Background and Purpose: Heart failure with preserved ejection fraction (HFpEF) is more prevalent in women after menopause but the role of estrogen is elusive. We characterized cardiac characteristics in female ZSF1 rats during HFpEF manifestation comparing ovariectomized and non-ovariectomized animals to determine the impact of estrogen.
Methods: 33 female ZSF1 rats in three experimental groups (n=11) underwent echocardiographic analysis at 11, 14, 17 and 20 weeks of age. 50% of the obese phenotype rats were randomly assigned to ovariectomy at an age of ~6 weeks (Ov-ZSF1) leading to estrogen depletion before HFpEF manifestation, while the other animals (O-ZSF1) stayed intact. Lean non-ovariectomized healthy ZSF1 rats (L-ZSF1) served as control group. Invasive hemodynamic analysis was done at 20 weeks of age. Right carotid artery was used to determine endothelial function in an organ bath system with myographic recording ex vivo. Expression of the estrogen receptors and tyrosinated tubulin was determined by Western Blot analysis. NT-proBNP was measured using ELISA. Cardiac fibrosis was histologically quantified using Picrosirius-Red staining.
Results: During the observation period, the echocardiographic E/e’ ratio increased in the obese rats compared to L-ZSF1 whereas a significantly higher ratio was detected in Ov-ZSF1 already at 14 weeks of age while in O-ZSF1 pathologic ratio increase was detected at 17 weeks. At 20 weeks of age Ov-ZSF1 rats were heavier (+150% vs L-ZSF1, +36% vs. O-ZSF1), had higher blood glucose (+33% vs L-ZSF1, +9% vs. O-ZSF1), heavier hearts (+61% vs L-ZSF1, +13% vs. O-ZSF1) and higher NT-proBNP (+87% vs L-ZSF1, +35% vs. O-ZSF1). Cardiac characteristics that were significantly altered in O-ZSF1 compared to L-ZSF1 were partly further aggravated in Ov-ZSF1: left ventricular (LV) wall thickness (+35% vs L-ZSF1, +11% vs. O-ZSF1), LV end-diastolic volume (+64% vs L-ZSF1, +31% vs. O-ZSF1) and LV deceleration time (+77% vs L-ZSF1, +33% vs. O-ZSF1). Effective arterial elastance was reduced in all obese rats compared to L-ZSF1 irrespective of estrogen presence. Carotid vessels from Ov-ZSF1 showed stronger contraction due to phenylephrine stimulation and a delayed acetylcholine induced vasodilation compared to L-ZSF1. No differences in cardiac fibrotic remodeling were observed between the experimental groups. Tubulin and its tyrosination pattern were significantly altered in O-ZSF1 compared to L-ZSF1 and Ov-ZSF1. The estrogen receptor α (Erα) -46kDa isoform was lower in Ov-ZSF1 (-12% vs. L-ZSF1) but lowest in O-ZSF1 (-50% vs. L-ZSF1).
Discussion and conclusion: The known protective properties of estrogen appear to actually slow down the manifestation of HFpEF in this diabetic and hypertensive animal model. Nevertheless, the complex involvement of estrogen in molecular regulatory pathways underlying HFpEF leads to some counterintuitive observations, such as reduced Erα-46kDa expression or altered tubulin homeostasis compared to healthy controls in non-ovariectomized O-ZSF1. Analysis of HFpEF animal models with regard to hormonal status will help to elucidate and understand previously unknown HFpEF pathomechanisms and could reveal new treatment targets.
