| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
MEMBRANE TRANSPORTERS, ION CHANNELS, AND PUMPS
1Department of Anatomy and Physiology and 2Department of Animal Science and Industry, Kansas State University, Manhattan, Kansas
Submitted 20 September 2005 ; accepted in final form 6 January 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
·cm2), confluent monolayers that responded to forskolin with an increase in short-circuit current (Isc; 8 ± 1 µA/cm2) that required Cl–, HCO3–, and Na+, and was partially sensitive to bumetanide. mRNA was expressed for a number of anion transporters, including CFTR, electrogenic Na+-HCO3– cotransporter 1b (NBCe1b), downregulated in adenoma, pendrin, and Cl–/formate exchanger. Both forskolin and isoproterenol caused an increase in cellular cAMP levels. In addition, PVD9902 cell monolayers responded to physiological (i.e., adenosine, norepinephrine) and pharmacological [i.e., 5'-(N-ethylcarboxamido)adenosine, isoproterenol] agonists with increases in Isc. Unlike their freshly isolated counterparts, however, PVD9902 cells did not respond to glucocorticoid exposure with an increase in amiloride-sensitive Isc. RT-PCR analysis revealed the presence of both glucocorticoid and mineralocorticoid receptor mRNA as well as mRNA for the 
- and 
-subunits of the epithelia Na+ channels (- and -ENaC), but not 
-ENaC. Nonetheless, PVD9902 cells recapitulated most observations in freshly isolated cells and thus represent a powerful new tool to characterize mechanisms that contribute to male reproductive function. male reproductive tract; cystic fibrosis; epithelial Na+ channel expression; glucocorticoid receptor; adrenergic; vasopressin
Cystic fibrosis (CF), a genetic disease of ion transport, is universally associated with male infertility. CF is the most common lethal recessive genetic disorder among Caucasians and is caused by mutations in a single gene, CFTR, an anion channel that is thought to regulate many cellular functions (2, 29, 32, 47; for reviews, see Refs. 21 and 52). In addition to infertility, CF symptomatology includes pancreatic insufficiency, recurrent pulmonary infection, steatorrhea, meconium ileus, and recurrent bowel obstruction (54). More than 97% of male patients with severe CF have congenital bilateral absence of the vas deferens (CBAVD) (18), although the duct appears to develop normally in utero (20). Patients with CF who are atypical or exhibit mild forms of the disease are equally affected by reproductive failure (39). Studies in which the frequency of CFTR mutations in infertile or subfertile men has been examined have suggested a strong link between CF and infertility (46). Otherwise healthy men who seek medical intervention for infertility have a higher incidence of mild mutations in at least one CF allele compared with the general population (3, 13, 51). The relationship between CF-associated ion transport abnormalities and male infertility requires a better understanding of the role of the male reproductive tract, including the vas deferens, in generating an environment that promotes both ductal maintenance and viable sperm delivery.
The limited knowledge of epithelial function in the ductal system beyond the epididymis arises from a lack of in vivo or in vitro studies, likely as a result of difficulties in accessibility and a lack of well-characterized immortal cell lines. It was previously shown that epithelial cells from porcine and human vas deferens can be isolated and grown in culture to yield electrically tight monolayers that are responsive to a variety of physiological and pharmacological agents (11, 12, 36, 41). This procedure for collecting data regarding vas deferens epithelial cell function, although valuable, is extremely labor intensive and expensive and also is susceptible to biological variation. Many epithelial cell lines derived from a variety of species and tissues [e.g., Madin-Darby canine kidney (MDCK) cell line; human airway epithelial cell line Calu-3; human colonic cell line T84] have been established as invaluable model systems with which to determine underlying mechanisms that account for tissue function. Model systems that have been reported are immortalized male mouse fetal vas deferens (MFVD) ductal cell lines (50), hamster ductus deferens (DDT1MF-2) cells (34), and human fetal cells (17). However, MFVD cells are stromal and DDT1MF-2 cells are smooth muscle, neither of which is appropriate for studies designed to elucidate ion transport function. Furthermore, there is no indication that cells derived from the human fetus can function as a confluent epithelium. A cell line derived from adult vas deferens epithelium that recapitulates in vivo functions is clearly needed but has not yet been described.
The objective of the current study was to characterize a spontaneously immortalized porcine vas deferens epithelial cell line that we have named PVD9902 and to verify its functional similarity to findings regarding freshly isolated tissue reported previously. The results indicate that after 180 passages, PVD9902 cells exhibit the ability to form electrically tight epithelial monolayers that are responsive to physiological and pharmacological modulators, including various neurotransmitters. The responses that we observed are similar to those in freshly isolated cells, with the exception of glucocorticoid responsiveness. PVD9902 cells express mRNA for a number of ion transport proteins that have been postulated to interact or work in conjunction with CFTR. Thus PVD9902 represents a model system that can readily be used to study mechanisms associated with epithelial function in the distal male reproductive duct system.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Porcine vas deferens epithelial cells were isolated and initially cultured using methods described in detail previously (41). Medium was refreshed every other day until the cells reached 
90% confluence in a 25-cm2 flask (Corning, Corning, NY). Expended medium was removed, and cells were washed with Mg2+- and Ca2+-free PBS for cell culture (PBScc; composition in mM: 140 NaCl, 2 KCl, 1.5 KH2PO4, and 15 Na2HPO4). Subsequently, cultured cells were exposed for 2 min to 1 ml of dissociation medium (PBScc containing 0.25 g/l trypsin and 2.65 mM EDTA; Life Technologies, Rockville, MD). Excess dissociation medium was removed, and the flask was returned to the incubator for 8–10 min at 37°C to detach the cells. Detached cells were suspended in 6 ml of DMEM (Invitrogen, Carlsbad, CA) supplemented with 10% FBS, penicillin (100 U/ml), and streptomycin (100 µg/ml; all from Invitrogen). The cell suspension was used to seed a new 25-cm2 flask (0.5-ml suspension plus 4.5-ml growth medium at 1:12 subculturing ratio), with the subculturing defining a passage. In addition, suspended cells were seeded onto 12-mm-diameter permeable supports (Snapwell Clear; Corning) or 6.5-mm-diameter permeable supports (Transwell Clear; Corning) and cultured for 14–20 days with media changed every other day. A single isolation of cells, which we designated as PVD9902, was passaged numerous times without any apparent loss of ability to proliferate in culture or form cell monolayers with high transepithelial electrical resistance (Rte). Beginning at passage 17, we considered that the cells had spontaneously immortalized. Thus stocks of cells were washed with PBScc, dissociated from the flask, suspended in 93% FBS-7% DMSO, and stored in liquid N2. Stocks of cells from later passages were generated and also stored in liquid N2. A continuously subcultured stock flask of PVD9902 cells was maintained in the laboratory for >4 yr with approximately weekly passaging (up to passage 230). The present report includes data from cells through passage 180. Parallel studies of cells from substantially different passages were performed by quickly thawing frozen stocks in a 37°C water bath and seeding them into prewarmed growth medium in 25-cm2 tissue culture flasks. For some experiments, as indicated, cells were cultured in the presence of dexamethasone (100 nM) for 1–3 days before being subjected to assay. Unless otherwise stated, three ranges of passage numbers were used in the present study, and they are designated as the low-passage (passages 17–30), medium-passage (passages 60–80), and high-passage groups (passages 140–180).
