CrossTalk proposal: Apical NKCC1 of choroid plexus epithelial cells works in the net inward flux mode under basal conditions, maintaining intracellular Cl- and cell volume.

Abstract

Choroid plexus epithelial cells (CPECs) secrete most of the cerebrospinal fluid (CSF) and regulate its ionic composition through poorly understood mechanisms. CPECs express Na+/K+-ATPase and Na+–K+–2Cl− cotransporter 1 (NKCC1) on their apical membrane (CSF facing), deviating from typical basolateral membrane location in Cl−-secreting epithelia (Quinton et al. 1973; Piechotta et al. 2002; Praetorius & Damkier, 2017; Gregoriades et al. 2019). Given this peculiar polarity, only shared with retinal pigment epithelial cells (Miller & Edelman, 1990; Gundersen et al. 1991; Joseph & Miller, 1991; Lobato-Alvarez et al. 2016), the direction of basal net ion fluxes mediated by apical NKCC1 and associated water fluxes in CPECs is controversial, and the cotransporter function unclear. There are two opposing viewpoints: one proposes that under basal conditions apical NKCC1 works in the net efflux mode, cotransporting ions and water, directly contributing to sustained CSF production (Steffensen et al. 2018). In contrast, data from our lab support the hypothesis that under basal conditions apical NKCC1 of CPECs is indeed continuously active, but it works in the net inward flux mode, transporting ions and associated water towards the cell interior, maintaining intracellular Cl− concentration ([Cl−]i), and cell water volume (CWV) needed for CSF secretion (Gregoriades et al. 2019). According to this viewpoint, apical NKCC1 has an absorptive function, contributing indirectly to CSF secretion, and probably working as the sensor and regulator of CSF K+ suspected by Husted & Reed (1976). NKCC1 transport is electroneutral, with a coupling stoichiometry of 1Na+:1K+:2Cl−. Thus, the direction and magnitude of net ion cotransport is determined by the vectorial sum of the chemical potential gradients of the three cotransported ions: ΔμNa+ + ΔμK+ + 2ΔμCl−. Like other membrane transporters, NKCC1 is reversible; it can work in the net inward- or net outward-flux mode, or near equilibrium. The net free energy driving NKCC1-mediated ion transport (ΔμNKCC1) is defined by the equation in Fig. 1. From this equation, it is evident that variations in the chemical gradients of any of the transported ions can alter the direction of transport, but this is particularly critical for the extracellular K+ concentration ([K+]o) and the intracellular Na+ concentration ([Na+]i). Thus, modest changes in [K+]o or [Na+]i have a major impact on the sign and magnitude of ΔμNKCC1 (Fig. 1). The hypothesis that NKCC1 works in the net efflux mode, directly contributing to CSF secretion, largely originated from two sets of observations. First, estimates of [Na+]i in CPECs using flame photometry measurements of whole choroid plexus (CP) tissue pieces from rodents yield an unusually high value (30–50 mM), compared to other epithelial cells (Johanson & Murphy, 1990; Murphy & Johanson, 1990; Keep et al. 1994; Steffensen et al. 2018). This high [Na+]i makes NKCC1-mediated net ion efflux thermodynamically feasible (Fig. 1A). Second, intracerebroventricular bumetanide at concentrations expected to block apical NKCC1 produced partial (50%) inhibition of basal CSF secretion in canines (Javaheri & Wagner, 1993) and mice (Steffensen et al. 2018), and 50% inhibition of unidirectional blood–CSF flux of radiolabelled Cl− (Johnson et al. 1987). Third, wide-field imaging on ex vivo CP with the fluorescent Na+ indicator SBFI showed an increase in the dye emission signal during tissue exposure to bumetanide (10 μM), which was attributed to an increase in [Na+]i of CPECs, resulting from NKCC1 inhibition. The rationale is that if NKCC1 transports Na+ outwardly, its inhibition with bumetanide should lead to an increase in [Na+]i (Steffensen et al. 2018). The hypothesis that apical NKCC1 works in the net inward flux mode, having an absorptive function (Gregoriades et al. 2019), originated from measurements of in vitro CP tissue showing that 86Rb+ uptake, as surrogate of K+, depended on the presence of external Na+ and Cl−, and was blocked by bumetanide (Bairamian et al. 1991). Further, bumetanide decreased intracellular water of in vitro CP. Later work in single CPECs from rat suggested that NKCC1 functions as a K+-reabsorption mechanism from CSF to blood (Wu et al. 1998). In addition, Inhibition of NKCC1 produced shrinkage of single CPECs. In these experiments, relative cell volume (CV) changes were estimated from measurement of cross-sectional areas obtained by video-enhanced differential interference contrast (DIC) microscopy. Bumetanide (100 μm) caused a 9% decrease in basal CV, whereas increasing [K+]o from 3 to 25 mm produced a 33% increase in CV that was blocked by bumetanide or external Na+ removal. These observations were consistent with the in vitro CP tissue measurements of Johanson's lab (Bairamian et al. 1991), but they were challenged because of the high concentration of bumetanide, which could have effects in other transporters expressed in CPECs (Brown et al. 2009; Hughes et al. 2010). However, high