| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Program in Membrane Biology and Renal Unit, Department of Medicine, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts 02114
b Unit of Molecular Toxicology, Institute for Medical Research and Occupational Health, 10001 Zagreb, Croatia
ABSTRACT
Fluid movement across epithelia lining portions of the male reproductive tract is important for modulating the luminal environment in which sperm mature and reside, and for increasing sperm concentration. Some regions of the male reproductive tract express aquaporin (AQP) 1 and/or AQP2, but these transmembrane water channels are not detectable in the epididymis. Therefore, we used a specific antibody to map the cellular distribution of another AQP, AQP9 (which is permeable to water and to some solutes), in the male reproductive tract. AQP9 is enriched on the apical (but not basolateral) membrane of nonciliated cells in the efferent duct and principal cells of the epididymis (rat and human) and vas deferens, where it could play a role in fluid reabsorption. Western blotting revealed a strong 30-kDa band in brush-border membrane vesicles isolated from the epididymis. AQP9 is also expressed in epithelial cells of the prostate and coagulating gland where fluid transport across the epithelium is important for secretory activity. However, it was undetectable in the seminal vesicle, suggesting that an alternative fluid transport pathway may be present in this tissue. Intracellular vesicles in epithelial cells along the reproductive tract were generally poorly stained for AQP9. Furthermore, the apical membrane distribution of AQP9 was unaffected by microtubule disruption. These data suggest that AQP9 is a constitutively inserted apical membrane protein and that its cell-surface expression is not acutely regulated by vesicular trafficking. AQP9 was detectable in the epididymis and vas deferens of 1-wk postnatal rats, but its expression was comparable with adult rats only after 3–4 wk. AQP9 could provide a route via which apical fluid and solute transport occurs in several regions of the male reproductive tract. The heterogenous and segment-specific expression of AQP9 and other aquaporins along the male reproductive tract shown in this and in our previous studies suggests that fluid reabsorption and secretion in these tissues could be locally modulated by physiological regulation of AQP expression and/or function.
epididymis, epithelial transport, male reproductive tract
INTRODUCTION
The composition of the luminal fluid in which sperm reside during their passage through the excurrent ducts (efferent ducts, epididymis, and vas deferens) is progressively modified after leaving the seminiferous tubules, and the concentration of sperm in the lumen is increased significantly [1–4]. In many tissues, water channel proteins known as aquaporins (AQPs) have been implicated in transmembrane water transport [5, 6]. We have shown previously that one member of this family, AQP1, is located in the efferent ducts where between 50% and 90% of the seminiferous fluid is absorbed after leaving the testis [7–11]. In this respect, the efferent ducts resemble renal proximal tubules that also express large amounts of AQP1 and that absorb up to 80% of the glomerular ultrafiltrate [8]. Fluid reabsorption in efferent ducts is under steroid hormone control [12, 13], and the level of AQP1 in these tubules is also dependent on estrogens and falls to almost zero in animals chronically treated with diethylstilbestrol (DES) [14, 15].
Up to now, 10 mammalian AQPs have been identified, and their expression in a number of different cells and tissues has been documented. Aquaporins are proteins of around 30 kDa in their nonglycosylated state, and both the N-terminus and the C-terminus are cytoplasmic. They span the lipid bilayer six times, and they exist as functional tetramers in the plasma membrane [6]. Some AQPs such as AQP2 and AQP4 are selective water pores, while others such as AQP3 and AQP5 are also permeable to small solutes such as glycerol and urea [16]. Most AQPs are constitutively expressed at the cell surface, a notable exception being AQP2 in the kidney, which is delivered to the cell surface by vesicle exocytosis only after stimulation of collecting duct epithelial cells with vasopressin, the antidiuretic hormone [17]. Many cell types express multiple AQPs, often spatially segregated on the apical or basolateral plasma membranes of epithelial cells. However, it is not unusual to detect the expression of different AQPs even within the same membrane domain—for example AQP3 and AQP4 on the basolateral plasma membrane of renal collecting duct principal cells [6]. The significance of such apparent duplication of function is not fully understood but may be related to the different selectivity properties of these AQPs to water and to other solutes. Membranes that express AQPs have an osmotic water permeability that can be up to two to three orders of magnitude greater that that seen in lipid bilayers lacking AQPs, and different AQPs have widely differing water permeabilities [16]. Finally, it should be noted that bulk water transport across membranes via AQP channels is not itself an active process. Water flow is dependent on the presence of an osmotic gradient across the membrane or tissue in question, the AQP being a transmembrane conduit through which osmotic flow down a concentration gradient occurs. In view of the relevance of AQPs to fluid transport in other tissues, it is likely that these proteins have a significant role to play in both reabsorption and secretion along the reproductive tract.
