Cl⁻ Influx into Rat Cortical Lens Fiber Cells Is Mediated by a Cl⁻ Conductance That Is Not ClC-2 or -3

Abstract

purpose. Exposure of organ-cultured lenses to Cl⁻ channel blockers under isotonic conditions induces a localized cortical zone of extracellular space dilations. The purpose of this study was to investigate whether elongated lens fiber cells from this zone contain an anion conductance that mediates Cl⁻ influx and whether two chloride channel isoforms known to be expressed in the lens (ClC-2 and -3) are responsible.

methods. Fiber cells were isolated by enzymatic dissociation in the presence of Gd³⁺ and Co²⁺ and their electrical properties analyzed by whole-cell patch clamping. Cells from the zone of extracellular space dilations were selected for analysis on the basis of cell length. RT-PCR and immunocytochemistry were used to determine whether ClC-2 or -3 channel isoforms are expressed in fiber cells located in the zone of extracellular space dilations.

results. Cells from the zone of extracellular space dilations were typically >120 μm in length and exhibited an outwardly rectifying Cl⁻ conductance that was blocked by DIDS (4,4′-diisothiocyanostilbene-2,2′-disulfonic acid) and displayed an anion selectivity sequence of I⁻ > Cl⁻ ≫ gluconate. ClC-2 and -3 were found to be expressed at the transcript and protein level in lens fiber cells, but subsequent immunocytochemical studies indicated that expressed proteins did not colocalize with cell membranes in the zone of extracellular space dilations, being predominately cytoplasmic in nature.

conclusions. Taken together, the data indicate that extracellular space dilations are due to the inhibition of a Cl⁻ channel(s) that normally mediates Cl⁻ influx into cortical lens fiber cells under isotonic conditions. The molecular identity of this channel remains to be determined.

The transparent properties of the mammalian lens are a direct result of its unique tissue architecture and cellular physiology, which eliminates light-scattering and improves the optical properties of the lens. Disruption of the crystalline packing of the fiber cells, by either cellular swelling or dilation of the normally tight spaces between the cells, increases intralenticular light scattering, and, in a variety of animal cataract models, can result in the formation of localized opacities. ^{1 2 3} Tissue disruption that mimics that in lens cataract can be induced experimentally by exposing isolated lenses under isotonic conditions to pharmacologic reagents that specifically inhibit either Na⁺ pumps ⁴ or Cl⁻ channels. ^{5 6 7} These findings suggest that the maintenance of lens tissue architecture, and hence lens transparency is not a purely passive process but relies on the active interplay of ion channels and transporters to maintain fiber cell morphology. 

Indeed, the existence of such an active transport system in the lens was indicated by the observations of Robinson and Patterson, ⁸ who showed that a standing flow of ionic current exists around the lens that is directed inward at the poles and outward at the equator. Mathias et al. ⁹ have suggested that this standing current underpins the generation of a unique internal microcirculation system that is responsible for nutrient delivery, maintenance of fiber cell ionic homeostasis and therefore lens transparency. Whereas it is not universally accepted, evidence in favor of the model is accumulating. ^{10 11} In particular, the morphologic analysis of tissue damage induced by organ culturing rat lenses under isotonic conditions in the presence of a variety of Cl⁻ channel blockers lends support to the notion that a circulating flux of Cl⁻ ions exists in the lens. ^{6 7}  

In lenses exposed to Cl⁻ channel blockers, two distinctive damage zones that exhibit different damage phenotypes are observed. ^{6 7} Tissue damage is initially evident as a localized band of equatorial extracellular space dilations located some 100 to 150 μm in from the capsule that act as the precursors to more extensive tissue breakdown observed in this zone at later time points. The development of extensive tissue damage is accompanied by the development of a secondary zone of cell swelling in peripheral fiber cells. ⁶ The fact that tissue morphology is disrupted under isotonic conditions in the presence of these Cl⁻ channel blockers suggests that the putative Cl⁻ channels are functional under isotonic conditions in the normal lens. Furthermore, the generation of these two distinct damage phenotypes can be explained with reference to the circulation model and experimental measurements of lens membrane potential. ⁹ By measuring radial differences in transmembrane potential and the concentration of Cl⁻ in the whole lens, it is possible to calculate the electrochemical gradient for Cl⁻ ion movement (E _Cl) as a function of radial distance. ^{12 13} Based on this analysis, it is predicted that Cl⁻ would move from the extracellular space into fiber cells in the inner lens, but would move from the cytoplasm of fiber cells to the extracellular space in the lens periphery. Therefore, one would expect that an inhibition of Cl⁻ channels in the inner lens would block the uptake of Cl⁻ from the extracellular space by fiber cells. This would cause an accumulation of Cl⁻ ions and water in the tortuous extracellular space and the subsequent formation of extracellular space dilations. In the lens periphery, the efflux of Cl⁻ ions from fiber cells would be blocked, thereby causing an intracellular accumulation of osmolytes and resultant fiber cell swelling. 

