Molecular Cloning and Expression of an Inwardly Rectifying K⁺ Channel from Bovine Corneal Endothelial Cells

Abstract

purpose. To determine the presence of a putative inwardly rectifying K⁺ channel in bovine corneal endothelial (BCE) cells and to characterize its molecular and electrophysiological properties.

methods. An RT-PCR strategy was used to clone an IRK1 channel sequence from BCE mRNA. Northern blot analysis was used to confirm expression of this sequence in cultured BCE cells. Two-electrode voltage-clamp and whole-cell patch-clamp recordings were used to characterize the cloned channel expressed in Xenopus oocytes and the native channels in cultured BCE cells, respectively.

results. A full-length (1284 bp) coding sequence that shares 99.7% nucleotide sequence and 100% amino acid sequence identity to bovine lens IRK1 (Kir2.1) was cloned. The authors designate this sequence BCE IRK1 or BCIRK1. Northern blot analysis indicated that BCIRK1 mRNA is expressed in cultured BCE cells with two major transcripts of 7.5 and 5.5 kb. BCIRK1 cDNA was subcloned into the vector, pcDNA3.1(−), and cRNA transcribed from the BCIRK1 cDNA clone was injected into Xenopus oocytes. Two-electrode voltage-clamp recordings from injected oocytes revealed inwardly rectifying K⁺ currents that were blocked by external Ba²⁺ and Cs⁺ in a concentration- and voltage-dependent manner. Whole-cell patch-clamp recordings from dissociated cultured BCE cells revealed strongly inwardly rectifying K⁺ currents with similar properties.

conclusions. Corneal endothelial cells express IRK1 (Kir2.1) inwardly rectifying K⁺ channels. Consistent with the properties of IRK1 channels, BCIRK1 is likely involved in regulating membrane potential and possibly other cellular functions in corneal endothelial cells.

The corneal endothelium lines the posterior surface of the cornea and is responsible for pumping salt and water from corneal stroma to aqueous humor, thereby maintaining corneal transparency. Although progress has been made in identifying the ion transport mechanisms that comprise what is called the “endothelial pump,” ¹ the constituent transporters and channels have not been characterized at the molecular level. One class of ion channels of current interest in ion transporting epithelia is K⁺ channels. These channels play a fundamental role in intracellular homeostasis and the generation of the resting membrane potential and have been implicated in the regulation of cell volume and transepithelial transport in many types of epithelial cells, ² including the corneal endothelium. ^{3 4} In a series of biophysical studies, Rae and colleagues ³ characterized a number of ion channels in corneal endothelial cells. Two types of K⁺ channels were identified: a transient, outwardly rectifying K⁺ channel present in rabbit endothelium ³ and an anion-stimulated, Ba²⁺-insensitive K⁺ channel expressed in endothelia from a variety of species. ^{5 6} In a subsequent study, however, Shepard and Rae ⁷ demonstrated that a rabbit corneal endothelial cDNA library contained a 614-bp sequence matching an inwardly rectifying, Ba²⁺-sensitive K⁺ channel, IRK1, initially cloned from the mouse macrophage ⁸ and now designated Kir2.1. The expression of a K⁺ current in corneal endothelial cells with properties consistent with IRK1 activity has not yet been described. 

We have used bovine corneal endothelial (BCE) cell cultures to study K⁺ channel expression. The BCE cell preparation, developed in our laboratory, ⁹ has been widely used for ion transport and other studies, including those of Jentsch et al. ¹⁰ and Bonanno and Giasson, ^{11 12} which defined a workable model for endothelial bicarbonate transport. As an initial step toward molecular identification of K⁺ channels in BCE cells, we generated a mouse IRK1 (Kir2.1) ⁸ probe and used it on Northern blot analysis of BCE cell mRNA to detect transcripts homologous to IRK1. These results are consistent with the reported presence of IRK1 cDNA sequence in rabbit corneal endothelial cDNA libraries. ⁷ Therefore, the aim of the present study was to clone the full-length inward rectifier K⁺ channel sequence from BCE cells using an RT-PCR strategy and confirm its sequence homology to IRK1, and to identify inwardly rectifying K⁺ channels in cultured BCE corneal endothelial cells and compare their electrophysiological properties to those of the cloned channel expressed in Xenopus oocytes. Our results indicate that BCE cells express an IRK1 inwardly rectifying K⁺ channel, which we designate BCIRK1. The channel exhibits near sequence identity with mouse macrophage IRK1, ⁸ and with IRK1 channels from other tissues, including lens ¹⁴ and, in a recent preliminary communication, rabbit corneal epithelium and endothelium. ¹⁵ Consistent with the presence of the BCIRK1 gene, and BCIRK1 message as determined by Northern blot analysis, cultured BCE cells exhibit a highly K⁺-selective, inwardly rectifying K⁺ current with properties similar to that of the expressed channel in Xenopus oocytes. A preliminary account of the BCIRK1 data has been reported in abstract form. ¹⁶  

