| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
From the Department of Anatomy and Cell Biology, McGill University, Montreal, Quebec, Canada.
| Correspondence to: Dr Louis Hermo, McGill University, 3640 University St, Montreal, Quebec, Canada H3A 2B2 (e-mail: lhermo{at}med.mcgill.ca ). |
| Received for publication September 28, 2001; accepted for publication November 20, 2001. |
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Key words: Water channels, Sertoli cells, Leydig cells, principal cells, clear cells, nonciliated cells
-helical
domains and has a distinct pore and molecular weight of approximately 30 kd in
the nonglycosylated state (King and Agre,
1996; Wintour,
1997; Verkman and Mitra,
2000). Thus far, 10 members of the AQP family (0-9) have been
identified (Deen and van Os,
1998; Echevarria and Ilundain,
1998; King et al,
2000; Verkman and Mitra,
2000). On the basis of different permeability properties, AQPs
have been subdivided into AQPs with real water sensitive pores and the
aquaglyceroporins with slightly less sensitive pores
(Borgnia et al, 1999; van Os et al, 2000;
Sansom and Law, 2001). AQPs have been analyzed extensively in tissues involved in fluid transport and show a wide range of distribution (Brown et al, 1995; Wintour, 1997; Deen and van Os, 1998; Echevarria and Ilundain, 1998; Verkman and Mitra, 2000). While some are tissue-specific, more than 1 AQP can be present in the same tissue and even cell type (King and Agre, 1996; Wintour, 1997; Echevarria and Ilundain, 1998). Some AQPs are highly selective for the passage of water, while others also permeate urea, glycerol, and other small noncharged solutes (Kuriyama et al, 1997; Echevarria and Ilundain, 1998; Borgnia et al, 1999; Tsukaguchi et al, 1999). AQPs may also be involved in membrane fluidity and structural integrity, as well as tumor growth and angiogenesis (Verkman and Mitra, 2000). Some AQPs are constitutively expressed, while others are regulated by hormones as well as by pH variations, phosphorylation, and binding of auxiliary proteins (Brown et al, 1995, 1998; Deen and van Os, 1998; Echevarria and Ilundain, 1998; Engel et al, 2000; Verkman and Mitra, 2000). Various disease states have been associated with alterations in AQP expression and targeting in cells (Lee et al, 1997; Connolly et al, 1998; King et al, 2000; van Os et al, 2000; Verkman and Mitra, 2000).
In the seminiferous epithelium of the testis, Sertoli cells continuously produce fluid in which the developing germ cells are bathed. Water moves from the interstitial space into the lumen via a standing osmotic gradient created by ionic pumps restricted to the apicolateral plasma membranes of Sertoli cells (Setchell et al, 1969; Byers and Graham, 1990; Hinton and Setchell, 1993). Water is important in creating the fluid environment for the passage of sperm toward the rete testis and eventually to the epididymis (Voglmayr et al, 1970; Waites and Einer-Jensen, 1974; Hinton and Setchell, 1993).
In the efferent ducts, the nonciliated epithelial cells reabsorb 50%-90% of the fluid coming from the testis, and water continues to be removed from the lumen of the epididymis (Crabo, 1965; Levine and Marsh, 1971; Wong and Yeung, 1978; Setchell and Brooks, 1988). The localization of Na+/K+-adenosine triphosphatase (ATPase) in the basolateral position of the epithelium of the efferent ducts and epididymis serves to create a standing osmotic gradient whereby water passes from the lumen into the interstitial space (Byers and Graham, 1990; Ilio and Hess, 1992).
Tissue distribution and regulation of AQPs have been examined in some detail in the male reproductive tract. AQP-1 and -9 have been localized to the efferent ducts of the rat, monkey, and human (Brown et al, 1993; Fisher et al, 1998; Pastor-Soler et al, 2001), with AQP-1 having been proposed to be regulated by estrogens (Fisher et al, 1998). AQP-9 has been localized to the Leydig cells of the rat testis (Elkjaer et al, 2000; Nihei et al, 2001) and, in the rat epididymis, to the principal cells (Elkjaer et al, 2000; Pastor-Soler et al, 2001). AQP-8 has been suggested to be expressed by germ cells, spermatocytes exclusively, or the luminal side of seminiferous tubules, corresponding to Sertoli cells of the rat testis (Calamita et al, 2001; Elkjaer et al, 2001; Kageyama et al, 2001; Tani et al, 2001). In the rat epididymis, AQP-8 either was not found to be expressed (Calamita et al, 2001) or was noted in basal cells (Elkjaer et al, in press). Aside from the controversy of AQP expression in the testis and epididymis, there has not been much work done on the distribution of AQPs during postnatal development (Kageyama et al, 2001; Pastor-Soler et al, 2001) and their regulation in different cell types of the rat testis, efferent ducts, and epididymis.
