Modulating mitofusins to control mitochondrial function and signaling

Abstract

Mitofusins reside on the outer mitochondrial membrane and regulate mitochondrial fusion, a physiological process that impacts diverse cellular processes. Mitofusins are activated by conformational changes and subsequently oligomerize to enable mitochondrial fusion. Here, we identify small molecules that directly increase or inhibit mitofusins activity by modulating mitofusin conformations and oligomerization. We use these small molecules to better understand the role of mitofusins activity in mitochondrial fusion, function, and signaling. We find that mitofusin activation increases, whereas mitofusin inhibition decreases mitochondrial fusion and functionality. Remarkably, mitofusin inhibition also induces minority mitochondrial outer membrane permeabilization followed by sub-lethal caspase-3/7 activation, which in turn induces DNA damage and upregulates DNA damage response genes. In this context, apoptotic death induced by a second mitochondria-derived activator of caspases (SMAC) mimetic is potentiated by mitofusin inhibition. These data provide mechanistic insights into the function and regulation of mitofusins as well as small molecules to pharmacologically target mitofusins.

Fig. 1. A small molecule activator of mitofusins.

a Domain model of MFN2 showing GTPase (green), transmembrane region (black), HR1 (purple), and HR2 (blue). Amino acid numbers corresponding to human MFN1 and MFN2. Cartoon depiction of MFN2 structure showing anti-tethering (closed) conformation and pro-tethering (open) conformation. b Cartoon depiction of conformational activation of MFNs and fusion of the OMM. MASMs promote the pro-tethering conformation of MFNs and subsequently mitochondrial fusion. c Ribbon presentation of the human full-length MFN2 structural model highlighting specific interactions between the HR1 (purple) and HR2 (blue) helical segments. d Pharmacophore hypothesis based on the sidechains of the HR1-amino acids: Val372, Met376, His380 interacting with HR2 as in (c) comprising of three hydrophobic points, one aromatic ring, and one hydrogen bond donor. e Screening of putative MASMs in cells using mitochondrial aspect ratio (Mito AR) as readout. Cells were treated with MASMs (1 μM, 2 h). Data represent mean of two independent biological replicates. f Chemical structure of MASM7. g Confocal micrographs of MEFs treated with MASM7 (1 μM, 2 h). Mitochondria were stained with Mitotracker green. Scale bar 20 μm. Micrograph is representative of n = 3 independent experiments. h MASM7 and 367–384Gly concentration-responsively increased Mito AR in MEFs. Cells were treated with MASM7 or 367–384 Gly at the indicated concentrations for 2 h. Data represent mean of two independent biological replicates. i Quantification of Mito AR of WT, Mfn1 KO, Mfn2 KO, and Mfn1/Mfn2 DKO MEFs treated with MASM7 (1 μM, 2 h). Data represent mean ± SEM of three independent biological replicates. Statistics were obtained using two-tailed unpaired t-test: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Source data are provided as a Source Data file.

Fig. 2. A small molecule inhibitor of mitofusins.

a Ribbon presentation of the human full-length MFN2 structural model highlighting specific interactions between the HR1 (purple) and HR2 (blue) helical segments. b Pharmacophore hypothesis based on the sidechains of the HR1-amino acids: Leu408, Ala412, Tyr415 interacting with HR2 as in (a) comprising 2 hydrophobic points, one aromatic ring, and one hydrogen bond donor or acceptor. c Screening of putative MFIs using Mito AR as readout. Cells were treated with MFIs (10 μM, 6 h). Data represent mean of two independent biological replicates. d Chemical structure of MFI8. e Confocal micrographs of MEFs treated with MFI8 (20 μM, 6 h). Mitochondria were stained with Mitotracker green. Scale bar 20 μm. Micrograph is representative of n = 3 independent experiments. f MFI8 and 398–418Gly concentration-responsively decreased Mito AR in MEFs. Cells were treated with MFI8 or 398–418Gly at the indicated concentrations for 6 h. Data represent mean of two independent biological replicates. g Quantification of Mito AR of WT, Mfn1 KO, Mfn2 KO, and Mfn1/Mfn2 DKO MEFs treated with MFI8 (20 μM, 6 h). Data represent mean ± SEM of three independent biological replicates. Statistics were obtained using two-tailed unpaired t-test: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. h Quantification of Mito AR of MEFs treated with MASM7 (1 μM), MFI8 (20 μM), and the combination of MASM7 and MFI8 for 6 h. Data represent mean ± SEM of three independent biological replicates. Statistics were obtained using one way ANOVA: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Source data are provided as a Source Data file.