Chromosome characterization and analysis.
To confirm that PVD9902 cells were indeed of male porcine origin, PCR analysis was conducted using primers specific to the SRY gene (GenBank accession no. U49860; kindly provided by Dr. Andrea Cupp, Animal Science Department, University of Nebraska, Lincoln, NE), the porcine homolog of the sex-determining region of the Y chromosome (pSRY). As a positive control for each PCR experiment, primers for the X- or Y-linked zinc finger gene (ZFX/ZFY) also were used (1). These primers were selected from a conserved segment of the human ZFY and ZFX genes and were expected to recognize porcine mRNA as well. Genomic DNA was extracted from 1–5 x 106 cells from each of the indicated cell sources and analyzed using the Qiagen DNeasy kit (Qiagen, Valencia, CA) according to the manufacturer's protocol for isolating DNA from cultured animal cells. PCR was performed using the GeneAmp PCR reagent kit (Applied Biosystems, Foster City, CA) and consisted of 250 ng of genomic DNA, 1x PCR buffer, a 200 nM concentration of each primer, 200 µM 2-deoxynucleotide 5'-triphosphate, 1.5 mM MgCl, and 2.5 U of AmpliTaq DNA polymerase. Thermocycler settings were as follows: 3 cycles at 98°C for 3 min, 65°C for 2 min, and 72°C for 1 min, followed by 35 cycles at 94°C for 15 s, 60°C for 30 s, and 72°C for 30 s.
CFTR, ENaC, anion exchangers, and corticosteroid receptors expression by RT-PCR.
PVD9902 cells were cultured on Snapwell tissue culture inserts as described above. Total RNA was isolated from confluent monolayers by lysing the cells in TRI Reagent (Sigma-Aldrich, St. Louis, MO) in accordance with the manufacturer's protocol. Some monolayers were treated with 100 nM dexamethasone before RNA isolation as indicated in RESULTS. RNA isolates from the fetal porcine jejunal intestinal epithelial cell line IPEC-J2 and the porcine kidney epithelial cell line LLC-PK1 were routinely used as positive controls for mRNA expression, and a murine Swiss 3T3 fibroblast (3T3) cell line served as a negative control. The quality of purified total RNA samples was confirmed using RNA Nano LabChip analysis (Agilent Technologies, Palo Alto, CA) as well as detection of expected bands on a denaturing gel corresponding to 18S and 28S ribosomal RNA. All RNA samples were subjected to DNase I treatment (Ambion, Austin, TX) according to the manufacturer's specifications. RT-PCR was performed (OneStep RT-PCR; Qiagen) using primer pairs specific for the coding sequences of porcine homologs of CFTR, glucocorticoid receptors (GR), mineralocorticoid receptors (MR), the -subunit of the epithelial Na+ channel (-ENaC), the electrogenic Na+-HCO3– cotransporter (NBCe1b) pancreatic splice variant (NBCe1b or SLC4A4), and three members of the SLC26 gene family; SLC26A3 (a protein that is downregulated in adenoma; DRA), SLC26A4 (pendrin) and SLC26A6 (a C1–/formate exchanger, CFEX that is also known as a putative anion transporter-1; PAT-1) in separate reactions. Primers for - and -ENaC were provided by Xiaofei Wang and Paul M. Quinton (University of California, San Diego, CA) and had been shown previously to amplify a product of the expected size from porcine airway epithelial mRNA (53). All PCR experiments were performed using 35 cycles, except for ENaC subunits, which were run using a variety of reaction conditions comprising 50 cycles. Primer sequences, accession numbers for sequences on which they are based, annealing temperatures, and expected product sizes are summarized in Table 1. RT-PCR products were analyzed by performing electrophoresis using a 1% agarose gel along with a 100-bp DNA ladder. Further analysis of RT-PCR-amplified transcripts was performed using sequence determination with an automated genetic analysis system (CEQ 8000; Beckman Coulter, Fullerton, CA).
