Activation of the Amiloride-Sensitive Epithelial Sodium Channel by the Serine Protease mCAP1 Expressed in a Mouse Cortical Collecting Duct Cell Line

Culture of mpkCCD_c14 Cells and Short-Circuit Current Measurements

Cells were grown in a hormonally defined Dulbecco's modified Eagle's medium:Ham's F12 (1:1) medium supplemented with 2% fetal calf serum as described by Bens et al. (10) in an atmosphere of humidified air/5% CO₂ at 37°C. To ensure high rates of basal transport, cells (between passages 24 and 35) were cultured for 5 consecutive days on collagen-coated Transwell permeable filters (Costar) with complete medium followed by a 4- to 8-d culturing in minimal medium lacking fetal calf serum, epidermal growth factor, and transferrin (10). Short-circuit current (I_sc) measurements were performed using a modified Ussing chamber (9). The mpkCCD_c14 cells were incubated with increasing concentrations (5, 50, 500, and 5000 μg/ml) of aprotinin in the luminal bath. After 18 h of incubation, I_sc was measured, then 5 μM amiloride was added to the apical side of the filter and I_sc was measured again after 30 min incubation. The amiloride-sensitive short circuit current was deduced using these two values.

Identification and Isolation of a Novel Full-Length Murine Serine Protease cDNA

Reverse transcription (RT)-PCR was performed on total RNA from mpkCCD_c14 cells. Briefly, 3 μg of total RNA was treated with DNase I (Boehringer Mannheim, Germany) followed by reverse transcription using Superscript II (Life Technologies) and random primers (Pharmacia). PCR was performed using 1/10 of the cDNA digested with RNase H (Life Technologies) at 55°C for 15 min. PCR was done in 50-μl reactions containing 1.5 mM MgCl₂, 50 mM KCl, 10 mM Tris-HCl, pH 8.3, 150 μM of each deoxynucleoside 5′-triphosphate, 5% DMSO, 0.5 μM of each primer, and 2.5 U of Taq DNA polymerase (Boehringer Mannheim). Forty cycles were run using primers D1 and D2 each consisting of 30 s at 94°C, 30 s at 48°C, and 1 min at 72°C. After the last cycle, elongation was allowed to proceed for 7 min at 72°C. A second PCR was performed using 1/20 of the first PCR reaction with nested primers D1 and D3 under the same conditions. The 512-bp amplified PCR product was separated on a 1.2% agarose gel, extracted, and purified (Qiagen, Chatsworth, CA), subcloned into pT7Blue Vector (Blunt Cloning Kit, Novagen, Madison, WI), and sequenced. The primers used were as follows: D1 (sense) 5′-AA(AG)TT(CT)CCITGGCA(AG)GT-3′, nt +118 to +134; D2 (antisense) 5′-CC(AG)CA(CT)TC(AG)TCICCCCA-3′, nt +743 to +727; D3 (antisense) 5′-CCIGC(AG)CA(AGT)ATCAT(AG)TC-3′, nt +629 to +613; according to xCAP1 (5′ and 3′).

Rapid Amplification of cDNA Ends

For 5′ rapid amplification of cDNA ends (5′-RACE), after reverse transcription of mpkCCD_c14 total mRNA (as described above), mCAP1-specific reverse primers RC1 (5′-CTTCACCTCATACGCTTCC-3′, nt position +297 to +279) and RC2 (5′-GTTGCCATCGTAGGTGATG-3′, nt position +201 to +183) were used sequentially with adapter-specific primer dC-5R (5′-GCATGCTCGAGCGGCCGCAACCCCCCCCCCCCCCCCCC-3′) and 5R (5′-GCATGCTCGAGCGGCCGCAAC-3′), according to the manufacturer's instructions (Life Technologies). For 3′-RACE, reverse transcription was performed using an oligo(dT) adapter primer (5′-CGAGATCTATGCGGCCGCTTTTTTTTTTTTTTTTTTTT-3′). mCAP1-specific primers RC3 (5′-TCAGTGAGCCTCCAGACC-3′, nt position +535 to +552) and RC4 (5′-CTGTAGCTGCCTGTACAA-3′, nt position +600 to +617) were used. To amplify the full-length cDNA from mpkCCD_c14 cells, primer 1 (sense 5′-TGCCTTCAAAACCAGCCTTC-3′, nt position -73 to -54) and primer 2 (antisense 5′-TGTCCTTGGGTGTGCTGTG-3′, nt position +1083 to +1065; mCAP1; accession no. AF 188613) covering the open reading frame of mCAP1 were used according to the manufacturer's instructions (High Fidelity Expand PCR System, Boehringer, Germany). The 1156-bp PCR product was purified by agarose gel electrophoresis, cloned into pT7Blue Vector (Blunt Cloning Kit, Novagen), and confirmed by sequencing.