Materials and Methods

Cell Culture

Primary cultures of BCE cells were prepared as described by MacCallum et al. ⁹ Adult bovine eyes were obtained from slaughtered cattle 2 to 5 hours after death, and adhering extraocular tissues were removed from the globes by dissection. The eyes were then vigorously rinsed and subsequently covered for 15 minutes with gauze wetted with an aqueous solution of chlortetracycline (50 μg/ml). The corneas were excised and placed, endothelial side up, in a plastic cup. The endothelial surface was covered with Earle’s balanced salt solution buffered to pH 7.3 with 15 mM HEPES, 10 mM TES and 10 mM BES (EBSS). Corneal endothelial cells were dislodged from Descemet’s membrane by rinsing and then subsequently incubating the cells at 37°C for 5 to 7 minutes in a solution of Ca²⁺-, Mg²⁺-free EBSS that contained 5 mM EDTA and 0.01% trypsin. Corneas were gently rubbed with a silicone rubber spatula, and the dislodged cells floating in the trypsin-EDTA solution were collected. Cells obtained from two eyes (∼0.5 × 10⁶) were added in 5 ml of culture medium (Dulbecco’s modified Eagle’s minimal essential medium with 1 g glucose/l, plus 10% iron-supplemented bovine calf serum and 50 μg/ml gentamicin sulfate) to 25-cm² culture flasks, gassed with 95% air/5% CO₂, and incubated at 37°C. Confluent endothelial cultures (∼2.5 × 10⁶ cells/flask) were subcultivated using a 5- to 7-minute incubation in EBSS containing 0.01% trypsin and 5 mM EDTA. The free cells were added to an equal volume of culture medium. After centrifugation, the cells were resuspended in sufficient medium to initiate between two and four subcultures per primary culture. 

Molecular Cloning of BCIRK1

Primers designed from the mouse macrophage IRK1 cDNA sequence (GenBank X73052) ⁸ were used for amplification and cloning of the entire coding sequence of IRK1 from BCE cells. Total RNA isolated from primary-cultured BCE cells was used to generate first-strand cDNA using Superscript Reverse Transcriptase (Gibco BRL Life Technologies, Gaithersburg, MD) with the procedures outlined in the manufacturer’s instructions. The oligonucleotides were synthesized by Gibco BRL Life Technologies. The sense primer was GCGCAGAATTCGGCTTATGGGCAGTGTGAGAACCAACCGCTACAGC from the region between 338 to 367 bp of the mouse macrophage IRK1 sequence (GenBank X73052), ⁸ and the antisense primer was GCGCTGGAATTCGGCTTTCATATCTCCGATTCTCGCC between 1624 and 1605 bp. Both the sense and anitisense primer had a 5′ EcoRI restriction site. The PCR products were generated by using AmpliTaq DNA polymerase and cycled 30 times (30 seconds at 94°C, 30 seconds at 55°C, 1 minute at 72°C, followed by a 10-minute extension at 72°C). The main product band was isolated from agarose gel and cloned into the EcoRI site of the pcDNA 3.1(−) expression vector (Invitrogen, Carlsbad, CA). The insert was flanked by a pair of unique restriction sites, NotI and BamHI. The sequence of the PCR product was confirmed by DNA sequencing analysis. 

Total RNA and PolyA⁺ RNA Preparations

Total RNA was isolated from primary cultures of BCE cells as described. ¹⁷ Briefly, cells were homogenized in 4 M guanidine isothiocyanate and β-mercaptoethanol to inactive RNases. The homogenate was layered over 6 M CsCl and ultracentrifuged, and the RNA pellet was resuspended in RNase-free water. BCE polyA⁺ RNA was prepared by double polyA⁺ selection using MessageMaker Reagent Assembly (Gibco BRL Life Technologies), and bovine heart polyA⁺ RNA was isolated by a modified guanidithiocyanate method, followed by polyA⁺ selection with two rounds of oligo(dT)-cellulose columns (Clontech, Palo Alto, CA). 

