Figure 5.5. Current-voltagerelations obtained from whole-cell patch clamp recordings reveal four types ofvoltage-regulated channels on salamander Müller cells. A. Current responses for the inward calcium currents (Ca) and A-type potassium currents (KA) are shown on the left hand scale of ordinates; the larger calcium-activated potassium currents (KCa) are on the right. Note that activation of these currents occurs when cells are depolarized to between -40 and -60 mV. B. Potassium currents through the inward rectifier (KIR) recorded from a cell shorn of its endfoot, and bathed sequentially in solutions containing 2.5, 16, and 80 mM K+. This voltagedependent channel is open under physiological conditions, and the channel conductance increases as the cell is hyperpolarized from the resting membrane potential. In addition, the I-V relation shifts to the right at increasing levels of [K+]o, producing a further rise in conductance in regions of the cell exposed to the high [K+]o (Newman, 1985b). (Copyright 1985 Macmillan Magazines Ltd., reprinted with permission.)
mammalian Müller cells. In addition to the currents mediated by rapidly inactivating (KA) and inwardly rectifying (KIR) channels, voltage-regulated K+ channels with markedly different conductances, as well as currents flowing through a delayed rectifier (KDR) channel have been reported (cf. Nilius and Reichenbach, 1988; Reichenbach et al., 1992; Reichelt et al., 1993). Each of these channels, as well as the Ca2+-activated K+ channel, could participate to some degree in the regulation of K+o, although it has proven difficult to distinguish their individual contributions to the overall process. In addition, there is reason to believe that other mechanisms are also engaged directly or indirectly in K+ buffering, all of which confirm the importance of K+ homeostasis to the functional integrity of neural tissues (cf. Amedee et al., 1997).
As already noted, establishing the functional significance of voltage-dependent channels can present a difficult challenge. However, KIR channels are clearly a significant component of the K+ buffering system of Müller cells. Strictly speaking, KIR channels are functionally distinct from voltage-gated K+ channels—i.e., they do not activate over a fixed range of membrane potential, and inward rectification is due mainly to the blockade of outward current by internal Mg2+ rather than a voltage-activated gating mechanism (Matsuda et al., 1987). In addition, unlike classical voltage-gated channels, KIR is typically open at or near the cell's resting potential, its current regulation is highly dependent upon [K+]o, and for the most part it is unidirectional, allowing a much larger K+ influx than efflux (cf. Fig. 5.5 B).
However, it is important to note that the outward K+ current carried by IK(IR) is not insignificant, and as will be described, the channel can serve to extrude K+ in regions of the cell membrane not exposed to high [K+]o. Fig. 5.5B shows that the current through KIR channels is determined by the electrochemical driving force on K+. Typically, the channel conductance is maximal at potentials below the potassium equilibrium potential (EK), and when [K+]o is raised, the resultant changes in (EK), and the voltage-dependence of the channel, serve to effectively preserve inward rectification and enhance K+ influx. Thus, a rise in [K+]o may reduce EK, but the membrane potential (EM) remains "clamped" by those regions of the membrane not exposed to high K+o, and K+ enters the cell.
There is now general consensus that KIR channels play a prominent role in the K+ buffering activity of Müller cells, but their properties and membrane distribution remain somewhat controversial. In mammalian (rabbit) retinae, two types of inward rectifier were identified: high conductance (105 pS) Kir channels localized in the proximal processes of the
Müller cells, and low conductance (60 pS) KIR channels in the membrane of the soma and distal processes (Nilius and Reichenbach, 1988). The endfoot membrane, on the other hand, was reportedly dominated by nonrectifying, large conductance channels (3.5 to 6-fold greater than the conductances of the inward rectifiers) which the authors contend are responsible for the time-and voltage-independent currents through K+ channels at the vitreal surface of the retina (see also Reichelt et al., 1993). This channel distribution would evidently provide an effective means for K+ siphoning to the vitreous, i.e., K+ entry through inward rectifying channels within the plexiform (synaptic) layers of the retina, and extrusion through passive ion channels at the Müller cell endfoot.
However, a comprehensive investigation of KIR channels of mammalian Müller cells with molecular biological, immunochemical, and electro-physiological techniques, failed to confirm these observations and brought into question the existence of multiple types of KIR channels in the Müller cell membrane (Ishii et al., 1997). Whole-cell and cell-attached patch-clamp recordings revealed only a single population of KIR channel currents in dissociated Müller cells from rabbit retina. This channel was identified as the Kir4.1 member of the superfamily of functionally diverse inward-rectifier channels (Takumi et al., 1995; Fakler and Puppersberg, 1996). Moreover, it exhibited properties that matched almost precisely those of rat Kir4.1 channels expressed in human embryonic kidney cells, and a glial-cell derived low-conductance KIR4.1 channel expressed in Xenopus oocytes (Takumi et al., 1995). The unitary conductance of the channel (~ 25 ps), and its dependence upon intracellular ATP, are properties found also in IK(IR) channels of monkey Müller cells (Kusaka and Puro, 1997). Surprisingly, recordings from more than 200 cell-attached membrane patches of rabbit Müller cells failed to detect conductances consistent with any of the variety of KIR channels reported previously (Nilius and Reichenbach, 1988). It will be interesting to see how this discrepancy is resolved, but it should be obvious that we have yet to identify the many factors that influence channel expression. Nevertheless, it has become increasingly apparent that the inward rectifying K+ channel is the principal mediator of K+ buffering currents in the cell membrane of amphibian and mammalian Müller cells.
In addition to their functional studies of rabbit Müller cells, Ishii et al. (1997) used a polyclonal antibody raised in rabbit against a C-terminal region of rat KIR4.1 to study the distribution of these channels in rat retina. Confocal light-microscopic immunocytochemistry of retinal wholemounts and immunogold EM of retinal sections revealed that the channels were present in the cell body, processes, and endfoot of the Müller cell. Moreover, Kir4.1 immunoreactivity in isolated Müller cells demonstrated that the channels tended to be grouped in clusters. This clustering, which probably enhances channel activity in regions where K+ fluxes are maximal, was shown to be induced by insulin and laminin, two endogenous proteins of the extracellular matrix. It is likely that channel clustering also accounts for the wide variability in the numbers of channels recorded in cell-attached patches from particular regions of the Müller cell membrane.
5.2.3. Kir Channels and K+ Buffering Currents
As already indicated, the potassium conductance in various regions of the Müller cell membrane is highly species-dependent (Fig. 5.4), suggesting that there is more than one avenue for K+-induced current flow. In amphibia, for example, there is an overwhelming preponderance of IK(IR) channels in the endfoot (Brew et al., 1986), whereas in rat, channel density is high both at the endfoot and at the distal ends (Ishii et al., 1997).
As shown in Fig. 5.5, a rise in extracellular K+ increases channel conductance and shifts the reversal potential in a depolarizing direction (in accordance with the Nernst equation). Because current flow through KIR channels is inward at resting potentials more negative than the potassium equilibrium potential (Ek), K+ will enter the Müller cell in regions where neuronal activity produces a local increase in [K+]o, e.g., in the plexiform layers. In those regions where K+ is unchanged (the vitreous) or reduced (the subretinal space) by neural activity, the resting potential is positive to Ek, and K+ can exit the Müller cell via KIR channels. Thus, potassium buffering currents will depend upon the membrane distribution of KIR (Brew and Attwell, 1985) and can lead to the deposition of K+ in the vitreous as well as in the distal retina (Fig. 5.6), where it may be actively transported across the RPE to the choroidal vasculature (Steinberg and Miller, 1979).
Was this article helpful?