A murine model of Barth syndrome recapitulates human cardiac and skeletal muscle phenotypes

Abstract

Barth syndrome is a mitochondrial disorder with hallmarks of cardiac and skeletal muscle weakness. It is caused by pathogenic variants in the X-linked gene tafazzin (TAZ), required for cardiolipin remodeling. Previously described germline and conditional Taz knockout models are not ideal for therapeutic development because they lack the combination of robust survival to adulthood, cardiomyopathy and skeletal muscle weakness. We characterized a cardiac and skeletal muscle-specific Taz knockout model (TazmKO) in which Cre recombinase is expressed from the muscle creatine kinase promoter (mCK-Cre). TazmKO mice survived normally. Cardiolipin composition was abnormal in both heart and skeletal muscle. TazmKO had reduced heart function by 2 months of age, and function progressively declined thereafter. Reduced treadmill endurance and diminished peak oxygen consumption were evident by 3 months of age, suggesting reduced skeletal muscle function. Electron microscopy showed abnormalities in mitochondrial structure and distribution. Overall, TazmKO mice display diminished cardiac function and exercise capacity while maintaining normal survival. This model will be useful for studying the effects of TAZ deficiency in striated muscles and for testing potential therapies for Barth syndrome.

Keywords: Barth syndrome; Cardiomyopathy; Skeletal myopathy; Tafazzin.

Fig. 1.

Taz knockout and growth of Taz^mKO and control mice. (A) Reverse transcription quantitative PCR (RT-qPCR) measurement of Taz and Gapdh mRNA levels. Unpaired two-tailed t-test. (B,C) Capillary western blot measurement of Cre, TAZ and GAPDH protein levels at postnatal day (P)7, P14, and P28. Representative blots are shown in B. Quantification is shown in C. Unpaired two-tailed t-test. (D) Serial measurement of the weight of Taz^mKO (mKO) and littermate control (WT) mice. Mixed-effects model with Sidak’s multiple comparison test. (D) Assessment of body composition by DEXA scan at 6 months of age. Unpaired two-tailed t-test. ns, not significant; *P<0.05, **P<0.01, ***P<0.001. Graphs show mean±s.d. Data points in A and C represent samples from different animals.

Fig. 2.

Systolic heart failure in Taz^mKO mice. (A,B) Echocardiographic measurements of heart systolic function [fractional shortening (%); A] and left ventricular internal diameter at end systole (LVID;s; B) were made monthly from 1 to 6 months of age. Mixed-effects analysis with repeated measures Sidak's multiple comparison test. (C) Heart weight (HW) normalized to body weight (BW) at 2 and 6 months of age. Unpaired two-tailed t-test. (D) Relative mRNA levels of the indicated cardiac stress markers, as measured by RT-qPCR from ventricular tissue at 6 months of age. Data are normalized to Gapdh. Unpaired two-tailed t-test with Holm-Sidak multiple testing correction. (E) Cardiac ATF4 protein levels. Top: representative capillary western blot of ATF4 and GAPDH. Lower and upper band are consistent with ATF4 and phosphorylated ATF4. Bottom: quantitative comparison of ATF4 normalized to GAPDH. n=5. (F) Cardiac sections stained with Picrosirus Red and Fast Green. Scale bars: 100 µm. (G) Relative mRNA levels of cardiac fibrosis markers. Unpaired two-tailed t-test. Data are presented as mean±s.d. Each point in C, D and G represents one mouse. ns, not significant; *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.

Fig. 3.

Treadmill endurance and skeletal muscle characterization of Taz^mKO mice. (A) Treadmill endurance of Taz^mKO and control mice over time. Mixed effects model with repeated measures. Comparisons between genotypes at each time point was corrected for multiple testing using Sidak's method. (B) Peak oxygen consumption rate (VO₂) in Taz^mKO and control mice aged 4 and 6 months. Unpaired two-tailed t-test. (C) Maximal respiratory exchange ratio (RER_max) in Taz^mKO and control mice aged 4 and 6 months. Unpaired two-tailed t-test. (D) Skeletal muscle weights in mice aged 2 and 6 months. Unpaired two-tailed t-test. (E) Representative Picrosirus Red/Fast Green staining of quadriceps and skeletal muscle at 6 months. Scale bars: 100 µm. Graphs show mean±s.d. Each point represents one mouse. ns, not significant; *P<0.05, ***P<0.001, ****P<0.0001.

Fig. 4.

Metabolic activity of Taz^mKO and control mice with normal cage activity. Eleven control and 13 Taz^mKO mice aged 4-5 months were individually housed, and VO₂, VCO₂, activity and food intake were monitored for 84 h. (A) Total energy over time. (B) Total energy expenditure versus mouse weight. Analysis using a generalized linear model (GLM) showed significant effects of mass and genotype, and a significant interaction between mass and genotype. (C) Respiratory exchange ratio (RER) over time. Arrows point to later-dark photoperiod with lower RER. (D) RER in dark or light photoperiods, or total. One-way ANOVA. (E,F) Food intake (E) and activity (F) in dark or light photoperiods, or total. One-way ANOVA. In A and C, heavy and light lines indicate mean and s.e.m., respectively. Arrows indicate difference in RER near the end of the dark cycle. Each point in B, D, E and F represents one mouse. White and gray shading indicate light and dark photoperiods, respectively. In B, D, E and F, the central line, box and whiskers indicate the median, 25th and 75th percentile, and 1.5× the interquartile range, respectively. NS, not significant.

Fig. 5.

Cardiolipin composition of Taz^mKO cardiac and skeletal muscle. Cardiolipin composition was measured using matrix-assisted laser desorption/ionization-time of flight mass spectrometry with an internal standard. (A-C) Total cardiolipin (CL; A), monolysocardiolipin (MLCL; B) and MLCL/CL ratio (C) of cardiac and skeletal muscle (quadriceps). (D-G) Distribution of CL (D,E) and MLCL (F,G) species in heart (D,F) and quadriceps muscle (E,G). Graphs show mean±s.d. Each point represents a separate mouse. Unpaired two-tailed t-test with false discovery rate (FDR) multiple-testing correction. FDR with values less than 0.05 are shown. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.

Fig. 6.

Taz^mKO mitochondrial morphology. (A) Representative electron micrographs of 6-month-old Taz^mKO and littermate control ventricular myocardium, quadriceps muscle (Quad) and gastrocnemius muscle (Gastroc). Dashed line boxed areas are enlarged in insets. (B) Mitochondrial cross-sectional area (CSA). Each point represents one mitochondrion. (C) Mitochondrial number density. Each point represents one field. (D) Clumping of cardiac mitochondria, scored as the percentage of mitochondria without myofiber contact. Each point represents one field. Mitochondrial clustering was not observed in skeletal muscles and was not quantified. (E) Number of mitochondria per sarcomere in skeletal muscles. Each point represents one field. Quantification in B-E was based on three images per mouse with three mice per group. B indicates median±95% c.i., while bar graphs in C and D show mean±s.d. Unpaired two-tailed t-test. (F) Representative capillary western blots of proteins from respiratory complexes I-V. (G) Quantification of F. In Quad, MTCO1 and UQCRC2 were not well resolved and were quantified together. Results were normalized to total protein, measured by intensity of Coomassie Blue staining across the entire capillary. n=5. Each point represents one mouse sample. Unpaired two-tailed t-test. ns, not significant (P>0.1); ***P<0.001, ****P<0.0001. In B-E, electron microscopy images were obtained from three separate mice.