Specific ATPases drive compartmentalized glycogen utilization in rat skeletal muscle

Abstract

Glycogen is a key energy substrate in excitable tissue, including in skeletal muscle fibers where it also contributes to local energy production. Transmission electron microscopy imaging has revealed the existence of a heterogenic subcellular distribution of three distinct glycogen pools in skeletal muscle, which are thought to reflect the requirements for local energy stores at the subcellular level. Here, we show that the three main energy-consuming ATPases in skeletal muscles (Ca2+, Na+,K+, and myosin ATPases) utilize different local pools of glycogen. These results clearly demonstrate compartmentalized glycogen metabolism and emphasize that spatially distinct pools of glycogen particles act as energy substrate for separated energy requiring processes, suggesting a new model for understanding glycogen metabolism in working muscles, muscle fatigue, and metabolic disorders. These observations suggest that the distinct glycogen pools can regulate the functional state of mammalian muscle cells and have important implications for the understanding of how the balance between ATP utilization and ATP production is regulated at the cellular level in general and in skeletal muscle fibers in particular.

Figure 1.

TEM analyses of subcellular glycogen distribution. (A) Fixed muscle segments were prepared for glycogen visualization by TEM and cut in longitudinal sections. Three fibers were photographed per muscle in a randomized systematic order including 12–16 images from the subsarcolemmal region and 12–16 images from the myofibrillar region. (B) In the myofibrillar images, the volumetric content of intermyofibrillar glycogen (long green arrow) and intramyofibrillar glycogen (short orange arrow) was estimated by point counting (Nielsen et al., 2011; Weibel, 1980). (C) In the subsarcolemmal images, the volume of subsarcolemmal glycogen (blue arrow) per fiber surface area was estimated by point counting and a fiber length measurement (Jensen et al., 2022). (D) After completion of analysis of the first 102 fibers, image-to-image variation showed a mean stereological coefficient of error of 0.14, 0.17, and 0.19 after analysis of 16 images for intermyofibrillar, intramyofibrillar, and subsarcolemmal glycogen, respectively. It was decided for the remaining fibers (n = 172) that 12 photographed images were sufficient per region. (E) Scatterplot of TEM-estimated total glycogen per muscle (mean of three fibers) versus the glycogen concentration determined from a homogenate. Concordance correlation coefficient (CCC) showed a moderate agreement between the two methods. R indicates Person’s correlation coefficient. Dotted red line indicates best linear fit. (F) The relative distribution (in percent) of the three pools to the total volumetric content. Values are mean and 95% confidence interval. n = 60 fibers from 20 rats.

Figure 2.

Effect of myosin ATPase inhibition during tetanic stimulation on subcellular compartmentalized glycogen metabolism. (A) Force production during repeated tetanic stimulations in control muscles (red) and in muscles with the myosin ATPase inhibited (BTS + Bleb; yellow) shown as mean and SD. n = 12–13 muscles. (B) Muscle homogenate glycogen concentration shown as mean with 95% confidence interval. n = 9–13 rats. (C) Muscle homogenate lactate concentration shown as geometric mean with 95% confidence interval. n = 9–13 rats. (D, F, and H) Single-fiber values of glycogen in three distinct subcellular pools: intermyofibrillar, intramyofibrillar, and subsarcolemmal. Data are shown as box plots displaying the first and third quartiles and split by the median. n = 27–39 fibers from 9 to 13 muscles. (E, G, and I) Point estimates with 95% confidence interval from linear mixed effect model on square root-transformed data from D. (J and K) Illustrations of the spatial association between SR Ca²⁺ ATPases situated at the SR membrane (orange in F) and intermyofibrillar glycogen, and between the myosin ATPases (orange in G) and inter- and intramyofibrillar glycogen. Interaction or main effects were tested using linear mixed-effect model on square root–transformed data with tetanic stimulation and myosin ATPase inhibitors as fixed effects and rat ID as random effect. P values for two-way interactions were 0.008, 0.001, 0.395, 0.039, and 0.045 for intermyofibrillar, intramyofibrillar, subsarcolemmal glycogen, homogenate glycogen, and homogenate lactate, respectively.

Figure 3.

Effect of Na⁺,K⁺-ATPase inhibition during salbutamol exposure on subcellular compartmentalized glycogen metabolism. (A, C, and E) Single-fiber values of glycogen in three distinct subcellular pools: intermyofibrillar, intramyofibrillar, and subsarcolemmal. Data are shown as box plots displaying the first and third quartiles and split by the median. n = 32–45 fibers from 11 to 16 muscles. (B, D, and F) Point estimates with 95% confidence interval from linear mixed effect model on square root–transformed data from (intermyofibrillar and intramyofibrillar glycogen) or log-transformed data (subsarcolemmal glycogen) from A. (G) Illustrations of the spatial association between Na⁺,K⁺ ATPases situated at the t-tubular membrane (orange) and intramyofibrillar glycogen. (H) Muscle homogenate glycogen concentration shown as mean with 95% confidence interval. n = 11–16 muscles. (I) Muscle homogenate lactate concentration shown as mean with 95% confidence interval. n = 11–16 muscles. Interaction or main effects were tested using linear mixed effect model on square root–transformed (intermyofibrillar and intramyofibrillar glycogen in A and C, respectively), log-transformed (subsarcolemmal glycogen in E), or nontransformed data (H and I) with salbutamol and ouabain as fixed effects and rat as random effect. P values for two-way interactions were 0.164, 0.010, 0.225, 0.281, and 0.052 for intermyofibrillar, intramyofibrillar, subsarcolemmal glycogen, homogenate glycogen, and homogenate lactate, respectively.