Several studies have shown that fluid reabsorption occurs in parts of the reproductive tract distal to the efferent ducts, notably the epididymis [10, 18–20]. We recently showed that AQP2 is highly expressed in the vas deferens of rats, but in contrast to its acute cell surface regulation by vesicle trafficking in the kidney [17], it is constitutively present on the apical plasma membrane of vas deferens epithelial cells [21, 22]. In our previous studies, neither AQP1 nor AQP2 were detected in cells lining the epididymis, despite the fact that fluid reabsorption is known to occur in this region of the excurrent duct system. Furthermore, we have also been unable to detect AQP3 and AQP4 protein expression in the epididymis using specific, affinity-purified antibodies that readily detect and distinguish these antigens in other tissues (unpublished observations). Therefore, we surmised that another member of the AQP family might be located in the epididymis, by analogy with other fluid-reabsorbing epithelia that contain AQPs. One possible candidate water channel, AQP9, was recently cloned from liver. It is a broadly selective neutral solute channel, in addition to being a water channel [23]. We show here that AQP9 is expressed not only in the epididymis but also in several other regions of the male reproductive tract including efferent ducts, vas deferens, prostate, and coagulating gland. Finally, AQP9 was detectable in the epididymis and vas deferens 1 wk after birth, but its expression reached levels comparable with adult rats only after 3–4 wk of postnatal development.
MATERIALS AND METHODS
Antibodies
An affinity-purified rabbit polyclonal antibody against an AQP9 C-terminal peptide from rat AQP9 (PSENNLEKHELSVIM—C) [23] was used for Western blotting and immunocytochemistry. The peptide was coupled to keyhole limpet hemocyanin (KLH), and antibodies were raised in rabbits using standard protocols. Antibodies were affinity purified from whole serum using the SulfoLink Kit according to the manufacturer's instructions (Pierce, Rockford, IL). Anti-AQP2 antibodies were against a C-terminal peptide and have been characterized previously [22]. A polyclonal antibody was raised in chicken against the C-terminal 14 amino acids of the 31-kDa subunit of the bovine kidney medulla proton pump, coupled to KLH, and was also affinity purified as previously described [24]. A previously characterized antibody raised against a C-terminal peptide from the Cl-/HCO3- exchanger AE2, but that cross-reacts with the AE1 anion exchanger, was provided by Dr. Seth Alper, Beth Israel Hospital, Boston [25].
Experimental Animals and Tissue Fixation
Experiments were conducted using postnatal (1–5 wk) and mature (300–350 g) male Sprague-Dawley rats (Charles River Laboratories, Wilmington, MA). Rats were anesthetized using a sodium pentobarbital (50 mg/ml) injection of 0.15 ml per 100 g body weight. The kidney and male reproductive organs were fixed via left ventricle cardiac perfusion with 150 ml Hank balanced salt solution followed by paraformaldehyde lysine periodate (PLP) fixation. Tissues from younger rats (1 and 2 wk postnatal) were fixed by immersion only. The original PLP recipe [26] was modified to increase the paraformaldehyde concentration from 2% to 4%. The kidney and reproductive organs, specifically the efferent ducts, epididymis, vas deferens, prostate, seminal vesicles, and coagulating gland were removed after 5 min perfusion in PLP and placed in PLP buffer at room temperature 4–6 h or overnight at 4°C. Tissue was washed three times in PBS and kept at 4°C in PBS containing 0.02% Na-azide prior to use.
Normal regions of human epididymis were obtained during surgery from three adults (age 50–65 yr) with testicular or bladder neck carcinoma or benign prostatic hypertrophy. The research was approved by the Zagreb Hospital Ethics Committee, and a written voluntary agreement was obtained from each patient before surgery was performed. Tissues were fixed in PLP by immersion.
Colchicine Treatment
Adult male Sprague-Dawley rats were injected with colchicine (0.5 mg/100 g body weight, i.p.) in PBS (0.9% NaCl in 10 mM sodium phosphate buffer, pH 7.4) as previously described [27]. Rats were anesthetized 12 h following colchicine treatment, and tissues were perfusion fixed in PLP as described above.
