The concentration of free Ca(2+) in the sarcoplasmic reticulum of frog cut twitch skeletal muscle fibers estimated with tetramethylmurexide

One aim of this article was to determine the resting concentration of free Ca²⁺ in the sarcoplasmic reticulum (SR) of frog cut skeletal muscle fibers ([Ca²⁺]_SR,R) using the calcium absorbance indicator dye tetramethylmurexide (TMX). Another was to determine the ratio of [Ca²⁺]_SR,R to TMX's apparent dissociation constant for Ca²⁺ (K_app) in order to establish the capability of monitoring [Ca²⁺]_SR(t) during SR Ca²⁺ release – a signal needed to determine the Ca²⁺ permeability of the SR. To reveal the properties of TMX in the SR, the surface membrane was rapidly permeabilized with saponin to rapidly dissipate myoplasmic TMX. Results indicated that the concentration of Ca-free TMX in the SR was 2.8-fold greater than that in the myoplasm apparently due to binding of TMX to sites in the SR. Taking into account that such binding might influence K_app as well as a dependence of K_app on TMX concentration, the results indicate an average [Ca²⁺]_SR,R ranging from 0.43 to 1.70 mM. The ratio [Ca²⁺]_SR,R/K_app averaged 0.256, a relatively low value which should not depend on factors influencing K_app. As a result, the time course of [Ca²⁺]_SR(t) in response to electrical stimulation is well determined by, and approximately linearly related to, the active TMX absorbance signal.     

Vanilloid receptor expressed in the sarcoplasmic reticulum of rat skeletal muscle

Vanilloid receptor subtype 1 (VR1) was cloned as a capsaicin receptor from neuronal cells of dorsal root ganglia. VR1 was subsequently found in a few non-neuronal tissues, including skeletal muscle [Onozawa et al., Tissue distribution of capsaicin receptor in the various organs of rats, Proc. Jpn. Acad. Ser. B 76 (2000) 68-72]. We confirmed the expression of VR1 in muscle cells using the RT-PCR method and Western blot analysis. Immunostaining studies with a confocal microscope and an electron microscope indicated that VR1 was present in the sarcoplasmic reticulum (SR), a store of Ca2+. The SR releases Ca2+ to cause a contraction when a muscle is excited. However, SR still releases a small amount of Ca2+ under relaxed conditions. We found that this leakage was enhanced by capsaicin and was antagonized by capsazepine, a capsaicin blocker, indicating that leakage of Ca2+ occurs through a channel composed of VR1.     

Glycogen synthase binds to sarcoplasmic reticulum and is phosphorylated by CaMKII in fast-twitch skeletal muscle

We investigated the subcellular localization of glycogen synthase (GS) in the adductor muscle of anesthetized rabbits injected intravenously with propranolol. Under these experimental conditions, glycogen content was about 10 mmol/kg of fresh tissue. Immunofluorescent and fractionation studies showed that GS associated with sarcoplasmic reticulum (SR) membranes. Glycogen and GS always co-sedimented, suggesting a predominant role of glycogen in targeting of GS to SR. SR-associated GS was phosphorylated in vitro by SR-bound Ca2+-calmodulin dependent protein kinase (CaMKII) and dephosphorylated by endogenous protein phosphatase 1 (PP1c). Based on measurements of GS activity ratio, in vitro phosphorylation of GS by CaMKII did not significantly affect GS activity per se. However, GS activity ratio was slightly reduced, when SR membranes were further incubated with ATP after prior phosphorylation by CaMKII, suggesting that CaMKII might act sinergistically with other protein kinases. We propose that SR-bound CaMKII plays a role in regulation of glycogen metabolism in skeletal muscle, when intracellular Ca2+ is raised.     

Cholesterol distribution and structural differentiation in the sarcoplasmic reticulum of rat cardiac muscle cells

Summary The polyene compound, filipin, was used as a probe to localize cholesterol in the membranes of the rat cardiac muscle cell, with particular reference to the sarcoplasmic reticulum (SR). Filipin binds specifically to cholesterol (and related 3--hydroxysterols) in membranes, producing distinct deformations which can be viewed by freeze-fracture and used as markers for the presence of cholesterol-rich regions in the membrane plane. In freeze-fracture replicas of filipin-treated rat myocardium, the muscle cells revealed abundant deformations in their plasma membranes, no deformations in mitochondrial membranes, and an intermediate response in the SR. These results are in agreement with the levels of cholesterol reported in isolated fractions of the different membrane types, and confirm the specificity of filipin action. Within the SR, the filipin-induced deformations were not randomly distributed but occurred more commonly in free SR at or near the Z-region of the sarcomere than in other parts of the free SR or the junctional SR. This finding is interpreted as evidence for a non-homogeneous distribution of cholesterol in cardiac muscle cell SR. The possible significance of cholesterol in relation to structural differentiation and function of the SR is discussed.     