|
PVD9902 cells (passage 149) were grown on permeable supports as described above and assessed in modified Ussing chambers (described below) to verify the presence of an electrically tight monolayer. Cells were recovered from the Ussing chamber and fixed in 10% buffered neutral formalin (BNF; Fisher Scientific International, Hampton, NH) and stored in PBS for histochemistry (PBSh; in mM: 150 NaCl, 5 KH2PO4, and 15 K2HPO4, pH 7.2–7.4). Immunocytochemical analysis was performed as previously described in detail (41). Briefly, small portions of permeable supports were separated and treated in parallel with each primary antibody or vehicle control. Nonspecific immunoreactivity was blocked by treating the cells with 5% normal goat serum (GIBCO-BRL, Grand Island, NY) with 0.2% Triton X-100 (Fisher Scientific International) before antibody exposure. Primary MAbs raised against zonula occludens (ZO)-1 (anti-ZO-1 rat MAb; Chemicon International, Temecula, CA) or anti-pan-cytokeratin (mouse MAb; Sigma-Aldrich) were diluted 1:200 in PBSh, 0.2% Triton X-100, and 1% BSA (Sigma-Aldrich) and then incubated for 1–2 h at room temperature. An appropriate secondary antibody (goat anti-rat IgG FITC for ZO-1 and goat anti-mouse IgG Texas Red for cytokeratin detection; Vector Laboratories, Burlingame, CA) was diluted 1:200 in PBSh and 1% BSA with or without 0.2% Triton X-100 and incubated for 1–2 h. Three 10-min washes with PBSh were performed between incubation steps. Cells were mounted on glass microscope slides in Vectashield mounting medium (Vector Laboratories) and observed using a Leica DM RX microscope (Leica, Solms, Germany) with appropriate filters for each fluorophore. In a separate experiment, a confluent monolayer (passage 76) was fixed in 10% BNF, processed as described above, and probed using a primary antibody raised against rat occludin (Zymed Laboratories, San Francisco, CA). The secondary antibody was goat anti-rabbit IgG conjugated to Alexa Fluor 488 (Molecular Probes, Eugene, OR). The cells were subsequently counterstained with TO-PRO-3 nucleic acid stain (Molecular Probes). Images were acquired using an LSM 510 META confocal microscope (Zeiss, Göttingen, Germany). Digital images were acquired using identical numerical aperture settings for treatment and controls and were prepared for publication in parallel using CorelDraw software (version 10.410; Corel, Ottawa, ON, Canada).
Cellular proliferation rate determination.
The cellular proliferation rate was determined in parallel for cells from each of the three passage groups. Stocks from each passage group were thawed and seeded into a 25-cm2 flask as described above. After reaching 90% confluence, cells were dissociated and manually counted using a hemocytometer to determine cell density in suspension medium. For each passage group, a 24-well plate was seeded with 1 x 105 cells/well and returned to the incubator. The next day, 12 of the 24 wells were treated with 0.2 U/ml insulin (Sigma-Aldrich) and allowed to grow for an additional 24 h. Forty-eight hours after initial seeding, four wells from each passage group (2 insulin-treated wells and 2 untreated wells) were dissociated, suspended, and counted to determine the number of cells present in the well, with the average of each pair of wells used in subsequent analysis. This procedure was repeated every 48 h for 240 h with daily media changes. The average number of cells observed at each time point was fitted using Eq. 1
 | (1) |
Electrophysiology.
Cells from the three passage groups were seeded onto permeable tissue culture supports as described above and maintained for 12–14 days, with media refreshed every other day. Cells formed an electrically resistive monolayer that was quantitatively assessed in a modified Ussing chamber to determine transepithelial potential difference (PDte), transepithelial electrical resistance (Rte, an indicator of epithelial barrier integrity), and short-circuit current (Isc, sensitive measure of net ion transport). Electrophysiological parameters were determined in Ringer solution (in mM: 120 NaCl, 25 NaHCO3, 3.3 KH2PO4, 0.83 K2HPO4, 1.2 CaCl2, and 1.2 MgCl2), held at 39°C, and continuously bubbled with 5% CO2-95% O2 to maintain pH and achieve continuous mixing. After PDte values were recorded, the monolayers were clamped at 0 mV and Isc was measured continuously using a voltage-clamp apparatus (model 558C; Department of Bioengineering, University of Iowa, Iowa City, IA). Selected neurotransmitters or ion transport modulators were added to the apical and/or basolateral medium as indicated. Data were digitally acquired at 1 Hz using a Macintosh computer equipped with an MP100A-CE interface and Acqknowledge software (version 3.2.6; BIOPAC Systems, Goleta, CA). A 5-mV bipolar pulse was administered every 100 s and, using Ohm's law, Rte was calculated.
Ion substitution studies were conducted in nominally Cl–-, Na+-, or HCO3–-free Ringer solution as described in detail previously (12). Briefly, Cl– was replaced by gluconate with supplemental Ca2+, Na+ was replaced by choline and N-methyl-D-glucosamine, and HCO3– was partially replaced by Cl– and partially replaced by HEPES, with pH adjusted to 7.4. Experiments in HCO3–-free Ringer solution were bubbled with room air. Osmolarity was similar in all Ringer solutions (280 ± 3 mosmol/l). All other conditions and equipment settings were maintained as indicated above.
cAMP measurement in cultured monolayers.
Total cellular cAMP was measured using a cAMP Biotrak enzyme immunoassay (EIA) system (Amersham Biosciences, Piscataway, NJ) for untreated PVD9902 cell monolayers and monolayers exposed to isoproterenol or forskolin. The assay was conducted in accordance with the manufacturer's instructions for the nonacetylation EIA procedure. Briefly, cells from early-passage PVD9902 cultures (passages 21–28) were grown to confluence on 6.5-mm-diameter Transwell tissue culture inserts with media changed every other day. On the assay day, apical and basolateral growth media were replaced with 200 and 800 µl, respectively, of PBScc 60 min before the assay. A noninterfering phosphodiesterase inhibitor (Ro 20-1724, 10 µM; Calbiochem/EMD Biosciences, San Diego, CA) was added to the basolateral compartment, with mixing, 6 min before agonist addition. Isoproterenol (3, 30, or 300 nM), forskolin (2 µM), or vehicle was added to the basolateral buffer and mixed. The agonist remained for 6 min, at which time the basolateral fluid was removed and the lysis buffer provided in the assay kit was added to the apical compartment and mixed. Aliquots of apical mixture from each treatment were transferred onto the supplied ELISA plate and analyzed using a Labsystems Multiskan plate reader (Thermo Electron, Woburn, MA). cAMP standards were included on each assay plate, and a standard curve was fitted to the data using SigmaPlot software. Quantitation of total cellular cAMP was performed with two dilutions of cell lysate from each monolayer.