Northern Blot Analysis

Total RNA (20 μg) extracted from various mouse tissues were run on a 0.8% denaturing glyoxal agarose gel and blotted onto nylon membranes (Hybond-N, Amersham). Membranes were hybridized with random-primed ³²P-labeled probes for mCAP1 (512 bp: nt +163 to +674), αrENaC, and rat GAPDH.

RT-PCR Analysis

Total RNA was prepared from whole mouse kidney, confluent mpkCCD_c14 cells, and microdissected tubules. Thin kidney slices from 1-mo-old mice (n = 3) fed with a standard diet were incubated in Dulbecco's modified Eagle's medium:Ham's F12 (1:1) supplemented with 0.1% collagenase (Boehringer Mannheim) for 1 h at 37°C. S2 (proximal convoluted tubule [PCT]) and S3 (pars recta [PR]) segments from proximal convoluted tubules, medullary and cortical ascending limbs of Henle's loop, distal convoluted tubules, and cortical (CCD) and inner medullary collecting ducts were microdissected as described (11). Single microdissected tubules (0.3 to 0.5 mm long) from each segment were pooled (five to eight single tubules) and stored at -80°C before use. Total RNA was extracted from whole kidney, cultured cells, and microdissected tubules using the RNA-PLUS extraction kit (Bioprobe Systems, Montreuil-sous-bois, France). The RNA were reverse-transcribed with Moloney murine leukemia virus reverse transcriptase (Life Technologies, Eragny, France) at 42°C for 45 min. Two hundred fifty nanograms of whole kidney and mpkCCD_c14 cell cDNA and non-reverse-transcribed RNA were amplified for 32 cycles in 100 μl of total volume containing 1.5 mM MgCl₂, 50 mM KCl, 20 mM Tris-HCl, pH 8.4, 40 μM dNTP, 1 U Taq polymerase, 30 pmol of mCAP1 primer 1 and RC1. Microdissected tubule cDNA were amplified for 38 cycles using the same buffer containing 4 pmol of mCAP1 primers and 0.45 pmol of β-actin primers (5′-CGTGGGCCGCCCTAGGCACCA-3′ and 5′-TTGGCCTTAGGGTTCAGGGGGG3′ as described previously) (10). The thermal cycling program was as follows: 94°C for 30 s, 57°C for 30 s, and 72°C for 1 min. Amplification products were run on 3.5% agarose gel, stained with ethidium bromide, and photographed, or were run on 4% polyacrylamide gel and autoradiographed. The identity of the amplified products was controlled by digestion with NciI (Life Technologies).

Electrophysiologic Measurements in Xenopus Oocytes

For functional expression studies, mCAP1 cDNA was subcloned into pSD5 expression vector and in vitro-transcribed (7). Expression studies were performed in stage V/VI occytes isolated from Xenopus laevis (Noerdhoek, South Africa). Routinely, oocytes were injected with 0.33 ng of each cRNA coding for the rat α-, β-, and γ-ENaC subunits in the presence or absence of 2 ng of cRNA coding for mCAP1 in a total volume of 100 nl. Oocytes were incubated overnight in modified Barth saline solution in the presence or absence of 100 μg/ml aprotinin. Electrophysiologic measurements were performed 24 to 48 h after cRNA injection using the two-electrode voltage-clamp technique, and the amiloride-sensitive current (I_Na) was measured in the presence of 120 mM of Na⁺ in frog Ringer's solution with 5 μM amiloride at a holding potential of -100 mV. The oocytes were perfused with 2 μg/ml trypsin during 2 to 3 min and I_Na was remeasured. Amiloride sensitivity of injected sodium channel subunits has been measured with 0.001, 0.01, 0.1, 1, and 10 μM amiloride in the presence of 120 mM Na⁺. Cell surface expression has been performed as described (12). Briefly, 1 ng of tagged rat α-, β-, and γ-ENaC cRNA were expressed in oocytes either coinjected with 2 ng of mCAP1 cRNA or water-injected. The density of the channel was then estimated by binding of iodinated anti-FLAG (M2) monoclonal antibody (Sigma, Buchs, Switzerland).

