Therapeutic efficacy of cardiosphere-derived cells in a transgenic mouse model of non-ischaemic dilated cardiomyopathy

Abstract

Aim: Cardiosphere-derived cells (CDCs) produce regenerative effects in the post-infarct setting. However, it is unclear whether CDCs are beneficial in non-ischaemic dilated cardiomyopathy (DCM). We tested the effects of CDC transplantation in mice with cardiac-specific Gαq overexpression, which predictably develop progressive cardiac dilation and failure, with accelerated mortality.

Methods and results: Wild-type mouse CDCs (10(5) cells) or vehicle only were injected intramyocardially in 6-, 8-, and 11-week-old Gαq mice. Cardiac function deteriorated in vehicle-treated mice over 3 months of follow-up, accompanied by oxidative stress, inflammation and adverse ventricular remodelling. In contrast, CDCs preserved cardiac function and volumes, improved survival, and promoted cardiomyogenesis while blunting Gαq-induced oxidative stress and inflammation in the heart. The mechanism of benefit is indirect, as long-term engraftment of transplanted cells is vanishingly low.

Conclusions: Cardiosphere-derived cells reverse fundamental abnormalities in cell signalling, prevent adverse remodelling, and improve survival in a mouse model of DCM. The ability to impact favourably on disease progression in non-ischaemic heart failure heralds new potential therapeutic applications of CDCs.

Keywords: Cardiomyopathy; Cell transplantation; Heart failure.

Figure 1

Functional benefits after cardiosphere-derived cell transplantation. (A) representative long-axis echocardiographic images at end-diastole (upper row) and end-systole (lower row) in control (CTL; wild-type) and in vehicle (Gαq+vehicle) and CDC (Gαq+CDC)-treated Gαq mice that were injected at 11 weeks of age at 3 months after treatment. Pooled data for left ventricular function show that cardiosphere-derived cell transplantation resulted in a sustained improvement of ejection fraction for 3 months in Gαq mice that received cardiosphere-derived cell at any of three different ages: 6 (B), 8 (C), and 11 (D) week old. Meanwhile, vehicle-treated Gαq mice showed significant deterioration of cardiac function. Data are means ± SEM; n = 6–8 in each group. The Mann–Whitney U test was applied. *P < 0.05 vs. Gq+CDC; **P < 0.01 vs. Gq+CDC; ***P < 0.005 vs. Gq+CDC.

Figure 2

Prevention of oxidative/nitrosative stress with CDC treatment. Representative western blots and pooled data (A) and representative immunohistochemical images [B; CTL (wild-type), vehicle and cardiosphere-derived cell-treated Gαq mouse hearts stained for gp91^phox, and 3-nitrotyrosine] from Gαq mice treated at 8 weeks of age. Membranes were stripped and GAPDH probed as a control for loading in each lane. Data are means ± SEM; n = 6–8 in each group. ^†P < 0.005 vs. Gαq+CDC and control (CTL; wild-type). Scale bars: 20 µm.

Figure 3

Suppression of inflammation with cardiosphere-derived cell treatment. Representative western blots and pooled data (A), and representative immunohistochemical images [B; CTL (wild-type), vehicle and cardiosphere-derived cell-treated Gαq mouse hearts stained for CD68, CD3, and CD20] from Gαq mice treated at 8 weeks of age. In CDC-treated Gαq mice, accumulation of CD68⁺ macrophages (B, upper row) and CD3⁺ T cells (B, middle row) was reduced. Membranes were stripped and GAPDH and histone H1 probed as a control for loading in each lane. Data are means ± SEM; n = 6–8 in each group. ^†P < 0.005 vs. Gαq+CDC and control (CTL; wild-type). Scale bars: 20 µm.

Figure 4

Restoration of the protein kinase C–protein kinase D–cAMP response element-binding protein remodelling pathway and suppression of apoptosis with cardiosphere-derived cell treatment. Representative immunohistochemical images [A and B; CTL (wild-type), vehicle and cardiosphere-derived cell-treated Gαq mouse hearts stained for p-cAMP response element-binding protein (Ser133) and cleaved caspase 3] and representative western blots and pooled data (C and D) from Gαq mice treated at 8 weeks of age. Cardiosphere-derived cell treatment restored the levels of relevant proteins in the protein kinase C–protein kinase D–cAMP response element-binding protein remodelling pathway to those in non-diseased controls (C). Increased protein density of PKCδ was associated with elevated nuclear contents of its downstream active effectors, phosphorylated protein kinase D (Ser744/748) and phosphorylated cAMP response element-binding protein (Ser133) in the vehicle-treated Gαq mice. (B and D) Increased abundance of active c-Jun N-terminal kinase, along with decreased active Akt protein density (Akt-p^T308) and markedly higher numbers of cells positive for cleaved caspase 3 in vehicle-treated Gαq mice. Activation of the protein kinase C–protein kinase D–cAMP response element-binding protein remodelling pathway and increased apoptosis were ameliorated by cardiosphere-derived cell therapy. Arrows in (B) point to apoptotic cleaved caspase 3⁺ cells. Membranes were stripped and GAPDH probed as a control for loading in each lane. Phosphorylated protein kinase C δ, protein kinase D, cAMP response element-binding protein, and Akt were normalized to total protein kinase Cδ, protein kinase D, cAMP response element-binding protein, and Akt, respectively. Quantification of JNK1 and JNK2 was performed using the same blots (D), taking advantage of the dual recognition of these two isoforms by the same antibody. Data are means ± SEM; n = 6–8 in each group. ^†P < 0.005 vs. Gαq+CDC and control (CTL; wild-type); scale bars: 10 µm (A), 20 µm (B).