Chemical sources.
Amiloride, adenosine, amphotericin B, bumetanide, dexamethasone, cortisol, histamine, insulin, isoproterenol, norepinephrine, 5'-(N-ethylcarboxamido)adenosine (NECA), and vasopressin were purchased from Sigma-Aldrich. Forskolin (Coleus forskohlii) was purchased from Calbiochem. N-[4-methylphenylsulfonyl]-N'-[4-trifluoromethylphenyl]urea (DASU-02) was synthesized de novo in the laboratory.
Stock solutions.
Solutions were prepared as follows: amiloride, 10 mM in distilled water; forskolin, 10 mM in ethanol; bumetanide, 20 mM in ethanol; histamine, 100 mM in distilled water; isoproterenol, norepinephrine, adenosine, and NECA, 10 mM in 10 mM ascorbic acid; and vasopressin, 1 mM in 10 mM ascorbic acid. Forskolin, bumetanide, and NECA stocks were stored at –20°C. Amiloride was stored at 4°C. All other compounds were freshly dissolved immediately before experiments.
Statistical analyses.
Regression analysis was performed as indicated above. Comparison of paired or unpaired results was performed using the t-test function of Excel software (version 10.45; Microsoft, Redmond, WA). Differences were considered statistically significant when the probability of a type I error was <0.05.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Experiments were conducted to verify unequivocally that PVD9902 cells were in fact derived from male porcine tissues. Thus genomic DNA was isolated from PVD9902 cultures and other defined cell sources for PCR evaluation using two sets of primers. The first primer set was specific for pSRY and was expected to produce a 362-bp product only from cells derived from male pigs. The second primer set was used to test for the presence of ZFX/ZFY, which is highly conserved across mammalian species. This primer set was designed to recognize a region of high homology between ZFX and ZFY (1) and was used as an internal control to verify that PCR occurred in each sample by the production of a 446-bp product in all isolation experiments. As shown in Fig. 1, PVD9902 (Fig. 1, lane A) and freshly isolated fibroblasts from a male pig (Fig. 1, lane B) were the only samples in which a pSRY-positive result could be visualized. The results presented in Fig. 1 (lane C, freshly isolated female porcine cells; lane D, female bovine cell line; and lane E, canine cell line) demonstrate that a PCR product of appropriate mobility for ZFX/ZFY was derived from all samples but that no pSRY-associated product was observed. These results demonstrate that PVD9902 cells contain DNA that is consistent with their tissue of origin, the porcine vas deferens epithelium.
|
Indirect immunofluorescence was used to test for the presence of cytokeratin, ZO-1, and occludin immunoreactivity. These markers have been used widely as indicators of epithelial cell lineage. Cells were cultured on permeable supports, and the presence of an electrically tight barrier was confirmed in modified Ussing chambers. Subsequently, monolayers were fixed and probed using the indicated primary antibodies. The results presented in Fig. 2A show that all cells were circumscribed by anti-ZO-1 immunoreactivity. Immunofluorescence was restricted to cell-cell contact points as expected for a tight junction-associated protein. Anti-pan-cytokeratin antibodies detected paranuclear cytosolic immunoreactivity in all cells (Fig. 2C). No immunofluorescence was observed when the respective primary antibody was omitted from the assay protocol (Fig. 2, B and D). The image shown in Fig. 2E represents occludin immunoreactivity and nuclear counterstaining observed using confocal microscopy. As with ZO-1, immunofluorescence was restricted to cell-cell contact points. Figure 2F shows an image of the same monolayer without primary antibody using the same filters. TO-PRO-3 nuclear counterstaining is clearly visible. Altogether, these data demonstrate immunoreactivity for three epitopes that are considered typical epithelial cell markers. The distribution of each epitope in the cell monolayer was consistent with the expected epithelial nature of the cells.
|
Tests were conducted to determine the proliferation characteristics of each passage in the presence and absence of insulin as previously conducted in freshly isolated epithelia (41). Cells (1 x 105) in the presence or absence of 0.2 U/ml insulin were seeded into each well of a 24-well tissue culture plate in duplicate. The average cell count from duplicate wells for each condition was determined every 48 h after seeding for up to 10 days. The entire protocol was conducted a total of three times, and typical data are plotted in Fig. 3, in which lines indicate the best fit of Eq. 1 to each data set. All passages displayed a similar growth pattern consisting of an initial lag phase (95 ± 6, 77 ± 10, 93 ± 4, 81 ± 8, 59 ± 21, and 45 ± 27 h for low, medium, and high passages and without and with insulin, respectively), followed by exponential growth and finally a sustained plateau. Each passage group showed a decreased lag phase in the presence of insulin, although late-passage cells exhibited a somewhat shorter lag time overall (Fig. 3). The difference in the derived lag phase may result from a net loss in total cell number that was observed at 48 h in the absence of insulin for low- and medium-passage cells. The short-lived decline in cell number in the absence of insulin was observed in two of the three protocols. No significant difference was observed regarding the final cell population density (range, 262,000–344,000 cells/well). We previously reported an insulin-associated increase in both proliferation rate and final population number in primary vas deferens epithelial cultures (41). The apparent reduction in the insulin response of PVD9902 cells is not understood but appears not to alter ion transport characteristics compared with freshly isolated cells (described below). Early-passage cells (passages 17–30) and medium-passage cells (passages 60–80) were seeded from frozen stocks that varied from several weeks to several months with regard to length of time in liquid N2. There was no obvious correlation between duration in liquid N2 and cellular responses.
