Skeletal muscle carnitine-acylcarnitine translocase deletion reveals vulnerability of oxidative muscle to fatty acid oxidation deficiency

Abstract

Fatty acid oxidation (FAO) is a critical bioenergetic source for skeletal muscle, with FAO impairments being linked to metabolic and contractile dysfunction. FAO is regulated by the carnitine shuttle, in which FAO-eligible fatty acids, in the form of acylcarnitines, are transported into the mitochondrial matrix by carnitine-acylcarnitine translocase (CACT); however, the role of CACT in muscle in vivo has remained unexplored. To determine the requirement of CACT in muscle FAO and its influence on muscle mitochondrial bioenergetics, lipid profile, and muscle contractility, a novel conditional skeletal muscle-specific CACT knockout mouse (Cact^Sk−/−) was generated. The requirement of CACT for long-chain FAO was confirmed by the complete abrogation of FAO flux in Cact^Sk−/− muscle mitochondria. CACT was also required for the oxidative flux of medium-chain octanoyl-carnitine and acetyl-carnitine. CACT loss disrupted the lipid profile of skeletal muscle, with long-chain acylcarnitine accumulation and a shift in the saturation profile of phospholipids away from saturated and highly unsaturated and toward di- and tri-saturated phospholipids. Elevated mitochondrial content was demonstrated by increased phospholipid content and mitochondrial staining in Cact^Sk−/− muscles, occurring to a greater extent in oxidative muscles. Loss of CACT reduced muscle-specific force production by ~70% in oxidative soleus muscle despite increased fiber size and compensatory mitochondrial accumulation that preserved muscle metabolic capacity. These findings demonstrate the crucial role of CACT in muscle FAO and show that oxidative muscles, in particular, undergo extensive lipid compositional, metabolic, and structural remodeling that coincides with impaired contractile function.

NEW & NOTEWORTHY

Novel model of CACT loss in muscle to demonstrate that CACT is required for long, medium, and short-chain fatty acid oxidation in skeletal muscle. Deletion of CACT drives oxidative and glycolytic muscle lipid remodeling and impairs oxidative, but not glycolytic, muscle contractile capacity. Oxidative soleus muscle mitochondria compensate for CACT loss by increasing mitochondrial content and muscle capacity for oxidation of nonfatty acid substrates.

INTRODUCTION

Skeletal muscle is a critical contributor to whole body physiology, enabling locomotion and serving as a major regulator of energy and glucose homeostasis. Its metabolic and functional capacity are tightly linked to mitochondrial oxidative capacity. Reflecting this, reduction in mitochondrial oxidative capacity, particularly fatty acid oxidation (FAO), is implicated as a common feature of many skeletal muscle physiological [e.g., aging (1, 2)] and pathological [e.g., obesity and type 2 diabetes (3, 4)] conditions. Mitochondrial FAO defects not only limit fatty acid-mediated energy production but also contribute to intramuscular lipid accumulation, which is considered causal in downstream aberrations in metabolism (e.g., insulin resistance) and function (e.g., contractile impairment and exercise intolerance) (5). Importantly, although the loss of mitochondrial FAO capacity is viewed as a uniform defect across skeletal muscle, it likely intersects with the distinct metabolic and functional features of different skeletal muscle fiber types; however, the distinct and potentially selective vulnerabilities of individual muscle types to FAO defects remain less well characterized.