The purpose of the present study was to localize AQP-1, -8, and -9 in cells of the testis, efferent ducts, and epididymis of the rat using Bouin-fixed, paraffin-embedded material for light microscope immunocytochemistry, with subcellular localization being assessed by electron microscope immunocytochemistry on frozen ultrathin sections. In particular, the cell-, region-, and tissue-specific distribution of these 3 AQPs was examined in the different areas of the adult male reproductive tract. In addition, the postnatal developmental pattern of the expression of AQP-1, -8, and -9 was examined, as well as their regulation by testicular factors in experimentally treated adult animals.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Immunoperoxidase Staining— Immunoperoxidase staining of sections was carried out according to the procedure of Oko and Clermont (1989). Polyclonal, affinity-purified anti-AQP antibodies were used at different dilutions in Tris-buffered saline (TBS), pH 7.4. The anti-AQP antibodies were obtained from Alpha Diagnostics International (San Antonio, Tex). The antibodies have been well characterized and were found to be specific to their respective peptides. The anti—AQP-1 antibody was raised against a 19 amino acid synthetic peptide within the carboxy terminus of the protein; this peptide contains most of the epitopes that are recognized by the polyclonal antibodies raised in rabbits against the entire protein (Denker et al, 1988). The anti—AQP-8 antibody was raised against a 16 amino acid synthetic peptide in the carboxy terminus of the protein (Koyama et al, 1997). The anti—AQP-9 antibody was raised against an 18 amino acid synthetic peptide within the carboxy terminus of the protein (Ishibashi et al, 1998). All antibodies were affinity-purified over a control peptide-Sepharose column and were found to be specific to their targets. They were supplied as a 1-mg/mL solution in phosphate-buffered saline (PBS), pH 7.4, with 0.1% bovine serum albumin as stabilizer. The antibodies also contained 0.1% sodium azide as a preservative.
Paraffin sections, 5 µm thick, were deparaffinized in Histoclear (Diamed Lab Supplies Inc, Mississauga, ON, Canada) and hydrated in a series of graded ethanol solutions. During hydration, residual picric acid was neutralized in 70% ethanol containing 1% lithium carbonate, and endogenous peroxidase activity was abolished in 70% ethanol containing 1% (vol/vol) H2O2. Once hydrated, the tissue sections were washed in distilled water containing glycine to block free aldehyde groups.
Before immunostaining, the sections were blocked for 15 minutes with 10% goat serum in TBS. This and subsequent treatments were accomplished by placing 100 µL of a solution onto a coverslip and overturning the tissue face of the slide onto the drop, thus ensuring that the entire tissue was treated with minimal fluid (Oko and Clermont, 1989). Coverslips were removed by dipping the slides in TBS containing 1% Tween-20 (TWBS). Sections were then incubated in a 37°C humidified incubation chamber for 1.5 hours with the primary antibody at a dilution of 1:100 (protein concentration of 0.01 mg/mL). After three 2-minute washes in TWBS, sections were once again blocked with 10% goat serum in TBS. They were then incubated for 30 minutes (at 37°C) with goat anti-rabbit immunoglobulin G conjugated to horseradish peroxidase (Sigma Chemical Co, St Louis, Mo) diluted 1:250 in TBS. This was followed by three 2-minute washes in TWBS.
The final product was achieved by incubating the sections for 10 minutes in 500 mL of TBS containing 0.03% H2O2, 0.1 M imidazole, and 0.05% diaminobenzidine tetrahydrochloride (Sigma), pH 7.4. Slides were then washed in distilled water and counterstained with 0.1% methylene blue. The tissue was dehydrated by passing the slides through a graded ethanol series, after which the sections were immersed in Histoclear and mounted with Permount. Specificity of the immunostaining was confirmed in tissues by incubation without the primary antibody and using normal rabbit serum.
