Activation of the sigma-1 receptor chaperone alleviates symptoms of Wolfram syndrome in preclinical models

Abstract

The Wolfram syndrome is a rare autosomal recessive disease affecting many organs with life-threatening consequences; currently, no treatment is available. The disease is caused by mutations in the WSF1 gene, coding for the protein wolframin, an endoplasmic reticulum (ER) transmembrane protein involved in contacts between ER and mitochondria termed as mitochondria-associated ER membranes (MAMs). Inherited mutations usually reduce the protein's stability, altering its homeostasis and ultimately reducing ER to mitochondria calcium ion transfer, leading to mitochondrial dysfunction and cell death. In this study, we found that activation of the sigma-1 receptor (S1R), an ER-resident protein involved in calcium ion transfer, could counteract the functional alterations of MAMs due to wolframin deficiency. The S1R agonist PRE-084 restored calcium ion transfer and mitochondrial respiration in vitro, corrected the associated increased autophagy and mitophagy, and was able to alleviate the behavioral symptoms observed in zebrafish and mouse models of the disease. Our findings provide a potential therapeutic strategy for treating Wolfram syndrome by efficiently boosting MAM function using the ligand-operated S1R chaperone. Moreover, such strategy might also be relevant for other degenerative and mitochondrial diseases involving MAM dysfunction.

Fig. 1.. Wfs1^ΔExon8 mice showed hypomobility in the circular open-field test.

Female and male Wfs1^WT (WT) and Wfs1^ΔExon8 (KO) animals were placed in the circular open field, and their behaviors were analyzed during 10 min: total locomotion (A) and time course for female (B) and male mice (C), immobility (D), walking speed (E), locomotion in the center (F), departure latency (G), and number of rearings (H) and groomings (I). n = 10 to 12 per group. Two-way ANOVAs: P > 0.05 for gender, P < 0.001 for genotype, and P > 0.05 for the interaction in (A); P > 0.05 for gender, P < 0.001 for genotype, and P > 0.05 for the interaction in (D); P < 0.05 for gender, P > 0.05 for genotype, and P > 0.05 for the interaction in (E); P > 0.05 for gender, P < 0.05 for genotype, and P > 0.05 for the interaction in (F); P > 0.05 for gender, P > 0.05 for genotype, and P > 0.05 for the interaction in (G); P > 0.05 for gender, P < 0.05 for genotype, and P > 0.05 for the interaction in (H); P < 0.05 for gender, P < 0.01 for genotype, and P < 0.01 for the interaction in (I). *P < 0.05 and **P < 0.01 versus WT mice; ##P < 0.01 versus female mice; Newman-Keuls test.

Fig. 2.. Wfs1^ΔExon8 mice showed learning and memory deficits more pronounced in males.