Northern Blot Analysis

Northern blot analysis prepared from 1.5 μg of polyA⁺ RNA isolated from primary cultures of BCE cells, and fresh bovine heart were hybridized with digoxigenin-labeled full-length antisense cRNA probes synthesized in vitro from linearized pcDNA3.1BCIRK1/V5HIS, which was generated by inversely inserting the BCIRK1 in frame into the pcDNA 3.1/V5HIS (Invitrogen), or from linearized pTRI-GAPDH-Mouse (Ambion, Austin, TX) by a RNA labeling kit (Boehringer Mannheim, Mannheim, Germany). Blots were hybridized for 2 hours at 65°C in ExpressHyb solution (Clontech) followed by high-stringency washes at 65°C in 0.1× SSC/0.5% SDS. Hybridized probe was detected by enhanced chemiluminescence (Ambion). Blots were exposed to x-ray hyperfilm (Amersham, Buckinghamshire, UK) and developed after 1 to 30 minutes. 

Expression of Transcipt in Xenopus Oocytes

cDNA templates for in vitro transcription were prepared by linearizing BCIRK1 cDNA clone, pcDNA3.1(−)BCIRK1, at the 3′ end with BamHI. The BCIRK1 cRNAs were synthesized using mMessage mMachine (Ambion) with T7 polymerase. Transcript aliquots were stored at −80°C. 

Xenopus oocytes (stages V and VI) were removed from anesthetized Xenopus laevis toads and manually defolliculated after incubation in a collagenase-containing bath for 2 hours. The following day, beveled pipettes were used to inject 50 nl of cRNA (∼0.1–0.5 ng/nl) into oocytes. Oocytes were incubated at 19°C for 3 to 7 days before electrophysiological recording. 

Two-Electrode Voltage-Clamp Recordings from Oocytes

Individual oocytes were placed in the recording chamber and continually perfused with frog Ringer (FR). The standard FR contained 98 mM NaCl, 2 mM KCl, 1.8 mM CaCl₂, 1 mM MgCl₂, and 5 mM HEPES (2.5 mM HEPES acid and 2.5 mM sodium HEPES), pH 7.5, 200 mOsm/l. In experiments testing the dependence of whole-cell currents on extracellular K⁺ concentration, Na⁺ was replaced with K⁺. Blocker sensitivity measurements were carried out using high K⁺ frog Ringer (KFR) containing 100 mM KCl, 1.8 mM CaCl₂, 1 mM MgCl₂, and 5 mM HEPES (2.5 mM HEPES acid and 2.5 mM sodium HEPES), pH 7.5, 200 mOsm/l. Ba²⁺ or Cs⁺ was added to this solution as a chloride salt to the indicated final concentrations. Microelectrodes were pulled (Sutter P-97; Sutter Instruments, Novato, CA) from borosilicate glass (1.2 mm OD, 0.69 mm ID) and had resistances of 0.5 to 1.5 MΩ when filled with 3 M KCl. Membrane potential (V) and current (I) were monitored online on a chart recorder (Kipp & Zonen, Bohemia, NY). Whole-cell currents were recorded from oocytes using a two-electrode voltage-clamp (TEV-200; Dagan Instruments, Minneapolis, MN, or Geneclamp 500; Axon Instruments, Foster City, CA). Voltage-step and -ramp commands were generated by computer control (Clampex; Axon Instruments). 

Whole-Cell Recording from Corneal Endothelial Cells

Whole-cell recordings were performed using standard techniques as described by Hughes and Takahira. ¹⁸ Patch-clamp recordings were done on freshly dissociated cultured BCE cells (passages 2 and 3) placed in a continuously perfused Lucite recording chamber. Cells selected for recording had a bright appearance under phase-contrast microscopy. All experiments were conducted at room temperature (23–25°C). The standard bath solution for patch-clamp experiments contained 135 mM NaCl, 5 mM KCl, 10 mM glucose, 1.8 mM CaCl₂, 1.0 mM MgCl₂, and 10 mM HEPES and was titrated to pH 7.4 with NaOH. In experiments testing the dependence of currents on extracellular K⁺ concentration, NaCl (or N-methyl-d-glucamine [NMDG]-Cl) was replaced with different concentrations of KCl to achieve[ K⁺] + [Na⁺] (or[ NMDG]) = 140 mM. For blocker experiments, the control bath solution contained 100 mM KCl, 40 mM NMDG, 10 mM glucose, 1.8 mM CaCl₂, 1.0 mM MgCl₂, and 10 mM HEPES, and Ba²⁺ or Cs⁺ (chloride salts) was added to the final concentrations indicated. The osmolality of all external solutions was 288 ± 5 mOsm/l. 