|
PVD9902 cultures were grown on permeable supports for 12–14 days and analyzed using a modified Ussing chamber as described previously for primary cultures (41). A typical Isc trace demonstrating basal Isc (zero or slightly positive), as well as Rte represented by vertical deflections, is presented in Fig. 4A. Apical amiloride has no effect on basal Isc. Exposure to forskolin results in an initial transient peak followed by a sustained plateau indicating ongoing anion secretion. Basolateral bumetanide inhibits a portion of this Isc, which suggests that a portion of the Isc is attributable to Cl– secretion that requires the activity of a basolateral Na+-K+-2Cl– cotransporter (NKCC). All passages exhibited this qualitatively identical pattern of responses. Isc before drug exposure and in the presence of amiloride, forskolin, and bumetanide are summarized in Fig. 4B for each passage group. Rte and PDte, along with the effects of ion transport modulators are summarized in Table 2. All passages exhibited an initial luminal negative PDte and basal Isc was <1 µA/cm. These results are consistent with a modest amount of net anion secretion. Initial Rte was typically <6,000 ·cm2, indicative of an extremely tight epithelium that functions to segregate fluid compartments of distinct compositions. Some variation between passage groups achieved statistical significance (e.g., higher passages displayed a more negative initial PDte and greater Isc) as indicated in Table 2. Amiloride, a selective blocker of ENaC, was uniformly without effect in each passage. Exposure to forskolin (2 µM) resulted in a rapid transient increase in Isc, followed by a sustained elevated current. Statistically significant differences between passage groups was observed for both peak and sustained Isc values, although a different pattern emerged for the two outcomes. The initial resistance and reduction of Rte after forskolin exposure were not statistically different between passage groups and were consistent with the activation of conductive pathways. Bumetanide inhibited a portion of this forskolin-induced anion transport, indicating the presence of NKCC, but the remaining current suggested that ions other than Cl– were involved. Previous reports provided evidence for HCO3– secretion in primary cultures of porcine vas deferens epithelial cells on the basis of ion substitution assay results (12). Similar experiments were conducted using PVD9902 cells in which selected ions were substituted with impermeant ions. Figure 5 summarizes the net Isc change in low-passage (Fig. 5A) and high-passage (Fig. 5B) in the absence of Cl–, Na+, or HCO3–. Although high-passage PVD9902 monolayers produced responses of higher magnitude (compared with ordinal scale), the pattern of ion transport between the passage groups for each condition was nearly identical. In normal Ringer solution, basal Isc is <1 µA, with the exception of high-passage cells in HCO3–-free Ringer solution, which demonstrated a slightly elevated basal current (2.04 ± 0.66 µA/cm). In both passage ranges, HCO3–-free Ringer solution supported a substantial forskolin-induced peak and sustained current, although it was not as pronounced as those in normal Ringer solution. The forskolin response in Cl–-free and Na+-free Ringer solutions, however, was greatly reduced compared with normal Ringer solution. The CFTR inhibitor DASU-02 had a modest effect in normal, Cl–-free and HCO3–-free Ringer solution but had no effect in Na+-free conditions. Bumetanide caused further reduction in Isc only in normal and HCO3–-free conditions. The profile of ion transport activity for PVD9902 monolayers in normal and ion-substituted Ringer solution was similar to that previously reported for primary epithelial cell cultures derived from pig vas deferens (12).
|
|
|
The phenomenological evidence presented above and in a previous report (12) suggests that vas deferens epithelia can secrete HCO3–, although the underlying mechanisms has not been defined fully. Thus experiments were conducted to detect HCO3– transporters known to be present in primary porcine vas deferens epithelial cultures and to detect exchangers reported to work in concert with CFTR (2, 27, 29). The results presented in Fig. 6 demonstrate that PCR products can be produced with the use of primers designed specifically to amplify the porcine homologs of four different HCO3– transporters and CFTR. As previously reported for primary vas deferens epithelial cells (12), PVD9902 cells expressed NBCe1b (Fig. 6A). Figure 6E demonstrates that CFTR mRNA is present in primary cultures of vas deferens epithelial cells, PVD9902, and the immortalized porcine jejunal epithelial cell line IPEC-J2. Importantly, Fig. 6, B–D, provides evidence that PVD9902 cells express mRNA for SLC26A3, SLC26A4, and SLC26A6. SLC26A3 (also known as DRA or congenital Cl– diarrhea) and SLC26A4 (also known as pendrin) are generally considered to be predominately Cl–/HCO3– exchangers, whereas SLC26A6 (also known as PAT-1 or CFEX) is capable of transporting sulfate and oxalate (33). Clearly, bands of the expected mobility for these transcripts were present in the PCR products derived using RNA isolated from PVD9902 monolayers and from other porcine epithelial cell lines (IPEC-J2 and LLC-PK1), but such products were not present in PCR residues derived using RNA isolated from 3T3 cells (data not shown). Gene specificity for the resolved bands was confirmed on the basis of sequencing of the amplified products (data not shown), which revealed >97% identity when aligned with the respective published sequences [National Center for Biotechnology Information (NCBI) accession numbers listed in Table 1]. These results demonstrate that vas deferens epithelia express mRNA for a repertoire of mechanisms that are capable of supporting anion transport, especially HCO3– transport. Additional experiments are required to test for the levels of protein expression and contributions to epithelial function. PVD9902 is likely to be a useful tool in addressing these questions directly.
|
Previous reports regarding freshly isolated porcine vas deferens epithelia revealed that this segment of the male reproductive tract is sensitive to neurotransmitter stimulation and responds with active ion transport (41). Examination of the ability of PVD9902 cells to respond to neurotransmitter stimulation confirmed sensitivity throughout the highest passage number tested. Typical responses to vasopressin and to the -adrenergic agonist isoproterenol are shown in Fig. 7, A–C and D–F, respectively, for each passage group. Most notable in each graph is the rapid transient increase in Isc upon exposure to the neurotransmitter, which was consistent across each passage group. The sustained response typically observed after this peak was abated to a much greater extent in the high-passage group, even though forskolin was consistently associated with a sustained plateau phase. Monolayers responded to forskolin in a nonadditive manner after vasopressin and -adrenergic stimulation, suggesting a convergence in the stimulatory pathway. Summarized values for the peak and sustained change in Isc are shown in Fig. 7, G and H, along with values for another -adrenergic agonist, norepinephrine; the adenosine agonists NECA and adenosine; and histamine. As these graphs demonstrate, each passage group responded to these modulators in a similar fashion (initial increase in Isc, followed by sustained elevated current), although the magnitude varied greatly with the use of some agonists. Most notable was the increasing magnitude of the response to NECA and adenosine in the high-passage group. For all passages, NECA, adenosine, and histamine exhibited little sustained Isc, whereas in each passage group, there was a sustained (or long-lasting residual) component of the response to isoproterenol, norepinephrine, and vasopressin. All passages of PVD9902 responded to the same repertoire of physiological and pharmacological stimulants with similar response profiles.
