Autophagy activation in COL6 myopathic patients by a low-protein-diet pilot trial

Abstract

A pilot clinical trial based on nutritional modulation was designed to assess the efficacy of a one-year low-protein diet in activating autophagy in skeletal muscle of patients affected by COL6/collagen VI-related myopathies. Ullrich congenital muscular dystrophy and Bethlem myopathy are rare inherited muscle disorders caused by mutations of COL6 genes and for which no cure is yet available. Studies in col6 null mice revealed that myofiber degeneration involves autophagy defects and that forced activation of autophagy results in the amelioration of muscle pathology. Seven adult patients affected by COL6 myopathies underwent a controlled low-protein diet for 12 mo and we evaluated the presence of autophagosomes and the mRNA and protein levels for BECN1/Beclin 1 and MAP1LC3B/LC3B in muscle biopsies and blood leukocytes. Safety measures were assessed, including muscle strength, motor and respiratory function, and metabolic parameters. After one y of low-protein diet, autophagic markers were increased in skeletal muscle and blood leukocytes of patients. The treatment was safe as shown by preservation of lean:fat percentage of body composition, muscle strength and function. Moreover, the decreased incidence of myofiber apoptosis indicated benefits in muscle homeostasis, and the metabolic changes pointed at improved mitochondrial function. These data provide evidence that a low-protein diet is able to activate autophagy and is safe and tolerable in patients with COL6 myopathies, pointing at autophagy activation as a potential target for therapeutic applications. In addition, our findings indicate that blood leukocytes are a promising noninvasive tool for monitoring autophagy activation in patients.

Trial registration: ClinicalTrials.gov NCT01438788.

Keywords: Bethlem myopathy; Ullrich congenital muscular dystrophy; autophagy; clinical trial; collagen VI; low-protein diet; muscular dystrophies.

Figures

Figure 1.

Analysis of autophagy in muscle biopsies of BM and UCMD patients before and after LPD trial. Muscle biopsies from 7 adult patients (P1-P7) were taken before (T0) and after 12 mo LPD (T12). (A) Western blot of MAP1LC3B (LC3B) and BECN1 protein levels in lysates of muscle biopsies at T0 and T12. GAPDH was used as a loading control. Data are representative of at least 3 technical replicates. (B) Densitometric quantification of western blots for LC3B and BECN1. LC3B-II and BECN1 protein levels were calculated as relative intensity with respect to GAPDH. Data represent the mean values (3 experiments) for each patient at T0 and T12. P = 0.018 and P = 0.028, respectively, for the comparison of LC3B-II and BECN1 mean values between T0 and T12 (Wilcoxon signed rank test for paired data). (C) Quantitative RT-PCR analysis of MAP1LC3B and BECN1 mRNA levels at T0 and T12. Data represent the mean value (2 experiments) for each patient at T0 and T12. P = 0.018 and P = 0.028, respectively, for the comparison of MAP1LC3B and BECN1 mean fold induction between T0 and T12 (Wilcoxon signed rank test for paired data). (D) Transmission electron micrographs of muscle biopsies after one-y LPD. Representative images of autophagosomes detected in muscle fibers of patient 1 (left panel) and patient 3 (right panel) containing degraded material (left panel, white arrowhead) and a degenerating mitochondrion (right panel). Scale bars: 200 nm. (E) Quantification of myofibers with autophagosomes/autolysosomes, as determined by transmission electron microscopy analysis of muscle biopsies from patients P1-P7 at T0 and T12. The graphs show the percentage of myofibers containing autophagosomes or autolysosomes from 5 different levels for each biopsy. P = 0.016, for the comparison between T0 and T12, following Wilcoxon signed rank test for paired data.

Figure 2.

TUNEL assay on muscle biopsies of BM and UCMD patients before and after LPD trial. Muscle biopsies from the 7 patients (P1-P7) were analyzed for apoptosis incidence before (T0) and after a 12-mo LPD (T12). (A) Representative micrographs of TUNEL assay on muscle biopsies of patient 2 at T0 (upper panels) and T12 (lower panels). Nuclei were stained in blue with Hoechst, TUNEL-positive nuclei are in red. Scale bar: 50 µm. (B) Quantification of TUNEL-positive nuclei per area in muscle biopsies from patients P1-P7 at T0 and T12. Mean values from 2 technical replicates of TUNEL assay experiments per condition are shown. P = 0.016, following Wilcoxon signed rank test for paired data, for the comparison between mean values at T0 and T12.

Figure 3.

Clinical parameters of the patients before and after LPD trial. (A-C) Motor function parameters in each of the 6 patients able to walk at T0 and T12. (A) Distance walked in 6 min (6MWD), measured in meters (m). (B) Timed 4-stair climb test (4 steps), measured in seconds (s). P = 0.034, following Wilcoxon signed rank test for paired data, for the comparison between mean values at T0 and T12. (C) Timed 10-min walk test (10 m), measured in seconds (s). (D) Forced vital capacity percentage (FVC, %) in each of the 7 patients. The comparison between the average FVC percentage at T0 and T12 shows a trend toward increase close to significance (P = 0.059) at T12, with respect to baseline visit (T0).

Figure 4.

Metabolic parameters of the patients before and after the LPD trial. (A) Histogram showing the respiratory quotient (RQ) calculated at T0 and T12 for each patient (P1-P7). RQ values of each patient at T0 and T12 are listed in Table S3. P = 0.031 for the comparison between T0 and T12, following Wilcoxon signed ranked test. (B) Representative 1H Carr-Purcell-Meiboom-Gill NMR spectra of serum of patient 4 at T0, T6 and T12, showing the methyl resonance of alanine (Ala), lactate (Lac) and acetate (Ac). Lac:Ala ratios in sera of all patients at T0, T6 and T12 are reported in Table 3.

Figure 5.

Autophagy activation after LPD is similar in muscles and in leukocytes of wild-type mice and col6a1 null mice. (A, B) Western blotting of MAP1LC3B/LC3B in protein lysates of tibialis anterior muscles and blood leukocytes (Leuko) from wild-type and col6a1 null (col6a1−/−) mice fed either with standard diet (SD) or low-protein diet (LPD) for 30 d. Each lane corresponds to one animal. For each condition, animals were injected with 0.4 mg/kg/d of colchicine at the end of the diet, to visualize the autophagic flux. RPS6 was used as a loading control. (C, D) Scatter plots showing linear association (dotted line) between LC3B-II relative intensity in tibialis anterior (TA) and leukocytes from western blot analysis as shown in (A and B). Relative protein levels of LC3B-II in muscles and leukocytes were quantified vs RPS6. Each point corresponds to one animal. Pearson coefficients (R) and P values are shown.

Figure 6.

Analysis of autophagy in blood leukocytes of BM and UCMD patients before and after the LPD trial. (A) Western blot of MAP1LC3B/LC3B and BECN1 levels in protein lysates of blood leukocytes at T0, T6 and T12. For patient 1, blood sample at T6 was not available. ACTB was used as a loading control. Arrowheads point at the specific band of BECN1. Data are representative of 2 technical replicates. (B) Densitometric quantification of western blot for LC3B and BECN1 in leukocyte fractions at T0 and T12. LC3B-II and BECN1 protein levels were calculated as relative intensity with respect to ACTB. Data represent mean value at T0 and T12 for each patient (2 experiments). P = 0.042 and P = 0.062, respectively for the comparison of LC3B-II and BECN1 values between T0 and T12 (Wilcoxon signed rank test for paired data).