If the model advanced in Young et al. ⁶ and Merriman-Smith et al. ⁷ is correct, fiber cells from the zone of extracellular space dilation should express Cl⁻ channels that mediate an influx of Cl⁻ ions. Unfortunately, definitive data on the molecular identity and functional properties of Cl⁻ channels in fiber cells in different regions of the rat lens are lacking. The Cl⁻ channel isoforms ClC-2 and -3 have been identified in cDNA libraries prepared from cultured human epithelial cells and the α-TN4 lens epithelial cell line. ¹⁴ Cl⁻ channels have also been identified in membrane vesicles derived from bovine lens fiber cells ¹⁵ and a Cl⁻ conductance that was activated by hypotonic solutions was identified in short fiber cells isolated from bovine lenses; however, the anatomic location of the cells was not specified and was likely to be very peripheral. ¹⁶  

Thus, to test our model we developed experimental protocols that allow the conductance properties of elongated fiber cells that originate from the zone of extracellular space dilations to be measured. In parallel with these experiments, we sought to determine at the molecular level whether the Cl⁻ channel isoforms ClC-2 and -3 are expressed in lens fiber cells and examined whether they represent suitable molecular candidates for the hypothesized Cl⁻ influx pathway. Our results support our model and indicate that fiber cells from the zone of dilations express an outwardly rectifying Cl⁻ channel that mediates Cl⁻ influx but that is probably not ClC-2 or -3. 

Materials and Methods

Reagents

A ClC-2 rabbit polyclonal antibody was purchased from Alomone Labs (Jerusalem, Israel), and a ClC-3 goat polyclonal antibody was obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). 5-Nitro-2-(3-phenylpropylamino) benzoic acid (NPPB) was obtained from Research Biochemicals, Inc. (Natick, MA). Wheat germ agglutinin conjugated to AlexaFluor 350 (Triticum vulgaris, WGA-350) was obtained from Molecular Probes (Eugene, OR). All other chemicals including: 4,4′-diisothiocyanatostilbene-2,2′-disulfonic acid (DIDS); tetramethylrhodamine-5-isothiocyanate-conjugated wheat germ agglutinin (WGA-TRITC) and phosphate-buffered saline (PBS tablets: phosphate buffer 10 mM, KCl 2.7 mM, NaCl 137 mM, pH 7.4) were obtained from Sigma (Sigma Chemical Company, St. Louis, MO) unless otherwise stated. 

Preparation of Isolated Fiber Cells

Three- to 4-week-old Wistar rats were killed by CO₂ asphyxiation and spinal dislocation, in accordance with protocols approved by the University of Auckland Animal Ethics Committee and in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Extracted lenses were placed into artificial aqueous humor (AAH: NaCl 149 mM, KCl 4.7 mM, CaCl₂ 2.5 mM, glucose 5 mM, and HEPES 5 mM [pH 7.4], with mannitol added to 320 mOsM · kg⁻¹) on the stage of a dissecting microscope. A pair of sharpened forceps was used to remove the capsule gently and the epithelial and fiber cells attached to it. The capsule with adherent cells was transferred to an Eppendorf tube and incubated for 30 minutes in 1 mL dissociation buffer (Na-gluconate 170 mM, KCl 4.7 mM, HEPES 5 mM, glucose 5 mM, and 0.125% [wt/vol] type 1A collagenase [Sigma-Aldrich]) at 37°C, with occasional gentle flicking to re-expose settled cell clumps to the enzyme solution. Cells were gently vortexed before being centrifuged at 1000 rpm for 2 minutes. Pelleted cells were resuspended in 200 μL of AAH, which contained 3 mM GdCl₃ and 4 mM CoCl₂ (AAH+GC) to block putative stretch-activated cation channel and connexin hemichannel currents activated by the dissociation procedure. ¹⁷ Cells were plated on a poly-l-lysine-coated glass coverslip, which formed the bottom of a recording chamber, which in turn was mounted on the stage of an inverted microscope (model IM35; Carl Zeiss Meditec, Frankfurt, Germany). Once adherent to the coverslip (∼5 minutes), the cells were overlaid with 1 mL AAH+GC. The continuous perfusion of the recording chamber was driven by a peristaltic pump (Minipulse 3; Gilson, Middleton, WI) connected to a solution-changing device (PCS-110; Dagan Corp., Minneapolis, MN). The bath was perfused at a rate of 0.4 to 0.6 mL · min⁻¹, and rapid switching between bath solutions was initiated via a TTL pulse delivered by an A/D board (Digidata 1200A; Axon Instruments, Inc., Union City, CA; triggered from within pClamp ver 8.1; Axon Instruments, Inc.). Drugs and blocking agents were either made up in AAH and introduced to the bath via the perfusion system or directly injected into the bath as a stock solution. Gluconate- and I⁻-based AAH solutions were prepared by equimolar replacement of NaCl with either Na-gluconate or NaI. All experiments were conducted at room temperature (∼20°C). 