Wfs1^WT (WT) and Wfs1^ΔExon8 (KO) mice were tested for spontaneous alternation in the Y-maze (A and B), object recognition (C and D), passive avoidance (E to G), and place learning in the water maze (H to O). (A) Spontaneous alternation performance and (B) total number of arm entries during the 8-min session in the Y-maze. (C) Preference for the object in position #2 during session 2 with two identical objects and (D) preference for the novel object in position #2 during session 3 of the object recognition test. (E) Step-through latency (STL) during passive avoidance training and (F) step-through latency and (G) escape latency (EL) during the retention test performed 24 hours after training. Violin graphs show individual data distribution with median and interquartile range. Acquisition profiles of the location of an invisible platform, placed in the northeast quadrant of the pool, for female (H) or male (I) mice. Training during 5 days consisted of three swims per day with 15-min intertrial time interval. (J) Acquisition slopes were calculated from individual acquisition profile. The swimming speed is shown in (K). On day 7, the platform was removed, and mice were submitted to a probe test. Typical paths are shown in (L). The presence in the training quadrant was analyzed (M). On day 8, the platform was moved to the southwest quadrant and rendered visible by placing a flag on it. Acquisition was analyzed with three swims with 15-min intertrial time interval for female (N) and male mice (O). n = 8 to 10 per group in (A) and (B), 5 to 9 in (C) and (D), 12 to 18 in (E) to (G), and 11 to 14 in (H) to (O). Two-way ANOVAs: P > 0.05 for gender, P < 0.001 for genotype, and P > 0.05 for the interaction in (A); P > 0.05 for gender, P < 0.0001 for genotype, and P > 0.05 for the interaction in (B); P > 0.05 for gender, P < 0.001 for genotype, and P > 0.05 for the interaction in (J); P > 0.05 for gender, P < 0.001 for genotype, and P > 0.05 for the interaction in (K). Kruskal-Wallis ANOVAs: P < 0.05 in (E); P > 0.05 in (F); P < 0.05 in (G). Friedman nonparametric repeated-measure ANOVAs: P < 0.0001; trial 4 versus trial 1: P < 0.01, trial 5 versus trial 1: P < 0.001 for WT in (H); P < 0.05; trial 5 versus trial 1: P < 0.05 for KO in (H); P < 0.0001; trial 3 versus trial 1: P < 0.05, trial 4 versus trial 1: P < 0.001, trial 5 versus trial 1: P < 0.01 for WT in (I): P > 0.05 for KO in (I); P < 0.05; trial 3 versus trial 1: P < 0.05 for WT in (N); P < 0.01; trial 3 versus trial 1: P < 0.01 for KO in (N); P < 0.01; trial 3 versus trial 1: P < 0.01 for WT in (O); P < 0.01; trial 2 versus trial 1: P < 0.05, trial 3 versus trial 1: P < 0.05 for KO in (O). *P < 0.05, **P < 0.01, and ***P < 0.001 versus WT mice; #P < 0.05 versus female mice; Newman-Keuls test in (A), (B), (J), and (K); Mann-Whitney test in (E) to (G). °P < 0.05 and °°P < 0.01 versus 50% level in (C) and (D) or 15-s level in (M), one-sample t test.

Fig. 3.. Wfs1^ΔExon8 mice showed increased anxiety particularly in females.

Wfs1^WT (WT) and Wfs1^ΔExon8 (KO) mice were tested in the black-and-white exploration box (A to D) and in the elevated plus maze (E to H). (A) Typical video-tracked patterns are shown for WT and Wfs1 KO mice in the black-and-white box. (B) The number of crossings between the light and dark compartments, (C) time in the light compartment, and (D) time per visit in the light compartment. (E) Typical time-course presence in the open (O) or enclosed (E) arms of the elevated plus maze, (F) total number of arm entries, (G) time in the open arms, and (H) time per visit in the open arms. The number of mice per group is indicated in the columns in (A) and (D). n = 8 to 12 per group in (A) to (D) and 11 to 14 in (E) to (H). Two-way ANOVAs: P > 0.05 for gender, P > 0.05 for genotype, and P > 0.05 for the interaction in (B); P > 0.05 for gender, P < 0.05 for genotype, and P > 0.05 for the interaction in (C); P > 0.05 for gender, P < 0.05 for genotype, and P > 0.05 for the interaction in (D); P > 0.05 for gender, P > 0.05 for genotype, and P > 0.05 for the interaction in (F); P > 0.05 for gender, P < 0.05 for genotype, and P > 0.05 for the interaction in (G); P > 0.05 for gender, P < 0.05 for genotype, and P > 0.05 for the interaction in (H). *P < 0.05 versus WT mice; Newman-Keuls test.

Fig. 4.. Expression and interaction analyses of WFS1, NCS1, and S1R.

(A and B) NCS1 and (C and D) S1R protein amounts in the hippocampus (A and C) and cortex (B and D) of Wfs1^WT (WT) and Wfs1^ΔExon8 (KO) mice. (E and F) NCS1 and (G and H) WFS1 protein amounts in the hippocampus (E and G) and cortex (F and H) of WT and S1R KO mice. n = 6 per group in (A) to (D) and 7 in (E) to (H). *P < 0.05 and **P < 0.01 versus WT mice; Newman-Keuls test. (I) Coimmunoprecipitation between S1R and WFS1 or NCS1 was examined in WT or SIR-V5 overexpressed Neuro2a cells. Total lysate showed immunoreactivity for WFS1, NCS1, and the S1R-V5. Immunoprecipitation (IP) was performed with WFS1, NCS1, and IgG as a control.

Fig. 5.. The S1R agonist PRE-084 attenuated the deficits of memory in male Wfs1^ΔExon8 mice and anxiety in female Wfs1^ΔExon8 mice.

