HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 280, Issue 27, 25871-25880, July 8, 2005

This Article Abstract Full Text (PDF) All Versions of this Article: 280/27/25871    most recent M502826200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Britton, F. C. Articles by Duan, D. Search for Related Content PubMed PubMed Citation Articles by Britton, F. C. Articles by Duan, D. Social Bookmarking                      What's this?

Functional Characterization of Novel Alternatively Spliced ClC-2 Chloride Channel Variants in the Heart^*

Fiona C. Britton¶, Guan-Lei Wang¶||, Z. Maggie Huang||, Linda Ye||, Burton Horowitz, Joseph R. Hume||, and Dayue Duan||**

From the Center of Biomedical Research Excellence and Departments of,^||Pharmacology and Physiology and Cell Biology, University of Nevada School of Medicine, Reno, Nevada 89557

Received for publication, March 15, 2005 , and in revised form, May 5, 2005.

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
A novel volume-regulated hyperpolarization-activated chloride inward rectifier channel (Cl.ir) was identified in mammalian heart. To investigate whether ClC-2 is the gene encoding Cl.ir channels in heart, ClC-2 cDNAs cloned from rat (rClC-2) and guinea pig (gpClC-2) hearts were functionally characterized. When expressed in NIH/3T3 cells, full-length rClC-2 yielded inwardly rectifying whole-cell currents with very slow activation kinetics (time constants > 1.7 s) upon hyperpolarization under hypotonic condition. The single-channel rClC-2 currents had a unitary slope conductance of 3.9 ± 0.2 picosiemens. A novel variant with an in-frame deletion at the beginning of exon 15 that leads to a deletion of 45 bp (corresponding to 15 amino acids in -helices O and P, rClC-2_509–523) was identified in rat heart. The relative transcriptional expression levels of full-length rClC-2 and rClC-2_509–523 in rat heart were 0.018 ± 0.003 and 0.028 ± 0.006 arbitrary units, respectively, relative to glyceraldehyde-3-phosphate dehydrogenase (n = 5, p = nonsignificant). A similar partial exon 15 skipping with a deletion of 105 bp (35 amino acids in -helices O-Q, gpClC-2_509–543) was also identified in guinea pig heart. Expression of both rClC-2_509–523 and gpClC-2_509–543 resulted in functional channels with phenotypic activation kinetics and many properties identical to those of endogenous Cl.ir channels in native rat and guinea pig cardiac myocytes, respectively. Intracellular dialysis of anti-ClC-2 antibody inhibited expressed ClC-2 channels and endogenous Cl.ir currents in native rat and guinea pig cardiac myocytes. These results demonstrate that novel deletion variants of ClC-2 due to partial exon 15 skipping may be expressed normally in heart and contribute to the formation of endogenous Cl.ir channels in native cardiac cells.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The ClC-2 Cl^- channels are members of the superfamily of voltage-gated Cl^- channels (1, 2). Originally, ClC-2 cDNA was cloned from a rat cDNA library (3), and several alternatively spliced forms were later cloned from several other tissues and species, including human (4–10). Expression of ClC-2 cDNA in Xenopus oocytes or mammalian cells resulted in hyperpolarization-activated inward-rectifying Cl^- currents that are sensitive to changes in cell volume and extracellular acidosis (7, 9, 11–13).

Despite the ubiquitous expression of ClC-2, the exact physiological role of this Cl^- channel remains uncertain and has not been thoroughly investigated in the heart (2, 14, 15). Although ClC-2 cDNA was cloned from rabbit heart and expressed in Xenopus oocytes (9), endogenous electrophysiological counter-parts for ClC-2 in rabbit heart have yet to be determined (9, 14). A novel volume-regulated chloride inward rectifier channel (Cl.ir), which is activated by hyperpolarization and cell swelling, was first discovered in native guinea pig and mouse cardiac atrial and ventricular myocytes (16). Subsequently, similar Cl.ir channels were found in rat cardiac myocytes (17, 18). The molecular properties of cardiac Cl.ir channels, however, are still not known. Cardiac Cl.ir channels share many common properties with ClC-2 Cl^- channels (3, 9), including inward rectification, time-dependent activation at hyperpolarizing voltages, activation by hypotonic cell swelling, blockade by cadmium (Cd²⁺), and a halide selectivity of Cl^- > I^-. The molecular expression of ClC-2 channels in cardiac myocytes at both the mRNA and protein levels has been confirmed in several species, including rabbit, rat, guinea pig, mouse, and dog (9, 16, 19, 20). We therefore proposed that ClC-2 may be a molecular candidate responsible for the native cardiac Cl.ir channels (16).

In this study, we first cloned a full-length cDNA for ClC-2 from rat heart (rClC-2) and characterized the expressed channels at both whole-cell and single-channel levels in a mammalian cell line. More importantly, we identified a novel alternative splice variant of ClC-2 in rat heart (rClC-2_509–523), which has a deletion at the beginning of exon 15 and results in a deletion of 15 amino acids in -helices O and P, without disruption of the ClC-2 protein reading frame. A similar partial exon 15 skipping with a deletion of 105 bp (35 amino acids in -helices O-Q, gpClC-2_509–543) was also found in guinea pig heart. When expressed in mammalian cells, both rClC-2_509–523 and gpClC-2_509–543 yielded functional channels with phenotypic activation kinetics and volume-sensitive properties different from those of the non-deletion form of rClC-2, but very similar to those of endogenous Cl.ir channels in rat and guinea pig heart, respectively. Therefore, our results suggest that endogenous Cl.ir channels in native cardiac cells may be formed by novel ClC-2 variants together with the full-length ClC-2.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Cloning of ClC-2 cDNA from Heart—ClC-2 cDNAs were cloned from rat and guinea pig cardiac muscle by PCR using specific primers (UP1/DN1) that span the entire coding region of the ClC-2 protein (GenBank^TM accession numbers X64139 [GenBank] and AF113529 [GenBank] ; Table I). Total RNA was isolated from heart using TRIzol reagent (Invitrogen). cDNA was prepared using Superscript II reverse transcriptase (Invitrogen). PCRs contained 1x Taq buffer (10 mM KCl, 20 mM Tris-HCl (pH 8.75), 10 mM (NH₄)₂SO₄, 2 mM MgSO_4, 0.1% Triton X-100, and 100 µg/ml bovine serum albumin), 1 mM deoxynucleoside triphosphates, 0.2 µM each primer, 5 µl of cDNA, and 1 unit of Taq polymerase (Promega). The amplification profile was as follows: 95 °C for 3 min (hot start); 30 cycles of 94 °C for 30 s, 68 °C for 20 s, and 72 °C for 2 min; followed by a final extension at 72 °C for 7 min, performed in a GeneAmp 2400 thermal cycler (PerkinElmer Life Sciences). PCR products were gel-purified and subcloned into the pcDNA3.1 vector (Invitrogen). Recombinant plasmid DNA was purified and sequenced at the Nevada Genomics Centre at the University of Nevada. Sequences were analyzed with Vector NTi 7.1 software (Infomax Inc., Bethesda, CA).

