A non-canonical pathway regulates ER stress signaling and blocks ER stress-induced apoptosis and heart failure

Endoplasmic reticulum stress is an evolutionarily conserved cell stress response associated with numerous diseases, including cardiac hypertrophy and heart failure. The major endoplasmic reticulum stress signaling pathway causing cardiac hypertrophy involves endoplasmic reticulum stress sensor PERK (protein kinase-like kinase) and eIF2α-ATF4-CHOP signaling. Here, we describe a non-canonical, AGGF1-mediated regulatory system for endoplasmic reticulum stress signaling associated with increased p-eIF2α and ATF4 and decreased sXBP1 and CHOP. Specifically, we see a reduced AGGF1 level consistently associated with induction of endoplasmic reticulum stress signaling in mouse models and human patients with heart failure. Mechanistically, AGGF1 regulates endoplasmic reticulum stress signaling by inhibiting ERK1/2 activation, which reduces the level of transcriptional repressor ZEB1, leading to induced expression of miR-183-5p. miR-183-5p post-transcriptionally downregulates CHOP and inhibits endoplasmic reticulum stress-induced apoptosis. AGGF1 protein therapy and miR-183-5p regulate endoplasmic reticulum stress signaling and block endoplasmic reticulum stress-induced apoptosis, cardiac hypertrophy, and heart failure, providing an attractive paradigm for treatment of cardiac hypertrophy and heart failure.

Statistical analysis was carried out by a Student's two-tailed t-test.

Supplementary Figure 5
The AGGF1 protein treatment increases the ventricular vascular density as shown by increased CD31-positive vessels in TAC and control sham mice . Scale bar, 50 μm. Data are presented as mean ± s.d. from at least three independent experiments (n=8/group, *P<0.05, **P<0.01). Statistical analysis was carried out by a Student's two-tailed t-test. Figure 6 Gene delivery by AAV9-AGGF1 viruses inhibits myocardial fibrosis induced by TAC with Masson trichrome staining. Scale bar, 50 μm. Figure 7 The AGGF1 treatment reduces ER stress activator Tunicamycin (TM)-induced expression of CHOP, Ero1, and DR5. Data are presented as mean ± s.d. from at least three independent experiments (n=6/group, **P<0.01). Statistical analysis was carried out by a Student's two-tailed t-test. Figure 8 Semi-quantitative RT-PCR analysis to characterize alternative splicing of XBP1. Total RNA samples were isolated from TAC mice or control sham mice treated with AGGF1 (0.25 mg/kg body weight) or control PBS, and reverse-transcribed into cDNA. PCR was then carried out using the cDNA and XBP1 primers (Supplementary Table 1). The RT-PCR products were then separated by 3% agarose gels. Two PCR bands were detected. The 211 bp and 185 bp bands represent the alternatively spliced XBP1 transcripts with or without the 26 nt intron 3. The ACTB (encoding -actin) was used as loading control ( 123 bp). The intensity of the 185 bp spliced XBP1 transcript (sXBP1) over the ACTB transcript was quantified and plotted at the bottom. (n=6/group, **P<0.01). Data are shown as mean ± s.d. from at least three independent experiments. Statistical analysis was carried out by a Student's two-tailed t-test. Data are shown as mean ± s.d. from at least three independent experiments. Statistical analysis was carried out by a Student's two-tailed t-test.

Supplementary Figure 10 Western blot analysis for CHOP expression in the hearts from TAC mice injected with Ago-miR-183-5p vs. Ago-miR-NC or
Antago-miR-183-5p vs. Antago-miR-NC. Data are shown as mean ± s.d. from at least three independent experiments (n=6/group, **P<0.01). Statistical analysis was carried out by a Student's two-tailed t-test. Supplementary Fig. 11 Real-time RT-PCR analysis for AGGF1 expression in H9C2 cells by AGGF1 siRNA (siAGGF1) vs. negative control scramble siRNA (siNC). Data are presented as mean ± s.d. from at least three independent experiments (n=3/group, **P<0.01). Statistical analysis was carried out by a Student's two-tailed t-test. Supplementary Fig 12 AGGF1 acts upstream of ZEB1 in the regulation of  miR-183-5p expression. (a) Relative miR-183-5p promoter luciferase activity in H9C2 cells after siAGGF1 with or without siZEB1 (n=3/group, **P<0.01). (b) Real-time RT-PCR analysis for miR-183-5p expression in H9C2 cells after siAGGF1 with or without siZEB1. (n=3/group, **P<0.01). Data are shown as mean ± s.d. from at least three independent experiments. Statistical analysis was carried out by a Student's two-tailed t-test.

Supplementary Fig 13 Real-time RT-PCR analysis for miR-183-5p expression levels in H9C2 cells transfected with ERK specific siRNA (siERK) or control siRNA (siNC).
Data are shown as mean ± s.d. from at least three independent experiments (n=3/group, **P<0.01). Statistical analysis was carried out by a Student's two-tailed t-test.

Supplementary Fig 14 Western blot analysis for the activation of ERK1/2 and ZEB1 expression in H9C2 cells treated with ISO with or without AGGF1 for 48 h.
Data are shown as mean ± s.d. from at least three independent experiments (n=6/group, **P<0.01). Statistical analysis was carried out by a Student's two-tailed t-test. Figure 15 MiR-183-5p mimics reduces ER stress activator Tunicamycin (TM)-induced expression of CHOP, Ero1 , and DR5. Data are shown as mean ± s.d. from at least three independent experiments (n=6/group, **P < 0.01). Statistical analysis was carried out by a Student's two-tailed t-test.