Therapeutic potential of deuterium-stabilized (R)-pioglitazone-PXL065-for X-linked adrenoleukodystrophy

Abstract

X-linked adrenoleukodystrophy (ALD) results from ABCD1 gene mutations which impair Very Long Chain Fatty Acids (VLCFA; C26:0 and C24:0) peroxisomal import and β-oxidation, leading to accumulation in plasma and tissues. Excess VLCFA drives impaired cellular functions (e.g. disrupted mitochondrial function), inflammation, and neurodegeneration. Major disease phenotypes include: adrenomyeloneuropathy (AMN), progressive spinal cord axonal degeneration, and cerebral ALD (C-ALD), inflammatory white matter demyelination and degeneration. No pharmacological treatment is available to-date for ALD. Pioglitazone, an anti-diabetic thiazolidinedione, exerts potential benefits in ALD models. Its mechanisms are genomic (PPARγ agonism) and nongenomic (mitochondrial pyruvate carrier-MPC, long-chain acyl-CoA synthetase 4-ACSL4, inhibition). However, its use is limited by PPARγ-driven side effects (e.g. weight gain, edema). PXL065 is a clinical-stage deuterium-stabilized (R)-enantiomer of pioglitazone which lacks PPARγ agonism but retains MPC activity. Here, we show that incubation of ALD patient-derived cells (both AMN and C-ALD) and glial cells from Abcd1-null mice with PXL065 resulted in: normalization of elevated VLCFA, improved mitochondrial function, and attenuated indices of inflammation. Compensatory peroxisomal transporter gene expression was also induced. Additionally, chronic treatment of Abcd1-null mice lowered VLCFA in plasma, brain and spinal cord and improved both neural histology (sciatic nerve) and neurobehavioral test performance. Several in vivo effects of PXL065 exceeded those achieved with pioglitazone. PXL065 was confirmed to lack PPARγ agonism but retained ACSL4 activity of pioglitazone. PXL065 has novel actions and mechanisms and exhibits a range of potential benefits in ALD models; further testing of this molecule in ALD patients is warranted.

FIGURE 1

PXL065 and Pioglitazone suppress very long chain fatty acids (VLCFA) levels in fibroblasts derived from two adrenoleukodystrophy (ALD)—C‐ALD and AMN—patients. (A) Dose–response effect of PXL065, pioglitazone on C26:0 levels measured by mass spectrometry in AMN fibroblasts following incubation with compounds at 0, 0.1, 0.5, 1, 2, 3.5, and 5 μM for 7 days. IC50s were calculated using nonlinear regression analysis. Results are mean ± SEM, n = 5 replicates. (B) Effect of PXL065 and pioglitazone on VLCFA levels measured by mass spectrometry in AMN and C‐ALD fibroblasts, following incubation at 10 μM for 7 days. Results are mean ± SEM, n = 3 replicates/condition/patient. Beneficial effects of the drugs on VLCFA levels were reproduced in fibroblasts derived from these donors in two other independent experiments. ** p < 0.01, *** p < 0.001, **** p < 0.0001 by One‐way analysis of variance (ANOVA) followed by Dunnett's multiple comparison versus untreated AMN/C‐ALD cells

FIGURE 2

PXL065 improves mitochondrial function and increases compensatory transporters, ABCD2 and ABCD3, mRNA levels in fibroblasts derived from two ALD—C‐ALD and AMN—patients. AMN (A) and C‐ALD (B) fibroblasts were exposed for 72 h to PXL065 or Pioglitazone at 10 μM. Bioenergetics analysis was performed using a Seahorse Analyzer and parameters were evaluated by sequential additions of: oligomycin (Oligo—1 μM), FCCP (0.25 μM) and Rotenone‐Antimycin A (Rot‐AA—1 μM). Basal is first three measurements, ATP‐linked is oxygen consumption rate (OCR) drop following oligo addition, Maximal Oxidative Capacity (MOC) is OCR following addition of FCCP. Results are mean ± SEM, n = 6 replicates/condition/patient. Cells were exposed for 72 h to PXL065 or pioglitazone at 10 μM prior to mRNA level analysis by RT‐qPCR in AMN (C) and C‐ALD fibroblasts (D). Results are mean ± SEM, n = 3–6 replicates/condition/patient. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001 by One‐way analysis of variance followed by Dunnett's multiple comparison versus untreated AMN/C‐ALD cells