509–523

Expression of Cardiac ClC-2 in NIH/3T3 Mammalian Cells—Full-length or spliced ClC-2 cDNAs cloned into the mammalian expression vector pDNA3.1 were either transiently or permanently transfected into NIH/3T3 cells by electroporation or Lipofectamine 2000 as described previously (22, 23). Briefly, for transient transfection, NIH/3T3 cells were harvested by trypsin-EDTA and resuspended at a density of 1 x 10⁶ cells/0.2 ml in an electroporation cuvette (Invitrogen) containing 20 µg of pcDNA3.1/ClC-2 vector and green fluorescent protein (GFP) plasmid (5 µg), as a marker for transfection. After a 10-min incubation on ice, cells were electroporated by applying a 330-V pulse using a pulse generator (Electroporator II; Invitrogen). For permanent transfection, ClC-2 cDNAs in the expression vector pcDNA3.1 were transfected into NIH/3T3 cells by Lipofectamine 2000 (Invitrogen) according to the manufacturer's protocol. Stable transfectants were selected with 1 g/liter G418 (Invitrogen). G418-resistant clones were isolated, expanded, and maintained. Transfected cells expressing ClC-2 were subcultured on glass coverslips for 24–48 h prior to electrophysiological recording. Cells expressing GFP were visualized using an inverted fluorescent microscope with a fluorescein isothiocyanate cube (Zeiss).

Preparation of Single Cardiac Myocytes—Single cardiac myocytes were isolated from left ventricles of adult rat (Sprague-Dawley, male, 150–200 g) or guinea pig (male, 150–200 g) using a well-established method as described previously (22, 24–27). Briefly, the animals were anesthetized by intraperitoneal injection of sodium phenobarbital (50 mg/kg) until a surgical level of anesthesia was confirmed by loss of withdrawal reflex to toe pinch. A thoracotomy was then performed, the heart was quickly removed, and the aorta was cannulated. The heart was retrogradely perfused first with a modified HEPES-buffered Tyrode solution at 37 °C, then perfused with a nominally Ca²⁺-free Tyrode solution (until the heart ceased to beat), and finally perfused with the same solution containing 0.04% collagenase (Type II; Sigma) and 1.0% bovine serum albumin for 10–15 min. The left ventricle was removed and further dissected into small pieces, and cell dissociation was achieved by gentle mechanical agitation in the cell storage solution. Only Ca²⁺-tolerant quiescent myocytes with a typical rod-shaped form and clear cross-striations were used for experiments.

Electrophysiological Measurements—Whole-cell currents were recorded from isolated NIH/3T3 cells or enzymatically isolated rat and guinea pig ventricular myocytes (16, 23) by the tight-seal whole-cell voltage clamp technique as described previously (16, 22, 23, 26, 28). Single-channel currents of expressed recombinant ClC-2 channels were recorded from NIH/3T3 cells using cell-attached patch clamp technique (22, 26). The voltage clamp protocols used in different experiments are described under "Results" and in the figure legends. Experiments were conducted at room temperature (22–24 °C).

Solutions and Drugs—The standard hypotonic external bath solutions used in the electrophysiological recordings in the NIH/3T3 cells contained 125 mM NaCl, 2.5 mM MgCl₂, 2.5 mM CaCl₂, 10 mM HEPES (pH 7.2, 250 mosmol/kg H₂O, [Cl^-]_o = 135 mM). The standard isotonic (300 mosmol/kg H₂O) and hypertonic (350 mosmol/kg H₂O) solutions were the same as hypotonic solution, except the osmolarity was adjusted using mannitol. The pipette solution for cell-attached recordings contained 135 mM N-methyl-D-glucamine (NMDG)-Cl, 5 mM HEPES, and 5.5 mM glucose (pH 7.2, pipette Cl^- concentration ([Cl^-]_p) = 135 mM). The pipette solution for whole-cell recordings contained 135 mM NMDG-Cl, 5 mM MgATP, 2 mM EGTA, 10 mM HEPES (pH 7.2, [Cl^-]_i = 135 mM, 300 mosmol/kg H₂O using mannitol). In the electrophysiological recordings in cardiac myocytes, nisoldipine (1 µM), 4-aminopyridine (2 mM), BaCl₂ (2 mM), and CsCl (10 mM) were consistently present in the extracellular bath solutions, and cations in the intracellular pipette solutions were replaced with the large non-permeable NMDG to prevent contamination from Ca²⁺ and K⁺ currents and the cationic inward currents I_K1 and I_f. The standard hypotonic extracellular bath solutions contained 100 mM NaCl, 10 mM CsCl, 1 mM MgCl₂, 1 mM CaCl₂, 2 mM BaCl₂, 0.33 mM NaH₂PO₄, and 5.0 mM HEPES (pH 7.2, 230 mosmol/kg H₂O, [Cl^-]_o = 118 mM). The standard isotonic (290 mosmol/kg H₂O) and hypertonic (360 mosmol/kg H₂O) solutions were the same as hypotonic solution, except the osmolarity was adjusted using mannitol. The standard pipette solution for whole-cell recordings contained 118 mM NMDG-Cl, 5 mM MgATP, 0.1 mM NaGTP, 5 mM EGTA, and 5 mM HEPES (pH 7.4, [Cl^-]_i = 118 mM, 290 mosmol/kg H₂O using mannitol). In some experiments in the rat ventricular myocytes, the bath solutions contained 100 mM NaCl, 5.4 mM KCl, 1.0 mM MgCl₂, 1.8 mM CaCl₂, 2.0 mM BaCl₂, 0.33 mM NaH₂PO₄, and 5.0 mM HEPES (pH 7.4, 230 mosmol/kg H₂O, [Cl^-]_o = 115 mM). The pipette solution contained 113 mM NMDG-Cl, 5 mM MgATP, 0.1 mM NaGTP, 5 mM EGTA, and 5 mM HEPES (pH 7.4, [Cl^-]_i = 113 mM, 290 mosmol/kg H₂O using mannitol). For intracellular dialysis experiments in NIH/3T3 cells and cardiac myocytes, anti-ClC-2 antibody (Ab) (Alomone Labs) was diluted in double-distilled water to 300 µg/ml and added into the pipette solution (final concentration, 3 µg/ml). The anti-ClC-2 Ab was generated against a peptide corresponding to amino acids 808–906 of rClC-2 protein and shares no homology to other ClC channel proteins. For pre-absorbed anti-ClC-2 Ab, Ab and antigen peptide (rClC-2_888–906) were dissolved separately, mixed in a ratio of 1:3, stored in the refrigerator overnight or at room temperature for >1 h, and added to the pipette solution to achieve a final concentration of Ab and antigen of 3 and 9 µg/ml, respectively. The osmolarity of the pipette dialysis solutions was not significantly altered by inclusion of either Ab alone or pre-absorbed Ab.