Electron Microscope Immunocytochemistry
Tissue Preparation—
Four adult male 90-day-old Sprague-Dawley
rats (350-450 g) were anesthetized with sodium pentobarbital, and their
efferent ducts were fixed by perfusion through the abdominal aorta with a
fixative containing 0.5% glutaraldehyde and 4% paraformaldehyde in 0.1 M
phosphate buffer, pH 7.4. Following its removal, the tissue was trimmed into
small pieces (0.5 mm3) and immersed for 2 hours in the above fixative at
4°C, followed by washing 2-3 times in 0.15 M PBS, pH 7.4, and next treated
with PBS containing 1.5 M sucrose. The tissue was then frozen in liquid
nitrogen until sectioned.
Immunogold Labeling of Frozen Sections— Ultrathin frozen sections of normal adult 90-day-old rat efferent ducts were mounted on 200-mesh, formvar-coated copper grids. Each grid was blocked for 15 minutes on a drop of 2% bovine serum albumin, 2% casein, and 0.5% ovalbumin (BCO) and then incubated for 1 hour on 15-µl drops of primary antibody diluted in BCO at a ratio of 1:1 (protein concentration of 0.5 mg/mL) for AQP-1 and a ratio of 1:10 (protein concentration of 0.1 mg/mL) for AQP-9. Grids were washed 3 times for 5 minutes each in Dulbecco phosphate buffered saline (DPBS). The labeled grids were incubated for 1 hour on goat anti-rabbit antibodies conjugated to 10- (AQP-1) and 15-nm (AQP-9) colloidal gold particles. The sections were then washed 3 times for 5 minutes each in DPBS, followed by several 5-minute washes in distilled water. In order to enhance membrane morphology, grids were stained with uranyl oxalic acid in water (0.3 M) for 5 minutes. Grids were protected with 2% methylcellulose prior to viewing. Electron micrographs were taken on a Philips 400 electron microscope. Specificity of the immunolabeling was confirmed by incubation without the primary antibody and using normal rabbit serum.
Regulation Studies Involving Orchidectomy and Efferent Duct Ligation— Adult male 90-day-old Sprague-Dawley rats (350-450 g) were obtained from Charles River Laboratories. The animals were subsequently subdivided into 5 groups. The first group consisted of 4 normal untreated animals. Bilateral ligation of the efferent ducts constituted the second group. After an intraperitoneal injection of sodium pentobarbital (Somnitol, MTC Pharmaceuticals), the testes and epididymides of 4 rats at each interval were exposed through an incision of the anterior abdominal wall. A ligature was placed around both right and left efferent ducts at a site close to the rete testis. The animals were sacrificed at 3, 7, 14, and 21 days following surgery. Bilateral orchidectomy constituted the third group. After anesthesia, both testes of 4 rats at each interval were cut away after a ligature was placed around the efferent ducts and testicular blood vessels. The animals were sacrified at 3, 7, 14, and 21 days after surgery. Bilaterally orchidectomized rats that received three 6.2-cm testosterone-filled implants constituted the fourth group. Testosterone-filled polydimethyl-siloxane (silastic) implants were prepared according to the method of Stratton et al (1973) and have well-characterized steroid release rates (Brawer et al, 1983). Subsequent to anesthesia, both testes were removed from 4 rats of each interval, and the implants were placed subcutaneously immediately after orchidectomy. The rats were sacrificed at 3, 7, and 14 days after surgery. The fifth group consisted of 4 sham-operated animals, 2 of which received 3 empty 6.2-cm-long implants, with all rats being sacrificed 14 days after initiation of the experiment.
Postnatal Development— Timed pregnant female Sprague-Dawley rats were obtained from Charles River Laboratories. Male pups were chosen from a number of litters and were maintained on a 14-hour dark, 10-hour light cycle. They were provided with food and water ad libitum. After birth, the normal development of the male pups was monitored by assessing body weight gain and by palpating their testes and epididymides. Only those pups showing normal trends in development, as reported by Hermo et al (1992c), were used. At each of the following days after birth (ie, 7, 21, 29, 39, 49, and 56), 6 rats were selected. At each interval, 2 rats were used to obtain their body weights and the weights of their testes and epididymides, while the other 4 rats were used to prepare the tissue for light microscope immunocytochemical analysis using anti-AQP antibodies as described above for normal adult 90-day-old animals.
