of the HFD/L-NAME group showed diastolic dysfunction with preserved fractional shortening, increased diastolic and systolic [Ca²⁺] and Ca²⁺ transient amplitude. In this group, the mitochondrial redox state was clearly oxidized, while mitochondrial membrane potential was more negative and stable and less ROS were produced compared to all other groups. L-NAME and LFD/L-NAME were comparable to the Con-group. The L-NAME group showed a mild oxidation of the mitochondrial redox state, while the membrane potential was more negative. In isolated mitochondria, mitochondrial respiration, Ca²⁺-retention capacity using Calcium-Green, mitochondrial membrane potential using TMRM and H₂O₂ production using AmplexRed was examined. In the presence of pyruvate/malate as substrates, but not with fatty acids, HFD/L-NAME showed reduced mitochondrial resipiration compared to L-NAME. Mitochondrial membrane potential was more stable in HFD/L-NAME vs. L-NAME and vs. LFD/L-NAME in the presence of pyruvate/malate, but not fatty acids. Ca²⁺-retention capacity was not different between groups. H₂O₂ production was not different between all groups, independent of the given substrate.

Conclusions Only the combined treatment of rats with HFD and L-NAME is sufficient to generate calcium mishandling and mitochondrial dysfunction typical for HFpEF. L-NAME induced a mild energetic deficit without hampering contraction and Ca handling. Interestingly, rats on LFD with L-NAME did not differ from healthy controls.