sensitivity (1%) live-cell-imaging microscopy (LCIM) measurement of changes in CWV in acutely dissociated single CPECs loaded with the fluorescent dye calcein definitely demonstrated that bumetanide (10 μm) produced a reversible decrease (16%) in basal CWV in CPECs from NKCC1+/+ mouse. In contrast, CPECs from NKCC1−/− did not respond to the same bumetanide concentration (Gregoriades et al. 2019), indicating that the bumetanide-induced cell shrinkage was specifically mediated by NKCC1. Methodological differences between labs are the most likely explanation for the divergent views concerning the apical NKCC1 net flux direction under basal conditions. In our lab, we measured CV changes following genetic or pharmacological inactivation of apical NKCC1, in single CPECs, both in situ and in vitro, using three different methods: transmission electron microscopy, DIC-LCIM, and real-time changes in CWV in single CPECs loaded with the fluorescent dye calcein. The three methods gave the same result, namely CPEC shrinkage following inactivation of NKCC1. This is the CV response predicted if apical NKCC1 is continuously active under basal conditions, inwardly transporting ions and associated water, thereby keeping basal CWV constant. We proposed that the sustained cell shrinkage observed upon genetic or pharmacological inactivation of NKCC1 results from the suppression of solute and water influx, and the presence of unbalanced net efflux pathways that continue working until the cells adopt a new shrunken steady state (Gregoriades et al. 2019). The identity of these net efflux pathways remains elusive. A critical issue in the debate about the directionality of NKCC1-mediated transport is the [Na+]i of CPECs. We measured [Na+]i in individually calibrated single CPECs, using the fluorescent dye ANG-2 (Gregoriades et al. 2019). We found that [Na+]i of NKCC1+/+ is 9.2 ± 0.6 mM, a value similar to that reported for amphibian CPECs using ion-sensitive microelectrodes (Saito & Wright, 1987), but significantly lower than that estimated from flame photometry of whole CP tissue, based on unproven assumptions about CP water and blood content (Delpire & Gagnon, 2019). Thus, the [Na+]i measured with ANG-2 and ion-sensitive microelectrodes favours NKCC1 working in the inward flux mode (Fig. 1). Interestingly, there were no differences in [Na+]i between CPECs from NKCC1−/− and NKCC1+/+, indicating that the mechanisms determining [Na+]i are not affected by NKCC1 deletion. Another observation supporting the net inward flux hypothesis is the basal [Cl−]i measured with the fluorescent indicator MQAE. Basal [Cl−]i of NKCC1+/+ CPECs was approximately 60 mm, whereas in CPECs from NKCC1−/− it was close to electrochemical equilibrium, indicating that NKCC1 mediates uphill Cl− transport, and maintains [Cl−]i above equilibrium. In contrast to the methods used in our lab, Steffensen and colleagues (Steffensen et al. 2018) did not measure single CPEC volume changes, but relative 2D changes in size of ex vivo pieces of whole CP loaded with calcein; however, this fluorescent dye was not used as a CWV indicator, but to stain the CP pieces. Images of square ‘regions of interest’ encroaching CP tissue and background were converted to black and white, respectively. Relative changes in CP tissue size caused by large osmotic challenges were inferred from changes in the black to white ratio. These measurements represent underestimates of 3D volume changes of the CP pieces, but even if those changes in whole tissue volume could be measured, the signals are difficult to interpret for various reasons. The most important is that CP tissue is a highly vascularized structure composed not only of CPECs, but also of connective tissue and immune cells, endothelial and vascular smooth muscle cells, and blood cells, mainly erythrocytes (Ghersi-Egea et al. 2018). The relative cellular composition of rat CP determined by differential counts show that only 25–40% of the total number of cells are epithelial (Quay, 1966). In the IV ventricle CP, where functional imaging experiments are typically done, only 37.7% of cells are epithelial; 48.6% are intravascular erythrocytes and 13.7% are stromal cells such as fibroblasts, smooth muscle and endothelial cells. Another concern in the interpretationation of measurements in whole CP tissue is the lack of consideration of the impact of unstirred layers and other diffusion barriers. Thus, inferences about NKCC1 dynamics in single CPECs from changes in volume of whole CP tissue are difficult to interpret, and conclusions are unwarranted. The SBFI measurements of Steffensen's group, asserting that inhibition of NKCC1 with bumetanide leads to an increase in [Na+]i, are also difficult to interpret for the following reasons. First, lack of proof that the recorded SBFI signals resulted from actual changes in intracellular Na+ (e.g. by transient exposure to ouabain). Second, SBFI was not calibrated. Third, bumetanide interferes with SBFI signals, and an apparent increase in the ratio 340/380 can be explained by bumetanide autofluorescence, and not by an increase in [Na+]i. When excited at 340–345 nm, bumetanide fluorescence increases in a