Howell Bryan, Medina Leonel E, Grill Warren M
Duke University, Department of Biomedical Engineering, Durham, NC, USA.
Duke University, Department of Electrical and Computer Engineering, Durham, NC, USA.
J Neural Eng. 2015 Oct;12(5):056015-56015. doi: 10.1088/1741-2560/12/5/056015. Epub 2015 Sep 8.
Models of excitable cells consider the membrane specific capacitance as a ubiquitous and constant parameter. However, experimental measurements show that the membrane capacitance declines with increasing frequency, i.e., exhibits dispersion. We quantified the effects of frequency-dependent membrane capacitance, c(f), on the excitability of cells and nerve fibers across the frequency range from dc to hundreds of kilohertz.
We implemented a model of c(f) using linear circuit elements, and incorporated it into several models of neurons with different channel kinetics: the Hodgkin-Huxley model of an unmyelinated axon, the McIntyre-Richardson-Grill (MRG) of a mammalian myelinated axon, and a model of a cortical neuron from prefrontal cortex (PFC). We calculated thresholds for excitation and kHz frequency conduction block, the conduction velocity, recovery cycle, strength-distance relationship and firing rate.
The impact of c(f) on activation thresholds depended on the stimulation waveform and channel kinetics. We observed no effect using rectangular pulse stimulation, and a reduction for frequencies of 10 kHz and above using sinusoidal signals only for the MRG model. c(f) had minimal impact on the recovery cycle and the strength-distance relationship, whereas the conduction velocity increased by up to 7.9% and 1.7% for myelinated and unmyelinated fibers, respectively. Block thresholds declined moderately when incorporating c(f), the effect was greater at higher frequencies, and the maximum reduction was 11.5%. Finally, c(f) marginally altered the firing pattern of a model of a PFC cell, reducing the median interspike interval by less than 2%.
This is the first comprehensive analysis of the effects of dispersive capacitance on neural excitability, and as the interest on stimulation with kHz signals gains more attention, it defines the regions over which frequency-dependent membrane capacitance, c(f), should be considered.
可兴奋细胞模型将膜比电容视为一个普遍存在且恒定的参数。然而,实验测量表明膜电容随频率增加而下降,即呈现出频散现象。我们量化了频率依赖性膜电容c(f)在从直流到数百千赫兹的频率范围内对细胞和神经纤维兴奋性的影响。
我们使用线性电路元件实现了一个c(f)模型,并将其纳入具有不同通道动力学的几种神经元模型中:无髓鞘轴突的霍奇金-赫胥黎模型、哺乳动物有髓鞘轴突的麦金太尔-理查森-格里尔(MRG)模型以及前额叶皮层(PFC)的皮质神经元模型。我们计算了兴奋阈值和千赫兹频率传导阻滞、传导速度、恢复周期、强度-距离关系以及发放率。
c(f)对激活阈值的影响取决于刺激波形和通道动力学。我们使用矩形脉冲刺激时未观察到影响,仅对于MRG模型,使用正弦信号时在10千赫兹及以上频率观察到阈值降低。c(f)对恢复周期和强度-距离关系的影响最小,而有髓鞘和无髓鞘纤维的传导速度分别最多增加7.9%和1.7%。纳入c(f)时阻滞阈值适度下降,在较高频率时影响更大,最大降幅为11.5%。最后,c(f)对PFC细胞模型的发放模式影响很小,使峰间间隔中位数减少不到2%。
这是对频散电容对神经兴奋性影响的首次全面分析,并且随着对千赫兹信号刺激的关注度不断提高,它定义了应考虑频率依赖性膜电容c(f)的区域。