Statistical Analyses

All results are reported as means ± SEM. Comparing independent sets of data, unpaired t tests were used to determine significance. In experiments in which oocytes were perfused with trypsin, paired t tests were used. P < 0.01 was considered significant. n represents the number of experiments performed.

Aprotinin Inhibits Amiloride-Sensitive Short Circuit Current (I_sc) in mpkCCD_c14 Cells

The mouse mpkCCD_c14 cell line has been derived from the mouse CCD (10). In these cells, the sodium transport (measured as a short circuit current [I_sc]) is 77 ± 12 μA/cm² (n = 20). When 5 μM amiloride was added to the luminal bathing solution, the I_sc dropped to 8 ± 2 μA/cm² (n = 20; P < 0.01), suggesting that most of the measured I_sc is due to an electrogenic ENaC-mediated Na⁺ transport. Overnight exposure of the apical side of mpkCCD_c14 cells with increasing concentrations of aprotinin, an inhibitor of serine proteases, reduced the amiloride-sensitive I_sc by about 50% (Figure 1). These results indicated that ENaC-mediated sodium transport in this mouse cortical collecting duct cell line might be controlled in part by serine proteases.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Aprotinin decreases I_sc of amiloride in mpkCCD_c14 cells. Measurements of the proportional amiloride-sensitive I_sc after overnight exposure of the apical side of mpkCCD_c14 cells with gradual concentrations of aprotinin (0 μg/ml, n = 20; 5 μg/ml, n = 8; 50 μg/ml, n = 14; 500 μg/ml, n = 14; 5000 μg/ml, n = 4; 5 μM amiloride, n = 20).

Identification of the Murine Homologue for xCAP1

RT-PCR analysis of mpkCCD_c14 cells using degenerated oligonucleotides to xCAP1 revealed a 512-bp PCR product. A 1768-bp cDNA (without polyA⁺) was obtained by 5′- and 3′-RACE (Figure 2). This clone contained an open reading frame of 339 amino acids. Alignment of the predicted amino acid sequence (Genetic Computer Group, Inc., Madison, WI) revealed that the deduced protein belongs to the serine protease family. It shows high homology with the trypsin-like protease family. It shared 50% homology with the channel activating protease xCAP1, 46% with the mouse trypsinogen precursor, and 80% with the human prostasin (Figure 2). The N terminus exhibits a putative cleavable signal peptide and the C terminus a hydrophobic domain with a presumed GPI membrane-anchoring site known to act as a dominant apical target signal in polarized epithelial cells (6). Therefore, this new mouse serine protease was named mCAP1 for mouse channel activating protease 1.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Mouse CAP1 (mCAP1) exhibits high homology with the trypsin-like protease family. Alignment of the amino acid sequence of mCAP1 with other serine proteases: human prostasin (1 to 343), Xenopus xCAP1 (1 to 329), and mouse trypsinogen precursor (1 to 246). Gray and white boxes indicate identical and similar residues, respectively. The catalytic triad (H₈₅, D₁₃₄, and S₂₃₈) is indicated by asterisks. Signal sequences in the N terminus and hydrophobic domain in C terminus of the protein are shown by black lines.