Electrophysiological Measurements

Patch pipettes were pulled from glass capillaries (model 150T-10; Clarke Electromedical Instruments, Reading, UK) with a horizontal pipette puller (model P-77, Sutter Instruments, Novato, CA). Pipettes were fire polished and filled with filtered pipette solution (SWCS: NaCl 10 mM, K-gluconate 130 mM, HEPES 10 mM, CaCl₂ 1.3 mM, EGTA 10 mM, and MgCl₂ 4.1 mM [pH 7.4], 320 mOsM · kg⁻¹ by freezing-point depression). When filled with this solution, pipettes had resistances of between 2 and 4 MΩ. Pipettes were mounted on the headstage of a patch-clamp amplifier (Axopatch 1-D; Axon Instruments, Inc.). The amplifier was controlled by a computer (500-MHz Celeron; Intel, Mountain View, CA, with Windows 2000; Microsoft, Redmond, WA) running the patch clamp software. Whole-cell currents in response to voltage steps and ramps were digitized and recorded directly to disc. Current records were analyzed both on- and off-line (pClamp suite of programs; Axon Instruments, Inc.). Cells were imaged with a charge-coupled device (CCD) camera (Panasonic, Auckland, New Zealand) and calibrated digital images were captured (PCTV-USB; Pinnacle Systems, Singapore) thereby allowing cell length to be analyzed off-line. Results of data analyses were statistically analyzed by multivariate ANOVA (Systat ver. 10; Systat Software Inc., Richmond, CA) and deemed significantly different at the P ≤ 0.05 level. 

RT-PCR Screening

Whole lenses were extracted from the eyes in sterile RNase-free, dimethyl-pyrocarbonate (DMPC)-treated PBS. Lenses for RNA work were first rolled on sterile filter paper to remove any adherent tissues and then decapsulated to remove the lens epithelium, and the remaining fiber cells were stored in an enzyme-inhibiting solution (RNAlater; Ambion, Austin, TX). Total RNA was isolated from the tissue (RNAqueous kit; Ambion) according to the manufacturer’s protocol. cDNA was synthesized from fiber cell total RNA with an RT-PCR system for first-strand synthesis (ThermoScript; Invitrogen-Gibco, Gaithersburg, MD). cDNA synthesis was performed at 60°C for 60 minutes according to the manufacturer’s instructions. Reverse transcriptase (15 U; ThermoScript; Invitrogen-Gibco) was added to the test reaction mix and DMPC-treated water was added to the negative (no reverse transcriptase) control. PCR amplification was performed in 20- to 100-μL reactions with final concentrations of 1× PCR buffer, 0.2 mM dNTPs, 0.05 U/μL DNA polymerase (Platinum Taq; Invitrogen, Carlsbad, CA), 1.5 mM MgCl₂, and 0.2 μM sense and anti-sense primers (Table 1) . Amplification was performed in 35 cycles of a three-step thermal cycling program (Hybaid Omnigene, Middlesex, UK): 30 seconds of denaturation at 94°C, 30 seconds of annealing at 55°C, and extension at 72°C for 60 seconds. Amplified PCR products were analyzed by electrophoresis on 0.8% agarose gels, extracted from the gel (QIAquick Gel Extraction Kit; Qiagen, Clifton Hill, Victoria, Australia), and subsequently sequenced by the dideoxy chain termination technique (AB 373A stretch DNA sequencer; Perkin Elmer, Boston, MA). 

