Dichloroacetate improves mitochondrial function, physiology, and morphology in FBXL4 disease models

Pathogenic variants in the human F-box and leucine-rich repeat protein 4 (FBXL4) gene result in an autosomal recessive, multisystemic, mitochondrial disorder involving variable mitochondrial depletion and respiratory chain complex deficiencies with lactic acidemia. As no FDA-approved effective therapies for this disease exist, we sought to characterize translational C. elegans and zebrafish animal models, as well as human fibroblasts, to study FBXL4–/– disease mechanisms and identify preclinical therapeutic leads. Developmental delay, impaired fecundity and neurologic and/or muscular activity, mitochondrial dysfunction, and altered lactate metabolism were identified in fbxl-1(ok3741) C. elegans. Detailed studies of a PDHc activator, dichloroacetate (DCA), in fbxl-1(ok3741) C. elegans demonstrated its beneficial effects on fecundity, neuromotor activity, and mitochondrial function. Validation studies were performed in fbxl4sa12470 zebrafish larvae and in FBXL4–/– human fibroblasts; they showed DCA efficacy in preventing brain death, impairment of neurologic and/or muscular function, mitochondrial biochemical dysfunction, and stress-induced morphologic and ultrastructural mitochondrial defects. These data demonstrate that fbxl-1(ok3741) C. elegans and fbxl4sa12470 zebrafish provide robust translational models to study mechanisms and identify preclinical therapeutic candidates for FBXL4–/– disease. Furthermore, DCA is a lead therapeutic candidate with therapeutic benefit on diverse aspects of survival, neurologic and/or muscular function, and mitochondrial physiology that warrants rigorous clinical trial study in humans with FBXL4–/– disease.

To reduce the chance of secondary site mutants we outcrossed the worm three times with wildtype N2 as suggested in Zuryn and Jarriault, 2013 (3). fbxl4 sa12470 zebrafish were homozygous for a point mutation c.813T>A that generated a premature stop codon in fbxl4 (4)  were obtained per condition. Plates were immediately scanned sequentially twice on a flatbed scanner (Epson V800) on day 4, and brood size analysis was obtained by analyzing the pixel difference of skeletonized images. The difference image score was normalized using a "percent of control" method to derive a normalized WormScan score, as previously described (9). Morphologic and locomotion analyses in fbxl4 sa12470 zebrafish stressed with chloramphenicol (CAP) and treated with DCA. AB WT and fbxl4 sa12470 zebrafish larvae were exposed to control media (0.1% DMSO, 10 mM Tris in embryo water) (5), 5 mM DCA diluted in control media, 2.5 mM CAP, or 2.5 mM CAP together with 5 mM DCA from 2 dpf until 7 dpf. CAP and DCA concentrations used in this study were chosen after testing 1.5, 2.5 and 5 mM CAP (according to (5)) and 0.5, 5 and 10 mM DCA on zebrafish larvae (data not shown). fbxl4 sa12470 larvae incubated with 2.5 mM CAP manifest a gross disease phenotype without high mortality that would prohibit neuromotor analysis, and incubation with 5 mM DCA showed greater efficacy to rescue the disease phenotype than 0.5 or 10 mM levels. Larval morphology and neuromotor response were analyzed from 5 to 7 dpf with statistical analysis performed on data obtained at 7 dpf. Morphology was analyzed using a stereomicroscope (Olympus MVX10). Muscle performance and function were evaluated in zebrafish larvae using touch and tap-evoked response as locomotion assay involving manual touching of larvae with a probe (touch response) or tapping the culture dish with a probe (tap response) (14) (5). Statistical analysis was undertaken with 3 independent biological replicates per condition, with 15 fish studied per biological replicate to yield a total of 45 larvae for each condition.

C. elegans, zebrafish larvae, and human fibroblast cell line sample preparation for biochemical analyses.
Except where noted, worms were analyzed at the 1-day adult stage. Eggs were transferred to NGM plates spread with OP50 (controls) alone or additionally treated with 25 mM DCA. One day adult worms were collected (~1,000 worms per tube) in S. basal solution.
After washing three times with S. basal, buffer was removed, worms were flash-frozen in liquid nitrogen, and then stored at -80 °C until later thawed for biochemical analysis. Thirty zebrafish embryos at 7 dpf were harvested per condition. After washing with E3 buffer twice, zebrafish larvae were flash-frozen in liquid nitrogen, and stored at -80 °C for biochemical analysis. Chemicals. All chemicals were purchased from Sigma (St. Louis, MO) unless otherwise stated.
All drugs used in this study were dissolved in the appropriate solvent and final concentrations reached through serial dilutions.
Statistical analyses. Statistical analyses were performed using GraphPad Prism 7.04. Data were summarized by mean ± standard deviation (SD), unless otherwise noted, e.g., as mean ± standard error of mean (SEM). Student's t-test was used for two-group comparisons of all experiments except for morphology and behavior analyses in zebrafish larvae. Cochran-Mantel-Haenszel test was used for comparisons of different conditions within AB WT or mutant zebrafish larvae, and Fisher's exact test for comparisons between AB WT and mutant zebrafish larvae within the same condition. Bonferroni correction was used to account for multiple testing issue by using a more stringent p-value cutoff, depending on the number of parallel tests.