Two splice variants of CaMKII-anchoring protein are present in the sarcoplasmic reticulum of rabbit fast-twitch muscle

Anchoring protein alphaKAP targets calmodulin kinase II (CaMKII) to the sarcoplasmic reticulum (SR), and in the rabbit is a substrate of CaMKII itself in fast-twitch, but not in slow-twitch muscle. This work was aimed at elucidating the molecular basis for differential phosphorylation of alphaKAP. Here we show that two, immunologically related, size forms (23 and 21 kDa) of alphaKAP are present in fast-twitch muscle SR in a 3:1 stoichiometry. Phosphorylation experiments identified the shorter form as the CaMKII specific substrate. Both forms are shown to be stably integrated into the holoenzyme. Two splice variants of alphaKAP were found in rabbit fast-twitch muscle and only one in slow-twitch muscle, using RT-PCR. Mobilities on SDS-PAGE are those expected. The shorter splice variants lacks the 33-nucleotide sequence inserted by alternative splicing present in full-length alphaKAP, akin to differences between variants A and B of brain alphaCaMKII. The absence of the 11-amino acid sequence creates a novel CaMKII phosphorylation site. Taken together our results show that alternative splicing regulates alphaKAP phosphorylation in a fiber-type specific manner.     

Scanning electron-microscopic studies on the three-dimensional structure of sarcoplasmic reticulum in the mammalian red,white and intermediate muscle fibers

Summary The three-dimensional structure of the sarcoplasmic reticulum (SR) in the red, white and intermediate striated muscle fibers of the extensor digitorum longus muscle of the rat was examined under a field-emission type scanning electron microscope after removal of cytoplasmic matrices by the osmium-DMSO-osmium procedure.In all three types of fibers, the terminal cisternae and transverse tubules form triads at the level of the A-I junction. Numerous slender sarcotubules, originating from the A-band side terminal cisternae, extend obliquely or longitudinally and form oval or irregular shaped networks of various sizes in front of the A-band, then become continuous with the tiny mesh (fenestrated collar) in front of the H-band. The A-and H-band SR appears as a single sheet of anastomotic tubules. Numerous sarcotubules, originating from the I-band side terminal cisternae, extend in threedimensional directions and form a multilayered network over the I-band and Z-line regions. At the I-band level, paired transversely oriented mitochondria partly embrace the myofibril. The I-band SR network is poorly developed in the narrow space between the paired mitochondria, but is well developed in places devoid of these mitochondria.The three-dimensional structure of the SR is basically the same in all three muscle fiber-types. However, the SR is sparse on the surface of mitochondria, so the mitochondria-rich red fiber has a much smaller total volume of SR than the mitochondria-poor white fiber. Moreover, the volume of SR of the intermediate fiber is intermediate between the two.     

Distribution of the endoplasmic reticulum and its relationship with the sarcoplasmic reticulum in skeletal myofibers

We have analyzed the distribution of the endoplasmic reticulum (ER) within isolated rat skeletal muscle flexor digitorum brevis myofibers. Studies with confocal microscopy indicated that the resident ER proteins displayed a perinuclear and cross-striated distribution that extended over the I band areas. Interestingly, two discrete distribution patterns were observed when different receptor or viral marker proteins were blocked in the ER. Accordingly, the vesicular stomatitis virus G protein that lost its efficient export through the Golgi apparatus during myogenesis preferentially marked the A-I junctional areas. The proteins that retained their Golgi processing after myogenesis, on the contrary, concentrated around the myonuclei and over the Z lines. Furthermore, the ER exit site marker sec23 located to Z lines but not to A-I junctions. To analyze the ultrastructural organization of the ER, we infected myofibers with recombinant virus expressing KDEL-tagged peroxidase that is translocated into the ER. With transmission electron microscopy, peroxidase activity was found in perinuclear and Z line-flanking tubular structures, but also within the terminal cisternae of the sarcoplasmic reticulum. The translocon-associated protein exhibited a similar localization. Taken together, the terminal cisternae contained unevenly distributed rough ER structures apparently lacking the export function. The exporting ER comprised perinuclear and Z line-flanking structures.     