FIGURE 3

PXL065 improves adrenoleukodystrophy (ALD) pathology in glial cells from Abcd1‐null mice. Cells were exposed to 10 μM PXL065, pioglitazone, or control media. (A) Very long chain fatty acids (C26:0) quantitative analysis performed by mass spectrometry after 7 days exposure. (B) Bioenergetics analysis performed using a Seahorse Analyzer after 72‐h incubation; parameters were evaluated by sequential additions of: oligomycin (Oligo), FCCP and Rotenone‐Antimycin A. Basal is first three measurements, ATP‐linked is oxygen consumption rate (OCR) drop following oligo addition, Maximal Oxidative Capacity (MOC) is OCR following addition of FCCP. (C) Compensatory transporter gene expression measured by RT‐qPCR (normalized by RLP27 expression) after 72‐h incubation. Results are mean ± SEM, n = 2–6 replicates/condition. ** p < 0.01, *** p < 0.001, **** p < 0.0001 by one‐way analysis of variance followed by Dunnett's multiple comparison versus untreated glial cells

FIGURE 4

PXL065 represses pro‐inflammatory gene expression in C‐ALD patient‐derived lymphocytes and in glial cells from Abcd1‐null mice. C‐ALD lymphocytes (A) were exposed for 72 h to PXL065 or Pioglitazone at 10 μM prior to mRNA level analysis by RT‐qPCR. Abcd1 null glial cells (B) were incubated with PXL065 or Pioglitazone at 10 μM for 2 h prior to stimulation by TNFα and IL1β for 70 h (total drug exposure of 72 h), followed by mRNA level analysis by RT‐qPCR. Results are mean ± SEM, n = 3 replicates/condition. Results were normalized by RLP27 expression. ** p < 0.01, *** p < 0.001, **** p < 0.0001 by One‐way ANOVA followed by Dunnett's multiple comparison vs untreated C‐ALD or Abcd1 null gial cells.

FIGURE 5

PXL065 reduces C26:0 levels in plasma, brain and spinal cord. Six‐8‐week‐old Abcd1‐null mice were treated with PXL065 or pioglitazone (15 mg/kg QD) for 8 weeks. C26:0 content was measured in plasma (A), brain (B) and spinal cord (C) by mass spectrometry. Results are mean ± SEM, n = 12 wild‐type, 15 untreated Abcd1‐null, 15 PXL065 treated Abcd1‐ null and 15 pioglitazone treated Abcd1‐null mice. * p < 0.05, ** p < 0.01, **** p < 0.0001 by one‐way analysis of variance followed by Dunnett's multiple comparison or by Kruskal–Wallis followed by Dunn's multiple comparison versus untreated Abcd1‐null mice

FIGURE 6

In older Abcd1‐null mice, PXL065 lowers spinal cord very long chain fatty acids, improves sciatic nerve morphology and enhances neurobehavioral functions. Older (age 13 months) Abcd1‐null mice were treated for 12 weeks with PXL065 or pioglitazone (15 mg/kg QD). (A) C26:0 content measured in spinal cord by mass spectrometry; results are mean ± SEM, n = 8 animals/condition. (B) Open‐field test monitoring results. Data are mean ± SEM, n = 8 animals/condition (seven animals for wild‐type in graph presenting number of rearings). (C) Axonal morphology (with representative images) of neurons determined by morphometric analysis of transversal slices of the sciatic nerve by electronic microscopy (800×). Data are mean ± SEM, n = 4 animals/condition. * p < 0.05, ** p < 0.01, **** p < 0.0001 by one‐way analysis of variance followed by Dunnett's multiple comparison or by Kruskal–Wallis followed by Dunn's multiple comparison vs untreated Abcd1‐null mice

FIGURE 7

PXL065 retains nongenomic actions—inhibition of ACSL4—without PPARγ agonist activity. Enzymatic activities were measured by ex situ biochemical assays. PPARγ agonist activity (A) was measured using recombinant enzyme with a cofactor recruitment fluorescence assay using recombinant ligand binding domain protein in the presence of increasing concentrations of PXL065 or pioglitazone from 0.03 to 100 μM. Results are expressed as a percent of the control response to 10 μM rosiglitazone. ACSL4 (B) and ACSL1 activity (C) assays were performed using recombinant enzymes and [14C] palmitate, in the presence of increasing concentrations of PXL065, pioglitazone and leriglitazone from 0.5 to 6 μM. Results are mean ± SEM, n = 3 experiments for ACSL and 2 experiments for PPARγ.

FIGURE 8

Diagram depicting the effects of PXL065 and Pioglitazone on PPARγ, the mitochondrial pyruvate carrier (MPC) and ACSL4, and the associated pathways. Potential mechanisms linking the targets of the compounds of interest to very long chain fatty acids lowering and their efficacy on the disease phenotype are represented in the blue tags.