|
Early-passage PVD9902 cells were exposed to isoproterenol (3, 30 and 300 nM) or forskolin, and the total cellular cAMP was measured in confluent monolayers on permeable supports (Fig. 8). Basal levels were established by analyzing cells with no stimulant added. A small amount of cellular cAMP (<2 pmol/well; 0.33 cm2) could be measured in basal conditions, whereas the addition of isoproterenol increased this level in a concentration-dependant fashion. The addition of forskolin resulted in a more than twofold increase in total cAMP. These results suggest that the cAMP second-messenger pathway is activated by exposure to adrenergic agonists or forskolin, which is expected to activate adenylyl cyclase directly and is a positive link to anion secretion.
|
Natural and synthetic glucocorticoids were shown previously to induce an amiloride-sensitive current in freshly isolated porcine vas deferens epithelial cells when included in the growth medium (36). Monolayers of PVD9902 cells from each passage group were exposed to dexamethasone (100 nM), a synthetic glucocorticoid, or cortisol (500 nM), a naturally occurring glucocorticoid, for 1–72 h to test for a similar response. More than 20 observations using cells from various passages were made, with none showing any amiloride-sensitive current after exposure to either compound (data not shown). Because cells were not routinely exposed to glucocorticoids at the time of isolation, it is unclear whether this particular cell line lost corticosteroid sensitivity before passage 17 or whether the cells lacked glucocorticoid responsiveness at the time of isolation. Nonetheless, the absence of response provided the basis for additional experiments to test for the presence of mRNA for the corticosteroid receptors and for each of the ENaC subunits. Primers were designed for porcine GR and MR and used to probe RNA isolated from PVD9902 (low passage), LLC-PK1, and 3T3 cells. Products of the expected mobility were observed for both primer sets for both porcine-derived cell lines, but not for murine cells (Fig. 9A). Thus the lack of response does not likely reflect the absence of corticosteroid receptors. Experiments were then conducted to test for the presence of -, -, and -ENaC mRNA. PVD9902 cell monolayers and monolayers of primary vas deferens epithelial cell cultures were exposed to 100 nM dexamethasone before RNA isolation to increase the likelihood of upregulating ENaC expression. Resolved PCR products for PVD9902 and primary cell cultures are presented in Fig. 9, B and C, respectively. The results suggest that PVD9902 cells lack mRNA for -ENaC. Clearly, additional studies are required verify this conclusion and to test for other cellular processes that corticosteroids would be expected to modulate.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
The reproductive tract plays an active role in male fertility. Fluid reaching the lumen of the vas deferens is highly modified compared with fluid in the epididymis, a change that is accomplished by fluid and electrolyte transport across the epithelium. More than 95% of the fluid secreted by the testes is reabsorbed in the efferent ducts (16). The epididymis is responsible for acidifying the environment via H+-ATPase located on the apical membrane surfaces of some cells (9) as well as a NHE (4, 5). Aquaporins are constitutively expressed along the epididymis and vas deferens and concentrate the seminal fluid by reabsorbing water (35, 45). This absorption is thought to be driven by the gradient created by the movement of Na+ across the apical membrane into the cytosol. Previously, we have suggested a possible role of the NBC in neutralizing the acidic fluid of the distal duct. We postulated that adrenergic agonists could acutely stimulate HCO3– transport to promote sperm activation (12). The present report offers a first analysis into HCO3– transporter presence in the vas deferens, revealing positive expression of the pSLC4A4 variant NBCe1b, pSLC26A3, pSLC26A4, and pSLC26A6 in PVD9902 cells. CFTR has been identified in the ductal epithelia of several species (7, 14, 17, 22, 49), and now we can securely place CFTR mRNA in primary epithelial cultures derived from both porcine vas deferens and PVD9902 cells. Furthermore, most functional responses are consistent with CFTR-mediated anion secretion. The CFTR channel blocker DASU-02 caused an obvious, significant reduction in Isc that was most pronounced in Cl–-free conditions. However, the modest inhibition in replete Ringer solution is surprising and may suggest the presence of additional transport pathways that remain to be determined. CFTR is also known to be associated with other ion transport proteins, such as ENaC (47) and K+ channels (32), and may be involved in processes related to HCO3– transport (2, 15, 29). CF-associated pathology, together with these observations, suggests that the epithelium lining the male reproductive tract and especially the vas deferens is complex and extremely sensitive to the loss of anion conductance and that the loss of this ion channel affects both sperm quality and duct maintenance. Thus ion transport regulation in the male reproductive tract is apparently important but not well understood. One could easily envision targeted therapeutic interventions designed to modulate male fertility. A better understanding of the role of ion transporters and associated neurotransmitters in regulating the luminal environment could lead to clinical therapies for male infertility or to the development of nonhormonal male contraceptives. The availability of a functionally competent vas deferens epithelial cell line would provide researchers with the means to study ion transport and associate regulatory processes in the distal male reproductive tract.