Tissue and Intrarenal mCAP1 mRNA Expression

Northern blot analysis revealed strong hybridization signals (about 1.8 kb) in kidney, lung, and salivary glands and a weak signal in distal colon, as well as in mpkCCD_c14 cells (Figure 3), and in skin, stomach, duodenum, small intestine, colon, bladder, and prostate (data not shown). There was no detectable hybridization signal in liver, heart, and brain (Figure 3), or in spleen, testis, ovary, pancreas, and muscle (data not shown). Hybridization with mouse α-ENaC subunit (lower band) revealed coexpression with mCAP1 in the relevant tissues (kidney, mpkCCD_c14, colon, lung, salivary gland). The upper band corresponds to a cross-hybridization to a 28S RNA (Figure 3).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Northern blot analysis of mCAP1 (1.8 kb) and αENaC (3.5 kb) mRNA expression in mpkCCD_c14 cells and various mouse organs revealed coexpression in the same tissues. RNA integrity was controlled with a rat GAPDH cDNA probe. After hybridization with αmENaC, a nonspecific band (upper band) appeared in all tissues.

To map precisely the distribution of mCAP1 along the nephron, we used RT-PCR on microdissected nephron segments. The sets of mCAP1 primers used permitted detection of amplified products in both whole mouse kidney and cultured mpkCCD_c14 cells. The specificity of the amplified products was confirmed by their digestion with the NciI restriction enzyme (Figure 4A). Because the intrarenal distribution of proteases is not known, we have analyzed the expression of mCAP1 along the mouse nephron (Figure 4B). The convoluted (PCT) and straight terminal portion (PR) of the proximal tubule exhibited the highest mCAP1 transcript abundance when compared with the levels of β-actin transcripts. Substantial amounts of mCAP1 transcripts were also detected in cortical ascending limbs of Henle's loop and CCD (Figure 4B). In contrast, very low levels of mCAP1 expression were detected in medullary ascending limbs of Henle's loop and inner medullary collecting duct as well as in distal convoluted tubules.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

mCAP1 expression in whole mouse kidney, mpkCCD_c14 cells, and along the renal tubule. (A) cDNA and non-reverse-transcribed RNA extracted from whole kidney (lanes 1 to 3) and cultured mpkCCD_c14 cells (lanes 4 to 6). Amplified products for mCAP1 (372 bp) were obtained in mouse kidney (lane 1) and cultured cells (lane 4). Digestion of the amplified products with NciI yielded two bands (231 bp and 141 bp) in both mouse kidney (lane 3) and cultured cells (lane 6). No amplified products were detected with non-reverse-transcribed RNA (lanes 2 and 5) or by omitting cDNA (lane 7). Fs, molecular weight marker. (B) Representative illustrations of three independent autoradiograms of ³²P-labeled amplified products from proximal tubules (PCT and PR), medullary ascending limb (mTAL) and cortical ascending limb (cTAL) of Henle's loop, distal convoluted tubule (DCT), cortical collecting duct (CCD), and inner medullary collecting duct (IMCD), using mCAP1 and β-actin primers. No labeled bands were detected by omitting cDNA (C).

Functional Analysis of mCAP1 in Xenopus Oocytes

Coexpression of mCAP1 with each of the ENaC subunits from Xenopus (xENaC), rat (rENaC), and human (hENaC) led to an approximately sixfold significant (P < 0.01, unpaired t test) increase in the macroscopic amiloride-sensitive current (xENaC I_Na: 0.2 ± 0.02 μA [n = 21] versus 1.01 ± 0.04 μA [n = 20]; rENaC I_Na: 0.75 ± 0.04 μA [n = 19] versus 5.03 ± 0.16 μA [n = 21]; hENaC I_Na: 2.41 ± 0.07 μA [n = 21] versus 12.12 ± 0.26 μA [n = 21]) (Figure 5A). In the absence of α-, β-, and γ-ENaC subunits, mCAP1 injected into oocytes did not generate measurable amiloride-sensitive current (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

Figure 5.