concentration-dependent manner, thereby overlapping with the SBFI excitation at 340 nm (Zhang et al. 1993; Robertson & Foskett, 1995; Rose & Ransom, 1996). When bumetanide is excited at 340 nm, at pH 7.4, fluorescence emission peaks between 400 and 450 nm (Fiori et al. 2003), the emission wavelength at which Steffensen's group recorded the signals. Another methodological difference, which was already mentioned, is that one group used whole CP (Steffensen et al. 2018) whereas the other used freshly isolated CPECs (Gregoriades et al. 2019). The advantages of the acutely isolated single-cell method include the following: (1) their optical properties, which make them ideal for real-time single-cell imaging microscopy measurements of ion concentrations and fluxes, using fluorescent probes combined with DIC optics; (2) the direction and magnitude of net fluxes through electroneutral transporters, like NKCC1, can be assessed by measuring changes in CWV, combined with pharmacological and molecular approaches (KO models, pharmacological inhibitors, external ions requirements); (3) minimization of fluid exchange artefacts resulting from diffusion barriers and unstirred layers; (4) cell viability can be readily assessed with calcein; and (5) preservation of epithelial cell structural and functional polarity (Reuss, 2001). The main disadvantage of acutely isolated epithelial cells is the loss of epithelial structure by detaching the cells from each other, and from the basal lamina, thereby eliminating paracellular pathways. A potential limitation of the studies of Steffensen et al. and Gregoriades et al. is that the experiments were performed in Hepes-buffered solutions. However, solutions were equilibrated with air and thus most likely contained some bicarbonate (HCO3−) resulting from the spontaneous hydration of CO2. This is important because two Na+ entry pathways in CPECs are HCO3−-dependent (Praetorius & Damkier, 2017): NCBE (SLC4A10) and NBCn1 (SLC4A7). It could be argued that decreased Na+ entry through these pathways reduces [Na+]i. However, inhibition of the Na+/K+-ATPase with ouabain resulted in an increase in [Na+]i in single CPECs and thus Na+ entry pathways were present in these cells (Fig. 7A in Gregoriades et al. 2019). Mechanisms of intracellular Na+ regulation have not been studied in CPECs but predictably, the ultimate determinant of [Na+]i will be the Na+/K+-ATPase that counteracts the Na+ ‘leaks’. Interestingly, early work in hippocampal neurons and astrocytes, which express Na+-coupled HCO3− transporters, demonstrated that basal [Na+]i measured with SBFI was not significantly different in Hepes-buffered compared with CO2/HCO3−-buffered saline (Rose & Ransom, 1996, 1997). The overwhelming evidence from both in situ and in vitro models shows that under basal conditions NKCC1 normally works in the net influx mode, maintaining both [Cl−]i and intracellular water volume of CPECs needed for CSF secretion. This explains the counterintuitive observation that inhibition of NKCC1 by intracerebroventricular bumetanide reduces CSF secretion by 50% (Javaheri & Wagner, 1993; Karimy et al. 2017; Steffensen et al. 2018). NKCC1 most likely works as a K+ absorption mechanism, transporting K+ from CSF to blood and contributing, together with the Na+/K+-ATPase, to maintenance of low (2.9 mm) and constant CSF [K+]. Several unanswered questions remain in this model, one of which is the identity of the basolateral exit pathways for K+. Readers are invited to give their views on this and the accompanying CrossTalk articles in this issue by submitting a brief (250 word) comment. Comments may be submitted up to 6 weeks after publication of the article, at which point the discussion will close and the CrossTalk authors will be invited to submit a ‘LastWord’. Please email your comment, including a title and a declaration of interest, to jphysiol@physoc.org. Comments will be moderated and accepted comments will be published online only as ‘supporting information’ to the original debate articles once discussion has closed. Francisco J. Alvarez-Leefmans MD, PhD, is Full Professor at the Department of Pharmacology and Toxicology, Boonshoft School of Medicine, Wright State University, Dayton, OH, USA. He received his PhD in physiology from University College London, where he continued his postdoctoral training with Professors Sir Bernard Katz and Ricardo Miledi. In 1997, he was awarded the Guggenheim Fellowship in Natural Sciences (Neuroscience) for his studies on chloride transport mechanisms in primary sensory neurons, where he first described and functionally characterized the Na+–K+−2Cl− cotransporter (NKCC1) in vertebrate nervous system. He developed the ‘calcein’ method to measure changes in cell water volume in single cells by live-cell imaging microscopy. He is interested in the molecular and cellular mechanisms used by choroid plexus epithelial cells to sense and regulate the cerebrospinal fluid potassium concentration, a fundamental problem of broad physiological and clinical significance for brain function. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. None. Sole author. Dayton Children's Hospital Foundation WSU grant 670180.