HDAC Inhibition Regulates Cardiac Function by Increasing Myofilament Calcium Sensitivity and Decreasing Diastolic Tension

Abstract

We recently established a large animal model that recapitulates key clinical features of heart failure with preserved ejection fraction (HFpEF) and tested the effects of the pan-HDAC inhibitor suberoylanilide hydroxamic acid (SAHA). SAHA reversed and prevented the development of cardiopulmonary impairment. This study evaluated the effects of SAHA at the level of cardiomyocyte and contractile protein function to understand how it modulates cardiac function. Both isolated adult feline ventricular cardiomyocytes (AFVM) and left ventricle (LV) trabeculae isolated from non-failing donors were treated with SAHA or vehicle before recording functional data. Skinned myocytes were isolated from AFVM and human trabeculae to assess myofilament function. SAHA-treated AFVM had increased contractility and improved relaxation kinetics but no difference in peak calcium transients, with increased calcium sensitivity and decreased passive stiffness of myofilaments. Mass spectrometry analysis revealed increased acetylation of the myosin regulatory light chain with SAHA treatment. SAHA-treated human trabeculae had decreased diastolic tension and increased developed force. Myofilaments isolated from human trabeculae had increased calcium sensitivity and decreased passive stiffness. These findings suggest that SAHA has an important role in the direct control of cardiac function at the level of the cardiomyocyte and myofilament by increasing myofilament calcium sensitivity and reducing diastolic tension.

Keywords: HDAC inhibitor; calcium; cardiomyocyte; contractility; heart failure; myofilament.

Figure 1

Effect of SAHA treatment on AFVM function. AFVM were isolated and incubated with SAHA or vehicle (2.5 μM) for 90 min, then incubated with Fluo-4am and Pluronic™ F-127 before functional measurements were acquired. (A) Representative contraction of SAHA treated vs. vehicle AFVM, which had an improvement in (B) fractional shortening, (C) Time to 50% baseline of contraction and (D) return velocity. (E) Representative calcium transient showing no difference between SAHA and vehicle-treated AFVM (F) peak calcium transient, but there was a decrease in (G) time to 30% baseline and (H) tau; n = 17–22 myocytes/parameter from 4 felines. Fractional shortening and tau were analyzed using two-sided Student’s t-test. Time to 50% BL, return velocity, calcium transient, and time to 30% BL were analyzed using Mann–Whitney test. NS stands for not significant. * p < 0.05, ** p < 0.01. Data shown are means ± SEM.

Figure 2

Effect of SAHA treatment on AFVM myofilament function. Skinned myocytes were isolated from AFVM to assess myofilament function. The (A) average force-calcium curves for SAHA and vehicle-treated skinned myocytes demonstrate a clear left shift with SAHA treatment. There was a decrease in the (B) EC₅₀ with SAHA treatment, indicating an increase in myofilament calcium sensitivity and an increase in (C) maximal calcium-activated force (F_max); n = 10 myocytes from 3 felines for each group. There was a decrease in (D) passive stiffness at increasing sarcomere lengths. n = 8 vehicle treated myocytes from 3 felines, n = 10 SAHA-treated myocytes from 3 felines. All analysis was performed using a two-sided Student’s t-test. * p < 0.05 Data shown are means ± SEM.

Figure 3

Effect of SAHA on myofilament acetylation. The myofilament acetylation was assessed in acutely treated AFVM (A–F). Western Blot analysis revealed that (A,B) total myofilament acetylation was not increased with SAHA treatment but there was significantly more (C) acetylation at a 20 kDa band (indicated with arrow in (A)); n = 6 felines per group. (D) Coomassie staining was performed, and the band was cut out for mass spectrometry analysis, which revealed that this 20 kDa band with (E) increased acetylation is the myosin regulatory light chain with one (F) specific residue (lysine 115) driving the increase in acetylation; n = 3 felines per group, 3 technical replicates per sample. All analysis was performed using a two-sided Student’s t-test. Data shown are means ± SEM

Figure 4

Effect of acute SAHA treatment on human trabeculae. Trabeculae were isolated from biopsies of non-failing human LV. There was an increase in (A) developed force and decreased (B) diastolic tension with SAHA treatment, with an increase in (C) dF/dt_max/diastolic tension; n = 5–13 trabeculae per parameter from 7 patients. Myofilaments were isolated from human LV trabeculae. There was a slight left shift in the (D) average force-calcium curve with SAHA treatment. SAHA-treated samples had a reduction in (E) EC₅₀, indicating an increase in myofilament calcium sensitivity but no difference in (F) F_max; n = 7 control myocytes from 3 patients, 9 SAHA from 3 patients. SAHA-treated samples had a decrease in (G) passive stiffness at increasing sarcomere lengths; n = 7 myocytes from 3 patients in each group. For trabeculae functional experiments (A–C), repeated measure two-way ANOVA was performed. For the myofilament functional experiments (D-G), analysis was performed using a two-sided Student’s t-test. * p < 0.05, ** p < 0.01. Data shown are means ± SEM.

Figure 5

Effect of acute SAHA treatment on abundance of calcium handling proteins. Western Blots (A) of right atrium trabeculae after being treated with vehicle or SAHA. There was no change in protein abundance with SAHA treatment for (B) phosphorylated calcium/calmodulin-dependent protein kinase (pCaMKII T286), (C) CaMKII, (D) sarcoendoplasmic reticulum calcium transport ATPase (SERCA), (E) sodium–calcium exchanger 1 (NCX1), (F) ryanodine receptor (Ryr), (G) Ryr serine 2808, (H) Cav1.2, (I) Cav1.2 150 kDa, and (J) Cav1.2 75 kDa; n = 7–8 samples per protein from 8 patients. All analysis was performed using a two-sided Student’s t-test. NS stands for not significant. Data shown are means ± SEM.

Figure 6

Effect of isoform selective HDAC inhibition on non-failing human trabeculae. Trabeculae were isolated from right atrial appendages (RAA) and treated with Rodin-A, IRBM-D, or vehicle at high or low concentrations. (A) Developed force, (B) dF/dt_max, and (C) dF/dt_min were all increased with high concentrations of Rodin-A and IRBM-D compared to vehicle treatment. There was no significant difference (D) developed force, (E) dF/dt_max, and (F) dF/dt_min in the lower concentration-treated groups compared to vehicle; high concentration: Rodin-A, n = 8 trabeculae; IRBM-D, n = 9 trabeculae; low concentration: Rodin-A, n = 7 trabeculae; IRBM-D, n = 13 trabeculae; vehicle, n = 8 trabeculae. RAA trabeculae were isolated from 30 patients. All analysis was performed using repeated measure two-way ANOVA. * p < 0.05, ** p < 0.01, *** p < 0.001 Rodin-A vs. vehicle, # p < 0.05, ## p < 0.01, ### p < 0.001 IRBM-D vs. vehicle. Data shown are means ± SEM.