Identification of nucleoside mimetics that selectively activate the ISR kinase HRI.

A. Screening pipeline used to identify selective ISR kinase activating compounds that promote protective mitochondrial elongation. B. Structures of the top 9 ISR activating compounds identified in our nucleoside mimetic screen. C,D. ATF4-FLuc activity in HEK293 cells stably expressing ATF4-FLuc11 treated for 8 h with the indicated concentration of 0357 (C) or 3610 (D) in the absence or presence of ISRIB (200 nM). E,F. Activation of the ATF4-FLuc ISR translational reporter (green), the XBP1-RLuc UPR reporter (red), the HSE-FLuc HSR reporter (blue) or the ARE-FLuc OSR reporter (purple) stably expressed in HEK293 cells treated with the indicated concentration of 0357 (E) or 3610 (F) for 16 h. G,H. ATF4-mAPPLE fluorescence in HEK293 cells stably expressing ATF4-mAPPLE and CRISPRi depleted of the indicated ISR kinase8 treated for 8 h with 0357 (G, 20 µM) or 3610 (H, 20 µM). **p<0.01, ***p<0.005 for one-way ANOVA.

Pharmacologic HRI activation induces ISR-dependent mitochondrial elongation.

A. Representative images of MEF cells stably expressing mtGFP (MEFmtGFP)58 treated for 6 h with vehicle (veh), halofuginone (HF, 100 nM), BtdCPU (10 µM), 0357 (10 µM) or 3610 (10 µM). The inset shows a 3-fold magnification of the region indicated by the yellow box. Scale bars, 10 µm (top) and 3.33 µm (bottom) B-D. Quantification of bounding box axis, ellipsoid principal axis, and sphericity from the entire dataset of representative images shown in (A). The number of individual measurements for each condition are shown above. E. Representative images of MEFmtGFP cells treated for 6 h with vehicle (veh), thapsigargin (Tg; 0.5 µM) halofuginone (HF, 100 nM), 0357 (10 µM) or 3610 (10 µM) in the presence or absence of ISRIB (200 nM). The inset shows 3-fold magnification of the image centered on the asterisks. Scale bars, 10 µm. F-G. Quantification of bounding box axis, ellipsoid principal axis, and sphericity from the entire dataset of representative images shown in (E). The number of individual measurements for each condition are shown above. *p<0.05, ****p<0.001 for Kruskal-Wallis ANOVA. Black asterisks indicate comparison to vehicle treated cells. Red asterisks show comparisons for ISRIB co-treatment.

Stress-independent activation of ISR kinases prevents ionomycin-dependent accumulation of fragmented mitochondria.

A. Representative images of MEFmtGFP cells pre-treated for 6 h with vehicle (veh), thapsigargin (Tg, 500 nM), halofuginone (HF, 100 nM), 0357 (10 µM) or 3610 (10 µM) and then challenged with ionomycin (1 µM) for the indicated time. The inset shows 3-fold magnification of the image centered on the asterisk. Scale bars, 10 µm. B-D. Quantification of bounding box axis length, ellipsoid principal axis length, and sphericity from the entire dataset of representative images shown in (A). The black dashed line shows the mean value of vehicle-treated cells prior to ionomycin treatment. The dashed red line shows the mean value of vehicle-treated cells following 15 min treatment with ionomycin. The number of individual measurements for each condition are shown above. *p<0.05, **p<0.01, ***p<0.005, ****p<0.001 for Kruskal-Wallis ANOVA. Black asterisks show comparison with vehicle-treated cells.

Pharmacologic activation of ISR kinases rescues basal mitochondrial morphology in patient fibroblasts expressing the disease-associated D414V MFN2 variant.