Immunocytochemistry

After removal from the globe, lenses were either immediately processed for immunocytochemistry or were placed an AAH solution that contained 1% penicillin-streptomycin at 37°C for 1 hour and screened visually for damage. Lenses that showed opacities during this incubation period were discarded. Lenses were then incubated for a further 6 hours in either AAH or AAH containing 100 μM NPPB dissolved in DMSO (0.1% vol/vol). DMSO, when added in the absence of drugs, had no effect on lens tissue architecture. ⁶ All lenses were then fixed in 0.75% paraformaldehyde in PBS for 24 hours in the dark at room temperature. Fixed lenses were cryoprotected for 1 hour at room temperature in each of 10% and 20% sucrose in PBS and then overnight in 30% sucrose at 4°C. Lenses were mounted onto chucks, embedded in optimal cutting temperature compound (Tissue-Tek; Sakura Finetek Europe, Loeterwoude, The Netherlands), frozen in liquid N₂, and cut in 10-μm-thick equatorial sections on a cryostat (model CM3050; Leica Microsystems, Wetzlar, Germany) at −17°C. Sections were adsorbed onto poly-l-lysine-coated coverslips, washed with PBS, and incubated for 1 hour at room temperature in blocking solution. For ClC-2 antibodies, the blocking solution was 3% bovine serum albumin and 3% normal goat serum in PBS, whereas for ClC-3 antibodies, it was 5% horse serum albumin in PBS. Primary antibodies were diluted in their respective blocking solutions (ClC-2, 1:150; ClC-3, 1:400). Primary antibodies were visualized with the appropriate secondary antibodies diluted in their respective blocking solutions and incubated 2 hours in the dark at room temperature, followed by 3× washes in PBS. ClC-2 antibodies were visualized with an FITC-conjugated swine anti-rabbit secondary antibody (1:40; Dako, Copenhagen, Denmark), and ClC-3 antibodies were visualized with a Cy3-conjugated donkey anti-goat secondary antibody (1:400, Jackson ImmunoResearch, Inc., West Grove, PA) or an AlexaFluor 488-conjugated donkey anti-goat antibody (1:400; Molecular Probes). Cell membranes were labeled with WGA-TRITC or WGA-Alexa 350 as a 1:50 dilution of 1 μg/mL in PBS for 2 hours in the dark and then washed 3 times in PBS. Labeled sections were mounted (Citifluor AF1; Agar Scientific, Stansted, UK) to reduce fading and examined with a confocal microscope (model TCS 4D; Leica Lasertechnik, Heidelberg, Germany) fitted with an argon-krypton mixed gas laser. All primary antibodies were tested for specificity of labeling by abolishment of signal after antibody preincubation with antigenic peptide in 50:1 excess for 2 hours at 37°C followed by 24 hours at 4°C (peptide control). All immunoreactive solutions were centrifuged at 14,000 rpm for 15 minutes at 4°C before use. All secondary antibodies were tested for specificity by labeling slides with secondary antibody without adding primary antibody (no-1° control). 

Results

Isolation of Viable Fiber Cells from the Zone of Extracellular Damage

Obtaining an isolated population of lens fiber cells for patch-clamp experiments that are viable and exhibit the low conductance properties expected of rat fiber cells in vivo ⁹ has long been problematic. In previous studies from our laboratory, we used isolated fiber cell preparations that were either very short in length and maintained in Ca²⁺ Ringer’s, ¹⁸ or very long but maintained in Ca²⁺-free solutions. ¹⁷ Both preparations exhibited the activation of a nonselective cation conductance and/or a large hemichannel-like conductance. With Ca²⁺ present in the bathing solution, activation of these conductances resulted in Ca²⁺-influx, the activation of Ca²⁺-sensitive proteases, and the subsequent vesiculation of the fiber cells. ¹⁹ This effect of extracellular Ca²⁺ on fiber cell morphology and membrane conductance is illustrated in Figure 1 . Thirty to 45 minutes after plating the cells in the presence of extracellular Ca²⁺ fiber cells appeared swollen and vesiculated (Fig. 1A) and the membrane currents were dominated by a large linear conductance that reversed at 0 mV (Fig. 1B) , suggesting the activation of a nonselective leak conductance(s). By including Gd³⁺, a known blocker of nonselective cation channels, ²⁰ and Co²⁺ a blocker of hemichannel currents, ²¹ in the bath solution, a preparation of cells was obtained that consisted of single fiber cells and rafts of multiple fiber cells that were amenable to patch clamping for several hours. Furthermore, not only did the morphology and viability of the cells improve (Fig. 1C) but the large linear conductance was abolished (Fig. 1D) . 