Localization of ank1.5 in the sarcoplasmic reticulum precedes that of SERCA and RyR: relationship with the organization of obscurin in developing sarcomeres

Ank1.5 is a muscle-specific isoform of ankyrin1 localized on the sarcoplasmic reticulum (SR) membrane that has been shown to interact with obscurin, a sarcomeric protein. We report here studies on the localization of obscurin and ank1.5 in embryonic and postnatal rodent skeletal muscles. Using two antibodies against epitopes in the N- and C-terminus of obscurin, two distinct patterns of localization were observed. Before birth, the antibodies against the N- and the C-terminus of obscurin stained the Z-disk and M-band, respectively. At the same time, ank1.5 was detected at the Z-disk, rising the possibility that obscurin molecules at M-band may not be able to interact with ank1.5. Localization of ank1.5 at Z-disks in E14 muscle fibers revealed that ank1.5 is among the earliest SR proteins to assemble, since its organization preceded that of other SR proteins, like SERCA and RyR. After birth, the antibody against the N-terminus of obscurin stained the M-band while that against the C-terminus stained both M-bands and the Z-disks. Starting from postnatal day 1, ank1.5 was found at the level of both M-bands and Z-disks. Altogether, from these results we infer that exposure of some obscurin epitopes changes during skeletal muscle development, resulting in distinct, antibody-specific, localization pattern. Why this occurs is not clear, yet these data indicate that the organization of obscurin at different locations in the sarcomere changes during muscle development and that this might affect the interaction with ank1.5.     

Three-dimensional electron microscopy of the sarcoplasmic reticulum and T-system in embryonic chick skeletal muscle cells in vitro

Summary The formation of the sarcoplasmic reticulum (SR) and the transverse tubular system (T-system) in embryonic chick skeletal muscle cells in vitro was studied by either the critical point drying-physical rupturing or physical rupturing-freeze drying together with rotary shadowing. In these cells, two membranous systems were observed. One was composed of flattened sacs which were either isolated or were connected to each other with slender processes to form mostly longitudinally oriented strands. Initially, these sacs had small granules at their surface and were found mainly under the sarcolemma. Later, they became smooth at their surface, extending throughout the cytoplasm to form irregular and dense networks. At later phases, the networks tended to be disposed at right angle to nascent myofibrils, exhibiting a characteristic honeycomb appearance. From the similarities in thin section images, they were identified as developing SR.The other membranous system were tubules with many enlargements. They were frequently associated with coated vesicles which appeared to take part in the formation, elongation, and anastomosing of developing tubules. These tubules could be impregnated with a tannic acid-glutaraldehyde-potassium ferrocyanide complex and, thus, were identified as T-tubules.Abbreviations CPD critical-point drying - ES exoplasmic surface of the sarcolemma - FD freeze-drying - PR physical rupture - PS protoplasmic surface of the sarcolemma - SR sarcoplasmic reticulum - TAGPF tannic acid-glutaraldehyde-potassium ferrocyanide - T-system transverse tubular system     

Skeletal muscle mitochondrial phospholipase A2 and the interaction of mitochondria and sarcoplasmic reticulum in porcine malignant hyperthermia