Functional analysis in Xenopus oocytes. (A) Effect of mCAP on I_Na in oocytes. Oocytes were injected with Xenopus (left), rat (middle), and human (right) ENaC subunits in the presence of mCAP1 (filled bars) or water (open bars). (B) Effect of perfusion of trypsin (closed bars, lanes 2, 4, 6, and 8) and preincubation with aprotinin (lanes 3, 4, 7, and 8) of oocytes injected with rat ENaC in the presence (lanes 5 to 8) or absence of mCAP1 (lane 1 to 4). The amiloride-sensitive current I_Na increases in the presence of mCAP1 in oocytes incubated in modified Barth saline solution (MBS) (lane 1 versus lane 5). Preincubation of the oocytes in MBS containing 100 μg/ml aprotinin prevented the effect of mCAP1 on I_Na (lanes 3 and 7), and perfusion of 2 μg/ml trypsin during the recording had no additive effect on the increase of the I_Na due to mCAP1 expression (lane 5 versus lane 6). Trypsin induced a large increase with maximum intensity after 2 to 3 min of perfusion in the absence of mCAP1 (lane 2), and this effect was not repressed by preincubation of the oocytes with aprotinin in the absence (lane 4) or presence (lane 8) of mCAP1. (C) I_Na (left) and cell surface expression of ENaC (right panel) in the presence (filled bars, n = 25) or absence (open bars, n = 29) of mCAP1.

We then compared the amiloride sensitivity of ENaC channel activity consisting of rat α-, β-, and γ-ENaC subunits coinjected with mCAP1 cRNA into Xenopus oocytes. The Ki_amil of ENaC activity alone (Ki_amil: 0.106 ± 0.008 μM; n = 18) did not significantly differ (P > 0.1, unpaired t test) from the Ki_amil in oocytes coexpressing all ENaC subunits and mCAP1 (Ki_amil: 0.91 ± 0.005 μM; n = 17). We conclude that the concentration of amiloride used in our assay is sufficient to block effectively ENaC-dependent sodium transport (over 95%) and that the effect of mCAP1 is therefore not underestimated.

Next, we tested the effects of trypsin and aprotinin on amiloride-sensitive ENaC-mediated sodium current in Xenopus oocytes injected with all three subunits of the rat (α-, β-, and γ-ENaC) in the presence (Figure 5B, lanes 5 to 8) or absence of mCAP1 cRNA (Figure 5B, lanes 1 to 4). In the absence of aprotinin, mCAP1 induced a five- to sixfold increase of the amiloride-sensitive sodium current (I_Na) (Figure 5B, lane 1 versus lane 5). Trypsin (2 μg/ml) was able to activate ENaC in the absence of mCAP1 (Figure 5B, lane 1 versus lane 2) but did not further increase the I_Na stimulated by mCAP1 (Figure 5B, lane 5 versus lane 6). When oocytes were incubated overnight with 100 μg/ml aprotinin, the increase of I_Na by mCAP1 was inhibited (Figure 5B, lane 5 versus lane 7). The inhibitory effect of aprotinin was only effective in the presence of mCAP1, as it was not observed in its absence (Figure 5B, lane 1 versus lane 3). This finding suggests that the oocyte has very little endogenous proteolytic activity able to activate ENaC. The effect of aprotinin on I_Na could be reversed by trypsin in noninjected or injected mCAP1 oocytes (Figure 5B, lane 3 versus lane 4 and lane 7 versus lane 8). The lack of additive effect of trypsin and mCAP1 and the inhibition of mCAP1 activation by aprotinin and its reversal by trypsin strongly suggested that mCAP1 and trypsin acted through a common proteolytic pathway. mCAP1 could increase the amiloride-sensitive sodium current mediated by ENaC by two mechanisms: either by changing the electrophysiologic properties of the channel (conductance, ion selectivity, gating kinetics) or by increasing its cell-surface expression.

Coexpression of FLAG-tagged α-, β-, and γ-ENaC subunits and mCAP1 cRNA in oocytes produced a fourfold increase (4.4 ± 0.7; n = 25) in I_Na (Figure 5C, left panel). This increase of I_Na using the tagged ENaC subunits was lower than the increase using nontagged subunits. Higher basal currents measured under those conditions might be responsible for this difference in decreasing the effect of the protease (13). We also observed a twofold decrease (1.8 ± 0.1; n = 25) in expression of the channel molecules at the cell surface (Figure 5C, right panel). These results suggested that the intrinsic channel activity was increased in the presence of mCAP1.