CFTR is widely accepted as an apical Cl–-conductive pathway and has been reported by many investigators to provide an exit route for HCO3– secretion (6, 19, 24, 38, 42, 43). However, an alternative ductal model for CFTR-associated secretion of a HCO3–-rich solution has been proposed by Ko et al. (27). According to this model, proximal duct cells express apical CFTR for Cl– permeation and SLC26A6, which Ko et al. proposed to have an electrogenic stoichiometry of 2 HCO3– to 1 Cl–. These authors further proposed that SLC26A3 is expressed in distal ductal epithelia and that it has a stoichiometry of 1 HCO3– to 2 Cl–. Electrochemical driving forces could favor increased concentration of HCO3– up to 140 meq/l with the use of this model. The observation that a basolateral HCO3–-loading mechanism is expressed in PVD9902 cells, along with both SLC26A3 and SLC26A6, perhaps offers some support for this model. More important, the expression of all components of the proposed scheme is under native control mechanisms in PVD9902 cells. Thus PVD9902 offer an excellent model with which to test systematically for direct or indirect interactions between CFTR and members of the SLC26 gene family.
PVD9902 cell monolayers fail to exhibit a glucocorticoid-induced increase in amiloride-sensitive Isc as shown previously for freshly isolated porcine vas deferens epithelial cells (36). Primary cultures of ovine duct cells also exhibit amiloride-sensitive Isc when exposed to dexamethasone in culture (7). Likewise, freshly excised human vas deferens exhibits amiloride-sensitive ion transport, and primary cultures of epithelial cells derived from this tissue exhibit an amiloride-sensitive basal current that is enhanced by glucocorticoid exposure (11). When exposed to glucocorticoids, PVD9902 lack any increment in basal current, although responses to forskolin, neurotransmitters, and various inhibitors remain qualitatively similar to responses from freshly isolated and cultured cells. PCR experiments were conducted using primers specific for GR and MR, and bands of the expected sizes and proven identity were produced for both sequences. Several possibilities exist regarding why mRNA is present for GR, whereas the typical amiloride-sensitive current is absent. Published data for other species have demonstrated that GR can produce different splicing variants that code different GR protein subunits, such as GR-, GR-, and GR-. It has been suggested that different protein subunits might exert different effects and might suppress each other's effects (28). The full-length coding sequence for porcine GR- has been determined on the basis of RNA isolated from porcine vas deferens epithelial cell primary cultures (GenBank accession no. AY779185). A partial sequence of porcine GR- has also been determined (unpublished data) and is present in PVD9902 (data not shown). The data presented in Fig. 9 represent amplification of a portion of the coding sequence that is common to both GR- and GR- such that the relative proportion of the GR splice variant mRNA remains to be determined. Currently, it is unknown whether other splice variants are present (e.g., GR-). In addition, it remains unknown whether the mRNA or protein expression levels for these distinct splice variants in PVD9902 are different from those in native vas deferens epithelial cells, a difference that might account for the observed outcome. An alternative and perhaps more likely explanation for the observed outcome is that PVD9902 cells express mRNA for the - and -ENaC subunits, but not for the -ENaC subunit. It has been reported that -ENaC or - and -ENaC overexpression can produce a modest amount of amiloride-sensitive current in some systems, but that ENaC channel activity is greatest when all subunits are expressed concurrently (10). The results from PVD9902 cells indicate that in this system, which depends on native regulators for ENaC subunit expression, - and -ENaC expression is not sufficient to observe an amiloride-sensitive Isc. The lack of -ENaC expression could result in either nonfunctional protein assembly or a rapid turnover of -ENaC and -ENaC in the cell membrane such that no functional channels are formed. More studies are needed to elucidate the full sequence of response mechanisms and determine the presence of each component at the protein level and in the apical membrane of PVD9902 cells.
As stated above, CFTR reportedly is associated with ENaC and the expression or function of these channels is reciprocally modulated (47). Thus one might be led to dismiss observations on CFTR regulation from this system that might have lost ENaC activity. However, it is important to note that PVD9902 may provide unique insights for inferences regarding ion transport in the pancreatic duct, which expresses CFTR, multiple bicarbonate transport mechanisms, and is not known to express ENaC. Furthermore, we contend that PVD9902 cells offer a unique tool with which to study ion transport, because comparative studies can be conducted with freshly isolated cells that express ENaC under glucocorticoid stimulation. In each cell type, the expression and regulation of cellular transport proteins are under the control of endogenous mechanisms. Thus comparisons can be made and one can rule out systematic differences that result from over- or underexpression of genes that are influenced by nonphysiological promoters. In this regard, PVD9902 cells are a unique research tool with which to study CFTR regulation in the context of a cell or tissue type that is affected adversely by CFTR mutations.
A porcine vas deferens epithelial cell line has many advantages compared with systems currently available for the study of male reproductive duct function. Review of the vast number and availability of cell model systems of rodent species suggests that the male reproductive tract could also be studied in this manner. However, in the case of the vas deferens, the mouse model is inadequate to relate epithelial cell function to the human counterpart. One salient example is that CFTR-knockout mice maintain an intact vas deferens throughout adulthood, whereas human males with either profound or mild mutations in this gene almost invariably exhibit CBAVD. This disease-associated difference is not surprising in light of reports that clearly have shown no CFTR immunoreactivity in either mouse or rat vas deferens (49), whereas CFTR immunoreactivity has been shown to be prominent in human tissues (48). These fundamental differences preclude rodent tissues from being used in studies related to the function of the most distal portions of the human reproductive tract that have implications for men. An animal model more closely related to humans is needed to study the ionic contribution of the vas deferens in male fertility. Our previous work suggests that the porcine model may be particularly well suited to such studies (11, 12, 36, 41). The present report demonstrates that the immortalized porcine cell line PVD9902 also may be particularly useful for studies concerning the male reproductive duct.