Vehicle solution (V) or PRE-084 (PRE; 0.3 mg/kg, intraperitoneally) was administered in male Wfs1^WT (WT) and Wfs1^ΔExon8 (KO) mice, 30 min before the Y-maze session (A); session 2 in the object recognition test (B) with session 3 results shown are in (C), or each training day in the water maze test, with data showing acquisition profiles (D and E) and slope (F) and the probe test performed 72 hours after the last training day (G), or the training session in the passive avoidance test, with step-through latency for retention shown in (H) and escape latency in (I). The drug was administered at the same dose in female WT and KO mice, 30 min before the black-and-white exploration box test or the elevated plus maze test. (J) The number of crossings between the two compartments, (K) duration spent in the light compartment, and (L) time per visit for the black-and-white exploration. (M) Total number of arm entries, (N) time in the open arms, and (O) time per visit in the elevated plus maze. n = 8 to 12 per group in (A), 6 to 10 in (B) and (C), 12 to 13 in (D) to (G), 10 to 14 in (H) and (I), 9 to 14 in (J) to (L), and 11 to 14 in (M) to (O). Two-way ANOVAs: P < 0.01 for treatment, P < 0.001 for genotype, and P < 0.05 for the interaction in (A); P < 0.05 for genotype, P > 0.05 for treatment, and P > 0.05 for the interaction in (F); P < 0.05 for treatment, P > 0.05 for genotype, and P > 0.05 for the interaction in (J); P < 0.01 for treatment, P > 0.05 for genotype, and P < 0.01 for the interaction in (K); P < 0.001 for treatment, P > 0.05 for genotype, and P > 0.05 for the interaction in (L); P > 0.05 for treatment, P > 0.05 for genotype, and P > 0.05 for the interaction in (M); P > 0.05 for treatment, P < 0.05 for genotype, and P > 0.05 for the interaction in (N); v > 0.05 for treatment, P > 0.05 for genotype, F_(1,47) = 1.304, and P > 0.05 for the interaction in (O). Kruskal-Wallis ANOVAs: P < 0.01 in (H); P < 0.001 in (I). Friedman nonparametric repeated-measure ANOVAs: P < 0.0001, trials 3 to 5 versus trial 1: P < 0.001 for WT/V and P < 0.05, trials 2 to 5 versus trial 1: not significant, for KO/V in (D); P < 0.0001, trials 3 and 4 versus trial 1: P < 0.01; trial 5 versus trial 1: P < 0.001 for WT/PRE and P < 0.0001, trial 3 versus trial 1: P < 0.05, trials 4 and 5 versus trial 1: P < 0.001 for KO/PRE in (E). *P < 0.05, **P < 0.01, and ***P < 0.001 versus V-treated WT mice; #P < 0.05, ##P < 0.01, and ###P < 0.001 versus V-treated KO mice; Newman-Keuls test in (A), (C), (D), (F), and (J) to (O); Mann-Whitney test in (H) and (I). °P < 0.05, °°P < 0.01, and °°°P < 0.001 versus 50% level in (C), versus 15-s level in (G); one-sample t test.

Fig. 6.. PRE-084 or overexpression of S1R attenuated the mobility deficits in wfs1ab^KO zebrafish larvae.

(A) Quantification of mRNA amounts for wfs1a, wfs1b, sigmar1, ncs1a, and ncs1b in wfs1ab^WT and wfs1ab^KO larvae by RT-PCR. (B) Typical blots and (C) quantification of protein amounts for S1r and Ncs1. (D, F, and I) Typical mobility patterns in the visual motor response test of zebrafish larvae and (E, G, and J) quantification of cumulated mobility during the OFF2 + OFF3 phases, expressed as percentage of control (V- or mCherry-treated wfs1^WT larvae). (H and K) Hypermobility ratio. (D and E) Dose-response effect of PRE-084 (0.1 to 10 μM). (F to H) Blockade of PRE-084 effect by the S1R antagonist NE-100, both drugs being administered at 3 μM. n = 10 to 12 per group in (A), 8 to 14 in (B) and (C), 24 to 36 in (D) and (E), 8 to 11 in (F) to (H), and 16 to 58 in (I) to (K). One-way ANOVAs: P < 0.0001 in (A); P < 0.05 in (G). Two-way ANOVA: P > 0.05 for S1R overexpression, P < 0.0001 for genotype, and P < 0.0001 for the interaction in (J). *P < 0.05, **P < 0.01, and ***P < 0.001 versus control wfs1ab^WT; ##P < 0.01 versus wfs1ab^KO larvae; Dunnett’s or Newman-Keuls test in (A), (C), (E), (G), and (I).

