Tumor necrosis factor-α reduces obesity and CpcPH-HFpEF in mouse models

L. Zimmermann (Köln)1, L. Gnatzy-Feik (Köln)1, E. M. Berghausen (Köln)1, M. Zierden (Köln)1, M. Vantler (Köln)1, S. Rosenkranz (Köln)1
1Herzzentrum der Universität zu Köln Klinik III für Innere Medizin Köln, Deutschland
Introduction:
Heart failure with preserved ejection fraction (HFpEF) is an age-related cardiometabolic disease characterized by vascular inflammation in the myocardium, pulmonary congestion, and right ventricular dysfunction. Patients with HFpEF often develop pulmonary hypertension (PH) associated with right ventricular (RV) hypertrophy. The structural remodeling of the pulmonary vessels contributes to the development of combined post- and precapillary PH (CpcPH).  Elevated tumor necrosis factor-α(TNFα) levels are considered a predictive marker for the risk of morbidity and mortality events in HFpEF. TNF-mediated metabolic processes could significantly influence both cardiac and pulmonary pathobiology. Therefore, the influence of TNF deficiency on the severity of CpcPH-HFpEF and associated consequences was investigated in a mouse model.
Methods: 
We used the established animal model of HFpEF, which is based on a “two-hit” approach in which diastolic dysfunction is induced over 12 weeks by metabolic (high-fat diet, HFD) and hypertensive stress (inhibition of constitutive NO synthase using Nω-nitro-L-arginine methyl ester, L-NAME). This model was extended to a “three-hit” CpcPH-HFpEF model by a subsequent 21-day hypoxia phase. In accordance with the experimental design, adult male TNF+/+ and TNF-/--mice were fed a high-fat diet (HFD; 60% fat content) for 12 weeks and treated with L-NAME in their drinking water. This was followed by three weeks of exposure to normobaric hypoxia (10% O₂) or normoxia while continuing the HFD and L-NAME treatment. TNF+/+ and TNF-/--mice that received a standard diet (chow diet) for 15 weeks served as control groups. Phenotypic changes were recorded by means of weight analyses, glucose tolerance tests, and morphometric, echocardiographic, and hemodynamic examinations of cardiopulmonary parameters. Right ventricular systolic pressure (RVSP) was measured using a Millar® pressure catheter inserted into the right ventricle.
Results:
Wild-type animals developed obesity, insulin resistance, and HFpEF under HFD+L-NAME, and additional PH under hypoxia. Under CpcPH-HFpEF conditions (HFD + L-NAME + hypoxia), TNF-/--mice showed significantly lower weight at the end of the experiment compared to TNF+/+-animals (28.7±1.7 vs. 37.9±4.4 g; p<0.01). In addition, the glucose tolerance test in the CpcPH-HFpEF group also revealed a significant difference between TNF-/- (433.3±45.0 mg/dl) and TNF+/+-mice (541.8±10.4 mg/dl, p<0.01). TNF+/+-animals, but not TNF-/--mice, showed echocardiographic signs of increased E/E' ratio under HFD + L-NAME + hypoxia (34.8±12.0 vs. 143.1±4.7; p<0.05). Both TNF+/+-mice (29.2±3.6 (normoxia) vs. 39.6±2.0 mmHg (hypoxia); p<0.001) and TNF-/--mice (27.8±1.6 (normoxia) vs. 36.5±0.1 mmHg (hypoxia); p<0.001) showed a significant increase in RVSP. Under CpcPH-HFpEF conditions, there is also a noticeable trend toward a lower-improved RVSP between TNF-/- and TNF+/+-mice (36.5±0.1 vs. 39.6±2.0 mmHg; p=0.1).
Conclusion: 
In summary, TNF-deficient mice were protected from obesity, insulin resistance, and CpcPH-HFpEF under conditions of meta-inflammatory stress and hypoxia. It can therefore be concluded that, under the given experimental conditions, TNFα has a protective effect on the severity of metabolic changes and CpcPH-HFpEF.