Highly specific ubiquitin-competing molecules effectively promote frataxin accumulation and partially rescue the aconitase defect in Friedreich ataxia cells

Abstract

Friedreich ataxia is an inherited neurodegenerative disease that leads to progressive disability. There is currently no effective treatment and patients die prematurely. The underlying genetic defect leads to reduced expression of the mitochondrial protein frataxin. Frataxin insufficiency causes mitochondrial dysfunction and ultimately cell death, particularly in peripheral sensory ganglia. There is an inverse correlation between the amount of residual frataxin and the severity of disease progression; therefore, therapeutic approaches aiming at increasing frataxin levels are expected to improve patients' conditions. We previously discovered that a significant amount of frataxin precursor is degraded by the ubiquitin/proteasome system before its functional mitochondrial maturation. We also provided evidence for the therapeutic potential of small molecules that increase frataxin levels by docking on the frataxin ubiquitination site, thus preventing frataxin ubiquitination and degradation. We called these compounds ubiquitin-competing molecules (UCM). By extending our search for effective UCM, we identified a set of new and more potent compounds that more efficiently promote frataxin accumulation. Here we show that these compounds directly interact with frataxin and prevent its ubiquitination. Interestingly, these UCM are not effective on the ubiquitin-resistant frataxin mutant, indicating their specific action on preventing frataxin ubiquitination. Most importantly, these compounds are able to promote frataxin accumulation and aconitase rescue in cells derived from patients, strongly supporting their therapeutic potential.

Keywords: Frataxin; Friedreich ataxia; Orphan drug development; Ubiquitin.

Fig. S1

Effect of UCM on cell viability. FRDA patients-derived lymphoblasts cell lines FRDA 798 (A) and FRDA 214 (B) were cultured in the presence of 10 μM of the indicated UCM, MG132 or DMSO alone (contr) for 5 days. Toxicity of the indicated compounds was measured with the XTT assay (Sigma-Aldrich) and expressed as relative cell viability compared to control untreated cells.

Fig. S2

UCM53 and UCM108 are more effective than the previously described compound. FRDA patients-derived lymphoblasts cell lines FRDA 798 (A) and FRDA 214 (B) or lymphoblasts derived from the corresponding unaffected carrier siblings, FRDA 241 or FRDA 215 respectively, were cultured in the presence of 10 μM of the indicated UCM, or DMSO alone (contr) for 5 days. Total cell extracts were resolved on SDS–PAGE and analyzed with anti-frataxin antibody, or anti-tubulin, as a loading control. The graphs represent the relative frataxin abundance as quantified by densitometric analysis and normalized with tubulin levels. Data represent the mean ± SEM from three different independent experiments. Tub: tubulin; mat: mature frataxin.

Fig. 1

Docking model of UCM71 to frataxin. (A) Crystal structure of frataxin (gray cartoon) with solvent accessible molecular surface around the W155 pocket. Overall structure as grey cartoon, molecular surface colored by lipophilicity (hydrophilic in magenta, lipophilic in green). (B) Putative model of the interaction between W155 pocket and the ligand UCM71 (ball and stick, CPK colors): the solvent accessible ligand surface (light blue mesh) fits perfectly with the naphthyl moiety buried by W155, P163, and S157. The phenothiazine ring recognizes the flat region formed by N146 and K147 (label not shown for clarity) and delimited by side chains of V131, S129, T149 and Q148. (C) Putative selected interactions of UCM71 with frataxin. Hydrogen bonds are formed between the hydroxyl substituent of the naphthyl group and the side chain of N146 and between the carbonyl group of the carbonyl-hydrazone scaffold and the hydroxyl of S157. These interactions induce minor rearrangements in the involved side chains of N146 (flip of terminal amide) and S157. (D) The flexibility of the carbonyl-hydrazone scaffold permits the flip of phenothiazine moiety to point its central amino group toward Q148 possibly inducing the flip of the amide group of its side chain resulting in the formation of a strong hydrogen bond.

