Differential sex-dependent susceptibility to diastolic dysfunction and arrhythmia in cardiomyocytes from obese diabetic heart failure with preserved ejection fraction model

Abstract

Aims: Sex differences in heart failure with preserved ejection fraction (HFpEF) are important, but key mechanisms involved are incompletely understood. While animal models can inform about sex-dependent cellular and molecular changes, many previous pre-clinical HFpEF models have failed to recapitulate sex-dependent characteristics of human HFpEF. We tested for sex differences in HFpEF using a two-hit mouse model (leptin receptor-deficient db/db mice plus aldosterone infusion for 4 weeks; db/db + Aldo).

Methods and results: We performed echocardiography, electrophysiology, intracellular Ca2+ imaging, and protein analysis. Female HFpEF mice exhibited more severe diastolic dysfunction in line with increased titin N2B isoform expression and PEVK element phosphorylation and reduced troponin-I phosphorylation. Female HFpEF mice had lower BNP levels than males despite similar comorbidity burden (obesity, diabetes) and cardiac hypertrophy in both sexes. Male HFpEF mice were more susceptible to cardiac alternans. Male HFpEF cardiomyocytes (vs. female) exhibited higher diastolic [Ca2+], slower Ca2+ transient decay, reduced L-type Ca2+ current, more pronounced enhancement of the late Na+ current, and increased short-term variability of action potential duration (APD). However, male and female HFpEF myocytes showed similar downregulation of inward rectifier and transient outward K+ currents, APD prolongation, and frequency of delayed afterdepolarizations. Inhibition of Ca2+/calmodulin-dependent protein kinase II (CaMKII) reversed all pathological APD changes in HFpEF in both sexes, and empagliflozin pre-treatment mimicked these effects of CaMKII inhibition. Vericiguat had only slight benefits, and these effects were larger in HFpEF females.

Conclusion: We conclude that the db/db + Aldo pre-clinical HFpEF murine model recapitulates key sex-specific mechanisms in HFpEF and provides mechanistic insights into impaired excitation-contraction coupling and sex-dependent differential arrhythmia susceptibility in HFpEF with potential therapeutic implications. In male HFpEF myocytes, altered Ca2+ handling and electrophysiology aligned with diastolic dysfunction and arrhythmias, while worse diastolic dysfunction in females may depend more on altered myofilament properties.

Keywords: Arrhythmia; CaMKII; Diastolic dysfunction; HFpEF; Sex differences.

Graphical Abstract

Figure 1

Sex differences in morphometric parameters in HFpEF mice. (A) HFpEF study protocol and assessment of extracardiac morbidities, heart function, and cardiomyocyte excitation–contraction coupling and electrophysiology. (B) Sex-dependent changes in body weight and blood glucose levels over time in db/db mice with chronic aldosterone infusion (db/db + Aldo) vs. vehicle-treated WT control (WT + vehicle). Two-way repeated measures ANOVA with Geisser–Greenhouse correction. (C) Sex-dependent cardiac hypertrophy, pulmonary oedema, and increased BNP plasma levels in db/db + Aldo vs. WT + vehicle (HW/TL, heart weight to tibial length ratio; BNP, B-type natriuretic peptide). Two-way ANOVA followed by Šídák’s multiple comparisons test. N = 12 animals/experimental group except for BNP measurements where N = 4.

Figure 2

More severe diastolic dysfunction in female HFpEF mice. (A) LV M-mode, flow, and tissue Doppler echocardiographic images in male and female db/db mice with chronic aldosterone infusion (db/db + Aldo) vs. vehicle-treated WT control (WT + vehicle), 4 weeks after minipump implantation. (B) Preserved EF, cardiac hypertrophy, and increased LVRI (a ratio between LV mass and LV internal diameter) in both sexes in db/db + Aldo mice (LVPWd, LV end-diastolic posterior wall thickness; LVM, LV mass; LVIDd, LV internal diameter in diastole). (C) More severe LV diastolic dysfunction and left atrial (LA) area enlargement in female db/db + Aldo mice. (E/A, ratio between mitral E-wave and A-wave; E/e′, ratio between mitral E-wave and e′-wave). Two-way ANOVA followed by Šídák’s multiple comparisons test. N = 12 animals in each group except for LA area where N = 8.

Figure 3

Sex differences in intracellular Ca²⁺ handling in HFpEF cardiomyocytes. (A) Intracellular Ca²⁺ signals in db/db mice with chronic aldosterone infusion (db/db + Aldo) vs. vehicle-treated WT control (WT + vehicle) cardiomyocytes paced at 1 Hz and during rapid caffeine application. (B) Intracellular Ca²⁺ transient (CaT) parameters. Diastolic [Ca²⁺] is the ratio of minimum F between beats at 1 Hz pacing and the resting F₀. (C) CaT decay tau and SR Ca²⁺ content (assessed by rapid local application of 10 mmol/L caffeine). n (cells)/N (animals) = 19/8 for WT + vehicle male, 19/7 for WT + vehicle female, 17/8 for db/db + Aldo male, and 19/7 for db/db + Aldo female. Hierarchical (nested) ANOVA followed by Šídák’s multiple comparisons test.

