Supplementary Figure 1: Preparation of recombinant OPA1 and OPA1-dependent fusion in vitro or in vivo.

(a) Procedures for the fractionation and purification of recombinant OPA1. (b,c) SDS-PAGE analysis and Coomassie blue staining of L-OPA1 (b) and S-OPA1 (c). Asterisk: OPA1 protein. Fractionation experiments were carried out more than 10 times for L-OPA1 and more than 3 times for S-OPA1 with similar results. (d) Calculation of fusion activity by fitting to the exponential curve with the trace from the experiments shown in Fig. 2a. In the 1:4,000 condition, ∼25 L-OPA1 molecules, which might form ∼2.5 complexes (Fig. 1h) in each liposome (the number of lipid molecules ∼100,000)³⁵, could mediate membrane fusion. The goodness of fit was evaluated by the R² value. (e) Preparation of the OPA1 KO cell line using the CRISPR/Cas9 system. Whole cell homogenates from OPA1 WT or KO HeLa cells were prepared and subjected to immunoblot analysis with the indicated antibodies (the experiment was performed one time). (f) OPA1 WT or KO HeLa cells were fixed, permeabilized, subjected to immunofluorescence analysis with antibodies to TOM20 (red), and examined by confocal microscopy (representative images of more than 3 independent experiments with similar results). Magnified images are shown in the boxed regions. Scale bars, 10 μm. (g,h) Heterotypic mitochondrial fusion in HeLa cells. Mitochondrial fusion was analyzed in cell hybrids from cells expressing mitGFP and mitRFP using OPA1-deficient or control HeLa cells. After 6 h culture, fusion efficiency was analyzed by fluorescence microscopy (g) and cell counting (h). n, number of cell hybrids analyzed from 3 independent experiments: WT × WT (n = 30), WT × CRISPR-OPA1 (n = 30), and CRISPR-OPA1 × CRISPR-OPA1 (n = 30). Scale bar: 10 μm. Statistics source data are available in Supplementary Table 1. Unprocessed original scans of blot is shown in Supplementary Fig. 3.