Targeting mAKAPβ expression as a therapeutic approach for ischemic cardiomyopathy

Abstract

Ischemic cardiomyopathy is a leading cause of death and an unmet clinical need. Adeno-associated virus (AAV) gene-based therapies hold great promise for treating and preventing heart failure. Previously we showed that muscle A-kinase Anchoring Protein β (mAKAPβ, AKAP6β), a scaffold protein that organizes perinuclear signalosomes in the cardiomyocyte, is a critical regulator of pathological cardiac hypertrophy. Here, we show that inhibition of mAKAPβ expression in stressed adult cardiomyocytes in vitro was cardioprotective, while conditional cardiomyocyte-specific mAKAP gene deletion in mice prevented pathological cardiac remodeling due to myocardial infarction. We developed a new self-complementary serotype 9 AAV gene therapy vector expressing a short hairpin RNA for mAKAPβ under the control of a cardiomyocyte-specific promoter (AAV9sc.shmAKAP). This vector efficiently downregulated mAKAPβ expression in the mouse heart in vivo. Expression of the shRNA also inhibited mAKAPβ expression in human induced cardiomyocytes in vitro. Following myocardial infarction, systemic administration of AAV9sc.shmAKAP prevented the development of pathological cardiac remodeling and heart failure, providing long-term restoration of left ventricular ejection fraction. Our findings provide proof-of-concept for mAKAPβ as a therapeutic target for ischemic cardiomyopathy and support the development of a translational pipeline for AAV9sc.shmAKAP for the treatment of heart failure.

Figure 1.. Targeting of mAKAPβ protects against isoproterenol-induced cardiomyocyte death in vitro.

a. Primary adult Sprague-Dawley rat ventricular myocytes were infected with adenovirus expressing mAKAP (shmAKAP) or control shRNA (shControl) and cultured in minimal medium for two days with and without 10 µmol/L isoproterenol (Iso) before TUNEL staining. TUNEL - green; Dapi nuclear stain – blue; bar - 100 µm. Arrowheads indicate nuclei with detectable TUNEL labeling. b. Quantification of TUNEL-positive cells (mean ± s.e.m.) for 3 independent experiments. * p < 0.05.

Figure 2.. Cardiomyocyte-specific mAKAPβ knock-out prevents systolic dysfunction following myocardial infarction.

Following oral administration of tamoxifen (125 mg/kg chow) for one week starting at 8 weeks of age to induce gene knock-out, mAKAP CKO mice and control mAKAP^f/f and MCM mice underwent permanent ligation of the left coronary artery survival surgery at ten weeks of age or sham operation. M-mode echocardiography was performed 4 weeks post-MI. a. Representative echocardiographic images. Left ventricular internal diameter indicated in diastole (LVID;d - orange) and systole (LVID;s - green). b-e. Left ventricular volume in diastole (b) and systole (c), ejection fraction (d), and anterior wall thickness in systole (e). Bars show mean ± s.e.m. Data were analyzed by two-way ANOVA with Tukey post-hoc testing. n: MCM-sham - 9; mAKAP^f/f-sham - 8; mAKAP CKO-sham – 15; MCM-MI- 12; mAKAP^f/f-MI - 10; mAKAP CKO-MI – 17. * vs. sham for the same genotype; ^† vs. MCM-MI; ^‡ vs. mAKAP^f/f-MI. ^‡ p < 0.05; ^††,^‡‡ p < 0.01; ***,^†††,^‡‡‡ p < 0.001. There were no significant differences among the 3 sham-operated cohorts.

Figure 3.. A new shRNA gene therapy vector inhibits mAKAPβ expression.

a. Alignment of mAKAP mRNA sequences from different vertebrate species. Conserved mAKAP shRNA target is in box. b. Design of an AAV9sc shRNA vector. ITR – AAV inverted terminal repeat (intact and deleted “Δ”); cTnT – chicken cardiac troponin T; beige blocks – human MIR30A genomic sequence; blue arrowheads – shRNA sequences. c. Western blot for mAKAPβ in whole heart extracts from individual adult mice 3 weeks after tail vein intravenous (IV) injection with 5×10¹¹ viral genomes (vg) AAV9sc.shmAKAP (n = 3) or AAV9sc.shControl virus (n = 2). Graph shows mAKAPβ expression (mean ± s.e.m.) normalized to total protein detected by Ponceau stain of the same blot (not shown). *** p < 0.001. See also Supplementary Figure 2. d. Normal human iPSC-CMs were infected 1 week before analysis by mAKAP western blot with AAV2sc vectors expressing the mAKAP and control MIR30A cassettes (as in b) under the control of the CMV promoter (multiplicity of infection = 1000). Representative blot shows technical replicates for single iPSC line. Graph shows average mAKAPβ expression (mean ± s.e.m. for mAKAP expression normalized to total protein detected by Ponceau stain of the same blot) for biological replicates performed with 3 different iPSC lines. * p < 0.05.

Figure 4.. AAV9sc.shmAKAP treatment attenuates systolic dysfunction and prevents heart failure following myocardial infarction in mice.

a. Schema for AAV9sc.shmAKAP treatment study. C57BL/6 male and female mice were subjected at 8–10 weeks of age to permanent ligation of the left coronary artery or sham survival surgery. Echocardiographic data obtained 2 days after surgery was used to identify mice meeting inclusion criteria, followed by randomization to generate two groups each of sham-operated and infarcted mice. The 4 mouse cohorts were injected IV with 5×10¹¹ vg of either AAV9sc.shmAKAP (highlighted in red) or AAV9sc.shControl the third day after survival surgery, followed by serial echocardiography and endpoint assessment 56 days post-operatively. This panel was drawn in part using pictures adapted from Servier Medical Art (http://smart.servier.com/), licensed under a Creative Commons Attribution 3.0 Unported License. b. Representative M-mode echocardiographic images. Left ventricular internal diameter indicated in diastole (LVID;d - orange) and systole (LVID;s - green). c-e, Left ventricular volume in diastole (c) and systole (d) and ejection fraction (e) by serial M-mode echocardiography. shControl-sham – black circles; shControl-MI – black filled circles; shmAKAP-sham – red triangles; shmAKAP- MI – red filled triangles. * vs. shControl–MI; ^† vs. sham for same virus (color-coded). f-h, Biventricular, biatrial, and wet lung weights at endpoint normalized to tibial length (TL). n for panels c-h as indicated in panel a. i, Images of hearts at the endpoint. Bar – 5 mm. j. Masson Trichrome-stained transverse sections. Bar – 1 mm. k. Infarct area for MI heart sections at the level of the papillary muscles, measured in sections stained as in j. n = 7,8. *,^† p < 0.05; **,^†† p < 0.01; ***,^††† p < 0.001.