Data Analysis—Group data are presented as mean ± S.E. Student's t test was used to determine statistical significance. A two-tailed probability (p) of 5% was considered significant.

View larger version (42K): [in this window] [in a new window]   FIG. 1. Analysis of ClC-2 splice variant transcripts in rat heart. A, model of ClC chloride channel topology with 18 -helical regions (helices A-R), derived from the three-dimensional crystal structure of bacterial ClC protein (adopted from Dutzler et al., Ref. 38). -Helices O and P (dark gray) are involved in the deletion variants of cardiac rClC-2. B, the amino acid sequences corresponding to -helices O-R (in dashed-line rectangles) of rat heart ClC-2 (rClC-2) were aligned with the splice variant cloned from rat heart. Partial skipping of exon 15 (starting from residue 509) resulted in an in-frame deletion of 15 amino acids in rat heart ClC-2 (rClC-2509–523) as indicated by Xs. C, PCR gel analysis of ClC-2 splice variants in rat heart. Primers F1, F2, and R1 (Table I) were designed to amplify across the deletion region. The template for PCR amplification was rat heart cDNA (lane 1) and the isolated ClC-2 cDNA plasmid clones rClC-2 (lane 2) and rClC-2509–523 (lane 3). PCR with F1/R1 primers generates either a 744- or 699-bp fragment, whereas PCR with primers F2/R1 generates either a 388- or 343-bp fragment (lane M, molecular marker). Both ClC-2 variants were amplified from rat heart cDNA by RT-PCR, verifying the authenticity of the deletion. D, quantitation of transcriptional expression by real-time RT-PCR on an ABI5700 sequence detector. The values are relative to GAPDH expression (n = 5).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Identification of ClC-2 Splice Variants in Rat Heart—When the isolated rat heart ClC-2 cDNA clones were sequenced, a short isoform of rClC-2 with a deletion of 45 bp (15 amino acids, 509–523), corresponding to transmembrane-spanning -helices O and P (2), was identified (Fig. 1). Fig. 1A shows a model of ClC channel topology with 18 -helical regions (helices A-R) derived from the three-dimensional crystal structure of bacterial ClC protein (adopted from Dutzler et al., Ref. 38). The -helices O-R (outlined in boxes) and the corresponding amino acid sequences of cardiac ClC-2 in this region are expanded in Fig. 1B. The 15-amino acid deletion present in rClC-2_509–523 is shown as Xs. PCR was performed on rat heart cDNA and on the isolated ClC-2 cDNA clones to further characterize the deletion in rat cardiac ClC-2. PCR primers were designed so that they bind in a region encompassing the 45-bp deletion in ClC-2, thus generating a 45-bp size difference in the amplification products. Primer pair F1/R1 amplifies either a 744- or 699-bp product, whereas primer pair F2/R1 amplifies either a 388- or 343-bp product (Fig. 1C). Both ClC-2 variants are present in rat heart cDNA, verifying the authenticity of the deletion. PCR performed on the isolated cDNA clones indicates the presence of one ClC-2 variant or the other (note the size difference of the PCR products in Fig. 1C). The PCR products were sequenced to confirm the presence of the deletion. The sequence of rClC-2_509–523 was compared with the genomic sequence of rat ClC-2 previously determined by Chu and Zeitlin (5). We found that the 45-bp deletion (509–523) occurred right at the beginning of exon 15. This deletion is not a complete "exon signature" alternative splicing event because it skipped only a portion of exon 15, which is 214 bp long. It may represent a new ClC-2 spliced isoform that has not been reported before in any other tissues of the rat (4, 5, 29). ClC-2 expression in rat heart was further analyzed by the use of real-time quantitative RT-PCR (Fig. 1D). The relative transcriptional expression levels of rClC-2 splice variants in rat heart relative to GAPDH (in arbitrary units) were calculated to be 0.018 ± 0.003 for rClC-2 and 0.028 ± 0.006 for rClC-2_509–523 (n = 5, p = nonsignificant).

Functional Expression of Cardiac rClC-2 Channels in Mammalian NIH/3T3 Cells—To examine the functional properties of the cloned cardiac ClC-2 channels, both full-length and deleted forms of ClC-2 cDNA were heterologously expressed in mammalian NIH/3T3 cells.