The standard pipette solution used in these experiments consisted of 30 mM KCl, 83 mM potassium gluconate, 5.5 mM EGTA-KOH, 0.5 mM CaCl₂, 4 mM MgCl₂, 10 mM HEPES, and 4 mM ATP (K⁺ salt) and was titrated to pH 7.2 with KOH. The osmolality of the standard pipette solution was 244 ± 5 mOsm/l. Patch pipettes were pulled from 7052 glass tubing (Garner Glass, Claremont, CA) using a multistage programmable puller (Sutter Instruments) and heat-polished to resistances in the range 3 to 5 MΩ just before use. Whole-cell currents were acquired using an Axopatch 1D amplifier (Axon Instruments) and filtered at 500 Hz to 2 kHz using an 8-pole Bessel filter. Data were acquired and analyzed using pCLAMP software (Axon Instruments). 

Results

Primary Structure of BCIRK1

We used RT-PCR to produce the entire coding sequence of IRK1 from total RNA isolated from cultured BCE cells. As shown in Figure 1A , this generated a main product band with an expected size of 1.3 kb. The band was isolated, cloned into pcDNA 3.1(−) vector, and sequenced. The nucleotide sequence of the coding region for the bovine corneal endothelium IRK1 (BCIRK1) channel is shown in Figure 1B . Sequence comparison between BCIRK1, and the bovine lens epithelium IRK1 (GenBank AF020792), ¹⁴ also cloned from primers derived from mouse IRK1 sequence (GenBank X73052), ⁸ revealed near perfect identity. Only four nucleotide differences (bold and underlined letters, Fig. 1B ) were present, and these all occurred at the third base of the codon and did not result in any amino acid change or translation frame shift. Like the bovine lens epithelium IRK1 sequence (GenBank AF020792), ¹⁴ the BCIRK1 clone sequence was 1284 bp in length, 3 bp shorter than the mouse IRK1 sequence (GenBank X73052). ⁸ This difference results in one extra amino acid (D390) in mouse IRK1. Therefore, the BCIRK1 clone encodes a 427 amino acid protein and shares 100% amino acid sequence identity with bovine lens epithelium IRK1 (GenBank AF020792). ¹⁴ This amino acid sequence is also identical with that of bovine aortic endothelium IRK1 (GenBank U95369). ¹⁹  

BCIRK1 Transcripts in BCE Cells

Two major transcripts corresponding to 7.5 and 5.5 kb were detected in Northern blot analysis of BCE cell polyA⁺ RNA using antisense BCIRK1 as a probe (Fig. 2) . Hybridization of bovine heart polyA⁺ RNA under identical conditions, however, yielded only a 5.5-kb transcript (Fig. 2) , similar to what has been observed for bovine aortic endothelial cell mRNA. ^{19 20} Rehybridization of the blot with a probe for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) indicated that these patterns were not due to differences in loading or transfering of polyA⁺ RNA to the nylon filter (Fig. 2) , although the transcript size for bovine heart GAPDH was found to be smaller than that of BCE cells. 

Expression of Inwardly Rectifying, K⁺-Dependent Currents in BCIRK1 cRNA-Injected Oocytes

Large inwardly rectifying currents were observed in BCIRK1 cRNA-injected Xenopus oocytes bathed with 100 mM K⁺ (Fig. 3 No significant inward currents were observed in uninjected- or water-injected oocytes under the same conditions (data not shown). Figure 3B summarizes the results of experiments in which the extracellular K⁺ concentration ([K⁺]_o) was varied (Na⁺ substitution). Increasing[ K⁺]_o caused the inward current to grow, with the inward slope conductance increasing roughly in proportion to the square root of[ K⁺]_o (Fig. 3C) . The zero current potential (V ₀) also varied with [K]_o, as expected for a K⁺ selective channel: V ₀ averaged −56.2 ± 4.2 (mean ± SEM; n = 5) in oocytes bathed in 10 mM K⁺, and it depolarized 45 mV to −9.4 ± 4.9 (n = 4) in 100 mM K⁺. This change in membrane potential is less than that predicted by the Nernst equation (58 mV), which is likely due to the presence of an endogenous Cl⁻ conductance. 

