The small compound Icerguastat reduces muscle defects in oculopharyngeal muscular dystrophy through the PERK pathway of the unfolded protein response

Abstract

Oculopharyngeal muscular dystrophy (OPMD) is an autosomal dominant disease characterized by the progressive degeneration of specific muscles. OPMD is due to a mutation in the gene encoding poly(A) binding protein nuclear 1 (PABPN1) leading to a stretch of 11 to 18 alanines at N-terminus of the protein, instead of 10 alanines in the normal protein. This alanine tract extension induces the misfolding and aggregation of PABPN1 in muscle nuclei. Here, using Drosophila OPMD models, we show that the unfolded protein response (UPR) is activated in OPMD upon endoplasmic reticulum stress. Mutations in components of the PERK branch of the UPR reduce muscle degeneration and PABPN1 aggregation characteristic of the disease. We show that oral treatment of OPMD flies with Icerguastat (previously IFB-088), a Guanabenz acetate derivative that shows lower side effects, also decreases muscle degeneration and PABPN1 aggregation. Furthermore, the positive effect of Icerguastat depends on GADD34, a key component of the phosphatase complex in the PERK branch of the UPR. This study reveals a major contribution of the ER stress in OPMD pathogenesis and provides a proof-of-concept for Icerguastat interest in future pharmacological treatments of OPMD.

Keywords: Drosophila model; GADD34; Icerguastat/IFB-088; OPMD; PEK/PERK; unfolded protein response.

Figure 1.

The UPR is activated in the OPMD Drosophila model. (a) Schematic representation of UPR pathway in Drosophila where the three branches are represented. (b) Immunostaining of wild-type and OPMD (Act88F-PABPN1-17ala/+) GFP:KDEL-expressing thoracic muscles with anti-GFP antibody. DNA was revealed with DAPI. The quantification of GFP:KDEL fluorescence was performed using imageJ and represented as the CTCF (corrected total cell fluorescence) in arbitrary units. Quantifications were from 6 wild-type muscles (n = 36) and 12 OPMD muscles (n = 36), from three biological replicates. The bars represent the means with standard deviations. ^#p < 0.0001, using the unpaired-t test. Scale bars: 10 µm. (c) Quantification of UPR mRNA levels in wild-type and OPMD (Act88F-PABPN1-17ala/+) thoraxes at day 2 and day 6, using RT-qPCR. mRNA levels were normalized to sop mRNA and set to 1 in the wild-type. Means are from three to five biological replicates quantified in triplicates, error bars represent standard deviations. (d) Western blots of protein extracts from wild-type and OPMD (Act88F-PABPN1-17ala/+) thoraxes revealed with anti-eIF2α and anti-phosphorylated eIF2α. α-Tubulin was used as a loading control. Quantification of the ratio phosphorylated eIF2α/total eIF2α is shown. Means are from four biological replicates, error bars represent standard deviations. (c, d) *p < 0.05, **p < 0.01, n.s.: non-significant, using the unpaired-t test.

Figure 2.

Half dosage of UPR genes reduces OPMD wing position defects. (a) Genotypes and abbreviations used in figures 2, 3 and 5. (b) Wing position defects were scored in OPMD (Act88F-PABPN1-17ala/+) flies in the absence or presence of UPR heterozygous mutants, each day between days 2 and 6. The numbers of scored flies are indicated (n). Quantifications were from of three biological replicates. **p < 0.01, ^#p < 0.0001, using the Chi2-test. (c) Western blots of protein extracts from OPMD thoracic muscles in the absence or presence of UPR hererozygous mutants, revealed with anti-eiF2α and anti-phosphorylated eiF2α. α-Tubulin was used as a loading control. Quantification of the ratio phosphorylated eiF2α/total eiF2α is shown. Means are from four biological replicates, error bars represent standard deviations. (d) Quantification of UPR mRNA levels in OPMD thoracic muscles in the absence or presence of UPR hererozygous mutants, at day 2 and day 6, using RT-qPCR. mRNA levels were normalized to sop mRNA and set to 1 in the wild-type. Means are from three biological replicates quantified in triplicates, error bars represent standard deviations. (c,d) *p < 0.05, ***p < 0.001, ^#p < 0.0001, n.s. or no significancy indicated: non-significant, using the one-way ANOVA test.

Figure 3.

PERK pathway heterozygous mutants reduce the size of PABPN1-17ala nuclear aggregates. (a) Percentage of nuclei with aggregates in OPMD thoracic muscles (Act88F-PABPN1-17ala/+) in the absence or presence of UPR heterozygous mutants. Adult thoracic muscles were dissected at days 2, 4 and 6 and stained with anti-PABPN1 and DAPI. Nuclear aggregates were visualized and scored using both staining. n.s. = non significant, using the chi² test. (b,c) Quantification of nuclear aggregate areas. Each nuclear aggregate was delimited in a focal plan and the surface area was calculated using Image J. Mean values of the surface areas are shown in arbitrary units (b). Distribution of nuclear aggregate surface areas are shown as boxplots (c). The boxes represent 50% of the values, horizontal lines correspond to the medians (50% of the values on each side of the line), and vertical bars correspond to the range. Extreme values are in open circles. **p < 0.01, ^#p < 0.0001, no significancy indicated: non-significant, using the one-way ANOVA test. (d) Confocal images of nuclear aggregates visualized with anti-PABPN1 staining. DNA was revealed with DAPI. In OPMD individuals, nuclear aggregates are large and can occupy most of the nucleus. An example of nucleus without aggregate is shown for comparison (no aggregate). Nuclear aggregates were smaller in the presence of UPR heterozygous mutants. Nuclei are delimited with a white dotted line in the merge. Scale bars: 5 µm. (e) Western blots of protein extracts from OPMD thoracic muscles in the absence or presence of UPR heterozygous mutants revealed with anti-PABPN1 antibody. α-Tubulin was used as a loading control. Quantification of PABPN1 levels is shown. Means are from three biological replicates, error bars represent standard deviations. *p < 0.05, n.s.: non-significant, using the one-way ANOVA test.