Muscle mitochondrial long-chain FAO requires the carnitine shuttle, which is mediated by the sequential action of carnitine palmitoyltransferase 1 (CPT1), generating acylcarnitine from acyl-CoA at the outer mitochondrial membrane, followed by transport of acylcarnitine into the mitochondrial matrix by carnitine-acylcarnitine translocase (CACT), and the retroconversion of acylcarnitine to acyl-CoA by CPT2 for subsequent beta-oxidation, producing reducing equivalents and acetyl-CoA for entry into the electron transport chain and citric acid cycle, respectively (6). Here, we sought to determine how deficits in skeletal muscle mitochondrial FAO affect muscle bioenergetics, structure, and contractile function using a novel conditional CACT skeletal muscle deletion mouse model, Cact^Sk−/−. In Cact^Sk−/− mixed muscles, mitochondrial oxidation of long-, medium- and short-chain acylcarnitines was completely lost, demonstrating the critical importance of CACT for transporting fatty acids across a range of lengths into the mitochondrial matrix for subsequent beta-oxidation. The consequences of CACT loss in muscle demonstrated that the oxidative soleus muscle was uniquely impacted, compared with more mixed/glycolytic muscles, with a higher degree of mitochondrial accumulation, mitochondrial metabolic remodeling, myofiber enlargement, and severe impairments in contractility. These findings indicate that CACT is essential for muscle mitochondrial FAO across chain lengths and that oxidative muscles are particularly vulnerable to FAO deficits.

METHODS

Animal Model

Slc25a20^{tm1a(EUCOMM)Wtsi} transgenic mouse strain, EM 07028 on the C57BL/6N-Atm1Brd background, was produced for European mouse mutant archive by the Wellcome Trust Sanger Institute (7–9); sperm was cryopreserved and shipped to Charles River Laboratories for rederivation. The transgenic Slc25a20, i.e., Cact, targeting cassette flanks exon 2 with loxP sites. A conditional-ready (floxed) allele was created by breeding with flpE recombinase-expressing mice (The Jackson Laboratory strain 005703 on C57BL/6J background). Resulting floxed for deletion Cact mice were bred with mice expressing Cre recombinase driven by the human alpha-skeletal actin promoter (The Jackson Laboratory strain 006149 on C57BL/6J background) to drive exon 2 excision, a translational reading frame shift, and CACT loss specifically in striated muscle cells. Floxed mice with (Cact^Sk−/−) and without (control) Cre were used herein, a mix of C57BL/6J and C57BL/6N. All mice had access to water and standard chow (PicoLab 5053, LabDiet) and were housed in a pathogen-free housing under 12-h light-dark cycle. Cact^Sk−/− were born at expected Mendelian ratios, were successful breeders, and displayed no overt phenotypes up to 1.5 yr of age, albeit were often leaner than control littermates, especially with age. All tissues were collected from mice in the fed state. All data are presented across an average age of 8 mo, ranging from 3 to 14 mo of age. No sex differences were detected, and data from both sexes were combined approximately equally, unless otherwise noted. All procedures were performed in adult male and female mice, with prior approval from the Institutional Animal Care and Use Committee at Wake Forest University (Assurance A3391-01) and East Carolina University (Assurance A3469-01).

Additional methodological details pertaining to tissue processing for western blotting, histology, gene expression, mitochondrial metabolism, lipidomics, metabolomics, and muscle contraction are provided in the Supplemental Methods and have been previously described (10–12).

Statistics

Data are presented as means ± SE. Statistical analysis and figures were generated using GraphPad Prism v. 8.0.0 for Windows (GraphPad Software). Data were compared using unpaired two-tailed Student’s t test. Significance level was set a priori at P ≤ 0.05.

RESULTS

CACT Deficiency Prevents Mitochondrial FAO of Long-, Medium-, and Short-Chain Acetylcarnitines in Mixed Skeletal Muscle