Blockage of BCIRK1 Currents by External Ba²⁺and Cs⁺

External Ba²⁺ produced a time- and voltage-dependent block of BCIRK1 currents (Fig. 4) . In the presence Ba²⁺ in the bath, voltage steps from a holding potential of 0 mV to more negative voltages produced a slow, time-dependent decrease in inward current (Fig. 4A) . The rate of this inactivation became faster with increasing hyperpolarization or[ Ba²⁺] (data not shown). At a given concentration of Ba²⁺, the fraction of current blocked in the steady state increased with membrane hyperpolarization (Fig. 4B) , suggesting the binding of Ba²⁺ to a site in the channel lying within the electric field across the membrane. We calculated the dissociation constant (K _d) for the Ba²⁺- induced block according to the first-order equation:  

\[I_{\mathrm{B}}{/}I_{\mathrm{0}}{=}1{/}(1{+}{[}B{]}{/}K_{\mathrm{d}})\]

where I _B/I ₀ is the ratio of current measured in the presence and absence of blocker and [B] is the blocker concentration. The K _d at −100 mV was 2.6 ± 0.5μ M (mean ± SEM, n = 5), which is similar to that reported for mouse macrophage IRK1. ²¹ BCIRK1 currents were also blocked by external Cs⁺ in a concentration and voltage-dependent manner (Fig. 5) . The time-dependent decrease of inward current was rapid and difficult to resolve because of the overlapping capacitative transient. The K _d for the Cs⁺-induced block at −100 mV averaged 69.9 ± 8.9 μM (n = 5). 

Inwardly Rectifying K⁺ Currents in BCE Cells

Isolated cultured BCE cells exhibited strong inwardly rectifying K⁺ currents that were dependent on the extracellular [K⁺]. Increasing extracellular[ K⁺] from 5 to 140 mM produced a positive shift in V ₀ from −40.3 ± 5.6 to− 7.8 ± 1.8 mV (mean ± SEM, n = 6). By comparing this change in membrane potential (ΔV ₀) to the change in K⁺ equilibrium potential (ΔE _K), we estimate that K⁺ channels comprise approximately 40% of the total membrane conductance (g _K/g _total= Δ V ₀/ΔE _K). Elevation of [K⁺]_o also dramatically increased the size of inward currents (Fig. 6 and enhanced the inward rectification of the I-V relationship (Fig. 6B) . A plot of the inward slope conductance as a function of [K⁺]_o showed that it varied roughly in proportion to the square root of[ K⁺]_o (Fig. 6C) . We therefore conclude that cultured BCE cells express inwardly rectifying K⁺ channels and that these channels may help set the resting membrane potential. 

To characterize further the properties of inwardly rectifying K⁺ currents in BCE cells, we examined their sensitivity to block by extracellular Ba²⁺ and Cs⁺. Superfusing BCE cells with Ba²⁺ reduced the amplitude of inward currents in a voltage- and time-dependent manner (Fig. 7) . The addition of 1 μM Ba²⁺ to the bathing solution produced a slow, time-dependent block of inwardly rectifying K⁺ current that was reminiscent of the kinetics of the Ba²⁺-induced block of BCIRK1 current (Fig. 7A) . The fraction of steady state current blocked increased with membrane hyperpolarization (Fig. 7B) , and at −100 mV, the apparent dissociation constant calculated by Equation 1 averaged 2.7 ± 0.4μ M (n = 5), a value that is indistinguishable from that obtained for the Ba²⁺-induced block of BCIRK1 currents at the same voltage (2.6 ± 0.5 μM). Extracellular Cs⁺ also blocked inwardly rectifying K⁺ currents in BCE cells in a voltage-dependent manner (Fig. 8B ), and like the cloned BCIRK1 channel, the kinetics of the block were rapid and difficult to resolve (Fig. 8A) . At −100 mV, the K _d of the Cs⁺-induced block averaged 96.8 ± 19.0 μM (n = 7), which is nearly identical with the value obtained for BCIRK1 currents (69.9 ± 8.9 μM). Thus, the inwardly rectifying K⁺ channel in BCE cells appears to have affinities for Ba²⁺ and Cs⁺ that are nearly identical with those of the cloned BCIRK1 channel. 