With this modified procedure, a viable population of isolated fiber cells that displayed a range of lengths that reflected their state of differentiation were obtained. To focus on cells originating from the appropriate area of the lens, we made use of the unique age gradient present in the lens cells, where young, short differentiating fiber cells are close to the capsule and older, longer fibers are positioned deeper toward the lens core (Fig. 2A) . In axial lens sections, the length of fiber cells can be measured and correlated with the distance from the capsule expressed as a function of number of cell layers. Because in the intact lens the capsule normally keeps the fiber cells under some tension, length measurements of isolated fiber cells could underestimate the length of fiber cells in the in vivo situation. Hence, a similar axial-length-to-cell-layer correlation was conducted on large isolated rafts of fiber cells obtained by reducing the amount of agitation used during the enzymatic dissociation procedure (Fig. 2B) . The two analyses yielded an essentially similar curvilinear relationship between consecutive cell layers and fiber cell length (Fig. 2C) . Finally, to identify the length of fiber cells that passed through the zone of extracellular space dilations, images of equatorial sections taken from lenses treated with Cl⁻ channel blockers in a previous study ⁶ were re-examined. In these images, it was found that the zone of extracellular space dilations routinely started ∼15 cell layers in from the capsule before extending deeper into the lens cortex. ⁶ Thus, by extrapolation from Figure 2C , it appears that isolated fiber cells from the zone of extracellular space dilations should be >120 μm in length in vitro. 

Characterization of Membrane Conductances in Cells from the Damage Zone

Cells that extended into the zone of extracellular space dilations that ranged from 130 to 420 μm in length were patch clamped, and their whole-cell currents recorded. At a holding potential of −40 mV, cells in this length range typically had a membrane conductance of 1.99 ± 3.1 pS/pF (n = 67), which agrees favorably with fiber cell conductances calculated from whole-lens impedance measurements. ⁹ Cells in this length range all displayed an outwardly rectifying membrane current which contained differing degrees of delayed activation (Figs. 3 4) . This suggests either the involvement of multiple channel types with different kinetics or a single channel with variable kinetics. Regardless of the mechanism, observed currents were unaffected by the addition of the K⁺ blockers Ba²⁺ or tetraethylammonium (data not shown), but were significantly reduced (65.2% ± 26.2%; n = 5) by the replacement of chloride with the impermeant anion gluconate (Figs. 3B 4B) . Furthermore, the peak outward current at +100 mV was significantly reduced (87.3% ± 7.5%; n = 5) by the addition of the chloride channel inhibitor 100 μM DIDS (Fig. 4D) . Attempts to inhibit this current using NPPB were problematic, because of the chelating and precipitation of the drug by the multivalent cations Gd³⁺ and Co²⁺ present in the bathing medium. These observations indicate that fiber cells isolated from the zone of extracellular space dilations express an outwardly rectifying Cl⁻ channel(s) that mediate an inward flux of Cl⁻ ions under isotonic conditions. 

Evaluation of ClC-2 and -3 as Molecular Candidates for the Cl⁻ Conductance

Because the Cl⁻ channel isoforms ClC-2 and -3 have been identified in lens epithelium cells at the molecular level, ¹⁴ we investigated whether these isoforms may be responsible for mediating Cl⁻ uptake into fiber cells located in the zone of extracellular space dilations. We first used RT-PCR to determine whether the two ClC transcripts are actually expressed in lens fiber cells. PCR products for both ClC-2 and -3 of the predicted size (Table 1) were obtained from fiber cell RNA prepared from decapsulated lenses (Fig. 5) . In all cases genomic contamination was excluded by the failure to amplify a PCR product from control RNA samples (RT minus) that lacked reverse transcriptase. All PCR products were sequenced and found to be identical with their respective ClC gene sequence (data not shown). Hence, it appears that, in addition to being expressed in lens epithelial cells, ¹⁴ ClC-2 and -3 are also expressed in rat lens fiber cells. 

If ClC-2 and -3 are involved in the generation of extracellular space dilations one would expect that the channels would localize in membranes of fiber cells located in this zone. To test this notion, the relative cellular locations of the two isoforms were first investigated by immunolabeling equatorial cryosections obtained from normal lenses (Fig. 6) . Because in other studies we have shown that the distribution of a variety of membrane proteins can change as a function of fiber cell differentiation, ^{22 23} the antibody labeling patterns from young peripheral fibers cells where compared with those of older fiber cells located toward the core of the lens (Fig. 6A) . ClC-2 antibody labeling in peripheral fiber cells (Fig. 6B) was predominately intracellular, whereas, in contrast, the labeling pattern observed in deeper fiber cells was exclusively associated with the membrane (Fig. 6C) . An essentially similar pattern of cytoplasmic labeling in peripheral fiber cells (Fig. 6D) and membrane-associated labeling in deeper fiber cells (Fig. 6E) was also observed for ClC-3. Labeling was dose-dependently abolished by preincubation with an antigenic peptide (data not shown), indicating specificity of the labeling. No-1° control slides were also dark. Taken together, our molecular studies indicate that ClC-2 and -3 are expressed in lens fiber cells and that the subcellular distribution of the two channels appears to change as a function of fiber cell differentiation, indicating that they play some role in lens physiology. Finally, to address specifically whether ClC-2 or -3 are associated with the formation of extracellular space dilations, images where obtained from lenses incubated in the presence of NPPB for 6 hours (Figs. 6F 6G) . It is clearly evident that neither ClC-2 (Fig. 6F) nor ClC-3 (Fig. 6G) colocalizes with membranes in the zone of extracellular space dilations. Although both channels are found in the membranes of deeper fiber cells, the lack of a membrane signal in the zone of extracellular space dilations argues against their involvement in the formation of this damage phenotype. 