Immunofluorescence
For 3- to 4-µm cryostat sections, PLP-fixed tissues were cryoprotected in a solution of 30% sucrose in PBS for approximately 2 h at room temperature. They were embedded in OTC Compound (Tissue-Tek, Sakura Fine Technical Co., Torrance, CA) mounted on a cutting block. After freezing in a Reichert Frigocut microtome, the tissue was cut and sections were picked up on Fisher Superfrost Plus microscope slides (Fisher Scientific, Pittsburgh, PA), as previously described [22, 28]. For 1-µm cryostat sections, tissues were cryoprotected in a solution of 2.3 M (79%) sucrose before freezing in liquid nitrogen and sectioning with a Leica FCS microtome. Sections were placed on Fisher Superfrost Plus microscope slides.
For immunostaining, sections were hydrated 5 min in PBS and were pretreated with an SDS antigen retrieval technique that we have previously described [29]. Slides were washed three times in PBS for 5 min each time, followed by preincubation in 1% BSA in PBS/0.02% sodium azide for 15 min. Sections were incubated in primary anti-AQP9 antiserum, diluted 1:400 in PBS/0.02% sodium azide for 90 min at room temperature and were washed twice for 5 min in high salt PBS (2.7% NaCl) to reduce nonspecific staining, and once in normal PBS. Sections were then incubated for 1 h with secondary antibody, goat anti-rabbit IgG coupled to CY3 (Jackson Immunologicals, West Grove, PA), and they were again washed as described above. Double labeling was performed by subsequent incubation of some sections with the chicken anti-H+ATPase antibody and donkey anti-chicken IgG conjugated to fluorescein isothiocyanate (FITC; Jackson Immunologicals). The slides were coverslipped, mounted in Vectashield (Vector Labs, Burlingame, CA), diluted 1:1 with Tris buffer, pH 8.5, and examined using a Nikon E800 epifluorescence microscope. Double-stained sections were digitally imaged using a Hamamatsu Orca charge-coupled device camera and IP Lab Spectrum software (Scanalytic, Vianna, VA). Final images were imported into and printed from Adobe Photoshop.
Some sections were double stained with anti-AQP2 or anti-AE2 and anti-AQP9 antibodies. Because these antibodies were all raised in rabbit, an amplification procedure was used to allow staining of sections with the first primary antibody without cross-reactivity with the second secondary antibody. Briefly, the first affinity-purified antibody, anti-AQP2 or anti-AE2, was applied at a dilution of 1:1000 or 1:16 000, respectively, a concentration that was too low to be detectable by conventional application of a secondary fluorescent antibody in these sections, as determined in preliminary experiments. The dilute anti-AQP2 (or anti-AE2) antibody was detected using a tyramide amplification kit (NEN Life Science Products, Boston, MA) with tyramide-FITC as a fluorescent reagent, according to the manufacturer's instructions. The sections were then incubated conventionally with anti-AQP9 and secondary goat anti-rabbit CY3, as described above. No cross-reactivity between the two sets of reagents was detectable under these conditions and the findings were confirmed using separate incubations on serial sections of tissue.
Control incubations were performed using antibodies that had been preabsorbed with their respective antigens prior to the first incubation step. All control incubations were negative, illustrating the specificity of the staining patterns reported here. In addition to the negative Western blot shown in Figure 1, anti-AQP9 antibodies gave no positive staining when applied to kidney sections (not shown), indicating that they did not cross-react in our immunocytochemical procedure with any AQP that is expressed in the kidney (e.g., AQP1, 2, 3, and 4).
|
Immunogold Staining for Electron Microscopy
The male reproductive tract was fixed by perfusion and then immersion in PLP (containing 4% paraformaldehyde), as described above. The fixed tissue was washed with PBS and stored at 4°C in PBS with 0.02% sodium azide until use. Pieces of epididymis were cryoprotected by immersion in 2.3 M sucrose, and ultrathin frozen sections were cut on a Leica FCS microtome, essentially as previously described [7]. Sections were incubated with affinity-purified anti-AQP9 antibody (1:100) for 2 h at room temperature, rinsed, and incubated for 1 h at room temperature with goat anti-rabbit IgG coupled to 10-nm colloidal gold (Ted Pella Inc., Redding, CA). Sections were postembedded and stained in a solution of 2% methylcellulose containing 0.5% uranyl acetate. Grids were examined using a Philips CM10 electron microscope (Philips Electronics Inc., Mahwah, NJ).
Immunoblotting (SDS-PAGE and Western Blotting)
Rats were anesthetized and perfused through the left ventricle with PBS, pH 7.4, containing protease inhibitors (Complete tablets; Boehringer Mannheim). Tissues were removed and snap frozen in liquid nitrogen before storing at -80°C prior to use. Frozen samples were thawed in iced-cold PBS containing protease inhibitors [22] and were placed in 750 µl of lysis buffer (10 mM Tris, pH 7.4, 160 mM NaCl, 0.5% IGEPAL CA-630, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, Complete protease inhibitor; Boehringer Mannhein). The samples were homogenized using a PRO 200 homogenizer (PRO Scientific Inc., Monroe, CT).
