Fig 2: PPAR signaling were significantly altered in MFN2 deficient spermatocytes. (A) mRNA expression levels of Ppara, Pparß, Ppar? and Pex3 were measured in MFN2 WT and MFN2 cKO spermatocytes via RT-qPCR. The data represent mean ± SEM and the cKO values were normalized to the WT controls. This data was quantified from n = 3 independent experiments. ***: p < 0.001.(B) The PPAR family protein expression levels were analyzed via western blot analysis using isolated spermatocytes. GAPDH was used as the loading control. (Ci-Cii) Testes sections were labeled with the anti-PEX3 antibody (in red) and co-labeled with DAPI (in blue) to label peroxisomes and cell nuclei, respectively. Scale bar is 50 µm. (Ciii) Bar chart to show the quantification of PEX 3 signal fluorescent intensity in MFN2 WT and cKO testes. (Di-Dii) Testis cells were immunolabeled with the anti-SYCP3 antibody to label spermatocytes (in red) and then co-labeled with BODIPY (in green) and DAPI (in blue) to label neutral lipid droplets and the cell nuclei, respectively. Scale bar is 10 µm. (Diii) Bar chart to show the quantification of BODIPY fluorescent intensity in MFN2 WT and cKO testicle cells. (Ciii and Diii) these data represent the mean ± SEM and the Students’ t-test was applied to analyze the data between each two groups, such that ***: p < 0.001.

Fig 3: The MFN2 cKO mouse exhibits abnormal spermatocyte differentiation. (A,B) Testes sections acquired at postnatal day (PD) 56 from (Ai–Aiii) MFN2 WT and (Bi–Biii) MFN2 cKO mice, were stained with hematoxylin. In the MFN2 cKO testes, the red asterisks indicate abnormal tubules in which the majority of the spermatogonia (SG), spermatocytes (SC) and post-meiotic spermatids (ST) were absent. SZ: spermatozoon. (Aiv,Biv) Epididymis histology sections were also stained with hematoxylin reagent. Scale bars are 100 µm. (C) Spermatozoa were extracted from the epididymis of (Ci,Cii) MFN2 WT and (Ciii–Cvi) MFN2 cKO mice at PD 35. They were then spread and fixed on glass slides, before being labeled with FITC-Periodic Acid-Schiff (PAS; in green) and DAPI (in blue). In the MFN2 cKO examples, the yellow arrowheads in (Ciii–Cvi) indicate abnormal spermatozoa containing a truncated or bent tail, and the white arrowhead in (Civ) indicates a spermatozoon with an abnormal nucleus. Scale bar is 20 µm. (Di) Bar graph to show the ratio of sperm abnormality in the MFN2 WT and MFN2 cKO groups, n = 223 and n = 300 spermatozoa were quantified in MFN2 WT and MFN2 cKO groups, respectively. (Dii) Individual dot plot to show the diameter of seminiferous tubules in MFN2 WT and MFN2 cKO testes. (E) Bar graphs to show the levels of (Ei) testosterone and (Eii) luteinizing hormone (LH) in the serum of MFN2 WT and MFN2 cKO mice, which were measured via ELISA from n = 3 mice in each group. In (Di–Eii), the data represent the mean ± SEM of three independent experiments and *p < 0.05, **p < 0.01. (F) Testis tissue sections from MFN2 WT and cKO groups were immunolabelled with various primary antibodies (?H2A.X labels double-strand breaks, SYCP3 labels meiotic spermatocytes, H1t labels mid-pachytene spermatocytes, PLZF labels undifferentiated cells, PCNA labels proliferative cells, cleaved-Caspase3 labels apoptotic cells and DAPI labels nuclei) and they were co-stained with DAPI (blue). Images were acquired via fluorescent microscopy. In (Fvi), the yellow arrowheads indicate PLZF-positive cells. Scale bars are 50 µm.

Fig 4: Comparison of meiotic chromosome spreads from MFN2 WT and MFN2 cKO spermatocytes. (A, D) Harsh and (C) mild chromosome spreads were prepared before immunolabeling procedures were conducted with the ?H2A.X, ?RF1, or MLH1 antibodies (in green) and the Sycp3 antibody (in red). The latter was used to label meiotic chromosomes. (A) Spermatocytes are shown at different stages. (B) The percentage of spermatocytes at these stages (i.e., L: leptotene, Z: zygotene, P: pachytene, and D: diplotene) were quantified in the MFN2 WT and MFN2 cKO groups. (C) The telomeres were immunolabelled with an anti-TRF1 antibody. (Ci,Cii) Images were captured by confocal microscopy. (Ciii) Three optical sections in the middle of the z-stack were used to quantify the TRF1 foci, which were expressed on the equator. 32 MFN2 WT spermatocytes and 27 MFN2 cKO spermatocytes were quantified in this experiment. (D) Pachytene stage spermatocytes in the (Di-Diii) MFN2 WT and (Div-Dvi) MFN2 cKO groups were immunolabelled with the anti-MLH1 (in green) and anti-SYCP3 (in red) antibodies. (Dvii) The number of MLH1 foci were quantified in the MFN2 WT and MFN2 cKO spermatocytes (n = 40 for each group). Scale bars are 5 µm *: p < 0.05; **: p < 0.01; ***: p < 0.001.

Fig 5: OLCs maturation in vitro. (A) The in vitro matured OLCs. Ctrl, human oocytes served as the positive control. (B) OLCs were stained with DAPI. (C) OLCs cultured in OLC-m were arrested at MII stage. Arrows indicate the first polar body. Nuclei were stained by DAPI. (D) Western blotting (left) and densitometry analysis (right) of SYCP3 expression in OLCs. Ctrl, human ovary tissue as the positive control. a-Tubulin was as internal control (E) Immunocytochemistry of a-Tubulin expression in MII OLC, the cell nuclei were stained with DAPI. Scale bar, 50 µm for B; 100 µm for A, C, and E