Figure 5

Cardiosphere-derived cell treatment markedly reduced cardiac collagen content and fibrosis. Representative Masson trichrome (A) and immunohistochemical images [B and C; CTL (wild-type), vehicle and cardiosphere-derived cell-treated Gαq mouse hearts stained for fibroblast-specific protein and collagen I] and representative western blots and pooled data (D) from Gαq mice treated at 8 weeks of age. Cardiosphere-derived cell treatment markedly reduced cardiac collagen content, whether total or co-localized with fibroblast-specific protein, and fibrosis (A–D). Membranes were stripped and GAPDH probed as a control for loading in each lane. Data are means ± SEM; n = 6–8 in each group. ^†P < 0.005 vs. Gαq+CDC and control (CTL; wild-type). ^‡P < 0.05 vs. Gαq+CDC and control (CTL; wild-type). Scale bars: 50 µm (A and B), 10 µm (C).

Figure 6

Cardiosphere-derived cell treatment restored expression of Krüppel-like factor 5 and angiotensin II receptor type I. Representative immunohistochemical images [CTL (wild-type), vehicle and cardiosphere-derived cell-treated Gαq mouse hearts stained for fibroblast-specific protein, KLF5, and AT-R1] and representative western blots and pooled data from Gαq mice treated at 8 weeks of age (A and B). Cardiosphere-derived cell treatment diminished co-localization of fibroblast-specific protein with KLF5 (A) and AT-R1 (B) and decreased AT-R1 and KLF5 protein levels (A and B). Arrows point to up-regulated KLF5 co-localized with fibroblast-specific protein in vehicle-treated Gαq mouse heart. Membranes were stripped and GAPDH probed as a control for loading in each lane. Data are means ± SEM; n = 6–8 in each group. ^†P < 0.005 vs. Gαq+CDC and control (CTL; wild-type). Scale bars: 10 µm.

Figure 7

Cardiosphere-derived cell treatment increased cardiomyocyte cycling and proliferation and augmented number of c-kit positive cells differentiating into cardiac lineage (c-kit⁺Nkx2.5⁺). Representative immunohistochemical images and pooled data [A–C; CTL (wild-type), vehicle and cardiosphere-derived cell-treated Gαq mouse hearts stained for Ki67 (A), aurora B (B), c-kit and Nkx2.5 (C)] from Gαq mice treated at 8 weeks of age. Arrows point to Ki67⁺ (A) and aurora B⁺ (B) cardiomyocytes and the cells positive for both c-kit and Nkx2.5 (C). Fractions of cycling (Ki67⁺) and proliferating (Aurora B⁺) cardiomyocytes are expressed as the number of Ki67⁺ and aurora B⁺ cardiomyocytes divided by the total number of cardiomyocytes per high-power field, respectively [pooled data (A) and (B)]. The portion of c-kit⁺Nkx2.5⁺ cells was calculated as the number of c-kit⁺Nkx2.5⁺ cells divided by the total number of cardiomyocytes per HPF [pooled data (C)]. Wheat germ agglutinin was applied for staining and delineation of cell membrane. Data are means ± SEM; n = 6–8 in each group. ^†P < 0.01 vs. Gαq+Vehicle and control (CTL; wild-type); scale bars: 10 µm.

Figure 8

Survival analysis, changes in antioxidant defence pathway, and schematic summary of key findings. (A) Kaplan–Meier approach was applied for assessment of survival rate (n = 8 in each group). Survival was significantly attenuated in vehicle-treated Gαq mice relative to either cardiosphere-derived cell-treated Gαq mice or wild-type controls after adriamycin insult; the latter two groups, however, were statistically indistinguishable (P < 0.001, log-rank test). Cardiosphere-derived cell treatment restored activity of antioxidant pathway, Nrf2-Keap1. Representative western blots and pooled data from Gαq mice treated at 8 weeks of age demonstrate restored protein content of nuclear Nrf2, cytoplasmic Keap1 and the downstream gene products of Nrf2, catalase, and copper-zinc superoxide dismutase (Cu-Zn SOD), with cardiosphere-derived cell treatment. Membranes were stripped and GAPDH and histone H1 probed as a control for loading in each lane. Data are means ± SEM; n = 6–8 in each group. ^†P < 0.005 vs. Gαq+CDC and control (CTL; wild-type). (C) Various deleterious pathways are up-regulated in the Gαq model of dilated cardiomyopathy. Treatment with cardiosphere-derived cell impacted favourably on key disease processes including oxidative stress, inflammatory and remodelling pathways, apoptosis, and fibrosis.