The conclusion from our molecular data that ClC-2 is not involved in the formation of extracellular space dilations is also supported by our electrophysiological data. In the present experiments, we found an outwardly rectifying Cl⁻ conductance in cells isolated from the zone of extracellular space dilations, whereas others have shown that ClC-2 forms an inwardly rectifying channel when expressed in cell lines. ²⁴ In contrast, expression of ClC-3 in cell lines produces an outwardly rectifying Cl⁻ channel, ²⁵ which, like other members of the ClC family, exhibits a characteristic Cl⁻ > I⁻ selectivity that distinguishes it from other classes of anion channel ^{25 26} (although conflicting reports exist for halide selectivity of ClC-3 in some preparations ²⁷ ). With this in mind, we performed additional anion-replacement experiments that used I⁻ in addition to gluconate (Fig. 7) . Equimolar replacement of extracellular Cl⁻ with I⁻ caused a significant negative shift in reversal potential from −17.5 ± 2.9 to −55.7 + 4.0 mV (n = 7) and an increase in peak positive current (114 ± 12.3%; n = 7) indicating that the channel(s) underlying the outwardly rectifying conductance prefer I⁻ over Cl⁻. Thus, the expression profiles and biophysical properties of both ClC-2 and -3 indicate that they are unlikely molecular candidates to mediate the outwardly rectifying Cl⁻ conductance that we observed in fiber cells isolated from the zone of extracellular space dilations. 

Discussion

Our previous observation that the exposure of organ-cultured lenses to a variety of Cl⁻ channel inhibitors produces two spatially and morphologically distinct damage phenotypes ⁶ supports the contention that the lens generates a circulating flux of Cl⁻ ions. ⁹ In this present study, we have shown that fiber cells isolated from one of these zones, the zone of extracellular space dilations, contain an outwardly rectifying anion channel(s) which could potentially mediate Cl⁻ ion influx under isotonic conditions. Blockade of this conductance by exposure to Cl⁻ channel inhibitors could lead to trapping of Cl⁻ and water in the extracellular space and the formation of extracellular space dilations. Our ability to characterize the electrophysiological properties of cells from this damage zone was critically dependent on the development of protocols that yielded viable isolated fiber cells and allowed us to identify cells that originate from the damage zone. 

Previous attempts by this laboratory ^{17 18} and others ¹⁵ to patch clamp isolated elongated fiber cells in Ca²⁺-containing Ringer solutions have been hampered by difficulties in obtaining a viable preparation of elongated fiber cells. This, of course, is not surprising, since these cells appear to lack the K⁺ channels and Na⁺ pumps necessary for generating a negative membrane potential. Thus, once isolated from the surface cells, the inner fiber cells have no means of generating a membrane potential and tend to depolarize. This depolarization initiates a cascade of events, which include ion influx, cell swelling, and the Ca²⁺-dependent activation of intracellular proteases and ultimately lead to fiber cell globulization. ²⁸ In the present study, our inclusion of Gd³⁺ and Co²⁺ in the external solution appears to have alleviated fiber cell globulization by blocking the influx of Ca²⁺ that initiates the cascade. 

Blocking this large linear conductance with Gd³⁺ and Co²⁺ revealed an outwardly rectifying anion conductance, which was inhibited by 100 μM DIDS and had permeability sequence of I⁻ > Cl⁻ ≫ gluconate, which we assume reflects the normal properties of lens fiber cells. This contention is supported by electrical measurements conducted by others in whole lenses. ⁹ The membrane conductance calculated from impedance analysis studies on rat lenses were in agreement with those obtain by patch clamping cells at the holding potential. In addition, membrane potential measurements in whole lenses after anion substitution indicate that fiber cells have a significant Cl⁻ conductance. ²⁹ In the present study, ion-replacement experiments with Na⁺ gluconate-based extracellular solutions resulted in a significant shift in reversal potential, indicating that in these elongated fiber cells Cl⁻ contributes to the resting membrane potential of these cells. 