All experimentation was carried out with minimal stress and discomfort being placed on the animals both during and after surgery as set up by the guidelines and approval of the University Animal Care Committee.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
In the epididymis, AQP-1 was expressed only over the myoid cells at the periphery of the epididymal tubules of the initial segment (Figure 3). No reaction was observed over the entire epithelium of any epididymal region (Figures 3 and 4). However, throughout the entire epididymis and efferent ducts, an intense reaction was noted over the endothelial cells of vascular channels located in the intertubular spaces (eg, Figure 4). AQP-1 was not expressed in any cell type of the testis.
|
|
In the testis, with anti—AQP-8 antibody, the reaction was noted exclusively over the seminiferous epithelium and not in the interstitial space (eg, Figure 5). The reaction appeared as a filamentous network stretching across the entire adluminal compartment of the epithelium, with no reaction being present over the basal compartment (Figure 5). Staining was similar at all 14 stages of the cycle of the seminiferous epithelium. The staining pattern corresponded to that of the stellate-shaped Sertoli cells. There was no staining of the tails of spermatids in the lumen, nor of residual bodies, remnants of the cytoplasm of step 19 spermatids (Figure 5). AQP-8 was not expressed in any cell type of the efferent ducts or epididymis.
|
AQP-9 expression in the testis was limited to the interstitial space. An intense immunoperoxidase reaction product in the form of a lacylike network outlined the periphery of Leydig cells (Figure 6). No reaction was seen over any cell type of the seminiferous epithelium.
|
In the efferent ducts, tufts of reaction product were noted over the apex of the nonciliated cells that corresponded to the staining of their microvilli (Figure 7). The cilia of ciliated cells were unreactive.
|
In the epididymis, AQP-9 expression was cell- and region-specific. In the initial segment, intense reactivity for anti—AQP-9 antibody was noted over the microvilli of the principal cells (Figure 8). However, the principal cells displayed only a weak to moderate reaction product over their microvilli in the caput and corpus regions of the epididymis, with no reaction being observed over the clear cells of these regions (data not shown). However, in the cauda epididymidis, intense reactivity was noted over the microvilli of the principal cells, and these cells also contained reactive spherical vesicles in their apical cytoplasm (Figure 9). Furthermore, the clear cells of the cauda epididymidis were intensely reactive (Figure 9). Sperm in the lumen of the efferent ducts and epididymis were consistently unreactive for anti—AQP-9 antibody (Figures 8 and 9).
|
|
Regulation of AQPs in the Efferent Ducts and Epididymis of Adult
Animals
At all time points after orchidectomy (up to 21 days), AQP-1 expression
over microvilli of nonciliated cells of the efferent ducts was comparable to
that of normal untreated animals (Figure
10), and there was no change in the expression over endothelial
cells of vascular channels (data not shown). Likewise, AQP-9 expression in the
efferent ducts was unaltered following orchidectomy (data not shown). However,
as early as 3 days after orchidectomy and at all later time points, the
principal cells of the initial segment of the epididymis exhibited little
reaction over their microvilli with anti—AQP-9 antibody
(Figure 11a). Likewise, the
reactivity of the clear cells of the cauda region was diminished, but only at
the 14- and 21-day time points after orchidectomy
(Figure 12). In contrast, the
principal cells of the caput, corpus, and cauda regions showed the same
pattern and intensity of reactivity at all time points after orchidectomy as
noted in normal animals (Figure
12). The administration of testosterone at high levels to
orchidectomized animals did not restore the expression to the control levels,
either in the case of the principal cells of the initial segment
(Figure 11b) or of the clear
cells of the cauda region (Figure
13).
|
|
|
|
Ligation of the efferent ducts did not affect expression in the nonciliated cells of the efferent ducts at any time point examined with either anti—AQP-1 or anti—AQP-9 antibody. In the efferent ducts and the epididymis, AQP-1 expression over the endothelial cells of vascular channels was also unaffected (data not shown). In contrast, efferent duct ligation at 3 days and all later time intervals abolished AQP-9 expression over the microvilli of the principal cells of the initial segment of the epididymis and diminished the expression over the clear cells of the cauda epididymidis (data not shown). However, there was no effect of ligation on AQP-9 expression over the principal cells of the caput, corpus, and cauda epididymidis.