Skeletal muscle-specific loss of CACT protein was verified in tibialis anterior (TA), extensor digital longus (EDL), and soleus muscles, with preserved expression in liver and heart (Fig. 1A). To confirm the loss of CACT’s impact on preventing mitochondrial FAO, assessment of isolated mitochondrial oxygen consumption (JO₂) rates for long-chain acylcarnitine was determined using a modified CK clamp protocol, which allows evaluation of substrate-specific respiration across a physiologically relevant range of energetic demand (ΔG_ATP) (e.g., ranging from rest to contraction). Cact^Sk−/− mixed muscle mitochondria were unable to oxidize the long-chain acylcarnitine, oleoyl(C18:1n9)-carnitine (Fig. 1B), confirming the requirement of CACT for long-chain acylcarnitine transport across the inner mitochondrial membrane and subsequent oxidation. Although CACT is known to be important for long-chain FAO, loss of CACT also completely prevented the oxidative flux of medium-chain octanoyl(C8:0)-carnitine and acetyl(C2)-carnitine across all energetic states, suggesting a critical role of CACT for transport and catabolism of all acylcarnitines, independent of length. For nonfatty acid metabolic substrates, pyruvate, glutamate, and a combined multisubstrate condition, mitochondrial metabolic capacity was unchanged between genotypes (Fig. 1C). These data suggest that despite the marked impairment in fatty acid-supported respiration, overall mitochondrial bioenergetic capacity in mixed muscle remained intact. Consistent with these findings, metabolomic analyses revealed minimal genotype-dependent differences in the abundance of glycolytic intermediates, tricarboxylic acid cycle metabolites, redox cofactors, adenine nucleotides, and amino acids (Fig. 1D, Supplemental Fig. S1A). Collectively, these data indicate that CACT deletion selectively disrupts FAO while preserving global energetic homeostasis in mixed skeletal muscle. Importantly, these data demonstrate the requirement of CACT for the oxidative metabolism of acylcarnitines of long-, medium-, and two-carbon in length.

Figure 1.

CACT loss ablates muscle FAO but preserves overall mitochondrial bioenergetics. A: CACT Western blot in TA, soleus, and EDL muscle, and heart and liver from control and Cact^Sk−/− mice, representative of n = 3 mice. Oxygen consumption (JO₂) of isolated mitochondria from mixed muscles energized with acylcarnitines (C18:1, C8, C2) (B) or pyruvate, glutamate, or multiple substrates across energy demands (ΔG_ATP) (n = 3–9 mice/group) (C). D: metabolites in Cact^f/f and Cact^Sk−/− gastrocnemius muscle (n = 5–6 mice/group). *P ≤ 0.05 by two-tailed t test; means ± SE. CACT, carnitine-acylcarnitine translocase; FAO, fatty acid oxidation.

Muscle Acylcarnitine Accumulation and Phospholipid Remodeling Due to CACT Loss

Muscle acylcarnitines were measured to determine if acylcarnitine exclusion from mitochondrial import, due to CACT loss, caused extramitochondrial acylcarnitine accumulation or if accumulation was thwarted by cellular acylcarnitine export. Acylcarnitine assessment revealed marked 3.3–12.6-fold accumulation of long-chain acylcarnitines across Cact^Sk−/− muscles (Fig. 2A, Supplemental Fig. S2A and Supplemental Tables S1 and S2). Although glycolytic, mixed, and oxidative muscles accumulated C18:0, C18:1, and C18:2, only the oxidative soleus muscle accumulated C16:0 (Fig. 2A), suggesting either selective generation or blocked efflux of palmitoyl-carnitine within oxidative muscle. Free carnitine is an antiporter with acylcarnitines by CACT; thus, loss of CACT may result in imbalances in free carnitine levels. Interestingly, free carnitine was significantly reduced in EDL, unchanged in gastrocnemius, and elevated in soleus, suggesting that free carnitine balance is likely regulated by additional transporters in a muscle type-specific manner (Supplemental Fig. S2B). In agreement with lower rates of acetylcarnitine oxidative flux in Cact^Sk−/− muscle (Fig. 1C), we observed accumulation of acetylcarnitine in soleus muscle and trending accumulation in EDL, as well as accumulation of short-chain acylcarnitines in EDL, gastrocnemius, and soleus (Supplemental Fig. S2D). Together, these data reveal that loss of CACT results in acylcarnitine accumulation, ranging from long to short due to the broad role of CACT in FAO across chain lengths, and demonstrate that export mechanisms to prevent extramitochondrial acylcarnitine accumulation are limiting.

Figure 2.

CACT loss induced acylcarnitine accumulation and membrane composition change across muscles. A: percent total ion count (% TIC) of long-chain acylcarnitines in EDL, gastrocnemius, and soleus muscle (n = 5 or 6 mice/group). Percent TIC for phosphatidylcholines (PC) species (B) and relative abundance of total PC and phosphatidylethanolamines (PE) (n = 5 or 6 mice/group) (C). D: Sudan Black stain, images taken at ×20 (n = 3 mice/muscle/group). *P ≤ 0.05 by two-tailed t test; means ± SE. CACT, carnitine-acylcarnitine translocase.