We first used the whole-cell patch clamp technique to examine the functional expression of full-length rClC-2 channels in NIH/3T3 cells (Fig. 2). To identify the rClC-2-transfected cells, the pcDNA3.1CT-GFP vector was co-transfected via electroporation. Cells expressing GFP were visualized at 24–48 h using a Zeiss inverted fluorescent microscope with a fluorescein isothiocyanate cube. Large inwardly rectifying currents were activated upon hyperpolarization and cell swelling in cells transfected with rClC-2/GFP (Fig. 2A). The activation of rClC-2 was fit by a bi-exponential time course with a fast component (₁) and a slow component (₂) (Fig. 2A, d). Cell swelling not only significantly increased the amplitude of the current but also accelerated the activation kinetics (Fig. 2A, d). These properties of the expressed rClC-2 current were generally similar to those of expressed brain rClC-2 current in oocytes (3, 30) or mammalian cells (12), but the activation kinetics (Table II) are significantly slower than endogenous Cl.ir current in the heart (16–18). No significant endogenous currents could be detected over the range of test potentials from NIH/3T3 cells transfected with GFP alone under isotonic, hypotonic, and hypertonic conditions (n = 6, Fig. 2B).

View this table: [in this window] [in a new window]   TABLE II Comparison of activation kinetics of ClC-2 and Cl.ir channels The activation process of the currents recorded at -140 mV was best fit to a bi-exponential function with a fast time constant (1) and a slow time constant (2) under hypotonic conditions. The values of the time constants were obtained from fit to the current traces with the curve-fitting program (Clampfit).

View larger version (25K): [in this window] [in a new window]   FIG. 2. Functional expression of full-length rClC-2. A, whole-cell currents recorded from cardiac rClC-2-transfected NIH/3T3 cells under (a) isotonic, (b) hypotonic, and (c) hypertonic conditions. Voltage protocol is shown in the inset. d, representative current recordings from one cell after hyperpolarization to -140 mV under isotonic and hypotonic conditions, respectively. The points represent current activation data, and the solid lines are least square curve fits obtained with the curve-fitting program (Clampfit). The activation process was best fit to a bi-exponential function with a fast time constant (1) and a slow time constant (2) under isotonic conditions. Hypotonic cell swelling increased the amplitude of the current and also accelerated the activation kinetics. Similar results were observed in three other cells tested. Hypotonic cell swelling accelerated the activation kinetics by reducing the mean fast time constant from 87 ± 6 to 67 ± 5 ms (at -140 mV, n = 4, p < 0.05) and reducing the slow time constant from 5419 ± 1091 to 1729 ± 212 ms (n = 4, p < 0.05). e, mean I-V curves of IClC-2 under isotonic, hypotonic, and hypertonic conditions (n = 6). Under isotonic conditions, ClC-2 currents appeared to "run up" after breaking into the cells upon hyperpolarization and usually reached the steady state within 5 min. When cells were exposed to hypotonic or hypertonic solutions, time-dependent changes in ClC-2 current amplitudes were also observed. When the changes reached steady state, the current amplitudes (nA) were measured at 2 s after the corresponding voltage step relative to the 0 current level and then normalized to the capacitance (pF, picofarads) of the cell. B, whole-cell currents recorded from NIH/3T3 cells transfected with GFP alone under (a) isotonic, (b) hypotonic, and (c) hypertonic conditions. The scale for current amplitude and time duration is the same as that in A.

View larger version (18K): [in this window] [in a new window]   FIG. 3. Single channel conductance and open probability of swelling-activated unitary ClC-2 channels. A, representative single-channel current traces of 5-min recording at RP -120 mV in a cell-attached membrane patch of a rClC-2-transfected NIH/3T3 cell. The sampling rate is 2 kHz, and current traces were filtered at 500 Hz for the analysis. B, all-points amplitude histogram. Solid curve shows the Gaussian fits, indicating two active channels (O1 and O2) in this patch. Numbers in parentheses are the proportion of the area under the curve of each conductance level (C, closed). C, mean I-V curves from three to five patches at various testing membrane potentials relative to the RP from RP -120 to +40 mV (in +20-mV increments). Because the current amplitude from -40 to +40 mV was very small and may not be accurate, the slope conductance was calculated from a linear fit of the inward current from RP -60 to -120 mV.

Functional Expression of Cardiac rClC-2_509–523 in NIH/3T3 Cells—The cardiac rClC-2_509–523 is an alternatively spliced isoform of ClC-2, which is predicted to result in ClC-2 proteins that have a deletion of 15 amino acids within conserved ClC protein -helices (helices O and P) (2). This putative cDNA may represent a new heart-specific alternatively spliced variant of ClC-2 because this variant has not been described in other tissues in either rat, guinea pig, rabbit, or pig (4, 5, 7, 9, 10). To test whether or not this deletion would lead to functional channels with phenotypic differences from the non-deletion ClC-2 or the endogenous cardiac Cl.ir channels in heart, we stably expressed rClC-2_509–523 cDNA in NIH/3T3 cells and examined the properties of expressed currents.

To eliminate contamination from any endogenous Cl^- current, parallel control experiments were carefully performed on untransfected cells under identical conditions. As shown in Fig. 5A, expression of rClC-2_509–523 in NIH/3T3 cells yielded large hyperpolarization-activated inwardly rectifying whole-cell currents that were sensitive to changes in cell volume, whereas non-transfected cells exhibited negligible hyperpolarization-activated inward currents (Fig. 5B). However, the slow component of the activation process of rClC-2_509–523 current upon hyperpolarization (Fig. 5A, d) appeared much faster than that of the non-deletion cardiac rClC-2 current (see Fig. 2B, d). At -140 mV, for example, under isotonic conditions, the slow time constant (₂) of rClC-2_509–523 activation was 2127 ± 223 ms (n = 4), whereas ₂ of rClC-2 was 5419 ± 1091 ms (n = 4, p = 0.025), and under hypotonic conditions, ₂ of rClC-2_509–523 was 402 ± 38 ms (n = 4), whereas ₂ of rClC-2 was 1729 ± 212 ms (n = 4, p < 0.001, see Table II). As seen in non-deletion rClC-2, intracellular delivery of anti-ClC-2 Ab (3 µg/ml) via the patch pipette also significantly inhibited the swelling-activated inwardly rectifying currents in rClC-2_509–523-transfected NIH/3T3 cells in a time-dependent manner (Fig. 6A). Intracellular delivery of the same amount of pre-absorbed sera was ineffective in suppressing the rClC-2_509–523 current in the NIH/3T3 cells (Fig. 6B), suggesting that anti-ClC-2 Ab specifically inhibits rClC-2_509–523 current. In addition, rClC-2_509–523 currents were also blocked by extracellular Cd²⁺ (0.5 mM, data not shown).