Expression of AQPs During Postnatal Development
In the efferent ducts, AQP-1 expression was noted as early as postnatal day
7, with staining being observed over the microvilli of the undifferentiated
epithelial cells and their basolateral plasma membranes
(Figure 14). This staining
pattern persisted as such into adulthood. The endothelial cells of the
vascular channels also displayed intense reactivity as early as postnatal day
7, with the expression continuing into adulthood (data not shown).
|
There was no expression of AQP-9 in the testis, efferent ducts, and epididymis by postnatal day 7. However, by postnatal day 21, Leydig cells displayed a lacylike reaction over their periphery that was maintained into adulthood (data not shown). Tufts of reaction product by day 21 appeared at the apex of nonciliated cells of the efferent ducts with anti—AQP-9 antibody (Figure 15) that were comparable to those seen in adult animals. While the principal cells of the initial segment showed a weak reaction product over their microvilli by day 21 (data not shown), the reaction became intense by day 29 (Figure 16), when it was comparable to that of normal adult animals. The principal cells of the caput and corpus regions of the epididymis displayed a weak to moderate reaction by day 21 with no expression in the clear cells of this region (data not shown), a pattern similar to that seen in normal adult animals. In the cauda epididymidis, the principal cells displayed an intense reactivity over their microvilli by postnatal day 21 comparable to that of normal adult animals (Figure 17). However, while the clear cells of the cauda region remained unreactive at day 21 (Figure 17), they showed intense reactivity by day 39 (Figure 18).
|
|
|
|
Electron microscope immunocytochemistry was performed on the efferent ducts with the anti—AQP-1 and -9 antibody. Numerous gold particles representing AQP-9 antigenic sites decorated the microvilli of the nonciliated cells, while there was little labeling over apical endocytic vesicles (Figure 19a). On the other hand, AQP-1 was expressed on the microvilli of nonciliated cells as well as over numerous apical endocytic vesicles (Figure 19b).
|
Control sections, where either the primary antibody was omitted or normal rabbit serum was used, revealed only an occasional gold particle in a given field, in comparison with the abundance of gold particles (approximately 50-fold) in tissues where the primary antibody was added. This finding is consistent with low background levels of labeling noted in the case of electron microscope immunocytochemistry. For light microscopy, an absence of reaction from tissue sections was noted comparable to that already demonstrated by us in several previous studies (Hermo et al, 1991; Veri et al, 1993).
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
|
In the testis, Sertoli cells continuously produce fluid in which the developing germ cells are bathed, which also serves as a medium to allow sperm to enter the efferent ducts (Setchell et al, 1969; Hinton and Setchell, 1993). Thus, AQP-8 in Sertoli cells would be involved in the transport of water from the interstitial space into the lumen. This appears to occur along a standing osmotic gradient, with Na+/K+-ATPase already being localized to Sertoli cells (Barham et al, 1976; Gravis et al, 1976). It is important to note that AQP-8 expression was restricted solely to the apicolateral plasma membranes of the Sertoli cells where these pumps are located.
In the present study, the distribution of AQP-9 was seen as an extensive lacylike network in the interstitial space. In the rat testis, lymphatic channels do not exist; instead, the interstitial space is considered a lymphatic sinusoid with Leydig cells and macrophages bathing in the lymph contained therein (Fawcett et al, 1973). Thus, the expression of AQP-9 in the interstitial space would not be related to lymphatics. Rather, AQP-9 expression appears to be related to the surface of Leydig cells, the major cell type of the interstitial space (Wing and Christensen, 1982), as also suggested by others (Tsukaguchi et al, 1999; Elkjaer et al, 2000; Nihei et al, 2001). However, why Leydig cells express AQP-9 is unclear. AQP-9 expression may maintain water equilibrium within these cells, but AQP-9 also allows the passage of solutes such as polyols, purines, and pyrimidines (Tsukaguchi et al, 1999). Another possibility may be the passage of steroids out of the cell, as a major function of Leydig cells is the production of testosterone. A role for AQPs in steroid transport has, however, not as yet been documented.
