The anti-leprosy drug clofazimine reduces polyQ toxicity through activation of PPARγ

Abstract

Background: PolyQ diseases are autosomal dominant neurodegenerative disorders caused by the expansion of CAG repeats. While of slow progression, these diseases are ultimately fatal and lack effective therapies.

Methods: A high-throughput chemical screen was conducted to identify drugs that lower the toxicity of a protein containing the first exon of Huntington's disease (HD) protein huntingtin (HTT) harbouring 94 glutamines (Htt-Q₉₄). Candidate drugs were tested in a wide range of in vitro and in vivo models of polyQ toxicity.

Findings: The chemical screen identified the anti-leprosy drug clofazimine as a hit, which was subsequently validated in several in vitro models. Computational analyses of transcriptional signatures revealed that the effect of clofazimine was due to the stimulation of mitochondrial biogenesis by peroxisome proliferator-activated receptor gamma (PPARγ). In agreement with this, clofazimine rescued mitochondrial dysfunction triggered by Htt-Q₉₄ expression. Importantly, clofazimine also limited polyQ toxicity in developing zebrafish and neuron-specific worm models of polyQ disease.

Interpretation: Our results support the potential of repurposing the antimicrobial drug clofazimine for the treatment of polyQ diseases.

Funding: A full list of funding sources can be found in the acknowledgments section.

Keywords: Chemical screening; Huntington's disease; Mitochondria; PPARγ; polyQ.

Fig. 1

A chemical screen to identify modifiers of polyQ toxicity. (a) Scheme of the Htt-Q₉₄ fusion protein expressed in U2OS^Q94 cells, which contains the first exon of human HTT, including the first 17 amino acids (N17), a stretch of 94 glutamines (Q94) and the proline-rich domain (PRD), fused to EGFP. (b) Representative image of Htt-Q₉₄ expression (monitored by EGFP, green) in U2OS^Q94 cells treated or not with dox (50 ng/ml) for 8 days. Hoechst (blue) was used to stain DNA and enable the quantification of nuclei. (c) High-Throughput Microscopy (HTM)-mediated quantification of nuclei numbers from the data presented in (a). The screening was only done once (n = 1), and at least 5000 nuclei were counter per condition. Error bars indicate SD. ∗∗∗p < 0.001, t-test. (d) Pipeline of the chemical screen. On day 0, U2OS^Q94 cells were seeded on 384-well plates. On the following day, cells were treated with dox (50 ng/ml) and with the compounds from the library at 1 μM. Nuclei numbers were quantified by HTM on day 9. Scattered controls of cells not treated with dox, or only treated with dox but without additional compounds were used for normalization. (e) Hit distribution of the screen described in (d). Compounds that led to an increase in nuclei numbers higher than 3SD when compared to the numbers founds on wells only treated with dox were taken for secondary validation (Fig. S1).

Fig. 2

Clofazimine and troglitazone alleviate polyQ toxicity in vitro. (a) Representative images from clonogenic survival assays performed in U2OS^Q94 cells, treated or not with dox (50 ng/ml) and the indicated drugs at 2 μM for 12 days. The full dose–response dataset from the clonogenic assays is available at (Fig. S2). (b) Representative images of Htt-Q₉₄ expression (detected by the EGFP signal, green) in U2OS^Q94 cells, treated or not with dox (50 ng/ml) and the indicated drugs at 2 μM for 8 days. Nuclei were stained with Hoechst (blue). Scale bar (white), represents 2 μm. (c) WB analysis of Htt-Q₉₄ expression levels, monitored both with an anti-EGFP antibody or an antibody against polyQ peptides, in the experiment defined in (b). Vinculin levels were assessed as a loading control. (d) Scheme of the competition assay using KBM7 cells expressing either mCherry (red, control) or a fusion protein between Htt-Q₉₄ and EGFP (green). When co-cultured, the percentage of Htt-Q₉₄ progressively declines. (e) Data from the KBM7 competition experiment defined in (d), in cultures treated with DMSO (control), TZD or CFZ (at 5 μM). Error bars indicate SD (n = 3).

