Identification of therapeutics that target eEF1A2 and upregulate utrophin A translation in dystrophic muscles

Abstract

Up-regulation of utrophin in muscles represents a promising therapeutic strategy for the treatment of Duchenne Muscular Dystrophy. We previously demonstrated that eEF1A2 associates with the 5'UTR of utrophin A to promote IRES-dependent translation. Here, we examine whether eEF1A2 directly regulates utrophin A expression and identify via an ELISA-based high-throughput screen, FDA-approved drugs that upregulate both eEF1A2 and utrophin A. Our results show that transient overexpression of eEF1A2 in mouse muscles causes an increase in IRES-mediated translation of utrophin A. Through the assessment of our screen, we reveal 7 classes of FDA-approved drugs that increase eEF1A2 and utrophin A protein levels. Treatment of mdx mice with the 2 top leads results in multiple improvements of the dystrophic phenotype. Here, we report that IRES-mediated translation of utrophin A via eEF1A2 is a critical mechanism of regulating utrophin A expression and reveal the potential of repurposed drugs for treating DMD via this pathway.

Fig. 1. Overexpression of eEF1A2 in skeletal muscles increases endogenous utrophin A protein levels and utrophin A IRES activity.

a Representative western blot comparing the expression profile of eEF1A2 protein in fast (EDL) and slow (soleus) skeletal muscles harvested from 6 to 7-week-old wild-type (WT) and mdx mice. β-actin was used as a loading control (N = 3). b eEF1A2 (eEF1A2-pcDNA) expression construct or control (pcDNA3.1) were electroporated into TA muscles of wild-type and mdx mice as well as in TA muscles from utrophin A 5′UTR (CMV/βGAL/UtrA/CAT) reporter transgenic mice. Representative western blots of endogenous utrophin A protein expression in wild-type and mdx mice following overexpression of eEF1A2 with the respective quantification. Ponceau staining was used as a loading control. Note the increase in utrophin A protein levels in skeletal muscles overexpressing eEF1A2 (N = 3). c Relative IRES activity as determined by a ratio of CAT: β-GAL activity in TA muscles of utrophin A 5’UTR (CMV/βGAL/UtrA/CAT) transgenic mice overexpressing eEF1A2 (N = 9). On the right, representative western blots demonstrating expression levels of the myc tag containing pcDNA-eEF1A2 expression vector and eEF1A2 expression levels in TA transgenic mouse muscle. Error bars represent SEM. *P < 0.05, versus control. Two-tailed Student's t-test were performed to determine statistical differences in these experiments. Source data are provided as a Source Data file.

Fig. 2. Pharmacological activation of eEF1A2 and utrophin A by FDA-approved drugs.

a An ELISA-based high-throughput drug screen was performed by treating C2C12 cells with 262 different FDA-approved drugs or vehicle control for 24 hours. Quantification of eEF1A2 and utrophin A protein levels from the 11 FDA-approved drugs considered as leads. A drug is considered a hit based on its ability to increase eEF1A2 and utrophin A protein levels over vehicle control (N = 3). b Activation of the utrophin 5′UTR IRES reporter construct by 24-h treatment of FDA-approved drugs in C2C12 cells. The treated muscle cell samples were subjected to a reporter assay to determine CAT and β-GAL activity representative of IRES activity. CAT activation was normalized to β-GAL and each drug activation was normalized to vehicle control (N = 6). c Relative quantification of utrophin A mRNA levels in C2C12 treated cells, determined by qRT-PCR (N = 3). The values were normalized to 18S mRNA levels. Error bars represent SEM. *P < 0.05, **P < 0.01, versus vehicle control. One-way ANOVA and Bonferroni post-hoc test was performed to determine statistical differences. Source data are provided as a Source Data file.

Fig. 3. Pharmacological stimulation of utrophin A in transgenic bicistronic utrophin 5’UTR IRES harboring mice.

