AMPKα1 deletion in myofibroblasts exacerbates post-myocardial infarction fibrosis by a connexin 43 mechanism

Fig. 1. Generation and characterisation of MF-specific AMPKα1 KO mice. a–f Cardiac fibroblasts (CFs) (a, d), myofibroblasts (MFs) (b, e) and cardiomyocytes (CMs) (c, f) were isolated from MF-specific AMPKα1 KO (cKO) and littermates WT (WT) mouse hearts 14 days after sham (a, d) or myocardial infarction (MI) (b, c, e, f) surgery. Western blotting was performed for analysing AMPKα1 (a–c) or AMPKα2 (d–f) expression (n = 3–4 mice/group). Dot plots represent data from individual mouse hearts, as well as mean ± SEM. Statistical significance was determined by unpaired t test. **p < 0.01 compared to WT. g–i Postn-Cre+/−;Rosa26-STOPfl/fl-enhanced yellow fluorescent protein (EYFP) (g, h) and Postn-Cre−/−;Rosa26-STOPfl/fl-EYFP (i) mouse hearts were collected 7 days following MI (h, i) or sham surgery (g), with EYFP expression assessed by immunohistofluorescence. Representative images of EYFP staining. Magnification: ×10. Scale bars: 100 µm

Fig. 2. MF-specific AMPKα1 deficiency aggravates LV adverse remodelling post-MI. a–c Left ventricular end-diastolic (LVEDV) (a) and -systolic (LVESV) (b) volumes and ejection fraction (EF) (c) of MF-specific AMPKα1 KO (cKO) and littermates WT (WT) mice measured by transthoracic echocardiography at the indicated time points post-myocardial infarction (MI) or sham surgery (n = 10–11 mice/group). Results are expressed as mean ± SEM. Statistical significance was determined by two-way ANOVA followed by Sidak’s multiple comparisons test. **p < 0.01, ***p < 0.001 compared to corresponding WT. d Kaplan–Meier survival curves of cKO and WT mice until 3 months after MI (n = 25–35 mice/group). Differences between groups were determined by the log-rank test, p = 0.25

Fig. 3. MF-specific AMPKα1 deficiency enhances LV fibrosis post-MI a representative images of MF-specific AMPKα1 KO (cKO) and littermates WT (WT) infarct areas stained by picrosirius red and visualised under bright-field light. Magnification: ×10. Scale bars: 100 µm. b–d Quantifications of fibrosis in the infarct (b), border (c) and remote (d) areas of cKO and WT mice (n = 5–10 mice/group). Results are expressed as % of measured tissue area. e Representative images of cKO and WT infarct areas stained by picrosirius red and visualised under polarised light. Magnification: ×10. Scale bars: 100 µm. f–h Quantification of picrosirius red polarised light analysis in the infarct (f), border (g) and remote (h) areas of cKO and WT mice (n = 4–5 mice/group). Results are expressed as red–orange on green–yellow pixel ratio in the measured tissue area. i Representative images of electron microscopy analysis in the infarct area of cKO and WT mice. Scale bars: 500 nm. j–k Quantification of collagen fibres density in the extracellular matrix of infarct (j) and border (k) areas of cKO and WT (n = 3 mice/group). Results are expressed as % of measured tissue area. l Proportional distribution of collagen…

Fig. 4. MF-specific AMPKα1 deficiency promotes CF activity. a, d Representative images of co-immunofluorescent staining of proliferating CFs (PCNA-positive: red fluorescence, vimentin-positive: green fluorescence) in the infarct area of MF-specific AMPKα1 KO (cKO) and littermates WT (WT) at 7 (a) and 14 (d) days post-myocardial infarction (MI). b, c Quantification of proliferating CFs in the infarct (b) or border (c) areas of cKO and WT at 7 days post-MI (n = 9 mice/group). e, f Quantification of proliferating CFs in the infarct (e) or border (f) areas of cKO and WT at 14 days post-MI (n = 5–7 mice/group). Results are expressed as % of PCNA-positive nucleus within the vimentin-stained tissue area. Magnification: ×20. Scale bars: 50 µm. Dot-plots represent data from individual mice, as well as mean ± SEM. Statistical significance was determined by unpaired t test (b, c, f) or Mann–Whitney test (e). *p < 0.05, **p < 0.01 compared to WT