Discussion

Inwardly rectifying K⁺ channels are widely distributed in mammalian cells. ^{22 23} The hallmark of these channels is their conduction of K⁺ more effectively in the inward than the outward direction, even when external [K⁺] is lower than internal[ K⁺]. Two articles appeared in 1993 reporting the identification of the first cDNA sequences encoding inwardly rectifying K⁺ channels. ^{8 13} Since then, seven subfamilies of inwardly rectifying K⁺ channels (Kir1.1a–Kir7.1) have been identified, ^{22 24 25 26 27 28 29 30} and within the Kir2 subfamily, four distinct Kir2 subfamily members, Kir2.1 (IRK1), ⁸ Kir2.2, ^{31 32} Kir2.3, ^{33 34 35 36} and Kir2.4, ^{29 37} have been cloned. Members of the IRK subfamily share several characteristics that permit their identification in native cells and expression systems by electrophysiological assays, such as dependence of conductance on extracellular [K⁺], strong inward rectification, and high sensitivity to block by Ba²⁺ and Cs⁺. In this article, we describe the coding sequence of BCIRK1 cDNA cloned from cultured BCE cells. Furthermore, we compare the functional properties of cloned BCIRK1 channels expressed in Xenopus oocytes to channels present in the cultured BCE cells. 

The deduced amino acid sequence of BCIRK1 is 100% identical with bovine arterial endothelial IRK1 ¹⁹ and bovine lens IRK1 ¹⁴ and 98% identical with mouse macrophage IRK1. ⁸ Thus, BCIRK1 is presumed to contain two transmembrane segments (M1 and M2), the pore-forming region H5, and putative phosphorylation sites for protein kinase C (S3, T6, S357, and T383), protein kinase A (S425), and tyrosine kinase (Y242 and Y366). ^{8 14 19 38} Although there are four nucleotide differences between the BCIRK1 and the bovine lens epithelium IRK1 clones (Fig. 1B) , all occurred at the third base of the codon and are unlikely therefore to be random mutations caused by PCR. 

Using antisense BCIRK1 cRNA as a probe, we detected major transcripts of 7.5 and 5.5 kb in Northern blot analysis of BCE cell mRNA (Fig. 2) . Our probe also hybridized to a single, 5.5-kb band in a blot of bovine heart mRNA, which is consistent with the results of Kubo et al., ⁸ who reported a major 5.5-kb mRNA for IRK1 in macrophage cells, forebrain, cerebellum, heart, and skeletal muscle. It is not clear why corneal endothelial cells have a 7.5-kb transcript in addition to a 5.5-kb transcript, nor whether both transcripts encode identical polypeptides. The murine Kcnj2 (IRK1/Kir2.1) gene isolated from 129Svj genomic DNA is comprised of two exons separated by a single 5.5-kb intron. Exon 1 encodes 168 bases of the 5′-UTR, whereas exon 2 encodes the remaining 235 bases of the 5′-UTR and the entire open reading frame and 3′-UTR. ³⁹ With regard to bovine tissue, one article reported isolation of a 5.1-kb IRK1/Kir2.1 cDNA (BIK) from a bovine aortic endothelial cell library. ¹⁹ The coding sequence of BIK shares 84% identity with the murine IRK1/Kir2.1, whereas both the 5′ and 3′ noncoding regions of BIK show low homology. ¹⁹ However, it is unknown how many exons and introns are present in the bovine IRK1/Kir2.1 gene. The molecular mechanism responsible for the different sized transcripts of IRK1/Kir2.1 (and for GAPDH transcripts) might be tissue-specific processing of the primary transcript such as alternative splicing, tissue-specific termination of transcription, or initiation of transcription from alternate promoters or different isoforms. 

The expression of BCIRK1 channels in Xenopus oocytes resulted in a macroscopic conductance with properties similar to those reported previously for mouse macrophage IRK1 ⁸ and bovine arterial endothelial cell IRK1. ¹⁹ BCIRK1 currents were strongly dependent on extracellular [K⁺], exhibited pronounced inward rectification, and were blocked by extracellular Ba²⁺ and Cs⁺ in a voltage-dependent manner. Whole-cell recordings from isolated BCE cells demonstrated the presence of an inwardly rectifying K⁺ conductance with similar properties. This current was activated by the elevation of [K⁺] in the bath, exhibited strong inward rectification, and was blocked by extracellular Ba²⁺ and Cs⁺, with apparent dissociation constants at −100 mV that were similar to those of the cloned BCIRK1 channel. Therefore, we conclude that cultured BCE cells express Kir2.1 channels in sufficient numbers to impact membrane properties. It will be of interest to confirm the expression of these channels in freshly dissociated corneal endothelial cells. 