Three epididymides from three rats were cut into small pieces and rinsed several times in PBS/protease inhibitors to remove most of the sperm. An enriched preparation of brush-border membranes was prepared using the Mg2+ precipitation technique previously described for kidney proximal tubules and small intestine [30–32]. Briefly, tissue was homogenized in 20 ml of a buffer containing 250 mM sucrose, 18 mM Tris-Hepes, 1 mM EDTA, Complete protease inhibitor (Boehringer Mannheim), pH 7.4, using a PRO 200 homogenizer followed by 20 strokes in a glass/Teflon Potter. The homogenate was incubated with 10 mM MgCl2 on ice for 20 min and was then centrifuged at 7700 x g for 15 min. The pellet was discarded, and the supernatant was further centrifuged at 20 000 x g for 15 min. The pellet was resuspended in a buffer containing 150 mM KCl and 5 mM Tris-Hepes at pH 7.4 by passing through a 25-5/8-gauge needle in a 1-ml syringe. The sample was diluted in 20 ml of resuspension buffer and centrifuged at 1900 x g for 15 min, the pellet discarded, and the supernatant was finally centrifuged at 30 900 x g for 30 min. This last pellet, enriched in brush-border membranes, was then mixed in 100 µl of the resuspension buffer using a 25-5/8-gauge needle and 1-ml syringe.
Protein concentration was determined using the bicinchoninic acid assay (Pierce, Rockford, IL). The samples were solubilized at 65°C for 15 min in NuPAGE LDS sample buffer (4x) with a final concentration of 10% NuPAGE Sample Reducing Agent (NuPAGE items from Novex, San Diego, CA). Protein was added at 20 µg per lane and was separated by SDS-PAGE and transferred to Immobilon membranes, as previously described [22]. Membranes were incubated overnight at 4°C in affinity-purified AQP9 antibody (1:2000 dilution). Goat anti-rabbit IgG conjugated to horseradish peroxidase (Sigma, St. Louis, MO) was applied to membranes for 1 h at room temperature. Proteins were detected using the Renaissance Western Blot Chemiluminescence Reagent (New England Nuclear, Boston, MA).
RESULTS
Immunoblotting (SDS-PAGE and Western Blotting)
By Western blotting, the affinity-purified anti-AQP9 antibody detected a broad band at an apparent molecular weight of 30 kDa in brush-border membranes isolated from the epididymis (Fig. 1). When total homogenates from the epididymis were used, the staining was considerably weaker, indicating that AQP9 is enriched in the brush-border membrane preparation. Fainter higher molecular bands were also detectable in the membrane preparation. The presence of differently glycosylated higher molecular weight bands has previously been described for several other AQPs [5, 6]. No stained bands were present when blots were performed using purified antibody that had been preabsorbed with the immunizing AQP9 peptide (Fig. 1B, peptide). Kidney samples (total homogenate and brush-border membranes) run as a negative control were unstained (Fig. 1A), consistent with the absence of AQP9 expression in this organ. Western blots using whole homogenates from other reproductive organs gave a weak positive staining (efferent ducts and vas deferens) or no detectable staining (all other tissues data not shown), presumably due to the low percentage of AQP9-containing membranes in the homogenates.
Immunocytochemistry on Control, Adult Rats
By indirect immunofluorescence, strong apical labeling was detected in many regions of the excurrent ducts and accessory glands. While apical staining was present in sections exposed to antibodies without prior antigen retrieval, the intensity of staining was increased considerably in sections pretreated with SDS [29]. Thus, all of the material presented in this study is taken from SDS-treated tissue sections. In efferent ducts, the staining was restricted to the apical brush border of nonciliated cells, whereas ciliated cells identified by positive basolateral AE2 staining [33] were negative (Fig. 2). In the initial segment of the epididymis, intense labeling of the apical stereocilia was observed in principal cells while proton-pump-rich apical/narrow cells were unstained (Fig. 3). In these initial segments, long apical stereocilia from principal cells often covered the apical region of adjacent apical or narrow cells, giving the impression that these cells were AQP9-positive, but absence of staining was clearly apparent in appropriate planes of section. The basolateral plasma membrane of all cell types was negative, and little or no intracellular vesicular staining was detectable. The fluorescence intensity was somewhat less intense in the caput epididymidis (Fig. 4A) but increased in intensity in the corpus and cauda (Fig. 4, B and C). The absence of AQP9 staining from proton-pump-rich cells was most evident in clear cells in the corpus and cauda epididymidis (Fig. 4, B and C) that have a large apical surface that can easily be distinguished from the surrounding principal cells. No staining was detectable in any tubule segment when antibody preabsorbed with the AQP9 immunizing peptide was used (not shown; see also Figs. 5, 7, and 8 for control incubations). As shown in Figure 5A, principal cells of the human epididymis also showed extensive apical staining for AQP9 in a pattern similar to that described for the rat tissue. While there are some differences between the C-terminal sequences of rat and human AQP9 [23, 34], the homology is sufficient (eight out of the nine C-terminal amino acids are identical) to allow the anti-rat AQP9 antibody to cross-react with the human protein. The staining of human tissue was abolished when sections were incubated with preabsorbed antibody (Fig. 5B).
