Taking intestinal anion secretion to TASK: a role for K2P channels in cyclic AMP-regulated anion secretion.

Abstract

Intestinal anion secretion has been the focus of numerous investigations ever since the discovery of its essential role in diarrheal disease. A study in this issue of The Journal of Physiology by Julio-Kalajzić et al. (2018) provides new insights into this process via the molecular identification and functional characterization of TASK-2 potassium channels as components of what appears to be a network of K+ channels that support cyclic AMP-activated anion secretion. The standard model of anion secretion involves activation of apical CFTR following protein kinase A phosphorylation and ATP binding to the channel. Earlier studies using rodent intestine and colon have shown that CFTR is capable of conducting either Cl− or HCO3− ions and therefore supports transepithelial Cl− or HCO3− secretion in response to elevations in intracellular cAMP (Tang et al. 2015). CFTR-mediated anion efflux would be expected to produce membrane depolarization that would reduce the driving force for further anion movement out of the cell if not for the activation of basolateral, cAMP-activated KCNQ1/KCNE3 K+ channels, which serve as a mechanism for charge compensation. These channels also support the functions of the Na+–K+-ATPase and NKCC1 cotransporters by providing a pathway for K+ recycling across the basolateral membrane that maintains concentration gradients for Na+ and K+. In the study by Julio-Kalajzić et al. (2018), tissue-specific expression of a dominant-negative mutation in KCNE3 (KCNE3-D90N) effectively reduced forskolin/3-isobutyl-1-methylxanthine (IBMX)-stimulated anion secretion in the colon relative to WT mice. However, KCNE3-D90N mice still exhibited residual cAMP-dependent anion secretion amounting to ∼40% of that observed in WT colon. Interestingly, attempts to inhibit the remaining anion secretion by employing KCa3.1-deficient mice expressing the KCNE3-D90N mutation were unsuccessful, although Ca2+-activated anion secretion evoked by carbachol was abolished. This result indicated that crosstalk between cAMP and Ca2+ signalling pathways, which could potentially activate KCa3.1 channels, does not account for residual anion secretion observed following suppression of KCNQ1/KCNE3 channel activity. However, when basolateral tetrapentylammonium [TPeA (a membrane-permeant, quaternary ammonium compound known to inhibit K2P channels, among other subtypes)] was tested on the colonic epithelium of KCNE3-D90N mice, most of the residual anion secretion was inhibited, but no effect was observed on carbachol-stimulated anion secretion. These results suggested that a K+ channel distinct from KCNQ1/KCNE3 and KCa3.1 was involved in sustaining the remaining anion secretion. Previous RT-PCR and presently reported immunocytochemistry experiments revealed the presence of TASK-2 K+ channels expressed in portions of the crypt and villous cells of the ileum and in the crypts and surface cells of the colon. Furthermore, when TASK-2-deficient mice were used to investigate the effects of forskolin/IBMX on anion secretion, no effect of TPeA was observed, indicating that TPeA-sensitive anion secretion was dependent on TASK-2 expression. To test whether a combination of TASK-2 and KCNQ1/KCNE3 channel activities completely accounted for forskolin/IBMX-stimulated anion secretion, the authors generated Kcne3−/−/Kcnk5−/− double knockout mice. Ussing chamber experiments with colons from these mice revealed a small amount of forskolin/IBMX-stimulated anion secretion that was not affected by basolateral treatment with C293B, although basolateral TPeA produced a further decrease in secretion. In contrast, carbachol elicited robust anion secretion in colons from Kcne3−/−/Kcnk5−/− mice. These results confirmed a role for both KCNQ1/KCNE3 and TASK-2 K+ channels in facilitating cAMP-dependent anion secretion and suggested that when both of these channels are absent, another TPeA-sensitive K+ channel is present that supports the remaining level of cAMP-dependent anion secretion. An issue that was not specifically addressed in the study by Julio-Kalajzić et al. (2018) concerns the relative amounts of Cl− and HCO3− secretion that constitute the Isc response following forskolin/IBMX stimulation. It is possible that TASK-2 channels do not contribute equally to these two components of the forskolin/IBMX response. Earlier studies have shown that forskolin-stimulated Cl− secretion in rodent colon depends on basolateral, bumetanide-sensitive NKCC cotransport activity (∼40–70% depending on the region of the colon) and 4-acetamido-4'-isothiocyano-2,2'-stilbene disulfonate (SITS)-sensitive, HCO3−-dependent Cl− uptake that appears to involve AE2 Cl−/HCO3− exchangers (Schultheiss et al. 1998; Gawenis et al. 2010). Cl− uptake by AE2 results in HCO3− efflux, likely causing localized alkalinization within the unstirred fluid layer adjacent to the basolateral membrane. In a previous study of HCO3− absorption in murine proximal tubules, basolateral alkalinization was proposed to activate TASK-2 channels resulting in membrane hyperpolarization necessary for sustaining electrogenic Na+–(HCO3−)3 cotransport activity (Warth et al. 2004). In the colon, increases in AE2 activity may provide a mechanism for coupling activation of TASK-2 K+ channels with Cl− uptake across the basolateral membrane and apical Cl− efflux through CFTR (Fig. 1). The author has no conflicts of interest to report. National Institute of Allergy and Infectious Disease. Grant Number: R01 AI128729-01