The BCIRK1 described here differs from K⁺ channels described previously in functional studies of corneal endothelium from rabbit and other species. In addition to several cation channels and a large conductance Cl⁻channel, ⁴¹ two K⁺ channels have been characterized in patch-clamp studies of corneal endothelial cells. ^{3 5 6 41} The first is a transiently activated, outward rectifying K⁺ channel present in freshly dissociated native rabbit corneal endothelial cells. ³ This voltage-dependent current has properties corresponding to the A-current, which is usually associated with excitable cells. This channel is insensitive to Ba²⁺ but is blocked by quinidine and 4-aminopyridine (4-AP), agents that also induce corneal swelling. ³ The second is a temperature- and anion-stimulated K⁺ channel located at the apical membrane of rabbit corneal endothelial cells. ⁶ Unitary currents carried by this channel exhibit mild inward rectification, but macroscopic currents are outwardly rectifying due to an increase in open probability with depolarization. ⁶ Although this channel is blocked by extracellular Cs⁺, it is insensitive to Ba²⁺. ⁴² In the present study on cultured BCE cells, we found that the predominant current in whole-cell recordings was strong inwardly rectifying K⁺ current that was sensitive to block by both extracellular Ba²⁺ and extracellular Cs⁺, with no obvious contribution from outwardly rectifying K⁺ currents of either type. The reason for the absence of these currents is not clear, but it is possible that anion-stimulated K⁺ channels were not expressed in cultured BCE cells or that they were present but inhibited by our use of nominally bicarbonate-free solutions and recording at room temperature. 

Taken together, our electrophysiological measurements on cultured BCE cells and Xenopus oocytes expressing BCIRK1 channels and molecular biological results indicate that cultured BCE cells express Kir 2.1/IRK1. Additional studies are required to identify and characterize K⁺ currents in native BCE cells and to analyze their contribution to the resting membrane potential and endothelial cell transport function as it relates to the maintenance of corneal clarity. 

 Supported by Grant EY11793 from the National Eye Institute (SAE).

 Submitted for publication January 31, 2000; revised April 8, 2000; accepted April 24, 2000.

 Commercial relationships policy: N.

¹  Dongli Yang and Fang Sun contributed equally to this work.

 

Figure 1.

 

RT-PCR analysis and primary structure of BCIRK1 channel. (A) Ethidium bromide–stained agarose gel showing RT-PCR products generated from total RNA isolated from cultured BCE cells using IRK1 specific-primers. MW1 and MW2, 1- and 100-kb DNA ladders (Gibco BRL), respectively. Size markers of representative bands are indicated. Arrow, the major BCIRK1 RT-PCR product at 1.3 kb. The faint lower band at 600 bp is likely due to nonspecific binding of the primers. (B) Nucleotide sequence of BCIRK1. Bold and underlined letters, nucleotide differences at corresponding positions compared with bovine lens epithelium IRK1 (GenBank AF020792). (C) Deduced amino acid sequence of BCIRK1. Boxed regions, the putative transmembrane regions (M1 and M2) and the pore forming region (H5).

Figure 2.

 

Expression of BCIRK1 mRNA (top) and GAPDH mRNA (bottom) in cultured BCE cells and bovine heart. Lanes from a single gel represent polyA⁺ RNA (1.5 μg/lane) isolated from bovine corneal endothelial cells (BCE) and bovine heart (BH). PolyA⁺ RNA was fractionated by 0.7% denaturing agarose gel electrophoresis, transferred to nylon membrane and subjected to hybridization using digoxigenin-labeled antisense BCIRK1 RNA probe. Subsequently, the membrane was reprobed with digoxigenin-labeled antisense GAPDH RNA as a loading control. The positions of RNA size markers (Gibco BRL) are indicated. Note that the BCIRK1 probe recognizes a 5.5-kb transcript in the BCE lane and a slightly larger transcript in the BH lane. A second transcript (7.5 kb) is also expressed in BCE. The GAPDH transcript is smaller in BH compared with BCE.

Figure 3.