View larger version (27K): [in this window] [in a new window]   FIG. 4. Inhibition of rClC-2 current by anti-ClC-2 antibody. NIH/3T3 cells transiently transfected with rClC-2/GFP and exposed to hypotonic solution for 15 min before whole-cell recordings. A, representative current traces recorded from GFP-positive cells when pipette solution contained anti-ClC-2 Ab at (a) 0, (b) 10, and (c) 20 min after formation of the whole-cell configuration. d, the average I-V curves (n = 7) recorded under these conditions. B, representative current traces recorded from GFP-positive cells when pipette solution contained pre-absorbed anti-ClC-2 Ab (control) at (a) 0, (b) 10, and (c) 20 min after formation of the whole-cell configuration.

509–523

509–523

509–523

509–523

509–523

View larger version (28K): [in this window] [in a new window]   FIG. 5. Functional expression of deletion form of rClC-2. A, whole-cell currents recorded from NIH/3T3 cells permanently transfected with rClC-2509–523 under (a) isotonic, (b) hypotonic, and (c) hypertonic conditions. Voltage protocol is the same as that described in Fig. 1A. d, the activation process of the channel under isotonic and hypotonic conditions was best fit to a bi-exponential function with a fast time constant (1) and a slow time constant (2) (see Fig. 1). Hypotonic cell swelling accelerated the activation kinetics by reducing 1 from 180 ± 6 to 64 ± 4 ms (at -140 mV, n = 4, p < 0.001) and reducing 2 from 2127 ± 223 to 402 ± 38 ms (n = 4, p < 0.05). e, mean I-V curves of rClC-2509–523 under isotonic, hypotonic, and hypertonic conditions (n = 6). B, whole-cell currents recorded from untransfected NIH/3T3 cells under (a) isotonic, (b) hypotonic, and (c) hypertonic conditions.

509–543

509–523

509–543

509–543

View larger version (31K): [in this window] [in a new window]   FIG. 6. Inhibition of rClC-2509–523 current by anti-ClC-2 antibody. Representative current traces of rClC-2509–523 stably expressed in NIH/3T3 cells. Currents recorded when pipette solution contained anti-ClC-2 antibody alone or pre-absorbed anti-ClC-2 antibody at (a) 0, (b) 10, and (c) 20 min after formation of the whole-cell configuration are shown in A and B, respectively. Cells were exposed to hypotonic solution for 15 min before whole-cell recordings. C, the average I-V curves recorded at 20 min under the same conditions as described in A (n = 6) and B (n = 5).

509–543

509–543

509–543

509–543

509–523

509–543

509–543

509–543

View larger version (36K): [in this window] [in a new window]   FIG. 7. Inhibition of endogenous Cl.ir current in rat ventricular myocytes by anti-ClC-2 antibody. Representative current traces of ICl.ir recorded in rat ventricular myocytes using the voltage protocols shown above A. Currents recorded when pipette solution contained anti-ClC-2 Ab alone or pre-absorbed anti-ClC-2 Ab at (a) 0 and (b) 20 min after formation of the whole-cell configuration are shown in A and B, respectively. Cells were exposed to hypotonic solution for 15 min before whole-cell recordings. The average I-V curves recorded at 20 min under the same conditions as described in A (n = 5) and B (n = 3) are shown in c.

509–543

509–543

509–543

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In this study, we cloned cardiac ClC-2 channels from rat and guinea pig hearts and identified a novel splice variant of ClC-2 channels that contains a deletion within exon 15, corresponding to the conserved -helices O-R, without disruption of the ClC-2 protein reading frame. We also determined the relative expression level of the novel ClC-2 splice variants in rat cardiac tissue. We functionally characterized the properties of cardiac ClC-2 channels in a mammalian cell line, and we have studied for the first time the single channel properties of cardiac ClC-2. The functional expression of the deletion-carrying variants of cardiac ClC-2 revealed phenotypic differences in the activation of the channel and may have functional consequences. These findings provide new molecular tools not only for further study of the physiological function of Cl.ir in heart but also for further understanding of the structure-function relationship of ClC-2 channels.

View larger version (30K): [in this window] [in a new window]   FIG. 8. Functional expression of gpClC-2509–543. A, the amino acid sequences corresponding to -helices O-R (in dashed-line rectangles) of the splice variant cloned from guinea pig heart were aligned with that of the non-deletion form of rClC-2 (see Fig. 1). Partial skipping of exon 15 (starting from residue 509) resulted in an in-frame deletion of 35 amino acids in gpClC-2 (gpClC-2509–543) as indicated by Xs. B, current-voltage (I-V) relationship and volume regulation of whole-cell currents recorded from cardiac gpClC-2509–543-transfected NIH/3T3 cells under (a) isotonic, (b) hypotonic, and (c) hypertonic conditions. Cells were clamped using the voltage protocols shown in the inset. Mean I-V curves (nA/picofarad (pF), n = 6) of IClC-2 under isotonic (Iso), hypotonic (Hypo), and hypertonic (Hyper) conditions. C, whole-cell currents recorded from untransfected NIH/3T3 cells under (a) isotonic, (b) hypotonic, and (c) hypertonic conditions. The scale for current amplitude and time duration is the same as that in B.

