The epithelial cells of the choroid plexus secrete cerebrospinal fluid (CSF), by a process that involves the movement of Na+, Cl− and HCO3− from the blood to the ventricles of the brain. This creates the osmotic gradient, which drives the secretion of H2O. The unidirectional movement of the ions is achieved due to the polarity of the epithelium, i.e., the ion transport proteins in the blood‐facing (basolateral) are different to those in the ventricular (apical) membranes. Saito and Wright (1983) proposed a model for secretion by the amphibian choroid plexus, in which secretion was dependent on activity of HCO3− channels in the apical membrane. The patch clamp method has now been used to study the ion channels expressed in rat choroid plexus. Two potassium channels have been observed that have a role in maintaining the membrane potential of the epithelial cell, and in regulating the transport of K+ across the epithelium. An inward‐rectifying anion channel has also been identified, which is closely related to ClC‐2 channels, and has a significant HCO3− permeability. This channel is expressed in the apical membrane of the epithelium where it may play an important role in CSF secretion. A model of CSF secretion by the mammalian choroid plexus is proposed that accommodates these channels and other data on the expression of transport proteins in the choroid plexus. Microsc. Res. Tech. 52:49–59, 2001. © 2001 Wiley‐Liss, Inc.

Whole cell patch-clamp studies were performed with tissue isolated from the cystic fibrosis (CF) transgenic Cftrm1cam mouse, to determine whether anion currents in choroid plexus epithelial cells require the expression of cystic fibrosis transmembrane conductance regulator (CFTR). Inclusion of 0.25 mM adenosine 3',5'-cyclic monophosphate (cAMP) and 375 nM protein kinase A (PKA) in the pipette solution caused a significant activation of a Cl(-)-selective, inward-rectifying conductance in cells from wild-type and CF mice. The small, outward currents observed in wild-type and CF animals, however, were not activated by cAMP-PKA. There were no significant differences in the size of currents between wild-type, heterozygote, and CF cells in the presence or absence of cAMP-PKA. A second whole cell conductance was activated when cells from wild-type mice were swollen. These volume-activated currents were Cl- selective and exhibited outward rectification. They were Ca2+ independent and ATP dependent and blocked by 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid and 5-nitro-2-(3-phenylpropylamino)benzoic acid. The volume-activated channels were also activated in CF mutant cells, and there was no significant difference in the size of the volume-activated currents between wild-type, heterozygote, and CF cells. It is concluded that CFTR neither contributes to the whole cell conductance nor regulates the other anion conductances in choroid plexus epithelial cells.

The contribution of ClC‐2 protein to the inwardly rectifying Cl− conductance in cultured porcine choroid plexus epithelial cells was investigated using Western analysis and whole‐cell current recordings. Inwardly rectifying currents were elicited by hyperpolarizing voltage at a potential more negative than −50 mV in the presence of intracellular protein kinase A (PKA). The relative halide selectivity estimated from the shift in the reversal potential (Erev) was I− > Br− > Cl− > F−. Extracellular vasoactive intestinal peptide (VIP) activated the same currents in a dose‐dependent manner with a half‐maximal concentration of 167·3 nM. H‐89 (a PKA inhibitor) interfered with the current activation by VIP. The Cl− channel was inhibited by external Cd2+, Ba2+or H+, but only weakly inhibited by known Cl− channel blockers including glibenclamide, NPPB, DIDS and anthracene‐9‐carboxylic acid (9AC). A specific antibody to ClC‐2 detected a 79 kDa protein in porcine choroid plexus cells, which was reduced in cells treated with antisense oligodeoxynucleotide for ClC‐2. Both PKA and VIP failed to activate the inwardly rectifying Cl− currents in cells transfected with the antisense oligodeoxynucleotide, while they activated the currents in cells transfected with GFP alone or the control oligodeoxynucleotide randomized from antisense oligonucleotide. It is concluded that ClC‐2 protein contributes to the inwardly rectifying Cl− conductance in porcine choroid plexus epithelial cells.

1. The sensitivity of the inward-rectifying Cl-channel in choroid plexus to changes in external pH (pHO) was examined.2. Cl-currents were recorded using whole-cell patch-clamp methods. The inward-rectifying channel was activated by 375 nm of the catalytic subunit of protein kinase A which was added to the electrode solution.3. Reducing pH. from 7-3 to 6-5 inhibited the inward-rectifying Cl-currents, whereas an increase in current was observed when pHo was elevated to 8-5. The inhibition of the conductance exhibited a sigmoidal relationship with decreasing pH over a range of 8-5 to 5-5. A half-maximal inhibition of the current was observed at pH 7-3. 4. The inhibition of the whole-cell current by reducing pH0 suggests that it is carried by channels which are distinct from other inward-rectifier Cl-channels, e.g. ClC-2, phospholemman and the channel in Xenopus oocytes.

The IsK (minK or KCNE1) protein is known to co-assemble with the KvLQT1 (KCNQ1) protein to form a channel underlying the slowly activating delayed rectifier K+ current (IKs). Controversy remains as to whether the IsK protein assembles with ERG (the ether-a-go-go-related gene) products to form or modulate the channel-underlying the rapidly activating delayed rectifier K+ current (IKr). We investigated the effects of antisense oligodeoxynucleotides (AS-ODN) against IsK and its mutant D77N [which underlies a form of long QT syndrome (LQT5) in humans] on the delayed rectifier K+ current (IK) of neonatal mouse ventricular myocytes in primary culture. Patch-clamp experiments on these cells showed that IK consists of IKs and IKr. IK was not recorded from ventricular cells transfected with AS-ODN, while it was recorded from cells transfected with the corresponding sense oligodeoxynucleotides (S-ODN). IK was not recorded from cells transfected with the D77N mutant, and the action potential duration was much longer than in cells transfected with wild-type IsK. Furthermore, HERG could not induce currents in COS-1 cells co-expressed with the D77N mutant and HERG (the human form of ERG). These results indicate that the IsK protein associates with both KvLQT1 and ERG products to modulate IKr and IKs in cardiac myocytes.