|
|
|
|
Aquaporin 9 staining was found throughout the entire length of the vas deferens (Fig. 6). As we have previously described, the initial portion is characterized by the presence of proton-pump-rich cells, and the principal cells do not express AQP2 [22, 35]. In contrast, AQP9 was abundantly expressed on apical stereocilia of principal cells in this region (Fig. 6A). Principal cells in the distal vas deferens expressed apical AQP2 as previously described [22] but also contain AQP9 in the same membrane domain (Fig. 6, C and D). No basolateral staining for AQP9 was detectable in the vas deferens, and intracellular staining was weak or absent.
|
Seminal vesicles did not contain detectable AQP9, while the epithelium of the coagulating gland showed a bright apical staining of all epithelial cells (Fig. 7A). The staining in the coagulating gland was completely abolished by prior incubation of the antibody with the immunizing peptide (Fig. 7B) The prostatic epithelium showed a variable level of staining that appeared to be intracellular (Fig. 8A), although a faint apical membrane staining was detected. This staining in the prostate was also abolished by preincubation of the antibody with AQP9 peptide (Fig. 8B).
|
|
Electron Microscopic Immunogold Localization of AQP9 in Epididymis
To confirm the apical membrane microvillar/stereocilia location of AQP9 in epididymal principal cells, immunogold staining was performed on ultrathin cryostat sections of the epididymis. Figure 9 shows that the apical microvilli were extensively labeled with gold particles, and that little or no staining was associated with cytoplasmic vesicles. Furthermore, apical membrane depressions corresponding to sites of clathrin-coated pits were unlabeled, indicating that AQP9 is not extensively internalized by a clathrin-mediated mechanism in these cells.
|
Aquaporin 9 Expression During Postnatal Development
To determine the pattern of AQP9 expression in the postnatal excurrent duct system, rat tissues were examined at different times after birth, from 1 wk to adult. In tissues from 1-wk-old rats, weak apical staining was detectable in some regions of the epididymis (Fig. 10A), while other regions remained unstained (not shown). By 2 wk, a stronger apical staining in tubules from the proximal region of the epididymis was detectable (Fig. 10B). After 3 and 4 wk, the apical staining had increased in intensity in all regions of the epididymis and was indistinguishable from the intensity of staining seen in adult tissue (Fig. 10C). All staining was completely abolished by preincubation of the antibody with AQP9 peptide. An example of the inhibition of staining in tissue from a 25-day-old rat epididymis is illustrated (Fig. 10D). A strong apical staining was seen in the vas deferens at all time points (not shown).
|
Aquaporin 9 Distribution in Colchicine-Treated Rats
We have previously reported that microtubule disruption by colchicine or low temperature causes extensive intracellular redistribution of membrane proteins that are recycling extensively between the cell surface and intracellular vesicles [27, 36–38]. In contrast, constitutive membrane proteins that have a longer residence time on the plasma membrane for several hours are less affected by microtubule disruption during the time frame of colchicine treatment (in this case, 12 h). As shown in Figure 11, AQP9 was still concentrated at the apical pole of principal cells in rats that had been treated with colchicine for 12 h. The staining was consistent with the apical microvillar/stereocilia staining seen in control rats. Little or no redistribution into intracellular vesicles was detected. In contrast, a rapidly recycling protein, the H+ATPase, is redistributed throughout the cytoplasm of clear cells in the same tissue after microtubule disruption (Fig. 11), as we have previously described [24, 38]. These data indicate that AQP9, like AQP1 in the efferent duct [7] and like AQP2 in the vas deferens [22], is not a rapidly recycling membrane protein.