Figure 4

Sex differences in myofilament alterations in HFpEF ventricles. (A) Titin isoform analysis in male (M) and female (F) db/db + Aldo HFpEF (HF) vs. vehicle-treated WT control (Ctrl) ventricular samples. Relative expression of the more compliant N2BA and the stiffer N2B titin isoform to total titin (TT) and myosin heavy chain (MHC) was assessed using gel electrophoresis. ANOVA followed by Šídák’s multiple comparisons test. (B) Western blot data showing increased phosphorylation of serine 170 of titin's PEVK domain normalized to titin Z1Z2 element (to assess total intact titin level) in db/db + Aldo females. Kruskal–Wallis ANOVA followed by Dunn’s multiple comparisons test. (C) pTnI normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH). ANOVA followed by Šídák’s multiple comparisons test. Three technical replicates (blots) were performed for each protein sample. N = 4 animals in each group.

Figure 5

Sex differences in arrhythmogenic APs in HFpEF cardiomyocytes. (A) Representative ventricular APs in male and female db/db mice with chronic aldosterone infusion (db/db + Aldo) vs. vehicle-treated WT control (WT + vehicle) cardiomyocytes paced at 1 Hz (above). Tachypacing (10 Hz) induced APD alternans (S, short; L, long) in db/db + Aldo (inset, below). (B) APD at 90% repolarization (APD₉₀). (C) Fifty consecutive APD₉₀ values at 1 Hz pacing. (D) Increased STV of APD₉₀ in db/db + Aldo. n (cells)/N (animals) = 20/7 for WT + vehicle male, 25/8 for WT + vehicle female, 24/8 for db/db + Aldo male, and 29/8 for db/db + Aldo female (for both APD₉₀ and STV). (E) Frequency dependence of APD₉₀. (F) Amplitude of APD₉₀ alternans at 10 Hz tachypacing. n (cells)/N (animals) = 13/7 for WT + vehicle male, 16/8 for WT + vehicle female, 11/7 for db/db + Aldo male, and 18/8 for db/db + Aldo female. Hierarchical (nested) ANOVA followed by Šídák’s multiple comparisons test.

Figure 6

Sex-dependent remodelling in K⁺ currents, L-type Ca²⁺ current, and late Na⁺ current in HFpEF cardiomyocytes. (A) Representative inward rectifier K⁺ current (I_K1) traces at −140 mV in male and female db/db mice with chronic aldosterone infusion (db/db + Aldo) vs. vehicle-treated WT control (WT + vehicle) cardiomyocytes. Increased cell capacitance and reduced I_K1 densities at −140 and −40 mV in db/db + Aldo in both sexes. n (cells)/N (animals) = 97/8 for WT + vehicle male, 112/8 for WT + vehicle female, 110/8 for db/db + Aldo male, and 109/8 for db/db + Aldo female. (B) Representative voltage-gated K⁺ current (I_Kv) traces. The net I_Kv current was reduced in both sexes in HFpEF. (C) Transient outward K⁺ current (I_to), slowly inactivating K⁺ current (I_K,slow), and sustained K⁺ current (I_sus) were separated by biexponential fitting to I_Kv traces. n (cells)/N (animals) = 15/5 for WT + vehicle male, 20/6 for WT + vehicle female, 17/6 for db/db + Aldo male, and 18/6 for db/db + Aldo female. (D) Representative L-type Ca²⁺ current (I_Ca,L) traces. I_Ca,L was reduced only in male db/db + Aldo cardiomyocytes. n (cells)/N (animals) = 25/7 for WT + vehicle male, 21/8 for WT + vehicle female, 19/8 for db/db + Aldo male, and 21/8 for db/db + Aldo female. (E) Representative late Na⁺ current (I_Na,L) traces. I_Na,L was markedly upregulated in db/db + Aldo, and the increase was larger in male than in female cardiomyocytes. n (cells)/N (animals) = 8/4 for WT + vehicle male, 10/5 for WT + vehicle female, 10/6 for db/db + Aldo male, and 10/6 for db/db + Aldo female. Hierarchical (nested) ANOVA followed by Šídák’s multiple comparisons test.

Figure 7

Sex-dependent electrophysiological responses to therapeutic interventions in HFpEF cardiomyocytes. (A) Representative ventricular APs at 1 Hz pacing in male and female db/db mice with chronic aldosterone infusion (db/db + Aldo) in control and following cell pre-treatments with empagliflozin, AIP, and vericiguat. (B) Attenuation of prolonged APD at 90% repolarization (APD₉₀) by empagliflozin, AIP, and vericiguat. (C) Fifty consecutive APD₉₀ values at 1 Hz pacing. (D) Increased STV of APD₉₀ in db/db + Aldo was reversed by empagliflozin and AIP in both sexes. Vericiguat reduced STV only in female db/db + Aldo cardiomyocytes. n (cells)/N (animals) = 24/8 for control male, 29/8 for control female, 16/6 for empagliflozin-treated male, 15/6 for empagliflozin-treated female, 14/4 for AIP-treated male, 11/4 for AIP-treated female, 12/4 for vericiguat-treated male, and 10/4 for vericiguat-treated female. (E) Representative APD₉₀ alternans during 10 Hz pacing (S, short; L, long). (F) APD₉₀ alternans was markedly reduced by empagliflozin and AIP in both sexes. Vericiguat reduced APD₉₀ alternans only in female db/db + Aldo cardiomyocytes. n (cells)/N (animals) = 11/7 for control male, 18/8 for control female, 12/5 for empagliflozin-treated male, 20/6 for empagliflozin-treated female, 6/4 for AIP-treated male, 8/4 for AIP-treated female, 9/4 for vericiguat-treated male, and 9/4 for vericiguat-treated female. Hierarchical (nested) ANOVA followed by Dunnett's multiple comparisons test.