Figure 4.

ICE reduces wing posture defects in OPMD Drosophila models. (a) Representation of GA and ICE molecules. (b–d) OPMD flies were transferred onto fresh instant Drosophila medium supplemented with ICE, GA or DMSO alone at day 2 after birth and then transferred to fresh medium with the drugs every day. Wing position defects were scored each day, between day 3 and day 6. Drug concentrations are indicated on the graphs. The numbers of scored flies are indicated (n). Quantifications were from one biological replicate in (b,c) and three biological replicates in (d). *p < 0.05, **p < 0.01, ***p < 0.001, ^#p < 0.0001, n.s.: non-significant, using the Chi2-test. The genotypes were UAS-PABPN1-17ala/+ ; Mhc-Gal4/+ in (b,c) and Act88F-PABPN1-17ala/+ in (d,e). (e) Western blots of protein extracts from thoracic muscles of OPMD flies fed with the drugs or DMSO alone, revealed with anti-eiF2α and anti-phosphorylated eiF2α. α-Tubulin was used as a loading control. Quantification of the ratio phosphorylated eiF2α/total eiF2α is shown. Means are from three biological replicates, error bars represent standard deviations. *p < 0.05, **p < 0.01, n.s.: non- significant, using the one-way ANOVA test.

Figure 5.

ICE acts on OPMD through its GADD34 target. (a,b) Wing position defects were scored for OPMD (Act88F-PABPN1-17alaR/+) and OPMD; GADD34^−/+ flies fed with drugs or DMSO alone. Flies were transferred onto fresh instant Drosophila medium supplemented with ICE, metixene or DMSO alone at day 2 after birth and then transferred to fresh medium with the drugs every day. Wing position defects were scored each day, from day 3 to day 6. Drug concentrations are indicating on the graphs. The numbers of scored flies are indicated (n). Quantifications were from three biological replicates in (a) and two biological replicates in (b). *p < 0.05, **p < 0.01, ***p < 0.001, ^#p < 0.0001, n.s.: non-significant, using the Chi2-test. (c) Immunostaining of wild-type and OPMD (Act88F-PABPN1-17ala/+) GFP:KDEL-expressing thoracic muscles with anti-GFP antibody. DNA was revealed with DAPI. OPMD flies were fed with 2 mM ICE or 2% DMSO alone. The quantification of GFP:KDEL fluorescence was performed using imageJ and represented as the CTCF (corrected total cell fluorescence) in arbitrary units. Quantifications were from 9 wild-type muscles (n = 40), 12 OPMD muscles from flies on 2% DMSO (n = 41) and 13 OPMD muscles from flies on 2 mM ICE (n = 39), from two biological replicates. The bars represent the means with standard deviations. **p < 0.01; ^#p < 0.0001, using the unpaired t-test. Scale bars: 10 µm.

Figure 6.

Oral treatment with ICE reduces PABPN1-17ala nuclear aggregates. (a) Percentage of nuclei with aggregates in OPMD thoracic muscles (Act88F-PABPN1-17ala/+). Adult flies were fed with the drugs that were provided in fresh medium every day from day 2. Adult thoracic muscles were dissected at days 3, 4 and 6 and stained with anti-PABPN1 and DAPI. Nuclear aggregates were visualized and scored using both staining. n.s.: non significant, using the Chi2-test. (b,c) Quantification of nuclear aggregate areas. Each nuclear aggregate was delimited in a focal plan and the surface area was calculated using ImageJ. Mean values of the surface areas are shown in arbitrary units (b). Distribution of nuclear aggregate surface areas are shown as boxplots (c). Legend is as in figure 3b, c. *p < 0.05, **p < 0.01, ^#p < 0.0001, no significancy indicated: non-significant, using the one-way ANOVA test. (d) Confocal images of nuclear aggregates visualized with anti-PABPN1 staining. DNA was revealed with DAPI. Examples of nuclei with or without an aggregate in OPMD individuals fed with DMSO alone. Nuclear aggregates were smaller when flies were raised on drug-supplemented medium. Examples of small aggregates at day 4 in the presence of 2 mM ICE or GA. Nuclei are delimited with a white dotted line in the merge. Scale bars: 5 µm. (e) Western blots of protein extracts from thoracic muscle of OPMD flies fed with the drugs or DMSO alone revealed with anti-PABPN1 antibody. α-Tubulin was used as a loading control. Quantification of PABPN1 levels is shown. Means are from three biological replicates, error bars represent standard deviations. n.s.: non-significant, using the one-way ANOVA test.