Fig 2: Caveolin-1 expression precludes DRP1 phosphorylation and redistribution during the early phase of ER stress. a Mock and CAV1 cells in control condition (con) or under early ER stress with tunicamycin (tun) were fractionated and processed for detection of phospho-DRP1 and DRP1 by western blotting using PDI as a loading control. C cytosol, LM light membranes (microsomes), HM heavy membranes (crude mitochondria). b Quantification of DRP1 phosphorylation analysed in (a) (n = 3). c Quantification of membrane-bound DRP1 distribution analysed in a (n = 3). d Experimental groups as in (a) were processed for immunofluorescence analysis. CAV1, mtHSP70 (mitochondrial marker, Mito) and DRP1 were stained with respective antibodies and then imaged using confocal microscopy. Merge between DRP1 and CAV1 images shows colocalized areas in yellow. Merge between DRP1 and Mito images shows colocalized areas in cyan. In merge images of Mito, DRP1 and CAV1, triple colocalization areas appear as white pixels. e CAV1 fluorescence per cell was quantified from images obtained in d (n = 3). f DRP1-CAV1 colocalization was quantified as Manders’ coefficients of images obtained in (d) (n = 3). g DRP1-Mito colocalization was quantified as Manders’ coefficients of images obtained in (d) (n = 3). h From images obtained in d, thresholded CAV1 (C) and Mito (M) images were generated. The intersection between M and DRP1 corresponds to mitochondrial DRP1 (DRP1mito). Merge between M and C images shows colocalized areas in magenta. The intersection between M and C corresponds to MAM. Merge between MAM and DRP1 or DRP1mito images shows colocalized pixels in white, which correspond to triple colocalization areas. i The fraction of DRP1 fluorescence that colocalizes with MAM was quantified from images obtained in h (n = 3). j The fraction of DRP1mito fluorescence that colocalizes with MAM was quantified from images obtained in h (n = 3). For each independent imaging experiment, 5–15 cells were analysed. Scale bars: 10 µm. Results are shown as mean ± s.e.m. &P < 0.05 compared with respective LM fraction. †P < 0.05 overall comparison between conditions. *P < 0.05 and **P < 0.05 compared with respective con condition. ##P < 0.01 and ####P < 0.0001 overall comparison between conditions. ns non-significant

Fig 3: Visualization of Ddit3, Eif4ebp1, Nupr1, Hspa9, and Atf5 mRNA in fixed, paraffin embedded sections of E18.5 femurs. Fluorescent in situ hybridization (FISH) of mRNA in WT and Hom tissue sections was performed at the same time, with the same reagents, and at identical conditions. The slides were imaged at identical fluorescence channel settings and digitally visualized with identical contrast enhancement to preserve relative fluorescence intensity. Cell nuclei labeled with DAPI are shown in grayscale. The other pseudo colors correspond to the gene symbol colors of the corresponding legends. Intense Col1a1 labeling identifies OBs while weaker Col1a1 labeling is observed in fibroblasts (mostly within periosteum). White boxes outline regions shown in zoomed panels.

Fig 4: Transcriptomic analysis of osteoblast enriched spot in frozen tissue sections. A) Mapping of OB-enriched spots for transcriptomic analysis based on marker gene expression. Top left: Frozen sections of growth plate and adjacent trabecular bone from proximal tibia of a wild type mouse (WT-1 sample). Top right: Osteoblast-enriched spots (yellow circles) selected for transcriptomic analysis based on above-threshold expression of Col1a1, Runx2, Sp7, and Ibsp (similar to Figure 1) as well as below-threshold expression of Col2a1 (<0.15% UMI) and Col10a1 (<0.03% UMI). Bottom left: Spots with high Ibsp (yellow), Col10a1 (magenta), and Col2a1 (cyan) expression. Bottom right: zoomed area of the left image illustrating Ibsp expression in trabecular area, Col10a1 expression in the hypertrophic zone and adjacent structures, and Col2a1 expression in the rest of the growth plate. B) Expression of ISR genes (top row), endoplasmic reticulum (ER) and mitochondrial HSP70 and HSP90 chaperones (second row), transcription factors regulating cellular response to stress (third row), and genes indicating mitochondrial distress (bottom row). Violin plots are shown for each of the 3 Het (yellow) and 3 WT (green) animals examined by srRNASeq. BIP encoded by Hspa5 is ER-HSP70, Hspa9 encodes mt-HSP70, Hsp90b1 and Trap1 encode ER-HSP90 and mt-HSP90, respectively. Cyb5r1, Gpt2, and Shmt2 encode mitochondrial proteins. Asparagine synthase gene (Asns) is a known marker of mitochondrial stress response regulated by ATF4.[ 33 ] More than 20% change in gene expression with p < 0.05 (Wilcoxon test) was considered significant. N.S. = not significant, ** p < 0.01, **** p < 0.0001.

Fig 5: Relative quantity (RQ) of key pathway/function markers in cultured primary OBs. A-F) RNASeq quantification of mRNA relative to geometric mean of housekeeping genes (Actg1, Actb, Mrfap1, and Sdha). G) Cell lysate protein quantification in Het relative to average WT values by Western blotting. H) Representative blots of the proteins and corresponding loading standards (each panel shows a separate blot). Short (3–5 h) treatment with 10 µg mL-1 tunicamycin (Tn) was used for validating effects of canonical, acute UPR on BIP and ATF4 (not expected to produce noticeable effects on mt-HSP70, ATF5, and EIF4EBP1). A) Hspa9 and Atf5. B) ISR genes. C) Endoplasmic reticulum (ER) UPR genes (Hspa5, Hsp90b1, Xbp1, Atf6) and Creb3l1. D) Markers of mitochondrial distress. E) Markers of mt-UPR. F) Markers of cellular malfunction: Bglap – OB differentiation, Lrp5 – Wnt signaling, Avil – cytoskeleton and cell motility, Rcc2 – cell cycle. Error bars show standard error of the mean in six replicates (mRNA) and 6–12 replicates (protein). N.S. = not significant, * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. More than 20% (see Table S1, Supporting Information) changes in relative expression with p < 0.05 were considered statistically significant. Two-tailed, heteroscedastic t-test and Mann–Whitney U-test produced consistent significance estimates for most markers, with the only exception of p = 0.03 in t-test and p = 0.1 in U-test indicated by # (EIF4EBP1 in G). For mRNA analysis, six wells per genotype were examined (3 at 14 and 3 at 21 days after seeding) from two separate cell preparations (two wells per genotype in the first and four wells per genotype in the second experiment). Similarly, six wells per genotype were examined by Western blotting in two separate cell preparations of three wells per genotype. The variation between the experiments was the same as between different wells within the same experiment. Technical variation between different Western blot lanes was comparable to or larger than the variation between different biological samples.