In the present study, an attempt was made to correlate the currents we observed in isolated fiber cells with the expression of the ClC family members ClC-2 and -3, known to be expressed in a cDNA library prepared from human lens epithelium. ¹⁴ Although we were able to show that ClC-2 and -3 are expressed at both the transcript and protein levels in fiber cells of the rat, our immunocytochemical localization experiments showed that neither isoform was found in the plasma membrane of cells from the zone of extracellular space dilations. Furthermore, it has recently become clear that the biophysical properties of these two channels render them unlikely molecular candidates for the outwardly rectifying current we have observed. ^{24 30} ClC-2 channels typically display an inward volume-sensitive current that is activated by hyperpolarization and low pH, that exhibits an anion selectivity of Cl⁻ > I⁻ and that is poorly inhibited by DIDS. In contrast, ClC-3 channels appear to be predominately intracellular channels that, when overexpressed in model systems exhibit outward rectification, an anion selectivity of Cl⁻ > I⁻, and poor inhibition by DIDS. 

The outward rectification, time-dependent activation on depolarization, anion selectivity (I⁻ > Cl⁻ ≫ gluconate), and inhibition by DIDS observed for the conductance in the present study is more characteristic of the Ca²⁺-activated Cl⁻ channel (CaCC) found in a variety of cells types. ²⁴ Because the time-dependent activation kinetics of these channels has been shown to decrease with increasing [Ca²⁺]_i, ³¹ the variable activation kinetics observed in the present study could be accounted for by the inherent variability in the [Ca²⁺]_i levels of isolated fiber cells. Whereas additional experiments are needed to test directly whether the observed Cl⁻ conductance is indeed mediated by CaCC, it raises the possibility that the Cl⁻ conductance in fiber cells may be regulated by intracellular signaling pathways that use Ca²⁺ as a second messenger. 

If not involved in the generation of extracellular space dilations what roles are ClC-2 and -3 performing in the lens, and what is the significance of the observed switch in labeling pattern from cytoplasmic in peripheral young nucleated cells to membranous in deeper mature anucleate fiber cells? Although the present study was not designed to address these questions directly, recent studies into the role of glucose transporters and the putative adhesion molecule MP20 suggest a possible explanation. These studies have shown that peripheral nucleated fiber cells contain a pool of intracellular vesicles that contain glucose transporters ²³ and MP20, ²² which at precise stages in fiber cell differentiation are inserted into the plasma membrane. We have interpreted these findings to mean that the young, nucleated lens fiber cells, which still contain protein synthesis machinery, manufacture the membrane proteins required by the mature denucleated fiber cells. The finding in other tissues that ClC-3 colocalizes with endosomes and synaptic vesicles ^{32 33} is therefore consistent with its playing a role in membrane trafficking in the lens. 

At present, we have no information on whether ClC-2 or -3 channels associated with the membrane in mature fiber cells are functional. If ClC-3 is involved with membrane trafficking, its subsequent appearance in the membranes of older fiber cells could simply reflect the loss of the endocytotic machinery involved in ClC-3 recycling, although we cannot exclude a functional role for ClC-3 in mature fiber cells. In terms of a role for ClC-2 in the lens, it is interesting to note that because of a reliance on anaerobic metabolism and subsequent lactate acid accumulation, mature fiber cells have a pH of 6.8 to 6.5. ^{34 35} Thus, the acidic environment of the lens core is ideally suited for the activation of pH-sensitive ClC-2 channels, suggesting that this channel is likely to play a more active role in mature fiber cells. Further, investigation of the role played by ClC-2 and -3 in mature fiber cells is dependent on obtaining isolated fiber cells or membrane vesicles from these very deep cells that are amenable to electrophysiological analysis. 

In summary, although neither ClC-2 nor -3 appears to be responsible for the outwardly rectifying current we have observed, we now have a biophysical fingerprint for this Cl⁻ conductance that can be used to facilitate the search for other molecular candidates. This conductance mediates Cl⁻ influx and appears to be involved in the operation of the lens internal microcirculation system. Further dissection of the molecular basis and regulation of this conductance therefore appears warranted, because its manipulation in vitro has been shown to affect lens hydration and transparency. ⁶ Whether dysfunction of this conductance(s) is also involved in the processes that lead to the fiber cell damage observed in osmotic cataract ³ remains to be determined. 

 Supported by the Health Research Council of New Zealand, the Lottery Grants Board, and the University of Auckland Research Committee.