To assess if excess long-chain acylcarnitines may be metabolically recycled toward complex lipid synthesis, the composition and total content of major complex lipid classes within each muscle were assessed. Analysis of the acyl chain composition profiles of the most abundant phospholipid, phosphatidylcholine (PC), demonstrated a pronounced genotype-driven shift in saturation, primarily in EDL and soleus, with limited differences in mixed gastrocnemius muscle (Fig. 2B). Specifically, PCs containing 0, 1, or 4 or more unsaturations of the acyl chains were reduced in Cact^Sk−/−, whereas species containing 2 or 3 unsaturations were highly enriched in both EDL and soleus (Fig. 2B). Given that acylcarnitines with 0, 1, and 2 unsaturations accumulated (Fig. 2A), the lack of a cohesive accumulation of these same species within membrane lipids suggests that accumulated acyl chains from acylcarnitines are not metabolically recycled into complex lipids. Collectively, these findings demonstrate that CACT loss drives marked membrane lipid acyl-chain remodeling in both oxidative and glycolytic muscles.

Although the acyl composition of membrane lipids was remodeled in Cact^Sk−/− mice across muscle types, the total content of complex lipids was regulated in a muscle-specific manner. Specifically, the total content of a highly abundant phospholipid, phosphatidylethanolamine (PE), was reduced in Cact^Sk−/− EDL, compared with controls, but increased in Cact^Sk−/− soleus (Fig. 2C). In agreement with this, there was a trend toward reduced PC in EDL and increased PC in soleus of Cact^Sk−/− (Fig. 2C). Sudan Black stain was used to visualize neutral and membrane lipid deposition across myofibers, revealing marked lipid accumulation throughout Cact^Sk−/−soleus muscle, with heightened lipid deposition in TA and EDL only evident in a subset of smaller-diameter fibers, characteristic of more oxidative fibers (Fig. 2D). Together, these data demonstrate that loss of CACT causes profound acyl composition remodeling of complex lipids, accumulation of extramitochondrial acylcarnitines across muscles, and membrane lipid accumulation occurring to a greater extent in oxidative myofibers and throughout the soleus muscle.

Soleus Muscle Contractile Deficit in Response to CACT Loss

Considering the striking lipid stain pathology of Cact^Sk−/−muscles, genotype effects on muscle function and structure were assessed. Muscle function was determined ex vivo on isolated EDL and soleus muscles by measuring force production with graded electrical stimulation. Cact^Sk−/− EDL showed a trending, but not significant, reduction in force production (Fig. 3A). However, Cact^Sk−/− soleus exhibited a pronounced functional deficit, generating significantly lower force at all stimulation frequencies, ~75% lower than controls (Fig. 3B), despite greater muscle mass (Fig. 3C). Total fiber type frequency, classified by MyHC expression as an indicator of muscle contractile remodeling and adaptation, was not different between genotypes for either EDL (Supplemental Fig. S3A) or soleus muscle (Fig. 3D), nor were overt genotype differences observed by hematoxylin-eosin (Supplemental Fig. S3B). The size of fibers, by myosin heavy-chain (MyHC)-defined fiber-type, within the EDL revealed no genotype-dependent differences (Supplemental Fig. S3C). In contrast, Cact^Sk−/− soleus had significantly larger cross-sectional area of myofibers (Fig. 3E), an effect observed across both type 2a and 2x fibers (Fig. 3F). These findings demonstrate that the loss of muscle FAO does not affect MyHC-defined fiber type distribution but is associated with larger, yet weaker myofibers.

Figure 3.

