Localization of the tandem pore domain K⁺ channel KCNK5 (TASK-2) in the rat central nervous system

Abstract

Tandem pore domain K⁺ channels (2P K⁺ channels) are responsible for background K⁺ currents. 2P K⁺ channels are the most numerous encoded K⁺ channels in the Caenorhabditis elegans and Drosophila melanogaster genomes and to date 14 human 2P K⁺ channels have been identified. The 2P K⁺ channel TASK-2 (also named KCNK5) is sensitive to changes in extracellular pH, inhibited by local anesthetics and activated by volatile anesthetics. While TASK-1 has been shown to be involved in controlling neuronal cell excitability, much less is known about the cellular expression and function of TASK-2, originally cloned from human kidney. Previous studies demonstrated TASK-2 mRNA expression in high abundance in human kidney, liver, and pancreas, but only low expression in mouse brain or even absent expression in human brain was reported. In this study we have used immunohistochemical methods to localize TASK-2 at the cellular level in the rat central nervous system. TASK-2 immunoreactivity is prominently found in the rat hippocampal formation with the strongest staining observed in the pyramidal cell layer and in the dentate gyrus, and the Purkinje and granule cells of cerebellum. Additional immunofluorescence studies in cultured cerebellar granule cells demonstrate TASK-2 localization to the neuronal soma and to the proximal regions of neurites of cerebellar granule cells. The superficial layers of spinal cord and small-diameter neurons of dorsal root ganglia also showed strong TASK-2 immunoreactivity. These results suggest a possible involvement of TASK-2 in central mechanisms for controlling cell excitability and in peripheral signal transduction.

Introduction

Ion channels that selectively pass potassium (K⁺) ions represent the largest family of ion channels. K⁺ channels play a role in the control of electrical responses in the central nervous system (CNS). They set and modulate the membrane potential, shape action potentials and regulate their frequencies, thus controlling neuronal firing and neurotransmitter release. K⁺ channels are organized into three main superfamilies according to their membrane topology. The largest superfamily comprises subunits with six transmembrane segments and one pore domain (for review, see Ref. [26]). The second superfamily comprises subunits with only two transmembrane segments and one pore domain (for reviews, see Refs. [20], [27]). The K⁺ channel subunits of these extensively characterized two superfamilies assemble as tetramers to form functional K⁺ channels [34], [60] and are further organized into families and subfamilies including the voltage-gated K⁺ channels, Ca²⁺-dependent K⁺ channels, ATP-sensitive K⁺ channels, G protein-coupled K⁺ channels, and inward-rectifying K⁺ channels. The third superfamily of K⁺ channel subunits was only recently discovered [21]. Mammalian members of this new class of K⁺ channels are comprised of subunits with four transmembrane segments and the unique feature of two pore domains in tandem [29]. This structural organization gave the new superfamily its name, i.e. tandem pore domain K⁺ channels (2P K⁺ channels). 2P K⁺ channels are also named KCNKx channels according to the Human Genome Organization Nomenclature Committee (HUGO; for more information see http://www.gene.ucl.ac.uk/nomenclature/). At present, 14 human 2P K⁺ channels have been identified [6], [10], [11], [15], [29], [32], [33], [42], [45], [46], [47], [50], [56], but functional expression has only been reported for eleven human 2P K⁺ channels (for reviews, see Refs. [16], [41]). 2P K⁺ channels are the most numerous K⁺ channels in both Caenorhabditis elegans [1], [58] and Drosophila melanogaster [49] genomes and data from the human genome sequencing program lead to the theoretical extrapolation that there may be more than 20 2P K⁺ channel genes expressed in humans [31]. Because 2P K⁺ channels are not gated by voltage, are non-inactivating, are resistant to classical K⁺ channel blockers, and pass currents over a wide range of voltages predicted by the Goldman–Hodgkin–Katz equation, they closely resemble physiological baseline K⁺ channels and are therefore believed to be the independent molecular entities of the long sought after K⁺ leak conductances [3], [19].

TASK-1, the TWIK-related acid-sensitive K⁺ channel, is the most extensively studied 2P K⁺ channel. Native baseline K⁺ currents resembling TASK-1 have been described in atrial cells [22], a human neuroepithelial body-derived cell line [40], carotid body type-I cells [4], adrenal glomerulosa cells [9], and cerebellar granule cells [39]. Recently, it has been shown that neurotransmitter modulation of TASK-1 is an important mechanism in the short and even long-term regulation of neuronal excitability. Interactions between TASK-1 and the thyrotropin-releasing hormone receptor 1 in hypoglossal motoneurons [55] and between TASK-1 and the γ-aminobutyric acid type A receptor (GABA_A) in cerebellar granule cells have been described [2]. In addition, activation of TASK-1 currents in central motoneurons and locus coeruleus neurons by volatile anesthetics suggests that TASK-1 may represent a molecular substrate for their clinical effects [52].

Despite sharing less than 30% amino acid identity in their sequences, TASK-2 has many physiological and pharmacological properties in common with TASK-1, such as the distinct sensitivity to changes in extracellular and intracellular acidity and activation by volatile anesthetics [11], [17], [28], [47]. However, there is much less information available about the functional role and expression pattern of TASK-2 compared with TASK-1. An mRNA distribution analysis reports the strongest expression of TASK-2 in human liver, kidney, and pancreas [38]. Previous studies in kidney localize TASK-2 to the cortical distal tubules and collecting ducts [47] and the proximal convoluted tubules [44]. However, the signals from native K⁺ channels recorded from cortical collecting tubules do not resemble the biophysical and pharmacological properties of those from TASK-2 [14], [57]. Although TASK-2 mRNA was not detected by Northern blot analysis and reverse transcription polymerase chain reactions (RT-PCR) in the human CNS with the exception of spinal cord [17], [38], RT-PCR analysis from mouse and rat tissues showed TASK-2 mRNA in cortex, cerebellum, and brain stem [17], [47].

In general, the function of K⁺ channels is determined by differences in their electrophysiological properties and modulation as well as by their expression pattern in different cell types. While an mRNA distribution analysis of all 2P K⁺ channels has recently been published [38], only limited data on the tissue distribution of 2P K⁺ channel proteins are available [7], [18], [24], [36], [48]. Interpretation of tissue distribution results of 2P K⁺ channels is further complicated by known cross-species and developmental differences [2], [24]. In this study we describe the immunohistochemical localization of TASK-2, previously cloned from human kidney [47], in the rat central nervous system at the cellular level with the aim of better understanding its expression, distribution, and function.