mCAP1 Has the Expected Properties for a Mammalian Serine Protease Regulating ENaC Activity in CCD Cells

mCAP1 shares 50% homology with Xenopus CAP1 (6) and 80% with prostasin (8). These three serine proteases share the same predicted structure, i.e., a cleavable signal sequence at the amino terminus and a consensus sequence for GPI membrane anchoring site, known to target proteins at the apical membrane of epithelial kidney cells (6). mCAP1 shares the same functional properties as xCAP1, i.e., its ability to activate ENaC from three different species (Xenopus, rat, and human). The activation by mCAP1 can be inhibited by aprotinin. Recently, Nakhoul et al. (14) have found that aprotinin added to the luminal side of cultured mouse collecting duct M-1 cells induces a 50% decrease of I_sc, similar to that found in the cultured mpkCCD_c14 cells (Figure 1). As aprotinin fully inhibits the I_sc measured in Xenopus A6 cells (6), mCAP1 appears to be less sensitive to aprotinin than its amphibian homologue, suggesting that the binding site for the serine protease inhibitor may be somewhat different. Alternatively, mpkCCD_c14 cells may express other proteolytic enzymes that would be resistant to aprotinin. Additional pharmacologic studies should clarify this point.

In a previous study (6), we showed that xCAP1 increased the amiloride-sensitive sodium current but induced a small but nonsignificant decrease (-20%) in the binding of anti-FLAG antibodies. Because the increase in I_Na could not be explained by a change in the single channel conductance or in the cell surface expression of ENaC, we suggest that the increase in I_Na was due to an increase in the overall open probability of the channel molecules present at the oocyte surface (12,13,15).

Like xCAP1, mCAP1 appears to activate preexisting channels located in the plasma membrane, because the number of molecules expressed at the surface is not increased by mCAP1 but is significantly decreased (by 50%). According to this assay, we propose that mCAP1 has a major effect on the overall open probability of the channel if one assumes that mCAP1 has no effect on ion selectivity and/or single channel conductance.

mCAP1 Is Probably the Mouse Counterpart of Human Prostasin

According to the data presented in this study, mCAP1 mRNA transcripts are expressed at high levels in epithelial tissues that are known to be the site of an amiloride-sensitive transepithelial sodium transport (CCD, colon, lung, salivary gland, skin, bladder). mCAP1 is also expressed at low levels in other epithelial tissues, such as the stomach, duodenum, and small intestine, where ENaC might be expressed under some special pathophysiologic conditions. It is not yet known whether human prostasin can activate ENaC in the kidney, the colon, or the lung. However, preliminary experiments have shown that it is the case in the Xenopus oocyte system (unpublished observation). On the basis of tissue-specific expression and functional studies, we would therefore like to propose that mCAP1, xCAP1, and prostasin are the corresponding genes for the three species studied so far.

Regulation of ENaC Activity and Sodium Transport

Using isolated microdissected tubules, the results from RT-PCR experiments clearly indicate that mCAP1 is expressed at high levels in the proximal tubule and to a lesser extent in the distal part of the nephron and the CCD (Figure 4). Although we have not analyzed the expression of mCAP1 at the protein level, the high expression of mCAP1 in the proximal tubule could indicate a local mode of action. In the rat, ENaC transcripts and ENaC activity are specifically found in the S3 segment of the proximal tubule, and the presence of an amiloride-sensitive sodium transport supports this finding (16). In the present study, mCAP1 appears to be expressed all along the proximal tubule (from S1 to S3). Another interesting and more intriguing possibility is that mCAP1 is secreted into the tubular fluid of the proximal tubule to activate ENaC expression in cells of the distal part of the nephron. Such a distant mode of action has also been proposed for other secreted proteases such as kallikrein (17). The second possibility predicts that mCAP1 is synthesized in cells of the CCD, then targeted to the apical membrane to control in situ the activity of ENaC. This is supported by: (1) the detection of mCAP1 mRNA transcripts in the collecting duct and in a highly differentiated cell line deriving from mouse CCD; (2) the presence of serine protease activity in the apical membrane of these cells, which can be inhibited by aprotinin; and (3) the regulation of ENaC activity in oocytes. Both modes of action are not mutually exclusive. The regulation of each of these two pathways could be different in function of various physiologic or pathophysiologic stimuli. Although it is not excluded that other serine protease may activate ENaC in vivo, the cloning of mCAP1 will allow us to study the physiologic role of this serine protease in vivo by gene targeting and to assess its physiologic relevance in the control of sodium balance, blood volume, and BP.