Fig. 2

UCM increase frataxin levels. (A) To detect frataxin precursor accumulation, 293 Flp-In cells stably expressing frataxin^1 -210 were treated for 24 hrs with 10 μM of the indicated UCM or 10 μM MG132 (MG). Total cell extracts were resolved on SDS–PAGE and analyzed with anti-frataxin antibody, or anti-tubulin, as a loading control. Pre: precursor; tub: tubulin. (B) To detect mature frataxin accumulation, 293 Flp-In cells stably expressing frataxin^1 -210 were treated for 72 hrs with 10 μM of the indicated UCM. Total cell extracts were resolved on SDS–PAGE and analyzed with anti-frataxin antibody, or anti-tubulin, as a loading control. Pre: precursor, int: intermediate, mat: mature; tub: tubulin.

Fig. 3

UCM prevent frataxin ubiquitination. (A) 293 cells were transiently co-transfected with HA-Ub and frataxin^1 -210. Twenty-four hours after transfection, cells were treated with 10 μM of the indicated UCM or with DMSO alone (contr). UCM57 was used as a non-effective control molecule. Cells were harvested 48 h after transfection. Where indicated (MG), cells were also treated with 10 μM MG132 and 50 ng/ml ubiquitin-aldehyde for the last 16 h. Total cell extracts were resolved on SDS–PAGE and revealed with anti-frataxin antibody. The arrows indicate the bands corresponding to frataxin precursor (Fxn-pre) and ubiquitin-conjugated frataxin (Ub-Fxn). (B) The graph represents the relative ubiquitination levels, quantified as the densitometric ratio between ubiquitinated frataxin bands and frataxin precursor bands for each MG132-treated lanes. Data represent the mean ± SEM from five different independent experiments. P-values were calculated with Student's t-test and were statistically significant (**P < 0.01) compared to non-treated control.

Fig. 4

UCM promote frataxin precursor accumulation by preventing K147-dependent degradation. (A) 293 Flp-In cells stably expressing frataxin^1 -210 (293-frataxin) or the lysine-mutant frataxin^K147R (293-frataxin^K147R) were treated for 24 hrs with 10 μM of the indicated UCM. Proteins were resolved on SDS–PAGE and revealed with anti-frataxin antibody or anti-tubulin, as a loading control. Pre: precursor; tub: tubulin. (B) The graph represents relative frataxin precursor levels as quantified by densitometric analysis. Data represent the mean ± SEM from five different independent experiments. P-values were calculated with Student's t-test and were statistically significant (*P < 0.05; **P < 0.01).

Fig. 5

UCM interact with frataxin. Fluorescence studies of the interaction of 2 μM recombinant frataxin precursor with different concentration of the indicated UCM (ligand). The graphs represent the fractional loss of the protein fluorescence intensity (ΔF/F₀), in the presence of different concentration of ligand, versus the ratio between ligand and frataxin. The interaction with UCM53 was also analyzed in the case of frataxin precursor previously denaturated in 3 M guanidinium hydrochloride (upper left panel, black square symbols). The half-saturation binding constant, L_1/2 (μM), of frataxin precursor with UCM53, UCM108 and UCM72 is indicated in each corresponding panel.

Fig. 6

UCM promote frataxin accumulation in FRDA cells and rescue the aconitase defect. FRDA patients-derived lymphoblasts cell lines FRDA 798 (A) and FRDA 214 (B) or lymphoblasts derived from the corresponding unaffected carrier siblings, FRDA 241 or FRDA 215, respectively, were cultured in the presence of 10 μM of the indicated UCM, or DMSO alone (contr) for 5 days. Total cell extracts were resolved on SDS–PAGE and analyzed with anti-frataxin antibody, or anti-tubulin, as a loading control. The graphs represent the relative frataxin abundance as quantified by densitometric analysis and normalized with tubulin levels. Data represent the mean ± SEM from four different independent experiments. P-values were calculated with Student's t-test and were statistically significant (*P < 0.05; **P < 0.01) compared to non-treated conditions. Tub: tubulin; mat: mature frataxin. (C) Patients-derived lymphoblasts FRDA 214 were cultured with DMSO alone (contr) or in the presence of 10 μM of UCM108 for 5 days. The unaffected carrier siblings FRDA 215 lymphoblasts were cultured with DMSO alone. Total aconitase activities were measured and normalized as described in the Methods section. Data represent the mean ± SEM from four different independent experiments. P-values was calculated with Student's t-test and was statistically significant (**P < 0.01) compared to non-treated condition.