|
DISCUSSION
Fluid and electrolyte transport across epithelial cells lining the excurrent duct system of the male reproductive tract significantly modifies the luminal environment in which spermatozoa mature and are stored. These processes begin in the rete testis, but most of the fluid reabsorption occurs in the efferent ducts and more distal segments of the excurrent duct system [8–12, 18, 20]. Our present data show that a recently identified member of the AQP family, AQP9, is abundantly expressed in different segments of the male reproductive tract, where it could represent an apical pathway for transepithelial water flow. Aquaporin 9 mRNA was first detected in hepatocytes and in some other cell types including immature spermatocytes and Leydig cells in rat testis [23]. Aquaporin 9, however, was not detected in human testis [34]. Expression studies in Xenopus oocytes revealed that it is a so-called promiscuous AQP that allows the passage not only of water, but also of other solutes such as urea and polyols including mannitol, as well as purines and pyrimidines [23]. Thus, AQP9 may have functions in addition to fluid transport across membranes. This may explain the presence of more than one member of the AQP family in the same membrane domain of some epithelial cells lining different portions of the male reproductive tract.
Nonciliated cells in the efferent ducts express abundant AQP1 on their apical, and to a lesser extent, their basolateral plasma membranes [7, 14]. We now show that nonciliated cells also express apical (but not basolateral) AQP9. Between 50% and 80% of the fluid produced in the seminiferous tubules is reabsorbed in efferent ducts, and it is likely that AQP1 plays a major role in this process, by analogy with the function of this protein in proximal tubule fluid reabsorption in the kidney [8, 39, 40]. Interestingly, the expression of AQP1 in efferent ducts, and fluid reabsorption in these tubules, depends on exposure to an appropriate level of estrogens [12, 14, 15]. Fluid reabsorption in the efferent ducts leads to a considerable concentration of sperm in this tubule segment, and it is surprising, therefore, that male AQP1 knockout mice seem to have a reproductive capacity that is not markedly different from wild-type mice (A.S. Verkman, personal communication), although definitive reproductive studies have not yet been performed. It now seems possible that the presence of AQP9 in the efferent ducts might compensate, at least partially, for the loss of AQP1 from this tissue. Permeability studies on isolated, perfused efferent ducts will be necessary to determine the relative contributions of AQP1 and AQP9 to transepithelial fluid movement in this epithelium.
While a considerable amount of fluid reabsorption occurs in the efferent ducts, significant transepithelial water reabsorption also occurs in more distal regions of the excurrent duct system, including the epididymis. In earlier studies, neither AQP1 nor AQP2 were detected in the epididymis [7, 22]. Two other basolateral water channels, AQP3 and AQP4 were also undetectable in the epididymis using specific anti-peptide antibodies that readily detect these proteins in the kidney and other organs (unpublished observations). Our data now show that AQP9 is an abundant apical membrane protein in all regions of the rat epididymis, and that it is also present in human epididymis. In these tissues, as well as other regions of the reproductive tract including the coagulating glands, it appears to be a constitutive apical membrane protein that may be responsible for apical membrane water and/or solute permeability of these epithelia.
Different members of the AQP water channel family follow distinct intracellular trafficking and targeting pathways. Most AQPs are delivered to either the apical or the basolateral plasma membrane of epithelial cells in a constitutive (i.e., nonregulated) manner, while in the kidney, AQP2 membrane insertion is stimulated by the antidiuretic hormone, vasopressin [17]. We have previously shown, however, that in the vas deferens, AQP2 follows a constitutive rather than a regulated pathway of membrane insertion [22]. Thus, putative targeting signals on a given protein can be interpreted in a cell-type specific fashion. Our data from colchicine-treated rats strongly suggest that in the epididymis, AQP9 also follows a constitutive pathway of plasma membrane insertion and that it does not rapidly recycle between the plasma membrane and intracellular vesicles. We have previously shown that microtubule disruption by colchicine (or by cold exposure of tissue) results in a marked redistribution of rapidly recycling membrane proteins from the cell surface into intracellular vesicles, whereas nonrecycling, or slowly recycling proteins are not affected by this maneuver [36, 37]. Aquaporin 9 clearly falls into the latter category as its distribution was similar in colchicine-treated and control rats. This conclusion is supported by light and electron microscopic immunocytochemical data showing a relative paucity of AQP9-positive cytoplasmic vesicles in principal cells of the epididymis, and by the absence of AQP9 immunostaining in apical endocytotic clathrin-coated pits in the epididymis. However AQP9 was located mainly on intracellular structures in the prostate, whereas we have previously reported that AQP1 is expressed on the plasma membrane of prostatic epithelial cells [7]. This intracellular distribution is not unique to AQP9, as another AQP, AQP6, has been found predominantly on intracellular vesicles in renal intercalated cells, where it may function as a chloride-conducting pathway [41].
