In skeletal muscle, Ca2+ release from the sarcoplasmic reticulum (SR) triggers contraction. In this study we develop a two compartment model to account for the Ca2+ dynamics in frog skeletal muscle fibers. The two compartments in the model correspond to the SR and the cytoplasm, where the myofibrils are placed. We use a detailed model for the several Ca2+ binding proteins in the cytoplasm in line with previous models. As a new feature, Ca2+ binding sites within the SR, attributed to calsequestrin, are modeled based on experimentally obtained properties. The intra SR Ca2+ buffer shows cooperativity, well represented by a Hill equation with parameters that depend on the initial [Ca2+] in the SR ([Ca2+]SR). The number of total sites as well as the [Ca2+]SR of half saturation are reduced as the resting [Ca2+]SR is reduced, on the other hand the Hill number is not changed. The buffer power remained roughly constant. The release process is activated by a voltage dependent mechanism that increases the Ca2+ permeability of the SR. We use the permeability time course and amplitude experimentally obtained during a voltage clamp pulse to drive the simulations. This model successfully reproduces the SR and cytoplasmic transients observed. Additionally, we simulate [Ca2+] SR transients in the case of high concentration of extrinsic Ca2+ buffers added to the cytoplasm to explore what properties of the permeability are necessary to account for the experimentally observed [Ca2+]SR transients. The main novelty of the model, the intra SR Ca2+ buffer, is crucial for reproducing the experimental observations and it would be of use in future theoretical studies of excitation contraction coupling in skeletal muscle.

[Ca2+] transients inside the sarcoplasmic reticulum (SR) were recorded in frog skeletal muscle twitch fibers under voltage clamp using the low affinity indicator Mag Fluo 4 (loaded in its AM form) with the purpose of studying the effect on Ca2+ release of extrinsic Ca2+ buffers (i.e. BAPTA) added at high concentration to the myoplasm. When the extrinsic Ca2+ buffer is added to the myoplasm, part of the released Ca2+ binds to it, reducing the Ca2+ signal reported by a myoplasmic indicator. This, in turn, hinders the quantification of the amount of Ca2+ released. Monitoring release by measuring [Ca2+] inside the SR avoids this problem. The application of extrinsic buffers at high concentration reduced the resting [Ca2+] in the SR ([Ca2+]SR) continuously from a starting value close to 400 μM reaching the range of 100 μM in about half an hour. The effect of reducing resting [Ca2+]SR on the Ca2+ permeability of the SR activated by voltage clamp depolarization to 0 mV was studied in cells where the myoplasmic [Ca2+] ([Ca2+]myo) transients were simultaneously recorded with Rhod2. The Ca2+ release flux was calculated from [Ca2+]myo and divided by [Ca2+]SR to obtain the permeability. Peak permeability was significantly reduced, from 0.026 ± 0.005 ms−1 at resting [Ca2+]SR = 372 ± 5 μM to 0.021 ± 0.004 ms−1 at resting [Ca2+]SR = 120 ± 16 μM (n = 4, p = 0.03). The time averaged permeability was not significantly changed (0.009 ± 0.003 and 0.010 ± 0.003 ms−1, at the higher and lower [Ca2+]SR respectively). Once the cells were equilibrated with the high buffer intracellular solution, the change in [Ca2+]SR (Δ[Ca2+]SR) in response to voltage clamp depolarization (0 mV, 200 ms) in 20 mM BAPTA was significantly lower (Δ[Ca2+]SR = 30.2 ± 3.5 μM from resting [Ca2+]SR = 88.8 ± 13.6 μM, n = 5) than in 40 mM EGTA (Δ[Ca2+]SR = 72.2 ± 10.4 μM from resting [Ca2+]SR = 98.2 ± 15.6 μM, n = 4) suggesting that a Ca2+ activated component of release was suppressed by BAPTA.

Raising the intracellular [Ca2+] ([Ca2+]i) was previously found to produce uncoupling between the electrical depolarization of the transverse tubules and contraction in skinned muscle fibers. Here we study the effect of elevated [Ca2+]i in voltage clamped cut fibers of frog skeletal muscle to establish how the charge movement, a measure of the activation of the dihydropyridine receptors (DHPR)-voltage sensors, and Ca2+ release, a consequence of the opening of the ryanodine receptor (RyR)-release channels, were affected. [Ca2+]i was raised by various procedures (pharmacological release from the sarcoplasmic reticulum, application of high [Ca2+]i intracellular solution, permeabilization of the plasma membrane by a Ca2+ ionophore) all of which produced impairment of excitation–contraction coupling. The charge movement was reduced from 20.2 ± 1.24 to 9.9 ± 0.94 nC/μF meanwhile the Ca2+ release flux was reduced from 13.5 + 0.7 to 2.2 ± 0.3 μM/ms (n = 33). This suggests that a significant fraction of the DHPRs that remained functional, could not activate RyRs, and were therefore presumably disconnected. These results are broadly consistent with the original reports in skinned fibers. Uncoupling was prevented by the addition to the intracellular solution of the protease inhibitor leupeptin. In approximately 40 % of the uncoupled cells we observed that the [Ca2+]i transient continued to rise after the voltage clamp pulse was turned off. This loss of control by membrane voltage suggests that the uncoupled release channels might have another mechanism of activation, likely by Ca2+.