Fig. 3. MASM7 and MFI8 interact specifically with the HR2 domain of MFN2.

a Plots of mean corrected normalized fluorescence (ΔF_norm: F_nomr bound − F_norm unbound) from MST signal analysis of titrations of HR2 with indicated compounds and peptides. Data represent mean ± SEM from three replicate experiments. b Quantification of Mito AR of Mfn1/Mfn2 DKO MEFs reconstituted with WT, L727A, and D725A/L727A Mfn2 and treated with MASM7 (1 μM, 2 h). Data represent mean ± SEM of three independent biological replicates. Statistics were obtained using two-tailed unpaired t-test: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. c Quantification of Mito AR of Mfn1/Mfn2 DKO MEFs reconstituted with WT, S685A and L692A Mfn2 and treated with MFI8 (20 μM, 6 h). Data represent mean ± SEM of three independent biological replicates. Statistics were obtained using two-tailed unpaired t-test: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. d Separation of MFN2 oligomers from isolated mitochondria treated with GTP, MASM7, and MFI8 as indicated using BN-PAGE electophoresis. Gray arrowhead marks 450 kD MFN2 oligomer and black arrowhead marks MFN2 dimer. e Quantification of the proportion of the MFN2 oligomer observed in the ∼450 kD band to the MFN2 dimer band after treatment with GTP (d). Data represent mean ± SEM of seven independent biological replicates. Statistics were obtained using one way ANOVA: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. f Cellular engagement (CETSA) of MFN2 by MASM7, MFI8, and the combination of both. A representative blot from three independent experiments is shown. g Quantification of temperature-dependent normalized MFN2 levels obtained by densitometry with corresponding fitted curves. Data represent mean ± SEM from n = 3 independent experiments. Source data are provided as a Source Data file.

Fig. 4. MASM7 and MFI8 alter mitochondrial functionality.

a A representative OCR trace of WT or Mfn1/Mfn2 DKO MEFs from three independent experiments is shown. OCR was normalized to Veh of WT MEFs for t = 1.31 min. Data represent mean ± SD from n = 6 replicates. b Bar graphs showing the quantification of basal, maximal respiration and ATP production of (a). Data represent mean ± SEM from n = 6 replicates. c A representative OCR trace of WT or Mfn1/Mfn2 DKO MEFs from three independent experiments is shown. OCR was normalized to Veh of WT MEFs for t = 1.31 min. Data represent mean ± SD from n = 5 replicates. d Bar graphs showing the quantification of basal, maximal respiration, and ATP production of (c). Data represent mean ± SEM from n = 5 replicates. e MASM7 concentration responsively increased membrane potential in MEFs. Cells were treated with the indicated concentrations of MASM7 for 6 h. Data represent mean ± SEM of four independent biological replicates. f MFI8 concentration responsively decreased membrane potential in MEFs. Cells were treated with the indicated concentrations of MFI8 for 6 h. Data represent mean ± SEM of four independent biological replicates. g Relative variance-stabilized expression levels of nuclear-encoded mitochondrial genes belonging to respiratory complexes. h Heatmap showing the alteration in the metabolites of the TCA cycle. Heatmap was generated by quantitative metabolomics from three independent biological replicates. i Heatmap showing the alteration in the metabolites of the TCA cycle. Heatmap was generated by quantitative metabolomics from three independent biological replicates. j NAD/NADH ratio. Data represent mean ± SEM of three independent biological replicates. k NAD/NADH ratio. Data represent mean ± SEM of three independent biological replicates. Statistics were obtained using one way ANOVA for panels (b, d) or two-tailed unpaired t-test for panel (j): *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Cells were treated with MASM7 (1 μM, 6 h) for panels (a, g, h, k) or MFI8 (20 μM, 6 h) for panels (c, g, i, j). Source data are provided as a Source Data file.

