Aberrant HSF1 signaling activation underlies metformin amelioration of myocardial infarction in mice

Figures

• 

  Figure 1. Metformin could reduce myocardial injury after myocardial infarction in mice (A) Simulation process diagram. (B) Masson staining of representative left ventricular section. The right panel shows infarct sizes. LV, left ventricle; INF, infarct area. Scale bar, 2,000 μm; n = 3; ∗∗∗p < 0.001. (C) Transmission electron microscopy photo of infarcted myocardial edge tissue. Scale bar, 1 μm.

• 

  Figure 2. HSF1 had protective effect on mouse myocardium (A) Expression levels of p-HSF1Ser326/HSF1 in tissues adjacent to infarcted myocardium in different treatment groups. (B) mRNA levels of HSF1 in tissues near infarcted myocardium in different treatment groups. (n = 3; SD) (C) Subcellular distribution of HSF1 in mouse cardiomyocytes. Red represents HSF1 fluorescence and blue represents DAPI. Scale bar, 50 μm. (D) Expression levels of HSF1 in the cytoplasm and nucleus, respectively. Cytoplasmic proteins were homogenized with β-actin as an internal reference, and nuclear proteins were homogenized with histone H3 as an internal reference. (E) Masson staining of representative left ventricular section. The right panel shows infarct sizes. LV, left ventricle; INF, infarct area. Scale bar, 2,000 μm; n = 3; ∗∗∗p < 0.001. (F) Representative pictures of transmission electron microscopy of tissue adjacent to infarcted myocardium in wild-type mice (WT) and HSF1-knockout homozygous mice (HSF1−/−). Scale bar, 1 μm.

• 

  Figure 3. Cardiomyocytes were treated with MET (100 μM) for 24 h Cells were fixed, permeabilized, and processed for immunofluorescence using antibodies against different SG markers (FMRP, G3BP, FXR1, and DDX3). DAPI is used as a marker for nuclei. The indicated percentage of SG-positive cells (>3 granules/cell) representing more than 1,000 cells of three independent experiments. UNT, untreated; MET, metformin. Scale bar, 20 μm.

• 

  Figure 4. HSF1 was the upstream regulator of AMPK/mTOR pathway (A) The mRNA expression levels of HSF1, AMPK, and mTOR in the cardiomyocytes of HSF1 overexpression group and its control group (Ad-HSF1 and Ad-NC), HSF1 knockdown group and its control group (si-HSF1 and si-NC) mice. (n = 3; SD; ∗∗∗p < 0.001) (B) Scatterplot of the correlation analysis between HSF1 and AMPK mRNA expression in mouse cardiomyocytes, measured by qPCR. (C) Punctate accumulation of GFP-LC3 in mouse cardiomyocytes at 48 h in different treatment groups (Ad-NC, Ad-HSF1, si-NC, and si-HSF1). (D) Expression levels of HSF1, AMPK/mTOR pathway, autophagy-related protein p62, and LC3II/I in mouse cardiomyocytes. (E) Expression levels of HSF1 and AMPK proteins in mouse cardiomyocytes after overexpression and knockdown of AMPK (Ad-NC, Ad-AMPK, si-NC, and si-AMPK). (F) Expression levels of mTOR and autophagy-related proteins p62, LC3II/I in mouse cardiomyocytes after overexpression and knockdown of AMPK or HSF1 (Ad-HSF1, si-HSF1, Ad-AMPK, and si-AMPK).

• 

  Figure 5. AMPKα2 promoter reporters were transactivated by HSF1 (A) Diagram shows the relative positions of full-length (FL) and fragments of AMPKα2 promoter reporters. (B and C) Reporter assay response of AMPKα2 promoter FL and the individual fragments was investigated. (D) Reporter assays of the P2 fragment of the AMPKα2 promoter containing a mutated TFBM as indicated. (E) A schematic illustrates the relative positions of qPCR probes to putative motif (pm) for ChIP-QPCR experiments. (F) Antibody-pulled down chromatins were analyzed by qPCR. RLU, relative light unit. Data were mean ± SD of three independent experiments. (n = 3; SD; ∗∗∗p < 0.001)

• 

  Figure 6. Metformin promoted autophagy in mouse cardiomyocytes through the AMPK/mTOR pathway (A) AMPK/mTOR pathway and its phosphorylated protein expression levels in tissues adjacent to infarcted myocardium in different treatment groups. (B) qPCR was used to detect the knockdown effect of si-AMPK in primary mouse cardiomyocytes. (n = 3; SD; ∗∗∗p < 0.001) (C) Comparison of different treatments on GFP-LC3 punctate aggregation of mouse cardiomyocytes in vitro (negative control, 1 μM met, 100 μM met and si-AMPK + 100 μM met). The left side shows representative photos of GFP-LC3 mouse cardiomyocytes. Scale bar, 20 μm. The percentage of cells with GFP-LC3 punctate aggregation is shown on the right. (n = 3) (D) Autophagy-related proteins LC3II/I and p62, AMPK/mTOR pathway, and their phosphorylated protein levels in mouse cardiomyocytes cultured in vitro. (E) Masson staining of representative left ventricular section. The right panel shows infarct sizes. LV, left ventricle; INF, infarct area. Scale bar, 2,000 μm; n = 3; ∗∗∗p < 0.001. (F) Transmission electron microscope representative pictures of the tissues adjacent to the infarcted myocardium in met + lv/sh-scramble group and met + lv/sh-AMPKα2…

• 

  Figure 7. The combination of metformin-HSF1 showed a stronger binding effect than the combination of metformin-AMPK (A and B) Schematic diagram of metformin ligand binding AMPK (A) and HSF1 (B) protein (pymol mapping); the protein is displayed in cartoon form. (C and D) Schematic diagram of the amino acid sites of metformin ligand binding to AMPK (C) and HSF1 (D) protein (pymol mapping). (E and F) Concentration gradient curve of compound metformin and protein AMPK (E) and HSF1 (F). (G) Binding curves of two proteins (AMPK and HSF1) at 3,200 nM and metformin, respectively. (H) Histogram of the maximum binding signal between the two proteins at 3,200 nM and metformin on the chip.

• 

  Figure 8. Mechanistic diagram of metformin (via HSF1 and AMPK) protecting mouse myocardium