CACT deletion alters oxidative soleus muscle contractile capacity and size. Ex vivo EDL (A) and soleus (B) muscle contraction normalized to muscle size and muscle mass (n = 3–6 mice/group) (C). D: representative cross-sectional images of a soleus stained for dystrophin (red) and myosin heavy chain isoforms [I (blue), IIa (green), and IIx (no stain)]; bar graph representing fiber type frequency distribution. Scale bar = 150 μm. E: average cross-sectional area (CSA) of individual fibers. F: distribution of soleus type I, IIa, IIx fiber CSA (n = 3/group). CSA data is representative of n = 3 mice, 3 images per mouse/fiber type/genotype. *P ≤ 0.05 by two-tailed t test; means ± SE. CACT, carnitine-acylcarnitine translocase.

CACT Loss Results in Mitochondrial Accumulation and Metabolic Compensation in Soleus Muscle

Mitochondrial biogenesis is a natural compensatory mechanism to improve energy homeostasis in situations of impaired energetics, such as that induced by CACT loss. The increase in Sudan Black signal, increase in myofiber size, and increased PE content, which is highly abundant in mitochondrial membranes, suggest possible mitochondrial accumulation due to CACT loss. To evaluate mitochondrial content, mitochondrial histological assessment was performed. Modified Gomori is a clinical stain used to show mitochondrial proliferation and myofiber damage that results from myopathies by staining heightened mitochondrial content red, with very minimal red signal detectable in normal healthy muscle. Cact^Sk−/− EDL red signal from Gomori was only evident in small-size fibers (Fig. 4A). However, in Cact^Sk−/− soleus, nearly every fiber showed greater red stain compared with controls, with mitochondrial aggregates distributed throughout both subsarcolemmal and intramyofibrillar regions (Fig. 4A). A similar pattern of mitochondrial abundance was observed using immunofluorescence for succinate dehydrogenase subunit A, a citric acid cycle and electron transport system representing protein (Fig. 4B). In agreement, higher protein abundance of mitochondrial oxidative phosphorylation (OXPHOS) complex subunits were observed by Western blot in Cact^Sk−/− soleus and EDL (Supplemental Fig. S4A). Significant increases in Pgc1α, or its isoforms 2 and 3, mRNA abundance in soleus and EDL suggest activation of mitochondrial biogenesis in response to CACT loss (Supplemental Fig. S4B). These data demonstrate mitochondrial accumulation as an adaptive response to CACT loss, particularly in oxidative soleus muscle.

Figure 4.

CACT loss induces mitochondrial accumulation and increased metabolic capacity in oxidative soleus muscle. A: Gomori trichrome staining of EDL and soleus (rep of n = 3 mice/group, taken at ×20, females). B: succinate dehydrogenase subunit A (SDHA, red) and DAPI (nuclei, blue) immunodetection of Cact^f/f and Cact^Sk−/− EDL and soleus muscle, males. Scale bar = 150 μm. C: oxygen consumption (JO₂) of EDL and soleus muscle homogenates energized by pyruvate (Pyr), glutamate (Glut), or palmitoylcarnitine (PC) (n = 4–7 mice/group, males). D: muscle metabolite levels in Cact^f/f and Cact^Sk−/− EDL (left) and soleus (right) muscle (n = 6 mice/group). *P ≤ 0.05 by two-tailed t test; means ± SE. CACT, carnitine-acylcarnitine translocase.

Next, high-resolution respirometry was used to determine if mitochondria accumulation in oxidative muscles altered muscle metabolic capacity in the face of CACT loss. In homogenates of Cact^Sk−/− soleus muscle, compared with controls, the respiration supported by pyruvate and glutamate were significantly increased by approximately twofold (Fig. 4C), whereas no genotype-dependent differences were observed in EDL (Fig. 4C). Metabolic profiling of EDL and soleus muscle failed to demonstrate differences in bioenergetic metabolite homeostasis between genotypes of EDL or soleus (Fig. 4D, Supplemental Fig. S4C), suggesting that the metabolic adaptations of the soleus muscle facilitate the preservation of the overall bioenergetics. Together, these findings show that FAO loss disproportionately affects oxidative muscle, eliciting compensatory increases in mitochondrial biogenesis and reliance on nonfatty acid substrates to preserve energetic homeostasis.