Fig. 5. AMPKα1 regulates Cx43 expression in CFs. a–h Human cardiac fibroblasts (HCFs) were transfected with AMPKα1-targeting siRNA or scramble for 48 h and differential genes expression was analysed by RNA-Seq (n = 5 biological replicates/group) (a–e) or protein expression was assessed by Western blotting (n = 3 biological replicates/group) (f–h). a Volcano plot representation of differentially expressed genes. b Heat map representation of differentially expressed genes relevant for fibrosis and CF activity modulation. c–h Expression of mRNA (c–e) and corresponding proteins (f–h) relevant for CFs proliferation modulation. i–j HCFs were transfected with Cx43-targeting siRNA or scramble for 48 h and proliferation was assessed by flow cytometry measurement of EdU incorporation (n = 3 biological replicates/group) (i) or αSMA expression wad assessed by RT-qPCR (n = 3 biological replicates/group) (j). k HCFs were treated with Gap19 100 µM for 48 h and proliferation was assessed by flow cytometry measurement of EdU incorporation (n = 3 biological replicates/group). Dot plots represent data from biological replicates as well as mean ± SEM. Statistical significance was determined using false…

Fig. 6. AMPKα1 regulates Cx43 transcription and Cx43 mRNA degradation by a mechanism involving miR-125b-5p. a Cx43 promotor activity measurement in AMPKα1-deficient or control HEK293 cells. HEK293 cells were co-transfected with a luciferase-reporter plasmid and AMPKα1-targeting siRNA or scramble (n = 4 biological replicates/group). Results are expressed as luciferase activity relative to scramble. b Human cardiac fibroblasts (HCFs) were transfected with AMPKα1-targeting siRNA or scramble for 48 h before being treated with actinomycin D 5 µg/mL. Cx43 expression was analysed at the different time points by RT-qPCR (n = 3 biological replicates/group). Results are expressed as the mean of fold change over corresponding basal state. c In silico determination of miR-125b-5p predicted target sites in the 3′UTR of Cx43 mRNA by Diana tools, Pictar and microRNA.org. d HCFs were transfected with AMPKα1-targeting siRNA or scramble for 48 h and miR-125b-5p expression was assessed by RT-qPCR (n = 3 biological replicates/group). Results are expressed as fold change over scramble. e HCFs were transfected with miR-125b-5p mimic or control for 48 h, with Cx43 expression analysed by Western blotting (n = 3…

Fig. 7. AMPKα1 regulates Cx43 expression in infarcted areas and isolated fibroblasts of MF-specific AMPKα1 KO mice. a–d qRT-PCR measurement of Cx43 mRNA (a, b) and miR-125b-5p (c, d) expression in the infarcted (a, c) and remote (b, d) myocardium of MF-specific AMPKα1 KO (cKO) and littermates WT (WT) mice at 7 days post-myocardial infarction (MI) (n = 5–8 mice/group). e, g qRT-PCR measurement of Cx43 mRNA (e) and miR-125b-5p (g) expression in isolated MFs from cKO and WT hearts at 3 days post-MI (n = 7 mice/group). f Correlation between Cx43 and miR-125b expressions in MFs isolated from cKO and WT hearts at 3 days post-MI (n = 14). h, i Western blot analysis of Cx43 protein expression in isolated MFs (h) or cardiomyocytes (CMs) (i) from cKO and WT hearts at 14 days post-MI (n = 4 mice/group). Dot plots represent data from individual mouse hearts, as well as mean ± SEM. Statistical significance was determined by unpaired t test (a–e, g, h) or Mann–Whitney test (i) and correlation was assessed using the Pearson coefficient r calculation (f). *p < 0.05 compared to WT