Fig. 7.. PRE-084 restored the alterations of Ca²⁺ transfer from the ER and attenuated mitochondrial respiration deficits observed in fibroblasts derived from patients with WS.

Representative traces of aequorin-based measurements of (A and B) cytosolic Ca²⁺ and (C and D) mitochondrial Ca²⁺ uptake, induced in fibroblasts from controls (Ct, black line) and patient with WS (red line). Ca²⁺ transfer was induced by stimulation with 1 μM Bradykinin and PRE-084 treatment: 0.1 μM (gray line), 1 μM (brown line), and 10 μM (blue line) on control (left) and patients with WS (right). Quantifications for (E and F) cytosolic Ca²⁺ and (G and H) mitochondrial Ca²⁺ uptake are expressed as means ± SEM of five independent experiments for each cell line. (I) Oxygen consumption rate (OCR) traces of control (Ct, black line) and patient fibroblasts (WS, red line), expressed as picomoles of O₂ per minute, under basal conditions and after the injection of oligomycin (1.5 μM), FCCP (1 μM), and ROT/AA (1 μM). PRE-084 was tested at 0.03 μM (gray line), 0.1 μM (purple line), and 1 μM (blue line). Quantifications of (J) basal, (K) ATP-related, and (L) maximal respiration rates were calculated from OCR traces and expressed as means ± SEM from five independent determinations. n = 5 to 6 per group in (A) to (H) and 4 to 6 in (I) to (L). One-way ANOVAs: P < 0.05 in (E); P < 0.01 in (F); P < 0.05 in (G); P < 0.01 in (H); P < 0.0001 in (J); P < 0.01 in (K); P < 0.05 in (L). *P < 0.05, **P < 0.01, and ***P < 0.001 versus V-treated Ct cells; #P < 0.05, ##P < 0.01, and ###P < 0.001 versus V-treated WS cells; Dunnett’s test. FCCP, carbonyl cyanide-4 (trifluoromethoxy) phenylhydrazone; ROT/AA, rotenone/antimycin A.

Fig. 8.. PRE-084 restored the autophagy and mitophagy defects observed in fibroblasts derived from patients with WS.

(A to C) Autophagy in fibroblasts of patients: (A) typical fluorescence microscopy images of fibroblasts from patients with WFS1-null mutation (KO), compared to WT cells, transfected with GFP-LC3 plasmid. (B) Quantification of LC3 vacuolated cells per field. (C) Representative immunoblot showing the increased LC3 lipidation (conversion of LC3-I to LC3-II) in fibroblasts from patients and control cells. Densitometry ratios of LC3-II over actin are shown and are relative to the control. In both experiments, the specific vacuolar H⁺-dependent adenosine triphosphatase inhibitor bafilomycin A1 (BafA1) was added at a concentration of 100 nM for 2 hours to mimic a blockage of the autophagic flux. (D and F) PRE-084 reestablished the normal autophagic rates in fibroblasts from patients with WFS1-null mutation (KO) to a value similar to those observed in WT samples, as demonstrated by fluorescence microscopy analysis: (D) typical images and (E) quantification and by immunoblot (F). In both experiments, PRE-084 was used at the concentrations of 0.03, 0.1, and 1 μM. Densitometry ratios of LC3-II over actin are shown and are relative to the control. (G and H) Mitophagy activation was analyzed by confocal microscopy with the specific mitochondrial and lysosomal markers, MitoTracker Green and LysoTracker Red, in fibroblasts from patients, with WFS1-null mutation (KO) and WT fibroblasts. (G) Representative images of the colocalization points and (H) quantification. Scale bars, 10 μm in (A), (D), and (G). n = 15 per group in (B), 11 in (E), and 7 to 8 in (H). (B) Two-way ANOVA: P < 0.0001 for the genotype, P < 0.0001 for the BafA1 treatment, and P < 0.05 for the interaction in (B). One-way ANOVAs: P < 0.0001 in (E); F_(4,32) = 20.21 and P < 0.0001 in (H). *P < 0.05 versus WT cells; Dunnett’s test.