Comparative studies were carried out on the Ca²⁺-transport systems of mitochondria and sarcoplasmic reticulum from longissimus dorsi muscle of genetically selected malignant hyperthermia-prone and normal pigs in order to identify the biochemical lesion responsible for the enhanced release of Ca²⁺ in the sarcoplasm occurring in porcine malignant hyperthermia. Mitochondria isolated from longissimus dorsi muscle of malignant hyperthermia-prone pigs contained a significantly (P < 0.001) higher amount of endogenous long-chain fatty acids. Similar amounts of endogenous mitochondrial phospholipase A₂ were observed in both types of pigs, but the total activity in malignant hyperthermia-prone pigs was at least twice that of normal. Spermine, a phospholipase A₂ inhibitor, lowered the activity in both types of mitochondria to a similar final level. Mitochondria of malignant hyperthermia-prone pigs showed a significantly (P < 0.001) higher oligomycin-insensitive (Ca²⁺ + Mg²⁺)-ATPase activity, but the Mg²⁺-ATPase and the (Ca²⁺ + Mg²⁺)-ATPase activities were similar in both types of pigs. Sarcoplasmic reticulum isolated from longissimus dorsi muscle of malignant hyperthermia-prone pigs showed a significantly higher (Ca²⁺ + Mg²⁺)-ATPase activity and a lower rate of Ca²⁺ uptake; the maximal amount and the rate of Ca²⁺ uptake by sarcoplasmic reticulum of malignant hyperthermia-prone pigs were half that of normal. Mitochondria from longissimus dorsi muscle of malignant hyperthermia-prone pigs inhibited the Ca²⁺-transport system of the sarcoplasmic reticulum of longissimus dorsi from both normal and malignant hyperthermia-prone pigs, but mitochondria from normal pigs had no influence on the sarcoplasmic reticulum from either type. Experimental evidence favours the concept that long-chain fatty acids released from skeletal muscle mitochondria by endogenous mitochondrial phospholipase A₂ are responsible for the enhanced release of Ca²⁺ from mitochondria (Cheah, K.S. and Cheah, A.M. (1981) Biochim. Biophys. Acta 634, 70–84), and also additional release of Ca²⁺ from sarcoplasmic reticulum into the sarcoplasm during porcine malignant hyperthermia syndrome.     

The gating of the sheep skeletal sarcoplasmic reticulum Ca2+-release channel is regulated by luminal Ca2+

The effects of changes in luminal [Ca²⁺] have been investigated in sheep skeletal sarcoplasmic reticulum (SR) Ca²⁺-release channels after activation of the channels by different ligands from the cytosolic side of the channel. Native heavy SR membrane vesicles were incorporated into planar phospholipid bilayers under voltage-clamp conditions. Experiments were carried out in symmetrical 250 mm Cs⁺. Lifetime analysis indicates that channels activated solely by cytosolic Ca²⁺ exhibit at least two open and five closed states. The open events are very brief and are close to the minimum resolvable duration. When channels are activated solely by cytosolic Ca²⁺, luminal Ca²⁺ does not appear to exert any regulatory effect. The P ₀ and duration of the open and closed lifetimes are unchanged. However, if channels are activated by ATP alone or by ATP plus cytosolic Ca²⁺, increases in luminal [Ca²⁺] produce marked increases in P ₀ and in the duration of the open lifetimes. Our results demonstrate that maximum activation of the skeletal SR Ca²⁺-release channel by ATP cannot be obtained in the absence of millimolar luminal [Ca²⁺].We are grateful to the British Heart Foundation for financial support.     

Ca2+-induced sarcoplasmic reticulum Ca2+ release in myotubularin-deficient muscle fibers

Skeletal muscle deficiency in the 3-phosphoinositide (PtdInsP) phosphatase myotubularin (MTM1) causes myotubular myopathy which is associated with severe depression of voltage-activated sarcoplasmic reticulum Ca²⁺ release through ryanodine receptors. In the present study we aimed at further understanding how Ca²⁺ release is altered in MTM1-deficient muscle fibers, at rest and during activation. While in wild-type muscle fibers, SR Ca²⁺ release exhibits fast stereotyped kinetics of activation and decay throughout the voltage range of activation, Ca²⁺ release in MTM1-deficient muscle fibers exhibits slow and unconventional kinetics at intermediate voltages, suggestive of partial loss of the normal control of ryanodine receptor Ca²⁺ channel activity. In addition, the diseased muscle fibers at rest exhibit spontaneous elementary Ca²⁺ release events at a frequency 30 times greater than that of control fibers. Eighty percent of the events have spatiotemporal properties of archetypal Ca²⁺ sparks while the rest take either the form of lower amplitude, longer duration Ca²⁺ release events or of a combination thereof. The events occur at preferred locations in the fibers, indicating spatially uneven distribution of the parameters determining spontaneous ryanodine receptor 1 opening. Spatially large Ca²⁺ release sources were obviously involved in some of these events, suggesting that opening of ryanodine receptors in one cluster can activate opening of ryanodine receptors in a neighboring one. Overall results demonstrate that opening of Ca²⁺-activated ryanodine receptors is promoted both at rest and during excitation-contraction coupling in MTM1-deficient muscle fibers. Because access to this activation mode is denied to ryanodine receptors in healthy skeletal muscle, this may play an important role in the associated disease situation.     

