Fluctuation analysis of actin in vitro sliding along native thick filaments from molluscan smooth muscles

1)我们从软体动物平滑肌中纯化了肌凝蛋白长丝，其中肌凝蛋白运动定向良好，我们在体外记录了许多不同长度的肌动蛋白丝沿着单个肌凝蛋白粗丝的滑动运动。我们用五种不同的肌凝蛋白粗丝获得了良好的记录。2)通过计算一定时间内肌动蛋白丝滑动距离的方差，分析了肌动蛋白丝滑动距离的波动。分析得出每个肌动蛋白丝的有效扩散系数Din值。然后，我们研究了Dm对肌动蛋白丝长度的依赖性，发现Dm不依赖于肌动蛋白长度，并且是恒定的，正如之前在体外随机定向的运动蛋白中发现的那样。3)通常认为，运动蛋白驱动细胞骨架丝单向滑动的作用在统计上是独立的和随机的。如果是这种情况，我们可以预期Dm与肌动蛋白丝长度的反比成正比，这是中心极限定理的直接结果。通过赫胥黎1957年的“弹簧肌凝”模型的计算机模拟，我们证实了这一理论预测，即如果运动是随机和统计独立的，那么滑动距离的波动与纤维长度的反比成正比。4)然而，上述关于Din不依赖于肌动蛋白丝长度的实验发现表明，这种观点并不适用于蛋白质马达产生的体外滑动运动，而是表明运动蛋白产生定向滑动运动的机制具有协同性。5)因此我们得出结论，肌凝蛋白马达在产生肌动蛋白丝的单向滑动运动中的作用不是随机的，它们的作用必须是协同的。目前，我们正在通过计算机模拟研究这种协同性的可能机制。

1) We purified long myosin thick filaments from molluscan smooth muscles, in which myosin motors are well oriented, and we recorded sliding movements of many actin filaments with various lengths along individual myosin thick filaments in vitro. We obtained good recordings with five different myosin thick filaments.2) We analyzed the fluctuation of the sliding distance of actin filaments, by calculating the variance of the sliding distance for a given period of time. The analysis yields a value of an effective diffusion coefficient, Din, for each actin filament. We then studied dependence of Dm on the actin filament length, and found that Dm does not depend on the actin length and is constant, as was previously found with randomly oriented motor proteins in vitro.3) It is often believed that the actions of motor proteins in driving a cytoskeletal filament to slide unidirectionally are statistically independent and random. If this is the case, we can expect that Dm is proportional to the inverse of the actin filament length, which is a direct consequence of the central limit theorem. By computer simulation with Huxley's 1957 model of 'spring myosins' we confirmed this theoretical prediction that the fluctuation of sliding distance is proportional to the inverse of the filament length if the motor actions are random and statistically independent.4) However, the above experimental finding that Din does not depend on the actin filament length indicates that this belief does not hold in the case of the in vitro sliding movement generation by protein motors, and instead indicates that there is a cooperativity in the mechanism of the directional sliding movement generation by motor proteins.5) We therefore conclude that the actions of myosin motors in generating the unidirectional sliding movement of actin filaments are not random and there must be cooperativity in their actions. We are currently studying possible mechanisms of the cooperativity by computer simulation.