Development of adeno-associated viral vectors targeting cardiac fibroblasts for efficient in vivo cardiac reprogramming

Abstract

Overexpression of cardiac reprogramming factors, including GATA4, HAND2, TBX5, and MEF2C (GHT/M), can directly reprogram cardiac fibroblasts (CFs) into induced cardiomyocytes (iCMs). Adeno-associated virus (AAV) vectors are widely used clinically, and vectors targeting cardiomyocytes (CMs) have been extensively studied. However, safe and efficient AAV vectors targeting CFs for in vivo cardiac reprogramming remain elusive. Therefore, we screened multiple AAV capsids and promoters to develop efficient and safe CF-targeting AAV vectors for in vivo cardiac reprogramming. AAV-DJ capsids containing periostin promoter (AAV-DJ-Postn) strongly and specifically expressed transgenes in resident CFs in mice after myocardial infarction (MI). Lineage tracing revealed that AAV-DJ-Postn vectors expressing GHT/M reprogrammed CFs into iCMs, which was further increased 2-fold using activated MEF2C via the fusion of the powerful MYOD transactivation domain (M-TAD) with GHT (AAV-DJ-Postn-GHT/M-TAD). AAV-DJ-Postn-GHT/M-TAD injection improved cardiac function and reduced fibrosis after MI. Overall, we developed new AAV vectors that target CFs for cardiac reprogramming.

Keywords: AAV; Fibrosis; Regeneration; Reprogramming.

Graphical abstract

Figure 1

Screening for AAV capsids efficient in transducing cardiac fibroblasts (A) Strategy for in vitro (B–E) and in vivo (F and G) screening to analyze cellular tropism of multiple AAV capsids (1–10 and DJ). The AAV vectors contained CMV promoter-driven GFP (AAV-CMV-GFP). (B–E) In vitro screening of AAV capsids. FACS analysis of GFP expression in CFs (B) and CMs (D) transduced with the indicated AAV-CMV-GFP capsids at an MOI of 2.5×10⁵ VGs/cell after one week. Quantitative data of GFP expression in CFs (C) and CMs (E) are shown (n = 3 independent triplicate experiments). (F and G) In vivo screening of AAV capsids. IHC of α-actinin, GFP, and DAPI in mouse hearts one week after MI. The indicated capsid AAV-CMV-GFP vectors (1×10¹¹ VGs/mouse) were directly injected into the mouse infarct hearts. Note that most GFP⁺ cells were α-actinin⁺ CMs. Quantitative data of GFP expression in nonmyocytes and CMs are shown in (G) (n = 5 independent biological experiments). All data are presented as mean ± SD. ^∗p < 0.05; ^∗∗p < 0.01 vs. the relevant control. Scale bars represent 50 μm. See also Figure S1.

Figure 2

Efficient and specific gene expression in cardiac fibroblasts with AAV-DJ-Postn vectors (A–D) The AAV expression cassette containing Postn promoter-driven Cre recombinase (AAV-Postn-Cre) was packaged into AAV6, 9, and DJ capsids (A). Schematic of the analysis of R26-tomato mice injected directly into the infarcted hearts with the AAV-Postn-Cre vectors (B). IHC of Tomato, cTnT (cardiac Troponin T), and DAPI in the Tomato hearts two weeks after MI with AAV6, 9, and DJ-Postn-Cre injection (C). High-magnification views show the border areas (C, lower panels). See also Figure S1G. Quantitative data of Tomato⁺ cells with AAV6, 9, and DJ-Postn-Cre injection are shown in (D) (n = 5 independent biological experiments). (E) Experimental scheme for injecting AAV-DJ-Postn-GFP into the ICR mouse infarcted hearts. Post-MI hearts were harvested at the indicated time points. (F and G) IHC of GFP, cTnT, COL1, SMMHC, CD31, and DAPI in the AAV-DJ-Postn-GFP-injected ICR mouse hearts one week after MI. Most GFP⁺ cells were immunopositive for COL1. The ratios of cTnT⁺ CMs, COL1⁺ CFs, SMMHC⁺ SMCs, and CD31⁺ ECs to GFP⁺ cells are shown in (G) (n = 5 independent biological experiments). See also Figure S2F. (H and I) IHC of GFP, cTnT, and DAPI in the AAV-DJ-Postn-GFP-injected mouse hearts after 1, 2, and 4 weeks of MI. Quantitative data of GFP⁺ cells at the border/infarct areas are shown in (I) (n = 5 independent biological experiments). (J) AAV-DJ-Postn-GFP was directly injected into healthy mouse hearts. IHC of GFP, cTnT, and DAPI expression after one week. High-magnification views show the injection site (lower panels). All data are presented as mean ± SD. ^∗∗p < 0.01 vs. the relevant control. ns, not significant. Scale bars represent 50 μm. See also Figures S1–S3.

