The discovery that glial cells possess voltage-gated ion channels (Mac-Vicar, 1984; Bevan and Raff, 1985) opened a new era in the study of glial cell physiology throughout the nervous system. This observation was particularly surprising in view of the results of earlier studies on invertebrate and vertebrate glia in situ, which led to the conclusion that glial cell membranes are entirely passive, and unresponsive to changes in transmembrane voltage (Kuffler, 1967). Indeed, it was reported that altering the transmembrane potential by as much as 200 mV produced no apparent change in membrane resistance (Kuffler et al., 1966).
We know now that the seeming lack ofvoltage-dependent ion channels was almost certainly due to shunting of current from the site of injection to neighboring glial cells via intercellular gap junctions. With the advent of patch-clamp technology (Hamill et al., 1981), advances in cell isolation and culture procedures, and development of the tissue slice preparation, it has become possible to study in detail the membrane properties of individual glial cells. The results have been remarkable in demonstrating the presence of almost every major class of voltage-dependent Na+, K+, Ca2+, and anion channel previously thought to be exclusively on excitable cells (cf. Mac-Vicar, 1984; Ritchie, 1987; Sontheimer, 1994).
Channels with similar properties have been shown to be present also on Müller cells (Newman, 1985b; Brew et al., 1986; Nilius and Reichenbach, 1988; Reichelt et al., 1993; Chao et al., 1993, 1994a, 1997; Ishii et al., 1997). Several examples are illustrated in the following sections. The striking similarities in channel properties between glial and neuronal voltage-activated
Figure 5.4. A. Intracellular voltage responses of an isolated rabbit Müller cell to focal ejections of K+ at the sites indicated. Note that the distribution of conductances is markedly different from that ofthe salamander Müller cell shown in Fig. 5.2. B. A comparison ofthe distribution of K+ conductances along the Müller cells of several vertebrate species. Values represent the mean ± SEM of individual voltage responses expressed as a percentage of the response recorded at the endfoot of each cell. Note that in mouse and monkey retina, responses recorded in the region of the cell soma exceed that at the endfoot, whereas in cat, the K+ conductance at the distal end of the cell is almost 2-fold greater than at the endfoot region (Newman, 1987). (Copyright 1987 The Journal of Neuroscience, reprinted with permission.)
channels have led to some provocative suggestions regarding their functional significance. However, a great deal of experimental work is necessary to provide a clear picture of the ways in which voltage-gated channels contribute to the activities of glial cells or to their interactions with neurons.
One problem in assigning a role to the voltage-sensitive channels is the high resting K+ permeability of glia, which tends to hold the membrane potential near the K+ equilibrium potential of -80 to -90 mV. For many of these ionic channels, there is little current flow until the cell is depolarized by 40 mV or more above its resting potential, a condition that is not thought to occur under normal physiological conditions. However, not all voltage-regulated channels require large membrane depolarizations, and those that do may be activated under pathological conditions.
5.2.1. Voltage-dependent Channels of Müller Cells
Newman (1984, 1985a) reported that the channels present in the end-foot membrane at the vitreous surface are primarily of one type, namely, a passive, time- and voltage-independent K+ channel. However, when shorn of their endfeet, it was possible to identify four types of voltage-dependent ion channel (Fig. 5.5) in the Müller cell membrane distal to the endfoot (Newman, 1985b): a Ca2+ channel, a Ca2+-activated K+ channel (KCa), a rapidly inactivating A-type K+ channel (KA), and an inward rectifying K+ channel (KIR), each of which had been identified previously both in excitable and inexcitable cells. The different voltage sensitivities and ionic properties of the various K+ channels are representative of but a few of the diverse potassium channel subtypes that have been characterized biophysically and pharmacologically (Rudy, 1988; Jan and Jan, 1990,1994).
Surprisingly, single-channel patch recordings from axolotl Müller cells revealed only one type of K+ channel: an inwardly-rectifying K+ channel, both at the endfoot where channel density is high, as well as at the cell body where few channels are located (Brew et al., 1986). One would not have expected the K+-channel population of Müller cells from salamander and axolotl to be so different, and indeed, more recent data from tiger salamander, in which the recording conditions were optimized to provide an adequate space clamp, confirmed the overwhelming preponderance of inward-rectifying K+ channels throughout the Müller cell membrane (Newman, 1993).
Although there is mounting evidence that the inward-rectifying K+ channels of Müller cells provide the principal avenue for K+ fluxes in mammalian as well as amphibian retinae, it would appear that the types of K+ channel expressed by Müller cells may vary considerably among species. An example of this diversity can be seen in the results obtained from
Was this article helpful?