Transgenic mice were treated with Betaxolol (Bet) (5 mg/kg), Pravastatin (Prava) (2 mg/kg) or vehicle for 7 days. a, b Western blots and quantification of eEF1A2 and utrophin A protein levels normalized to ponceau using protein extracts from TA muscles from the treated transgenic mice. c Activation of utrophin 5′UTR IRES reporter construct after a 7-day treatment with Bet and Prava in transgenic mice. Representative western blots of CAT and β-GAL protein levels from TA muscles from the treated transgenic mice. The ratio of CAT/β-GAL level of each drug normalized to vehicle control represents the IRES activity. Recombinant CAT is used as a positive control. d Utrophin A mRNA levels in Bet-, Prava- or vehicle-treated transgenic mouse TA muscles, standardized to 18S. N = 8, error bars represent SEM. *P < 0.05, **P < 0.01, ***P < 0.001, vs vehicle control. Two-tailed Student's t-test (a, b) and one-way ANOVA, accompanied by Bonferroni post-hoc test (c, d), were performed to determine statistical differences in these experiments. Source data are provided as a Source Data file.

Fig. 4. The effects of Betaxolol and Pravastatin treatment on muscle strength of mdx mice.

Mdx mice were treated with Betaxolol (Bet), Pravastatin (Prava) or vehicle for 4 weeks. a mdx and wild-type (WT) mouse body weight (g). b Forelimb and c Hindlimb grip strength analysis of WT, vehicle-treated, Bet- or Prava-treated mdx mice normalized or not normalized to body weight. N = 6 for WT and N = 8 for mdx + veh or treated, error bars represent SEM. *P < 0.05, **P < 0.01 vs vehicle treated mdx mice. One-way ANOVA and Bonferroni post-hoc test was performed to determine statistical differences for (a–c). A split-plot ANOVA was used for force drop analysis (d). Source data are provided as a Source Data file.

Fig. 5. Increase of sarcolemmal localization and protein levels of utrophin A by Betaxolol and Pravastatin treatment in mdx mice.

a Representative examples of cross-sections obtained from TA muscles of wild-type (WT) mice and mdx mice treated with Betaxolol (Bet), Pravastatin (Prava) or vehicle control (saline) for 4 weeks, were immunostained with utrophin A (UTR-A) antibody. Scale bars, 50 mm. b Quantification of sarcolemmal utrophin A positive fibers to total muscle fibers. c, d Western blots and quantification of eEF1A2 and utrophin A protein levels normalized to ponceau using protein extracts from TA muscles from the treated mdx mice. e Utrophin A mRNA levels in Bet-, Prava- or vehicle-treated mdx mouse TA muscles, standardized to 18S. N = 6 for WT and N = 8 for mdx + veh or treated. Error bars represent SEM. *P < 0.05, **P < 0.01, vs vehicle control. One-way ANOVA and Bonferroni post-hoc test was performed to determine statistical differences. Source Data for b–e are provided in the Source Data File.

Fig. 6. Morphological features of mdx muscle fibers treated with Betaxolol and Pravastatin.

a Representative examples of cross-sections of TA muscles from wild-type (WT) and mdx mice treated with Betaxolol (Bet), Pravastatin (Prava) or with vehicle (saline) that were stained using hematoxylin and eosin. b Representative examples of cross-sections of TA muscles from WT mice and mdx mice treated with Bet, Prava or vehicle that were immunostained with goat-anti-mouse IgM Alexa 594. c Percentage of central nucleation in TA muscle fibers. d Quantification of IgM positive fibers vs total muscle fibers. N = 6 for WT and N = 8 for mdx + veh or treated. Error bars represent SEM, *P < 0.05, **P < 0.01, ***P < 0.001, significantly different from mdx vehicle control. Scale bars, 50 mm. One-way ANOVA and Bonferroni post-hoc test was performed to determine statistical differences. Source Data for c and d are provided in the Source Data File.

Fig. 7. Pravastatin does not induce a utrophin A increase in eEF1A2-null mice.

eEF1A2-null mice (wasted mice—wst) and wild-type mice (WT) were treated with Pravastatin (Prava) (2 mg/kg) or saline for 5 days. a, b Western blots of utrophin A and eEF1A2 protein levels using protein extracts from TA muscles from wild-type and wasted mice treated with pravastatin or vehicle control. c Quantification of utrophin A and eEF1A2 protein levels normalized to ponceau in treated WT and wasted mice. N = 4 for WT or wst + veh and N = 6 for WT or wst + Prava. Error bars represent SEM, *P < 0.05, significantly different from vehicle control. Two-tailed Student's t-test was performed to determine statistical differences. Source data are provided as a Source Data file.