A. Representative images of control human fibroblasts expressing MFN2WT treated for 6 h with vehicle (veh), halofuginone (HF, 100 nM), 3610 (10 µM), 0357 (10 µM), and/or ISRIB (200 nM). The inset shows 3-fold magnification of the image centered on the asterisks. Scale bars, 15 µm (top) and 5 µM (bottom). B-D. Quantification of bounding box axis (B), ellipsoid principal axis (C), and sphericity (D) from the entire dataset of images described in panel A. The number of individual measurements for each condition are shown above. E. Representative images of patient fibroblasts expressing MFN2D414V treated for 6 h with vehicle (veh), halofuginone (HF, 100 nM), 3610 (10 µM), 0357 (10 µM), and/or ISRIB (200 nM). The inset shows 3-fold magnification of the image centered on the asterisks. Scale bars, 15 µm (top) and 5 µM (bottom). F-H. Quantification of bounding box axis (F), ellipsoid principal axis (G), and sphericity (H) from the entire dataset of images described in panel E. The number of individual measurements for each condition are shown above. *p<0.05, ***p<0.005, ****p<0.001 for Kruskal-Wallis ANOVA. Black asterisks show comparison with vehicle-treated cells. Red asterisks show comparisons for ISRIB co-treatment.

Identification of nucleoside mimetics that selectively activate the ISR kinase HRI.

A. The ATF4-FLuc and ATF4-mAPPLE reporters containing the 5’UTR of ATF4.8,11 B. ATF4-Gluc activity, measured by luminescence, in HEK293 cells stably expressing ATF4-FLuc treated for 8 h with vehicle, thapsigargin (Tg, 0.5 µM), or oligomycin A (OA, 50 ng/mL). C. ATF4-FLuc activity, normalized to vehicle, in HEK293 cells stably expressing ATF4-FLuc treated for 8 h with the indicated dose of the indicated compound. The signals observed in veh or thapsigargin (Tg, 0.5 µM) cells is shown by the dashed lines. Error bars show SEM for n=9 replicates. D. Expression, measured by qPCR, of the ISR target genes ASNS and CHAC1 in HEK293 cells treated for 8 h with vehicle, thapsgargin (Tg; 0.5 µM), 0357 (20 µM), or 3610 (20 µM). Error bars show 95% confidence intervals. E. Expression, measured by qPCR, of the ISR target genes Asns and Chac1 in MEF cells treated for 8 h with vehicle, thapsgargin (Tg; 0.5 µM), 0357 (20 µM), or 3610 (20 µM). F. Expression, measured by qPCR, of the UPR target gene BiP, the HSR target gene HSPA1A, and the OSR target gene NQO1 in HEK293 cells treated for 8 h with vehicle, thapsgargin (Tg; 0.5 µM), 0357 (20 µM), or 3610 (20 µM). Error bars show 95% confidence intervals. *p<0.05, ***p<0.005 for one-way ANOVA.

Pharmacologic HRI activation induces ISR-dependent mitochondrial elongation.

A. Image processing and analysis workflow to quantify several parameters that define mitochondrial shape. B. Representative images of MEFmtGFP cells treated for 6 h with vehicle (veh), thapsigargin (Tg, 500 nM), or CCCP (10 µM). The inset shows a 3-fold magnification of the region indicated by the yellow box. Scale bars, 10 µm (top) and 3.33 µm (bottom) C-E. Quantification of bounding box axis, ellipsoid principal axis, and sphericity from the entire dataset of representative images shown in (B). The number of individual measurements for each condition are shown above. ****p<0.001 for Kruskal-Wallis ANOVA. Black asterisks show comparison with vehicle-treated cells.

Stress-independent activation of ISR kinases prevents ionomycin-dependent accumulation of fragmented mitochondria.

A-C. Quantifications of bounding box axis, ellipsoid principal axis, and sphericity of MEFmtGFP cells pre-treated for 6 h with vehicle and then challenged with ionomycin (1 µM) for the indicated time. Representative images are shown in Fig. 3A. The number of individual measurements for each condition are shown above. ****p<0.001 for Kruskal-Wallis ANOVA. Black asterisks show comparison with vehicle-treated cells at time 0.

Pharmacologic activation of ISR kinases rescue basal mitochondrial morphology in patient fibroblasts expressing the disease-associated D414V MFN2 variant.

A-C. Bounding box length (A), ellipsoid principal axis length (B), and sphericity (C) in control human fibroblasts expressing MFN2WT or patient fibroblasts expressing MFN2D414V. Representative images are shown in Fig. 4A,E. ****p<0.001 for Mann-Whitney t-test.