In summary, the vas deferens is a major component of the duct system in males; yet, little is known about its function. The PVD9902 cell line, which is described for the first time herein, provides a readily available immortalized cell model for studying these issues. The similar responsiveness of PVD9902 cells to neurotransmitters and pharmacological modulators to freshly isolated cells, as well as their functional longevity and simple culture requirements, suggests that this cell line will be useful in the study of male reproductive tract function and that inferences derived from such studies might be extended to other species, including humans.
|
GRANTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
2. Ahn W, Kim KH, Lee JA, Kim JY, Choi JY, Moe OW, Milgram SL, Muallem S, and Lee MG. Regulatory interaction between the cystic fibrosis transmembrane conductance regulator and HCO3– salvage mechanisms in model systems and the mouse pancreatic duct. J Biol Chem 276: 17236–17243, 2001.
3. Anguiano A, Oates RD, Amos JA, Dean M, Gerrard B, Stewart C, Maher TA, White MB, and Milunsky A. Congenital bilateral absence of the vas deferens: a primarily genital form of cystic fibrosis. JAMA 267: 1794–1797, 1992.[Abstract]
4. Au CL and Wong PY. Luminal acidification by the perfused rat cauda epididymitis. J Physiol 309: 419–427, 1980.
5. Bagnis C, Marsolais M, Biemesderfer D, Laprade R, and Breton S. Na+/H+-exchange activity and immunolocalization of NHE3 in rat epididymis. Am J Physiol Renal Physiol 280: F426–F436, 2001.
6. Ballard ST, Trout L, Bebök Z, Sorscher EJ, and Crews A. CFTR involvement in chloride, bicarbonate, and liquid secretion by airway submucosal glands. Am J Physiol Lung Cell Mol Physiol 277: L694–L699, 1999.
7. Bertog M, Smith DJ, Bielfeld-Ackermann A, Bassett J, Ferguson DJ, Korbmacher C, and Harris A. Ovine male genital duct epithelial cells differentiate in vitro and express functional CFTR and ENaC. Am J Physiol Cell Physiol 278: C885–C894, 2000.
8. Breton S, Smith PJ, Lui B, and Brown D. Acidification of the male reproductive tract by a proton pumping H+- ATPase. Nat Med 2: 470–472, 1996.[CrossRef][ISI][Medline]
9. Brown D, Lui B, Gluck S, and Saboli
I. A plasma membrane proton ATPase in specialized cells of rat epididymis. Am J Physiol Cell Physiol 263: C913–C916, 1992.
10. Canessa CM, Schild L, Buell G, Thorens B, Gautschi I, Horisberger JD, and Rossier BC. Amiloride-sensitive epithelial Na+ channel is made of three homologous subunits. Nature 367: 463–467, 1994.[CrossRef][Medline]
11. Carlin RW, Lee JH, Marcus DC, and Schultz BD. Adenosine stimulates anion secretion across cultured and native adult human vas deferens epithelia. Biol Reprod 68: 1027–1034, 2003.
12. Carlin RW, Quesnell RR, Zheng L, Mitchell KE, and Schultz BD. Functional and molecular evidence for Na+-HCO3– cotransporter in porcine vas deferens epithelia. Am J Physiol Cell Physiol 283: C1033–C1044, 2002.
13. Casals T, Bassas L, Ruíz-Romero J, Chillón M, Giménez J, Ramos MD, Tapia G, Narváez H, Nunes V, and Estivill X. Extensive analysis of 40 infertile patients with congenital absence of the vas deferens: in 50% of cases only one CFTR allele could be detected. Hum Genet 95: 205–211, 1995.[CrossRef][ISI][Medline]
14. Chan HC, Ko WH, Zhao W, Fu WO, and Wong PY. Evidence for independent Cl– and HCO3– secretion and involvement of an apical Na+-HCO3– cotransporter in cultured rat epididymal epithelia. Exp Physiol 81: 515–524, 1996.[Abstract]
15. Choi JY, Muallem D, Kiselyov K, Lee MG, Thomas PJ, and Muallem S. Aberrant CFTR-dependent HCO3– transport in mutations associated with cystic fibrosis. Nature 410: 94–97, 2001.[CrossRef][Medline]
16. Clulow J, Jones RC, and Hansen LA. Micropuncture and cannulation studies of fluid composition and transport in the ductuli efferentes testis of the rat: comparisons with the homologous metanephric proximal tubule. Exp Physiol 79: 915–928, 1994.[Abstract]
17. Coleman L and Harris A. Immortalization of male genital duct epithelium: an assay system for the cystic fibrosis gene. J Cell Sci 98: 85–89, 1991.
18. Cuppens H and Cassiman JJ. CFTR mutations and polymorphisms in male infertility. Int J Androl 27: 251–256, 2004.[CrossRef][ISI][Medline]
19. Devor DC, Singh AK, Lambert LC, DeLuca A, Frizzell RA, and Bridges RJ. Bicarbonate and chloride secretion in Calu-3 human airway epithelial cells. J Gen Physiol 113: 743–760, 1999.
20. Gaillard DA, Carré-Pigeon F, and Lallemand A. Normal vas deferens in fetuses with cystic fibrosis. J Urol 158: 1549–1552, 1997.[CrossRef][ISI][Medline]
21. Guggino WB and Banks-Schlegel SP. Macromolecular interactions and ion transport in cystic fibrosis. Am J Respir Crit Care Med 170: 815–820, 2004.
22. Harris A and Coleman L. Ductal epithelial cells cultured from human foetal epididymis and vas deferens: relevance to sterility in cystic fibrosis. J Cell Sci 92: 687–690, 1989.
23. Huang Y, Chung YW, and Wong PY. Potassium channel activity recorded from the apical membrane of freshly isolated epithelial cells in rat caudal epididymis. Biol Reprod 60: 1509–1514, 1999.
24. Ishiguro H, Steward MC, Sohma Y, Kubota T, Kitagawa M, Kondo T, Case RM, Hayakawa T, and Naruse S. Membrane potential and bicarbonate secretion in isolated interlobular ducts from guinea-pig pancreas. J Gen Physiol 120: 617–628, 2002.
25. Jones RC and Murdoch RN. Regulation of the motility and metabolism of spermatozoa for storage in the epididymis of eutherian and marsupial mammals. Reprod Fertil Dev 8: 553–568, 1996.[CrossRef][Medline]