Fig. 3

Clofazimine restores mitochondrial function in polyQ-expressing cells. (a) Volcano plot representing RNA-seq data illustrating the impact of CFZ treatment (5 μM; 8 days) in dox-induced U2OS^Q94 cells. Genes above dotted line are differentially regulated (p < 0.05). Blue dots highlight mitochondria-related genes. (b) GSEA analyses from the experiment defined in A, illustrating the overall increase in mitochondria-related pathways upon CFZ treatment in dox-induced U2OS^Q94 cells. (c) Representative images the JC-1 signal (red) in U2OS^Q94 cells treated or not with dox (50 ng/ml) and CFZ (5 μM). Nuclei were stained with Hoechst (blue). Scale bar (white), represents 5 μm. (d) HTM-dependent quantification of mean JC-1 signal per cell from the experiment defined in (c). Error bars indicate SD and black lines indicate media values (n = 3). ∗∗p < 0.01, n.s., non-significant, t-test. (e) Representative images from transmission electron microscopy of U2OS^Q94 cells treated or not with dox (50 ng/ml) and CFZ (5 μM). Arrows indicate mitochondria, which are significantly altered upon expression, and improved upon a concomitant treatment with CFZ. Scale bar (white), represents 0, 5 μm.

Fig. 4

Clofazimine activates PPARγ. (a) The transcriptional signature of CFZ-treated cells was used as input to search for drugs exerting a similar transcriptional signature at the Connectivity Map database from the Broad Institute at MIT. The panel indicates an enrichment of PPARγ agonists among the drugs showing a transcriptional signature resembling that of CFZ. (b) Molecular docking illustrating the fitting of CFZ (yellow) in an allosteric pocket of PPARγ (red). The interaction occurs through hydrophobic forces and the formation of a hydrogen bond with gln-470 (length of the bond, 4.0 Å). CFZ has hydrophobic interactions with tyr-473, val-450, gln-454, ile-456, lys-457, met-463, ser-464 and leu-465. (c) Binding affinities of CFZ and several PPARγ agonists towards PPARγ, based on the molecular docking experiment shown in (b). (d) Cellular thermal shift assay (CETSA) measuring the effects of TZD and CFZ on PPARγ levels at increasing temperatures. Both compounds increased the thermal stability of PPARγ when compared to the DMSO control. (e) Quantification of the CETSA studies shown in (d). Error bars indicate SD.

Fig. 5

Effect of CFZ in neurons in vitro. (a) Representative images of SH-SY5Y^Q94 cells differentiated with RA (10 μM, 5 days), and subsequently treated with dox (35 ng/ml) with or without CFZ (1 μM) for 3 additional days. Levels of Htt-Q₉₄ (measured by the CFP signal), TUBB3 (yellow) and MitoTracker (red) are shown. Hoechst (blue) was used to stain DNA and detect nuclei. An image of the entire well for this dataset, as well as the quantification of cell numbers is shown in Fig. S3. Scale bar (white), represents 15 μm. (b) HTM-dependent quantification of the cytoplasmic MitoTracker signal per cell from the experiment defined in (a). Error bars indicate SD and dashed lines indicate median values (n = 3). ∗p < 0.05, ∗∗∗p < 0.001, t-test. (c) Percentage of transdifferentiated neurons obtained from adult human dermal fibroblasts through direct reprogramming, in the presence or absence of the indicated doses of CFZ. Each dot represents the average value for an individual donor cell line (n = 7 CTRL lines, 84 wells analyzed in total). (d) Normalized neurite counts (% of control) in induced neurons treated with CFZ (n = 7 CTRL lines, 84 wells analyzed in total). (e) Normalized Mitotracker signal (% of control) in the neurites of induced neurons, in the presence of CFZ. Each dot represents the average values of individual donors. (n = 7 CTRL lines, 42 wells were analyzed in total) ∗∗p < 0.01, One-way ANOVA. All data are shown as mean ± SEM.

Fig. 6

CFZ rescues polyQ toxicity in worms and developing zebrafish. (a) Scheme illustrating the pipeline followed to evaluate Htt-Q₉₄ in developing zebrafish. Viability was monitored 5 days after microinjection with an Htt-Q₉₄-CFP expressing plasmid. (b) Representative images of zebrafish embryos 24 h after microinjection of the Htt-Q₉₄-CFP expressing plasmid or DMSO. Note the accumulation of dead embryos (black asterisk) upon Htt-Q₉₄-CFP expression, which was significantly rescued by CFZ (12.5 μM). Scalebar (white) represents 1 mm. (c) Quantification from the experiment defined in (a,b). Error bars indicate SC (n = 3). ∗p < 0.05, ∗∗p < 0.01, t-test. (d) Scheme illustrating the pipeline used to evaluate the effect of CFZ in a worm model of polyQ toxicity. L1 larvae Q97 worms, presenting pan-neuronal expression of Q67-YFP, were grown in the presence of DMSO or 5 μM CFZ. When animals reached the adult stage (1d), their motility was quantified by measuring the number of bends per 30 s. (e) Quantification of data from (d). The experiment was done in triplicate, and a representative one is shown. Dashed lines indicate median values. ∗∗∗∗p < 0.0001, t-test.