DISCUSSION

FAO is not only a critical energetic source but also known as a driver of long-chain acylcarnitine accumulation. Long-chain acylcarnitines transiently increase during physiological states such as fasting or acute exercise (13, 14). Moreover, the chronic accumulation of acylcarnitines is characteristic of different pathologies, including inherited mitochondrial FAO disorders (15, 16) and acquired metabolic diseases linked to mitochondrial oxidative capacity decrements, including obesity, insulin resistance, and type 2 diabetes (17, 18). Despite divergent disease etiologies, these conditions converge clinically on skeletal muscle myopathies marked by reduced contractile capacity, weakness, and exercise intolerance (15, 16, 19, 20). Mechanistically, two principal drivers have been typically considered across these metabolic myopathies, including energetic insufficiency and toxicity from accumulated lipid intermediates, including acylcarnitines. Indeed, acylcarnitine elevations in plasma are routinely used diagnostically in inherited FAO defects (15, 16) and, importantly, correlate with disease severity in various metabolic disorders (19, 21). Although energetic insufficiency is often presumed to underlie metabolic myopathies, our work suggests changes to contractile architecture might be more influential on muscle functional decline. This conclusion is supported by minimal alterations in global metabolomic profiles, preserved body weight (Supplemental Fig. S1B), organ mass (10, 12), reproductive health, lifespan, and maintained or even augmented mitochondrial bioenergetic capacity to use nonfatty acid substrates in mice with complete loss of skeletal muscle FAO due to CACT loss or the loss of the subsequent enzyme of the carnitine shuttle, carnitine palmitoylcarnitine transferase 2 (CPT2) (10–12). Given the absence of clear energetic failure, these findings further strengthen the notion that lipid intermediates and alterations in skeletal muscle morphology likely drive contractile dysfunction. We and others have shown long-chain acylcarnitines, but most specifically the C16:0 palmitoylcarnitine, inhibit calcium uptake and release at the sarcoplasmic reticulum (10, 22–24), which is essential for excitation-contraction coupling and whose disruptions substantially reduce muscle force production. In the current study, we observed that Cact^Sk−/− oxidative soleus muscle exhibited a marked approximately fourfold elevation in palmitoylcarnitine, paralleled with contractile deficits. In contrast, glycolytic Cact^Sk−/− EDL muscle failed to accumulate palmitoylcarnitine, despite similar increases in total and C18 acylcarnitine species, and did not demonstrate comparable functional impairment. This work further implicates the potential effect palmitoylcarnitine has on contractility, specifically in oxidative muscles where palmitoylcarnitine exclusively accumulates.

In addition to direct acylcarnitine-mediated contractile decline, excessive mitochondrial accumulation may contribute to contractile impairments in oxidative skeletal muscle. The mitochondrial network in skeletal muscle is tightly configured around the myofilaments, enabling rapid ATP delivery to the contractile machinery (25). However, aberrant mitochondrial biogenesis, in light of FAO deficiency, must be carefully regulated to avoid encroaching on myofilament space, which could disrupt sarcomeric organization and compromise contractile architecture and function (10, 11, 25). Herein, excessive mitochondrial accumulation could explain contractile decrements present in CPT2 (10–12) and CACT-deficient mice (Fig. 3A). However, FAO-specific deficits are not likely a major driver of excessive mitochondria in conditions such as type 2 diabetes where decreases in mitochondrial oxidative capacity are accompanied with reduced mitochondrial size and number, particularly in the intermyofibrillar pool—a phenomenon that occurs in parallel with increased acylcarnitine accumulation and reiterates the direct interference of contractility by accumulated FAO metabolic intermediates (3, 26). Ultimately, the role of mitochondrial displacement of myofibril structure and of acylcarnitines on muscle contraction requires further investigation.

Supplementary Material

Supplemental Table S1: https://doi.org/10.6084/m9.figshare.32190228.

Supplemental Table S2: https://doi.org/10.6084/m9.figshare.32190261.

Supplemental Figs. S1–S4: https://doi.org/10.6084/m9.figshare.32190312.

Supplemental Methods: https://doi.org/10.6084/m9.figshare.32319087.

DATA AVAILABILITY

Data generated/analyzed during the current study are available upon request from the corresponding or lead author.