Two drugs, 2-APB and SKF-96365, commonly used to block Store Operated Ca2+ Entry (SOCE) were found to have inhibitory effects at different levels of the Excitation Contraction Coupling (ECC) process in frog skeletal muscle fibers. Treatment with either drug suppressed Ca2+ release from the Sarcoplasmic Reticulum, but this effect was not due to inhibition of SOCE as it occurred in Ca2+-free conditions. 2-APB applied extracellularly at 100 μM, the usual concentration to suppress SOCE, reversibly reduced the charge movement elicited by pulses in the range between −45 and −35 mV from 7.99 ± 0.73 nC/μF (N = 17) before drug application to 6.27 ± 0.68 nC/μF in the presence of 2-APB. This effect was mostly on the delayed Qγ component. In fibers treated with the SERCA ATPase inhibitor CPA the Qγ component disappeared, under this condition the application of 2-APB did not suppress the remaining charge movement. Thus the effect of 2-APB on charge movement currents seemed to be secondary to the suppression of Ca2+ release, likely occurring directly on the release channels. No significant suppression of ECC was observed for concentration below 20 μM. 2-APB also inhibited the L-type Ca2+ current (20 ± 4%, N = 8). On the other hand SKF-96365 had a direct effect on the voltage sensor promoting its voltage dependent inactivation. Applied at 20 μM, a typical concentration used for inhibiting SOCE, to fibers held at −80 mV inhibited the charge moved in response to pulses ranging −45 to −30 mV from 7.95 ± 2.59 nC/μF to 3.48 ± 0.9 nC/μF (N = 12). A parallel reduction of Ca2+ release was observed. Wash out was drastically increased by hyperpolarization of the holding potential to −100 mV. SKF-96365 also inhibited the L-type Ca2+ current (41 ± 8%, N = 4) and increased its rate of inactivation.

Two procedures to inhibit Ca2+ release designed to differentiate between local and common pool mechanisms for the Ca2+ dependent, fast inactivation of Ca2+ release in skeletal muscle of the frog were compared. Inhibition by voltage dependent inactivation of Ca2+ release, without modification of the single channel current of the Ryanodine Receptor (RyR) and the [Ca2+] close to the open pore, produced a reduction in the rate of inactivation linearly related to the reduction in the peak of Ca2+ release flux. Linear fits in the individual fibers were performed, giving average values (±SEM, N = 8) of the best fit parameters of 5.75 × 10−3 ± 7.35 × 10−4 μM−1 for the slope and 0.07 ± 0.015 ms−1 for the ordinate intercept. Inhibition of Ca2+ release by reducing the Ca content of the sarcoplasmic reticulum (SR) involves reduction of the Ca2+ current through the single RyR. The reduction in rate of inactivation also followed linearly the reduction in Ca2+ peak release flux. The average values (±SEM) of the best fit parameters of linear fits were 14 × 10−3 ± 3.76 × 10−3 μM−1 and 0.019 ± 0.006 ms−1 (N = 7) for slope and ordinate intercept respectively. The differences between both parameters were statistically significant (by t test, at P = 0.05). The extent of inactivation, measured by the peak/final Ca2+ release flux ratio, was differentially affected by the two procedures. Inhibition by voltage dependent inactivation, despite slowing down the fast inactivation, increased the peak/final Ca2+ release flux ratio. In contrast, depletion of the SR reticulum reduced it. If the fast inactivation is driven by the high [Ca2+] attained locally, close to the open pore of the RyR, the inhibition of Ca2+ release due to voltage dependent inactivation should not modify the rate of inactivation while inhibition by SR Ca2+ depletion should reduce it. A process driven by [Ca2+] in a common pool should depend on the overall Ca2+ release independently of how it was modified. In this case both inhibitory procedures should reduce the inactivation rate similarly. Our findings are generally consistent with a common pool process. The differences between the two protocols could be understood if the organization of RyR in junctional and parajunctional release units is considered.

BACKGROUND:The complexity of our contemporary medical practice has impelled the development of different decision-support aids based on artificial intelligence and neural networks. Distributed associative memories are neural network models that fit perfectly well to the vision of cognition emerging from current neurosciences.METHODS:We present the context-dependent autoassociative memory model. The sets of diseases and symptoms are mapped onto a pair of basis of orthogonal vectors. A matrix memory stores the associations between the signs and symptoms, and their corresponding diseases. A minimal numerical example is presented to show how to instruct the memory and how the system works. In order to provide a quick appreciation of the validity of the model and its potential clinical relevance we implemented an application with real data. A memory was trained with published data of neonates with suspected late-onset sepsis in a neonatal intensive care unit (NICU). A set of personal clinical observations was used as a test set to evaluate the capacity of the model to discriminate between septic and non-septic neonates on the basis of clinical and laboratory findings.RESULTS:We show here that matrix memory models with associations modulated by context can perform automatic medical diagnosis. The sequential availability of new information over time makes the system progress in a narrowing process that reduces the range of diagnostic possibilities. At each step the system provides a probabilistic map of the different possible diagnoses to that moment. The system can incorporate the clinical experience, building in that way a representative database of historical data that captures geo-demographical differences between patient populations. The trained model succeeds in diagnosing late-onset sepsis within the test set of infants in the NICU: sensitivity 100%; specificity 80%; percentage of true positives 91%; percentage of true negatives 100%; accuracy (true positives plus true negatives over the totality of patients) 93,3%; and Cohen's kappa index 0,84.CONCLUSION:Context-dependent associative memories can operate as medical expert systems. The model is presented in a simple and tutorial way to encourage straightforward implementations by medical groups. An application with real data, presented as a primary evaluation of the validity and potentiality of the model in medical diagnosis, shows that the model is a highly promising alternative in the development of accuracy diagnostic tools.