 Submitted for publication February 24, 2004; revised June 9 and August 4, 2004; accepted August 13, 2004.

 Disclosure: K.F. Webb, None; B.R. Merriman-Smith, None; J.K. Stobie, None; J. Kistler, None; P.J. Donaldson, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

CIC Channel Primer Sets

Isoform (Accession Number)	Primer Pairs	Expected PCR Product Size (bp)
CIC-2 (X64139) 36	Sense (18 bp, position 889) CGGAACACAGAGATGCTA Antisense (20 bp, position 1122) AGGTCAAAGGGGAAGTCA	223
CIC-3 (D17521) 37	Sense (21 bp, position 1246) GCTAAAAAGAGGGAGGTG Antisense (21 bp, position 2288) CGAGAACTGCCAACAACA	1273

Figure 1.

 

Viability of isolated fiber cell preparations. (A, C) Differential interference contrast images of isolated fiber cell preparations bathed in AAH containing physiological [Ca²⁺]_o, in either the absence (A) or presence (C) of the Gd³⁺/Co²⁺ blocking solution (AAH+GC). (B, D) Whole-cell currents recorded from a single fiber cell in the absence of Gd³⁺/Co²⁺ (B) and after the addition of 3 mM Gd³⁺ (D). Note the absence of large linear leak currents in the cells incubated in the blocking solution enabling the detection of an outwardly rectifying current. The voltage protocol consisted of sequential steps from a holding potential of −40 to ±100 mV in 20-mV increments for 1000 ms each. Scale bar, 100 μm.

Figure 2.

 

Identification of fiber cells from the zone of extracellular space damage. (A) An axial cryosection through the lens modiolus region showing elongating fiber cells at the modiolus. (B) Bright-field image of an isolated fiber cell raft containing a columnar array of newly differentiated fiber cells. (C) Plots of fiber cell length versus cell layer in from the lens capsule obtained from fiber cell rafts (•) or axial sections (○) shows that isolated lens fiber cells are fractionally shorter after dissociation than at the equivalent depth in the intact lens. To ensure that current recordings were obtained from cells in the zone of extracellular space damage, only cells >120 μm in length were used for patch-clamp studies. Scale bar, 50 μm.

Figure 3.

 

Electrical properties of isolated fiber cells. (A) Image of a single isolated fiber cell ∼176 μm in length with a patch pipette attached. (B) Representative current traces from a cell that exhibited an outwardly rectifying current that had minimal delayed activation. Replacement of extracellular Cl⁻ with the impermeant anion gluconate reduced the magnitude of outward current in a partially reversible manner. The voltage protocol consisted of sequential steps from a holding potential of −40 to ±100 mV in 20-mV increments for 1000 ms each.

Figure 4.

 

Electrical properties of isolated fiber cells. Representative current traces from an isolated fiber cell that exhibited an outwardly rectifying current that had substantial delayed activation (A). This current was again reversibly reduced by replacement of extracellular Cl⁻ with the impermeant anion gluconate (B, C) and was blocked by 100 μM DIDS (D). The voltage protocol consisted of sequential steps from a holding potential of −40 to ±100 mV in 20-mV increments for 1000 ms each.

Figure 5.

 

ClC-2 and -3 expression in rat lens fiber cells. Agarose gel showing RT-PCR products. Lane 1: the DNA ladder. Lanes 2 and 4: RT-PCR products derived from lens fiber RNA with primers specific for ClC-2 and -3, respectively. Lanes 3 and 5: the RT-minus controls for ClC-2 and -3, respectively.

Figure 6.

 

Immunolocalization of ClC-2 and -3 in normal rat lenses. Images of equatorial sections double labeled with the membrane marker WGA (red) and antibodies (green) against either ClC-2 (B, C, F) or ClC-3 (D, E, G). (A) Overview image showing the areas from which high-power images from the lens periphery (B, D), zone of extracellular space dilations (F, G), and core (C, E) were obtained. (F, G, arrowheads) The location of extracellular space dilations within the field imaged. Scale bars: (A) 100 μm; (B–E) 10 μm.

Figure 7.

 

Anion selectivity of membrane currents. (A) Fiber cell membrane parameters recorded during a continuous perfusion experiment. One-second repetitive voltage ramps (−100 to +100 mV) were recorded every 5 seconds as the anion composition of the bath solution was systematically varied and the average current at +100 mV (•) and −100 mV (○) plotted against time. (B) Representative current-voltage plots of steady state current recorded for Cl⁻, I⁻, and gluconate solutions indicate a shift in reversal potential.

The authors thank Reiner Eckert for constructive discussions and Juliette Bell and Miriam Young for technical assistance.