Structural and functional lability induced by diethyl ether on the sarcoplasmic reticulum membrane

Structural and functional changes occuring in sarcoplasmic reticulum vesicles following exposure to low concentrations (5–7%, v/v) of diethyl ether in aqueous media, were studied by electron microscopy and by kinetic measurements of Ca²⁺ transport and ATPase activity. Electron microscopy of thin sectioned and freeze-fractured sarcoplasmic reticulum vesicles provided detailed resolution of Ca-ATPase amphiphilic molecules displaying ‘lollipop’ portions on the outer surface of the vesicle, and non-polar moieties penetrating the membrane's hydrophobic interior. This asymmetric disposition of ATPase molecules was disrupted in vesicles exposed to ether and then centrifuged and/or resuspended in aqueous media. Such vesicles had a tendency to undergo fragmentation, and the distribution of ATPase molecules was markedly altered. The continuous fuzzy layer of lollipops became discontinuous, and the intramembranous particles became randomly distributed over both the concave and the convex freeze-fracture membrane faces. Functionally, the vesicles lost their ability to accumulate calcium in the presence of ATP, although high rates of ATPase activity were maintained. Vesicles which were simply exposed to ether, without being subjected to centrifugation and/or homogenization, did not appear altered ultrastructurally, and retained their ability to accumulate calcium. In fact, the enzyme turnover and the maximal levels of calcium uptake were increased. It is concluded that diethyl ether interferes with lipid-lipid and protein-lipid interactions in the sarcoplasmic reticulum vesicle membrane, thereby facilitating molecular motions which may be a limiting factor in the transport mechanism. On the other hand, these weakened interactions permit structural denaturation and loss of the ability to maintain a transmembrane Ca²⁺ gradient when the vesicles are subjected to mechanical perturbations which are harmless in the absence of ether.     

Membrane-surfactant interactions The effect of triton X-100 on sarcoplasmic reticulum vesicles

The effect of Triton X-100 on purified sarcoplasmic reticulum vesicles has been studied by means of chemical, ultrastructural and enzymic techniques. At low detergent/membrane ratios (about 1 Triton X-100 per 60 phospholipid molecules) the only effect observed is an increase in vesicle permeability. Higher surfactant concentrations, up to a 1:1 detergent/phospholipid ratio, produce a large enhancement of ATPase activity. Membrane solubilization occurs as a critical phenomenon when the surfactant/phospholipid molar ratio reaches a value around 1.5:1, corresponding to 2 μmol Triton X-100/mg protein. At this point, the suspension turbidity drops, virtually all the protein and phospholipid is solubilized and every organized structure disappears. Simultaneously, a dramatic increase in the specific activity of the solubilized ATPase is observed. The sudden solubilization of almost all the bilayer components at a given detergent concentration is attributed to the relative simplicity of this membrane system. Solubilization takes place at the same surfactant/membrane ratio, at least between 0.5 and 4 mg membrane protein/ml. The non-solubilized residue seems to consist mainly of delipidized aggregated forms of ATPase.     

ATP-sensitive voltage- and calcium-dependent chloride channels in sarcoplasmic reticulum vesicles from rabbit skeletal muscle

Chloride channels in the sarcoplasmic reticulum (SR) are thought to play an essential role in excitation-contraction (E-C) coupling by balancing charge movement during calcium release and uptake. In this study the nucleotide-sensitivity of Cl⁻ channels in the SR from rabbit skeletal muscle was investigated using the lipid bilayer technique. Two distinct ATP-sensitive Cl⁻ channels that differ in their conductance and kinetic properties and in the mechanism of ATP-induced channel inhibition were observed. The first, a nonfrequent 150 pS channel was inhibited by trans (luminal) ATP, and the second, a common 75 pS small chloride (SCl) channel was inhibited by cis (cytoplasmic) ATP. In the case of the SCl channel the ATP-induced reversible decline in the values of current (maximal current amplitude, I _max and integral current, I′) and kinetic parameters (frequency of opening F _O , probability of the channel being open P _O , mean open T _O and closed T _c times) show a nonspecific block of the voltage- and Ca²⁺-dependent SCl channel. ATP was a more potent blocker from the cytoplasmic side than from the luminal side of the channel. The SCl channel block was not due to Ca²⁺ chelation by ATP, nor to phosphorylation of the channel protein. The inhibitory action of ATP was mimicked by the nonhydrolyzable analogue adenylylimidodiphosphate (AMP-PNP) in the absence of Mg²⁺. The inhibitory potency of the adenine nucleotides was charge dependent in the following order ATP⁴⁻ > ADP³⁻ > > > AMP²⁻. The data suggest that ATP-induced effects are mediated via an open channel block mechanism. Modulation of the SCl channel by [ATP] _cis and [Ca²⁺] _cis indicates that (i) this channel senses the bioenergetic state of the muscle fiber and (ii) it is linked to the ATP-dependent cycling of the Ca²⁺ between the SR and the sarcoplasm. Received: 4 September 1996/Revised: 6 December 1996     