In addition to being expressed in the epididymis, AQP9 is present along the entire length of the vas deferens. In contrast, AQP2 is expressed in the distal portion of the rat vas deferens exclusively and is not detectable in the proximal portion closest to the epididymis [21, 22]. Therefore, in the proximal portion, AQP9 is the only AQP so far detected on principal cells, whereas in the distal portion it is coexpressed with AQP2 on principal cell apical membranes. As is the case for other cell types that express more than one AQP in the same membrane domain, the functional significance of this apparent redundancy of AQP expression is unknown. For example, renal principal cells in some parts of the collecting duct have high levels of both AQP3 and AQP4 on their basolateral plasma membranes [42–44]. In this case also, one of the expressed channels (AQP4) is conductive only to water, while the other (AQP3) is a promiscuous water channel with a high urea and glycerol permeability [16, 45]. This already complex picture is confounded further in the ampulla of the vas deferens and the efferent duct by the additional expression of yet another water-selective AQP, AQP1, in both apical and basolateral plasma membrane domains [7]. Once again, the relative contributions of these proteins to transepithelial transport will require functional measurements on single- and double-AQP knockout mice, as we have recently described for kidney functional parameters in AQP1 and AQP4 double knockouts [46].
Interestingly, AQP9 was readily detectable on the apical membrane of cells of the coagulating gland but was not detectable in the seminal vesicle. The seminal vesicle produces a large amount of protein-rich secretion [47], and it is conceivable that expression of apical AQP9, which would increase apical membrane water permeability and allow fluid reabsorption to occur, might be incompatible with the production and storage of its secretory product. The coagulating gland produces a smaller volume of secretion, and it is believed that apical membrane blebbing or apocrine secretion is a major (but not the exclusive) pathway for protein export from this gland [48, 49]. The formation, swelling, and eventual detachment of these apical blebs might require localized fluid transport that would be increased by the presence of AQP9 in the apical membrane. However, until a more complete study using reagents that can detect the entire spectrum of mammalian AQPs is performed, the functional role of increased or decreased apical membrane water permeability in these accessory glands remains a matter of speculation.
In developing rats, AQP9 was detectable in some cells of the developing excurrent ducts 1 wk postnatally. However, expression levels were already high in the vas deferens at this early time point. A progressive increase in AQP9 expression in the epididymis was seen during postnatal development, both in terms of the number of tubule segments expressing AQP9 and the intensity of AQP9 staining in individual cells within these segments. In 3- to 4-wk postnatal rats, the expression levels and AQP9 distribution were indistinguishable from the adult pattern. Because androgens are present throughout the perinatal period and their levels increase during puberty, it is possible that AQP9 expression is modulated by androgens.
In summary, our data show that AQP9 is a major apical water channel that is expressed throughout the efferent ducts, epididymis (including human epididymis), and vas deferens, as well as in other regions of the male reproductive tract. This AQP could represent an important apical pathway for transmembrane water movement, and it could also serve as a conduit for other solutes, based on its known promiscuity [23]. The nature of the presumptive basolateral water channel in the epididymis that would complete the transepithelial water transport pathway in this part of the reproductive tract remains to be determined, because we were unable to detect either AQP3 or AQP4 (both basolateral proteins in the kidney) protein expression in epididymal epithelial cells (data not shown). The heterogenous and segment-specific expression of AQP9 and other AQPs along the male reproductive tract shown in this and in our previous studies suggests that fluid reabsorption and secretion in these tissues could be locally modulated by physiological regulation of AQP expression and/or function.
FOOTNOTES
First decision: 26 January 2001.
1 This work was supported by NIH grant DK38452 (D.B. and S.B.), DK55864 (A.V.H.), and an NIH Fogerty International Research Collaboration Award TW01057 (I.S. and D.B.). The Microscopy Core Facility of the MGH Program in Membrane Biology is additionally supported by a CSIBD Center Grant DK43351 and a DERC award DK57521. Research fellowships from INSERM, the Arthur Sachs Foundation (as part of the Fullbright Program), L'Assistance Publique, and the Association des Femmes Diplômées des Universités supported C.B. N.P.-S. was supported by an NIH NRSA HD08684. 