Fig. 5. MFI8 induces minority MOMP and cytochrome c release leading to caspase 3/7 activation.

a Caspase 3/7 assay in WT MEFs treated with MFI8 for 6 h. Data represent mean ± SEM of three independent biological replicates. b Caspase 3/7 assay in WT MEFs treated with MASM7 (1 μΜ, 6 h). Data represent mean ± SEM of three independent biological replicates. Caspase 3/7 assay in WT (c), Mfn1/Mfn2 DKO (d) and Apaf-1 KO (e) MEFs. Cells were treated with MFI8 or MFI22 (20 μM, 6 h). Data represent mean ± SEM of three independent biological replicates. f Western blot analysis of cytochrome c release from MEFs treated with MFI8 for 6 h. Blot is representative of n = 2 independent experiments. g Western blot analysis of cytochrome c release from Mfn1/Mfn2 DKO MEFs treated with MFI8 for 6 h. ABT-737 and S63845 were used as positive controls for the assay. Blot is representative of n = 2 independent experiments. h Mitochondrial membrane potential assay in Mfn1/Mfn2 DKO MEFs treated with MFI8 and ABT-737/S63845 for 6 h. Data represent mean ± SEM of three independent biological replicates. i Evaluation of dead cells using Annexin V/PI staining via flow cytometry. Cells positive to Annexin V, PI, or both, were considered as dead. Data represent mean ± SEM of five independent biological replicates. j Viability assay in MEFs treated with MASM7 or MFI8 for 72 h. Data represent mean ± SEM of three independent biological replicates. k U2OS cells expressing CytoGFP/MitoCherry were treated with MFI8 (20 μM, 6 h) or ABT-737 (10 μM, 3 h) and minority MOMP was quantified. Data represent mean ± SEM of three independent biological replicates. l Caspase 3/7 assay in U2OS cells treated with MFI8 (20 μM, 6 h) or MASM7 (1 μM, 6 h). Data represent mean ± SEM of three independent biological replicates. m Quantification of Mito AR in U2OS cells treated with MFI8 (20 μM, 6 h). Data represent mean ± SEM of n = 65 mitochondria. Statistics were obtained using one way ANOVA for (panels c, e, h, i, k, l) or two-tailed unpaired t-test for panels (b, d, m): *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Source data are provided as a Source Data file.

Fig. 6. MFI8 induces DNA damage.

a MEFs and Mfn1/Mfn2 DKO MEFs were immunostained for γH2AX. Cells were treated with MFI8 (20 μM, 6 h) or MASM7 (1 μM, 6 h). Scale bar 20 μm. b Quantification of γH2AX foci of WT MEFS (a). Error bars represent mean ± SEM of three independent biological replicates. c Quantification of γH2AX foci of Mfn1/Mfn2 DKO MEFs (a). Error bars represent mean ± SEM of three independent biological replicates. d Relative variance-stabilized expression levels of genes that are involved in DNA damage response. e U2OS cells were immunostained for γH2AX. Cells were treated with MFI8 (20 μM, 6 h). Scale bar 20 μm. f Quantification of γH2AX foci of (e). Error bars represent mean ± SEM of three independent biological replicates. g Quantification of γH2AX foci from U2OS cells treated with MFI8 or the combination of MFI8 with Q-VD-OPh. Error bars represent mean ± SEM of three independent biological replicates. Statistics were obtained using two-tailed unpaired t-test for panel (f) or one way ANOVA for panels (b, c, g): *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Source data are provided as a Source Data file.

Fig. 7. MFI8 induces cell death in combination with BV6 SMAC mimetic.

Evaluation of dead cells using Annexin V/PI staining as readouts via flow cytometry. Cells that were positive to Annexin V, PI, or both, were considered as dead. Cells were treated with MFI8 or MFI22 (20 μM, 6 h) and BV6 (20 μM, 6 h). a Percentage of dead cells in WT and Mfn1/Mfn2 DKO MEFs upon treatment with BV6. Data represent mean ± SEM from n = 5 independent experiments for WT MEFs and n = 4 independent experiments for Mfn1/Mfn2 DKO MEFs. b Percentage of dead cells in WT MEFs upon treatment with BV6, MFI8, or the combination of both. Data represent mean ± SEM from n = 5 independent experiments. c Percentage of dead cells in WT MEFs upon treatment with BV6, MFI22, or the combination of both. Data represent mean ± SEM from n = 4 independent experiments. d Percentage of dead cells in Mfn1/Mfn2 DKO MEFs upon treatment with BV6, MFI8, or the combination of both. Data represent mean ± SEM from n = 4 independent experiments. e Percentage of dead cells in Apaf-1 KO MEFs upon treatment with BV6, MFI8, or the combination of both. Data represent mean ± SEM from n = 3 independent experiments. Statistics were obtained using one way ANOVA for panels (a, b): *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Source data are provided as a Source Data file.

Fig. 8. Schematic representation of the mechanism of action of MASM7 and MFI8 and their effects in mitochondrial fusion, mitochondrial functionality, and signaling.

The figure was created with BioRender.com.