Muscle-specific GSTM2-2 on the luminal side of the sarcoplasmic reticulum modifies RyR ion channel activity

We show that a glutathione transferase (GST) protein, which is recognised by an antibody against the muscle-specific human GSTM2-2 (hGSTM2-2), is associated with the lumen of the sarcoplasmic reticulum (SR) of cardiac muscle, but not skeletal muscle. We further show that hGSTM2-2 modifies both cardiac and skeletal ryanodine receptor (RyR) activity when it binds to the luminal domain of the RyR channel complex. The properties of hGSTM2-2 were compared with those of the calsequestrin (CSQ), a Ca²⁺ binding protein also present in the lumen of the SR which, like GSTM2-2, contains a thioredoxin-fold structure and modifies RyR activity (Wei, L., Varsanyi, M., Dulhunty, A. F., Beard, N. A. (2006). The Biophysical Journal, 91, 1288–1301). The glutathione transferase activity of hGSTM2-2 is strong, while CSQ is essentially inactive. Conversely CSQ is a strong Ca²⁺ binder, but hGSTM2-2 is not. The effects of luminal hGSTM2-2 on RyR activity differ from those of CSQ in that hGSTM2-2 activates RyRs by increasing their open probability and conductance and the effects are independent of luminal Ca²⁺ concentration. The results suggest that GSTM2-2 can interact with specific luminal sites on the RyR complex and that the interaction is likely to be within the pore of the RyR channel. The differences between the effects of CSQ and hGSTM2-2 suggest that the thioredoxin fold is not a major determinant of the luminal actions of either protein. The results indicate that GSTM2-2 is a novel luminal regulator of the RyR channels in the heart.     

Disturbances of the sarcoplasmic reticulum and transverse tubular system in 24-h electrostimulated fast-twitch skeletal muscle

Chronic low-frequency stimulation of rabbit tibialis anterior muscle over a 24-h period induces a conspicuous loss of isometric tension that is unrelated to muscle energy metabolism (J.A. Cadefau, J. Parra, R. Cusso, G. Heine, D. Pette, Responses of fatigable and fatigue-resistant fibres of rabbit muscle to low-frequency stimulation, Pflugers Arch. 424 (1993) 529-537). To assess the involvement of sarcoplasmic reticulum and transverse tubular system in this force impairment, we isolated microsomal fractions from stimulated and control (contralateral, unstimulated) muscles on discontinuous sucrose gradients (27-32-34-38-45%, wt/wt). All the fractions were characterized in terms of calcium content, Ca²⁺/Mg²⁺-ATPase activity, and radioligand binding of [³H]-PN 200-110 and [³H]ryanodine, specific to dihydropyridine-sensitive calcium channels and ryanodine receptors, respectively. Gradient fractions of muscles stimulated for 24 h underwent acute changes in the pattern of protein bands. First, light fractions from longitudinal sarcoplasmic reticulum, enriched in Ca²⁺-ATPase activity, R₁ and R₂, were greatly reduced (67% and 51%, respectively); this reduction was reflected in protein yield of crude microsomal fractions prior to gradient loading (25%). Second, heavy fractions from the sarcoplasmic reticulum were modified, and part (52%) of the R₃ fraction was shifted to the R₄ fraction, which appeared as a thick, clotted band. Quantification of [³H]-PN 200-110 and [³H]-ryanodine binding revealed co-migration of terminal cisternae and t-tubules from R₃ to R₄, indicating the presence of triads. This density change may be associated with calcium overload of the sarcoplasmic reticulum, since total calcium rose three- to fourfold in stimulated muscle homogenates. These changes correlate well with ultrastructural damage to longitudinal sarcoplasmic reticulum and swelling of t-tubules revealed by electron microscopy. The ultrastructural changes observed here reflect exercise-induced damage of membrane systems that might severely compromise muscle function. Since this process is reversible, we suggest that it may be part of a physiological response to fatigue.