Transcript variants of the ClC-2 gene have been identified by a number of other investigators (4, 5, 7, 10). Chu et al. (4) reported the genomic structure of rat ClC-2 and identified that alternative mRNA splice variants of the rat ClC-2 gene are generated by exon skipping. One variant, expressed in many rat tissues, is due to the complete deletion of exon 20 and results in a short ClC-2 transcript (4). Another variant expressed in rat lung is the result of the loss of exon 13 and the first 6 bases of exon 14 (5). This atypical intron-exon splice site was also found to be present in pig, although tissue-specific mRNA expression patterns showed a variable degree of exon 13 skipping (10). In addition, Cid et al. (7) found two alternatively spliced ClC-2 variants in guinea pig intestinal epithelium, testis, and colonic epithelium, which differ in a 10-amino acid deletion located near the N terminus and in 7 amino acids upstream of the first transmembrane domain. These guinea pig splice variants exhibited different functional characteristics when expressed in mammalian HEK-293 cells (7). A short form of ClC-2 truncated at the N terminus was also identified in rabbit heart (19), but this cDNA appears to have been an artifact (9). In this study, we cloned two ClC-2 transcript variants from rat heart. One cardiac rClC-2 transcript was identical to the brain form rClC-2 (3), and the other (rClC-2_509–523) contains a novel deletion of 15 amino acid residues in the -helices O and P. Alignment of our transcripts with the published genomic structure of ClC-2 (5) reveals that this deletion is due to an atypical partial skip in exon 15. Even more interesting, a similar deletion within exon 15 was also observed in guinea pig heart (gpClC-2_509–543). This deletion of 105 bp corresponding to 35 amino acids also occurred at the beginning of exon 15 but was 20 amino acids longer than that observed in rClC-2_509–523. These splicing events thus appear to be due to the failure of the splicing machinery to recognize the acceptor site at the beginning of exon 15 and false recognition of different sequences within exon 15 as an acceptor site. The presence of this deletion was predicted not to disrupt the ClC-2 protein reading frame.

View larger version (42K): [in this window] [in a new window]   FIG. 9. Inhibition of gpClC-2509–543 current in NIH/3T3 cells and endogenous Cl.ir current in guinea pig ventricular myocytes by anti-ClC-2 antibody. Cells were exposed to hypotonic solution for 15 min before whole-cell recordings. A, representative current traces of gpClC-2509–543 stably expressed in NIH/3T3 cells. Currents were recorded when pipette solution contained pre-absorbed anti-ClC-2 Ab (a) or anti-ClC-2 Ab alone (b) at 0 and 20 min after formation of the whole-cell configuration, respectively. B, representative current traces of ICl.ir recorded in guinea pig ventricular myocytes using the voltage protocols shown above A. Currents recorded when pipette solution contained pre-absorbed anti-ClC-2 Ab or anti-ClC-2 Ab alone at 0 and 20 min after formation of the whole-cell configuration are shown in a and b, respectively.

509–523

509–523

Functional Expression of Cardiac ClC-2 Variants—Although the amino acid deletions occurred within conserved -helices O/P and O/P/Q in rClC-2_509–523 and gpClC-2_509–543 respectively, it is interesting that the heterologous expression of both rClC-2_509–523 and gpClC-2_509–543 yielded functional channels. The phenotypic activation kinetics of these deletion variants were similar to those of endogenous Cl.ir channels in native cardiac myocytes (16–18) but differed from those of the full-length cardiac rClC-2 channels expressed in NIH/3T3 cells (see Table II). The activation kinetics of endogenous Cl.ir channels in guinea pig cardiac myocytes (16) and rat cardiac myocytes (17, 18) are much faster than those of full-length rClC-2 but are very similar to those of gpClC-2_509–543 and rClC-2_509–523 expressed in NIH/3T3 cells, suggesting that the deletion variants of ClC-2 may produce the predominant functional patterns of Cl.ir channels in both guinea pig and rat hearts. These results, therefore, provide new sequence and functional evidence for ClC-2 transcript variation in the heart, although it is not known at this point whether the deletion variants in both rat and guinea pig heart represent a heart-specific spliced variant of ClC-2. It is also possible that the splice variants of ClC-2 may combine with full-length ClC-2 or perhaps other protein(s) to form functional Cl.ir channels in native cardiac tissues (11).

The x-ray crystallographic images of bacterial ClC protein show that each ClC subunit contains 18 -helical regions (designated A-R), and those that lie within the membrane are of variable length and are often oblique, with most not spanning the whole membrane (38). Although the crystal structure of bacterial ClC channel proteins allows identification of the amino acid residues lining the conduction pathways and forming the selectivity filter of the ClC channel, all the regions important for channel gating and pore and dimer formation are not definitely known (39). According to the crystal structure (38), the orientation of the ClC subunits of the dimer is such that the P and Q helices of one monomer are in close proximity to the H and I helices of the other. There are currently no experimental data suggesting that mutations in these helices would yield nonfunctional channels. A recent publication by Duffield et al. (40) examined the involvement of the -helices H, I, P, and Q at the dimer interface in ClC-1 common gating. In this thorough study of a series of site-specific mutations within transmembrane helices P and Q, they found that mutations within these helices actually caused changes only in the activation kinetics (particularly in the common gating), but not in the formation of functional ClC-1 channels. Furthermore, it has been recently demonstrated that the ClC-2 ortholog in Caenorhabditis elegans, CLH-3, also has two spliced isoforms (CLH-3a and CLH-3b). One of the shorter variants lacks 261 C-terminal amino acids yet still forms a functional channel with different activation kinetics than the longer variant (41). These independent studies are consistent with our results in that deletion of a part of the O, P, and Q helices, possibly involved in the dimer interaction of ClC-2 channels, changed the activation kinetics (slow gating process) but did not prevent expression of functional channels of rClC-2_509–523 and gpClC-2_509–543. The activation time constants of the deletion ClC-2 variants are very similar to those of endogenous Cl.ir channels in native cardiac myocytes (Table II) and close to the action potential duration, suggesting that the deletion variants may normally contribute to the regulation of cardiac action potential duration (14, 15). Further detailed comparison of the channel properties of different ClC-2 variants from different tissues or species is needed because it may be important for a better understanding of channel function and regulation. The identification of possible functionally distinct phenotypes of ClC-2 channels will certainly provide new insights into channel behavior and regulation in the native environment in heart.

The gating properties of the whole-cell currents of the expressed cardiac rClC-2 channels strongly resemble those of ClC-2 channels expressed in oocytes (7, 9, 11–13, 30) and endogenous I_Cl.ir in cardiac cells (16–18). Anti-ClC-2 Ab blocks all expressed ClC-2 channels and endogenous Cl.ir channels in both rat and guinea pig ventricular myocytes. These results further support the hypothesis that Cl.ir channels in the heart may be encoded by ClC-2.