2 Correspondence: Dennis Brown, Program in Membrane Biology/Renal
Unit, Massachusetts General Hospital East, 149 13th Street, Charlestown,
MA 02129. FAX: 617 726 5669; brown{at}receptor.mgh.harvard.edu
Accepted: March 14, 2001.
Received: December 8, 2000.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  A. Wellejus, H. E. Jensen, S. Loft, and T. E. Jonassen Expression of Aquaporin 9 in Rat Liver and Efferent Ducts of the Male Reproductive System After Neonatal Diethylstilbestrol Exposure J. Histochem. Cytochem., May 1, 2008; 56(5): 425 - 432. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Pietrement, N. Da Silva, C. Silberstein, M. James, M. Marsolais, A. Van Hoek, D. Brown, N. Pastor-Soler, N. Ameen, R. Laprade, et al. Role of NHERF1, Cystic Fibrosis Transmembrane Conductance Regulator, and cAMP in the Regulation of Aquaporin 9 J. Biol. Chem., February 1, 2008; 283(5): 2986 - 2996. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. M. Pastor-Soler, K. R. Hallows, C. Smolak, F. Gong, D. Brown, and S. Breton Alkaline pH- and cAMP-induced V-ATPase membrane accumulation is mediated by protein kinase A in epididymal clear cells Am J Physiol Cell Physiol, February 1, 2008; 294(2): C488 - C494. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Dube, L. Hermo, P. T.K Chan, and D. G Cyr Alterations in Gene Expression in the Caput Epididymides of Nonobstructive Azoospermic Men Biol Reprod, February 1, 2008; 78(2): 342 - 351. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Thimon, O. Koukoui, E. Calvo, and R. Sullivan Region-specific gene expression profiling along the human epididymis Mol. Hum. Reprod., October 1, 2007; 13(10): 691 - 704. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M Hagedorn, R. W Carlin, and B. D Schultz Oxytocin and Vasopressin Stimulate Anion Secretion by Human and Porcine Vas Deferens Epithelia Biol Reprod, September 1, 2007; 77(3): 416 - 424. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Da Silva, W. W. C. Shum, J. El-Annan, T. G. Paunescu, M. McKee, P. J. S. Smith, D. Brown, and S. Breton Relocalization of the V-ATPase B2 subunit to the apical membrane of epididymal clear cells of mice deficient in the B1 subunit Am J Physiol Cell Physiol, July 1, 2007; 293(1): C199 - C210. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Rojek, M. T. Skowronski, E.-M. Fuchtbauer, A. C. Fuchtbauer, R. A. Fenton, P. Agre, J. Frokiaer, and S. Nielsen Defective glycerol metabolism in aquaporin 9 (AQP9) knockout mice PNAS, February 27, 2007; 104(9): 3609 - 3614. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-F. Huang, R.-H. He, C.-C. Sun, Y. Zhang, Q.-X. Meng, and Y.-Y. Ma Function of aquaporins in female and male reproductive systems Hum. Reprod. Update, November 1, 2006; 12(6): 785 - 795. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. W. Carlin, R. L. Sedlacek, R. R. Quesnell, F. Pierucci-Alves, D. M. Grieger, and B. D. Schultz PVD9902, a porcine vas deferens epithelial cell line that exhibits neurotransmitter-stimulated anion secretion and expresses numerous HCO3- transporters Am J Physiol Cell Physiol, June 1, 2006; 290(6): C1560 - C1571. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Da Silva, C. Silberstein, V. Beaulieu, C. Pietrement, A. N. Van Hoek, D. Brown, and S. Breton Postnatal Expression of Aquaporins in Epithelial Cells of the Rat Epididymis Biol Reprod, February 1, 2006; 74(2): 427 - 438. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Hermo, D. L. Chong, P. Moffatt, W. S. Sly, A. Waheed, and C. E. Smith Region- and Cell-specific Differences in the Distribution of Carbonic Anhydrases II, III, XII, and XIV in the Adult Rat Epididymis J. Histochem. Cytochem., June 1, 2005; 53(6): 699 - 713. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. C. Branes, B. Morales, M. Rios, and M. J. Villalon Regulation of the immunoexpression of aquaporin 9 by ovarian hormones in the rat oviductal epithelium Am J Physiol Cell Physiol, May 1, 2005; 288(5): C1048 - C1057. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.-h. Cheung, G. P.H. Leung, M. C.T. Leung, W. W.C. Shum, W.-l. Zhou, and P. Y.D. Wong Cell-cell Interaction Underlies Formation of Fluid in the Male Reproductive Tract of the Rat J. Gen. Physiol., April 25, 2005; 125(5): 443 - 454. [Abstract] [Full Text] |