Figure 3

AAV-DJ-Postn vectors expressing activated MEF2C improved in vivo cardiac reprogramming (A) MEF2C and MEF2C-TAD structures. The AAV-DJ-Postn vectors encoding GATA4, HAND2, TBX5, MEF2C, and MEF2C-TAD (M-TAD) were generated. (B and C) Generation of Postn^MCM/Tomato mice (B). The Postn^MCM/Tomato mouse hearts were directly injected with PBS (Ctrl), AAV-DJ-Postn-GHT/M, and AAV-DJ-Postn-GHT/M-TAD after MI. TAM-containing food pellets were continuously administered for two weeks (C). (D–F) IHC of Tomato, α-actinin, and DAPI in Postn^MCM/Tomato mouse hearts four weeks after MI. High-magnification insets show the sarcomeric organization (D). z stack image of Tomato and α-actinin double-positive iCMs in the AAV-DJ-Postn-GHT/M-TAD hearts (E). Quantitative analyses of α-actinin⁺/Tomato⁺ cells in border areas (F, n = 5 independent biological experiments). A total of 36,000–60,000 cells were counted for each mouse, and five mice were analyzed per group. (G and H) Generation of Tcf21^iCre/Tomato mice (G). Tcf21^iCre/Tomato mice were pre-treated with TAM via intraperitoneal (i.p.) injection for five days, and one week later MI and AAV gene delivery were performed. Four weeks after MI, the hearts were harvested for IHC (H). (I) Section of the Tcf21^iCre/Tomato mouse hearts in Ctrl four weeks after MI. IHC images of Tomato, α-actinin, and DAPI. (J–L) IHC of Tomato, α-actinin, and DAPI in Tcf21^iCre/Tomato mouse hearts four weeks after MI (J). High-magnification insets show the sarcomeric organization (J, lower panels). z stack image for Tomato and α-actinin double-positive iCMs in AAV-DJ-Postn-GHT/M-TAD hearts (K). Quantitative analyses of α-actinin⁺/Tomato⁺ cells in the border area (L, n = 5 independent biological experiments). A total of 36,000–60,000 cells were counted for each mouse, and five mice were analyzed per group. All data are presented as mean ± SD. ^∗p < 0.05; ^∗∗p < 0.01 vs. the relevant control. Scale bars represent 50 μm.

Figure 4

AAV-based in vivo cardiac reprogramming improved cardiac function and reduced fibrosis after MI (A and B) Three-dimensional principal-component analysis (A) and hierarchical clustering (B) in the RNA sequencing data from the Ctrl- and AAV-Postn-GHT/M-TAD-injected ventricles four weeks after MI (n = 5 independent biological replicates). (C) Gene set enrichment analysis of genes upregulated and downregulated in the GHT/M-TAD-injected mouse hearts compared with Ctrl. (D) Heatmap of cardiac, fibrosis, and heart failure/inflammation-related gene expressions in the Ctrl and AAV-Postn-GHT/M-TAD hearts. (E) Representative M-mode tracing images of cardiac function evaluated via echocardiography. Arrows indicate the left ventricular internal diameter in diastole (LVIDd) and systole (LVIDs). (F) The left ventricle ejection fraction (LVEF) and fractional shortening (LVFS) in Ctrl, AAV-DJ-Postn-GFP, and AAV-DJ-Postn-GHT/M-TAD-injected mice were determined via echocardiography four weeks after MI (n = 15 independent biological experiments). (G and H) Comparison of fibrosis among the Ctrl, AAV-DJ-Postn-GFP, and AAV-DJ-Postn-GHT/M-TAD hearts four weeks after MI. Fibrosis was evaluated at Base, Mid, and Apex levels using Masson’s trichrome staining. Representative histology data (G) and quantitative analyses (H) of the fibrotic area at each level are shown (n = 5 independent biological experiments). All data are presented as mean ± SD. ^∗p < 0.05; ^∗∗p < 0.01 vs. the relevant control. ns, not significant. Scale bars represent 500 μm. See also Figure S4.