Single Channel Properties of Cardiac ClC-2—Generally, the unitary channels underlying ClC-2 have not been conclusively identified (2, 8, 35, 42). Inside-out patch recordings of ClC-2-ClC-2 concatemers expressed in Xenopus oocytes revealed pores with a single channel conductance of 2.6 ± 0.1 pS, which showed no bursting kinetics with a slow gating and behaved as a double-barrel-like channel, although the pore stoichiometry of the homomeric channel was not determined (35). ClC-2G expressed in Xenopus oocytes, however, has been reported to be a protein kinase A- and acidosis-activated channel with a linear I-V curve and a conductance of 29 pS at 800 mM CsCl (8). The ClC-2G channels do not behave as a double-barrel channel. In this study, we examined the single channel properties of cardiac rClC-2 channels transiently expressed in NIH/3T3 cells. The channel has a small conductance (3–4 pS) and an inwardly rectifying I-V relationship. The ensemble-averaged current shows a time-dependent slow activation. These properties are very similar to those of endogenous single-channel I_Cl.ir in cultured rat cortical astrocytes (34). The three conductance levels also fit the double-barrel model as proposed for cloned ClC-2 (2, 35, 43) and native Cl.ir channels in rat cortical astrocytes (34), although the possibility that two channels were activated in the same patch cannot be excluded at this point (8, 37). The gating properties of the whole-cell and ensemble-averaging single-channel currents of the expressed cardiac rClC-2 channels strongly resemble those of ClC-2 channels expressed in oocytes (7, 9, 11–13, 30) and endogenous I_Cl.ir in native cardiac cells (16–18), mouse parotid acinar cells (36), and cultured rat cortical astrocytes (34). These results further support the hypothesis that Cl.ir channels in the heart may be encoded by ClC-2.

	FOOTNOTES

 
^* The work was supported in part by National Center of American Heart Association Grant GIA 9950153N; National Heart, Lung and Blood Institute Grant HL63914; and National Center for Research Resources Grant P-20 RR-15581. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ Supported by American Heart Association Western Affiliate Fellowships.

Deceased.

^** To whom correspondence should be addressed: Dept. of Pharmacology, University of Nevada School of Medicine, Manville Medical Bldg., Rm. 9, 1664 N. Virginia St., MS 318, Reno, NV 89557-0270. Tel.: 775-784-4738; Fax: 775-784-1620; E-mail: dduan{at}med.unr.edu.

¹ The abbreviations used are: RT, reverse transcription; pS, picosiemens; GFP, green fluorescent protein; Ab, antibody; NMDG, N-methyl-D-glucamine; RP, resting potential; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