 

Dependence of BCIRK1 currents on extracellular K⁺ concentration. Whole-cell currents recorded from BCIRK1 cRNA-injected Xenopus oocytes superfused with the concentrations of K⁺ indicated. (A) Voltage-clamped currents recorded from a representative oocyte bathed in 2 mM K⁺ solution (top) and 100 mM K⁺ solution (bottom). The membrane potential was stepped from a holding potential of 0 mV to voltages ranging from −150 to +40 mV in 10-mV increments. Zero current is indicated by the horizontal line to the left of the current records. (B) Current–voltage relationships of BCIRK1 cRNA-injected Xenopus oocytes bathed in different extracellular K⁺ concentrations. Data were obtained by performing voltage ramps of 2 seconds’ duration from −120 to +60 mV at a rate of 100 mV/sec. Data represent mean ± SEM for five oocytes. (C) Normalized slope conductance plotted as a function of extracellular K⁺ concentration ([K⁺]). The slope conductance (G) from −120 to −100 mV was measured at each[ K⁺] and normalized with respect to G in 100 mM K⁺. Data were fitted to the relation G = 0.06[K⁺]^0.61 − 0.05. Data represent mean ± SEM for five oocytes.

Figure 4.

 

BCIRK1 currents are blocked by external Ba²⁺. (A) Whole-cell currents recorded from the same oocyte bathed in high potassium (100 mM K⁺) frog Ringer (KFR) and KFR plus 1 μM Ba²⁺. The membrane potential was stepped from a holding potential of 0 mV to voltages ranging from− 150 to +40 mV in 10-mV increments. (B) Current–voltage relationships for five oocytes in the presence of the indicated concentrations of Ba²⁺ (means ± SEM).

Figure 5.

 

BCIRK1 currents are blocked by external Cs⁺. (A) Whole-cell currents recorded from the same oocyte bathed in high potassium (100 mM K⁺) frog Ringer (KFR) and KFR plus 100 μM Cs⁺. The membrane potential was stepped from a holding potential of 0 mV to voltages ranging from −150 to +40 mV in 10-mV increments. (B) Current–voltage relationships for five oocytes in the presence of the indicated concentrations of Cs⁺ (means ± SEM).

Figure 6.

 

Dependence of inwardly rectifying currents in cultured BCE cells on extracellular K⁺ concentration. (A) Whole-cell currents recorded from a representative cultured BCE cell bathed in 5 mM K⁺ solution (top) and 140 mM K⁺ solution (bottom). The membrane was held at a 0 mV and stepped for 2 seconds to voltages ranging from −150 to +50 mV in 10-mV increments. (B) Current–voltage relationships of whole-cell currents measured in the presence of the indicated external K⁺ concentrations. Data represent the mean ± SEM for six cells. (C) Normalized slope conductance plotted as a function of extracellular K⁺ concentration ([K⁺]). The slope conductance (G) from −120 to −100 mV was measured at each [K⁺] and normalized with respect to G in 140 mM K⁺. Data were fitted to the relation G = 0.05[K⁺]^0.58 + 0.08. Data represent mean ± SEM for six cells.

Figure 7.

 

Ba²⁺-induced block of inwardly rectifying K⁺ currents in cultured BCE cells. Ba²⁺ was added to bath medium containing 100 mM K⁺ and 40 mM NMDG to the indicated final concentrations. (A) Whole-cell currents recorded in the same cell bathed in 100 mM K⁺ Ringer and in 100 mM K⁺ Ringer plus 1 μM Ba²⁺. The membrane potential was stepped from a holding potential of 0 mV to voltages ranging from −150 to +50 mV in 10-mV increments. (B) Current-voltage relationships of BCE cells in the presence of the indicated concentrations of Ba²⁺. Data represent the mean ± SEM for five cells.

Figure 8.

 

Cs⁺-induced block of inwardly rectifying K⁺ currents in cultured BCE cells. (A) Whole-cell currents recorded in the same cell bathed in 100 mM K⁺ Ringer and in 100 mM K⁺ Ringer plus 100 μM Cs⁺. The membrane was held at 0 mV and pulsed to voltages ranging from −150 to +50 mV in 10-mV increments. (B) Current–voltage relationships of BCE cells in the presence of the indicated concentrations of Cs⁺. Data represent the mean ± SEM for seven cells.

The authors gratefully acknowledge the help of Kenneth Crawford, Stephen S. Smith, and Xuehong Liu during the early phases of the oocyte expression studies. The authors thank Yukun Yun for his technical assistance.