	ACKNOWLEDGMENTS

 
We thank Dr. Jim Kenyon for helpful advice and Horace R. Goff for technical support and assistance during the completion of this study.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Estevez, R., and Jentsch, T. J. (2002) Curr. Opin. Struct. Biol. 12, 531-539[CrossRef][Medline] [Order article via Infotrieve]
2. Jentsch, T. J., Stein, V., Weinreich, F., and Zdebik, A. A. (2002) Physiol. Rev. 82, 503-568[Abstract/Free Full Text]
3. Thiemann, A., Grunder, S., Pusch, M., and Jentsch, T. J. (1992) Nature 356, 57-60[CrossRef][Medline] [Order article via Infotrieve]
4. Chu, S., Murray, C. B., Liu, M. M., and Zeitlin, P. L. (1996) Nucleic Acids Res. 24,3453 -3457[Abstract/Free Full Text]
5. Chu, S., and Zeitlin, P. L. (1997) Nucleic Acids Res. 25,4153 -4159[Abstract/Free Full Text]
6. Cid, L. P., Montrose-Rafizadeh, C., Smith, D. I., Guggino, W. B., and Cutting, G. R. (1995) Hum. Mol. Genet. 4,407 -413[Abstract/Free Full Text]
7. Cid, L. P., Niemeyer, M. I., Ramirez, A., and Sepulveda, F. V. (2000) Am. J. Physiol. Cell Physiol. 279,C1198 -C1210[Abstract/Free Full Text]
8. Malinowska, D. H., Kupert, E. Y., Bahinski, A., Sherry, A. M., and Cuppoletti, J. (1995) Am. J. Physiol. 268,C191 -C200[Medline] [Order article via Infotrieve]
9. Furukawa, T., Ogura, T., Katayama, Y., and Hiraoka, M. (1998) Am. J. Physiol. 274,C500 -C512[Medline] [Order article via Infotrieve]
10. Loewen, M. E., MacDonald, D. W., Gaspar, K. J., and Forsyth, G. W. (2000) Biochim. Biophys. Acta 1493,284 -288[Medline] [Order article via Infotrieve]
11. Jordt, S. E., and Jentsch, T. J. (1997) EMBO J. 16,1582 -1592[CrossRef][Medline] [Order article via Infotrieve]
12. Park, K., Arreola, J., Begenisich, T., and Melvin, J. E. (1998) J. Membr. Biol. 163, 87-95[CrossRef][Medline] [Order article via Infotrieve]
13. Stroffekova, K., Kupert, E. Y., Malinowska, D. H., and Cuppoletti, J. (1998) Am. J. Physiol. 275,C1113 -C1123[Medline] [Order article via Infotrieve]
14. Hume, J. R., Duan, D., Collier, M. L., Yamazaki, J., and Horowitz, B. (2000) Physiol. Rev. 80, 31-81[Abstract/Free Full Text]
15. Duan, D., Liu, L. L., Bozeat, N., Huang, Z. M., Xiang, S. Y., Wang, G. L., Ye, L., and Hume, J. R. (2005) Acta Pharmacol. Sin. 26,265 -278[CrossRef][Medline] [Order article via Infotrieve]
16. Duan, D., Ye, L., Britton, F., Horowitz, B., and Hume, J. R. (2000) Circ. Res. 86, E63-E71[Medline] [Order article via Infotrieve]
17. Komukai, K., Brette, F., and Orchard, C. H. (2002) Am. J. Physiol. Heart Circ. Physiol. 283,H715 -H724[Abstract/Free Full Text]
18. Komukai, K., Brette, F., Pascarel, C., and Orchard, C. H. (2002) Am. J. Physiol. Heart Circ. Physiol. 283,H412 -H422[Abstract/Free Full Text]
19. Furukawa, T., Horikawa, S., Terai, T., Ogura, T., Katayama, Y., and Hiraoka, M. (1995) FEBS Lett. 375, 56-62[CrossRef][Medline] [Order article via Infotrieve]
20. Britton, F. C., Hatton, W. J., Rossow, C. F., Duan, D., Hume, J. R., and Horowitz, B. (2000) Am. J. Physiol. Heart Circ. Physiol. 279,H2225 -H2233[Abstract/Free Full Text]
21. Britton, F. C., Ohya, S., Horowitz, B., and Greenwood, I. A. (2002) J. Physiol. 539,107 -117[Abstract/Free Full Text]
22. Duan, D., Winter, C., Cowley, S., Hume, J. R., and Horowitz, B. (1997) Nature 390,417 -421[CrossRef][Medline] [Order article via Infotrieve]
23. Duan, D., Cowley, S., Horowitz, B., and Hume, J. R. (1999) J. Gen. Physiol. 113, 57-70[Abstract/Free Full Text]
24. Duan, D., Fermini, B., and Nattel, S. (1992) Am. J. Physiol. 263,H1967 -H1971[Medline] [Order article via Infotrieve]
25. Duan, D., Fermini, B., and Nattel, S. (1995) Circ. Res. 77,379 -393[Abstract/Free Full Text]
26. Duan, D., Hume, J. R., and Nattel, S. (1997) Circ. Res. 80,103 -113[Abstract/Free Full Text]
27. Nagasaki, M., Ye, L., Duan, D., Horowitz, B., and Hume, J. R. (2000) J. Physiol. 523, Pt 3,705 -717[Abstract/Free Full Text]
28. Hamill, O. P., Marty, A., Neher, E., Sakmann, B., and Sigworth, F. J. (1981) Pfluegers Arcv. Eur. J. Physiol. 391,85 -100
29. Varela, D., Niemeyer, M. I., Cid, L. P., and Sepulveda, F. V. (2002) J. Physiol. 544,363 -372[Abstract/Free Full Text]
30. Grunder, S., Thiemann, A., Pusch, M., and Jentsch, T. J. (1992) Nature 360,759 -762[CrossRef][Medline] [Order article via Infotrieve]
31. Ehring, G. R., Osipchuk, Y. V., and Cahalan, M. D. (1994) J. Gen. Physiol. 104,1129 -1161[Abstract/Free Full Text]
32. Luckie, D. B., Krouse, M. E., Harper, K. L., Law, T. C., and Wine, J. J. (1994) Am. J. Physiol. 267,C650 -C658[Medline] [Order article via Infotrieve]
33. Pedersen, S. F., Beisner, K. H., Hougaard, C., Willumsen, B. M., Lambert, I. H., and Hoffmann, E. K. (2002) J. Physiol. 541,779 -796[Abstract/Free Full Text]
34. Nobile, M., Pusch, M., Rapisarda, C., and Ferroni, S. (2000) FEBS Lett. 479, 10-14[CrossRef][Medline] [Order article via Infotrieve]
35. Weinreich, F., and Jentsch, T. J. (2001) J. Biol. Chem. 276,2347 -2353[Abstract/Free Full Text]
36. Arreola, J., Begenisich, T., and Melvin, J. E. (2002) J. Physiol. 541,103 -112[Abstract/Free Full Text]
37. Fahlke, C., Rhodes, T. H., Desai, R. R., and George, A. L. J. (1998) Nature 394,687 -690[CrossRef][Medline] [Order article via Infotrieve]
38. Dutzler, R., Campbell, E. B., Cadene, M., Chait, B. T., and MacKinnon, R. (2002) Nature 415,287 -294[CrossRef][Medline] [Order article via Infotrieve]
39. Dutzler, R., Campbell, E. B., and MacKinnon, R. (2003) Science 300,108 -112[Abstract/Free Full Text]
40. Duffield, M., Rychkov, G., Bretag, A., and Roberts, M. (2003) J. Gen. Physiol. 121,149 -161[Abstract/Free Full Text]
41. Denton, J., Nehrke, K., Rutledge, E., Morrison, R., and Strange, K. (2004) J. Physiol. 555, 97-114[Abstract/Free Full Text]
42. Fahlke, C. (2001) Am. J. Physiol. Renal Physiol. 280,F748 -F757[Abstract/Free Full Text]
43. Middleton, R. E., Pheasant, D. J., and Miller, C. (1996) Nature 383,337 -340[CrossRef][Medline] [Order article via Infotrieve]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				H. F. Bao, L. Liu, J. Self, B. J. Duke, R. Ueno, and D. C. Eaton A synthetic prostone activates apical chloride channels in A6 epithelial cells Am J Physiol Gastrointest Liver Physiol, August 1, 2008; 295(2): G234 - G251. [Abstract] [Full Text] [PDF]

				D. T. McCloskey, L. Doherty, Y.-P. Dai, L. Miller, J. R. Hume, and I. A. Yamboliev Hypotonic Activation of Short ClC3 Isoform Is Modulated by Direct Interaction between Its Cytosolic C-terminal Tail and Subcortical Actin Filaments J. Biol. Chem., June 8, 2007; 282(23): 16871 - 16877. [Abstract] [Full Text] [PDF]

				L. P. Cid, M. I. Niemeyer, and F. V. Sepulveda ClC-2 channels get new partners. Focus on "Association between Hsp90 and the ClC-2 chloride channel upregulates channel function" Am J Physiol Cell Physiol, January 1, 2006; 290(1): C42 - C44. [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 280/27/25871    most recent M502826200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Britton, F. C. Articles by Duan, D. Search for Related Content PubMed PubMed Citation Articles by Britton, F. C. Articles by Duan, D. Social Bookmarking                      What's this?