In the efferent ducts, AQP-1 and -9 have been demonstrated to be expressed by nonciliated cells (Brown et al, 1993; Fischer et al, 1998; Pastor-Soler et al, 2001). However, in the present study, the expression of these AQPs was noted to be cell-, subcellular-, and tissue-specific (Table 1; Figure 20). While AQP-1 was expressed in both nonciliated and ciliated cells, AQP-9 was solely expressed in nonciliated cells. In addition, AQP-1 was intensely expressed on the microvilli and basolateral plasma membranes of nonciliated cells, while AQP-9 was noted exclusively on the microvilli of nonciliated cells. AQP-1 was also prominent on the plasma membranes of endosomes residing in the apical region of the cell. Furthermore, AQP-1, but not AQP-9, was expressed on the endothelial cells of large vascular channels of the intertubular space (Table 1).
|
While it is still unclear why nonciliated cells express 2 different AQPs on their microvilli, it may be related to the fact that AQPs are selective not only for water permeation but for small uncharged solutes as well (Tsukaguchi et al, 1999; Verkman and Mitra, 2000). Furthermore, as noted in the electron microscope, gold particles representing AQP-1 antigenic sites were localized not only over the microvilli of the nonciliated cells but also over the subsurface endocytic vesicles. These vesicles have been identified as endosomes, components of the endocytic apparatus, involved in the uptake of substances from the lumen (Hermo and Morales, 1984; Hermo et al, 1988). Endosomes are fluid-filled organelles and gradually evolve in a temporal and sequential manner into multivesicular bodies and lysosomes (Hermo et al, 1988, 1994). AQP-1 expression related to endosomes may serve to remove water to allow for a reduction in their size and a concentration of their content as they evolve to become smaller dense lysosomes (Hermo et al, 1994).
The efferent ducts are well recognized as a major site of reabsorption of water entering the lumen from the seminiferous tubules. In fact, Crabo (1965) reported that between 50% and 90% of the fluid coming from the testis is removed from the lumen of the efferent ducts, as also demonstrated by others (Levine and Marsh, 1971; Clulow et al, 1998). Water resorption in the nonciliated cells requires an apically located Na+/H+ exchanger, isoform NHE3, and an Na+/K+-ATPase that has been localized to their basolateral plasma membranes (Ilio and Hess, 1992, 1994; Hansen et al, 1999; Leung et al, 2001). The standing osmotic gradient created by these pumps and the expression of both AQP-1 and -9 as noted in the present study and by others (Brown et al, 1993; Fischer et al, 1998; Pastor-Soler et al, 2001) would then allow the rapid passage of water from the lumen into the intertubular spaces. Once in the intertubular space, AQP-1 expression on vascular channels, presumably lymphatics or large venous channels, would serve to remove water from this site and thus maintain water equilibrium in this tissue. The removal of water from the efferent ducts would serve to concentrate sperm in the initial segment of the epididymis to provide for better interactions of the sperm surface with the secretory products of their epithelial cells, especially as this is the region where sperm begin to acquire their maturational properties (Cooper, 1995).
In the epididymis, AQP-9 has been reported to be localized on the microvilli of the principal cells of all regions (Elkjaer et al, 2000; Pastor-Soler et al, 2001). However, in the present study, AQP-9 expression on the microvilli of the principal cells was noted to be region-specific, with the most intense reaction being noted in the initial segment and cauda regions (Table 2; Figure 20). In addition, in the cauda epididymidis, the subsurface apical vesicles of the principal cells were also reactive, unlike those reported by Pastor-Soler et al (2001). These vesicles, as judged by their size, correspond to endosomes, shown to incorporate tracers injected into the lumen and reside in the apical region of the cell (Hermo et al, 1988). This suggests the removal of water from endosomes as they mature into smaller dense lysosomes (Hermo et al, 1994).
|
In addition, in the present study, AQP-9 was expressed in the clear cells but of the cauda region only, unlike that reported by Pastor-Soler et al (2001) (Table 2; Figure 20), suggesting region specificity, as noted for the expression of several other proteins by these cell types (Hermo et al, 1994; Cornwall and Hann, 1995; Kirchhoff, 1999). The reaction over the clear cells is not indicative of the endocytosis of AQP-9 from the lumen. In fact, AQP-9 expression by the principal cells is not suggestive of secretion, as there was no staining of the Golgi apparatus of these cells or any visible reaction for anti—AQP-9 antibody in the epididymal lumen. The cytoplasmic reaction recorded for the clear cells, however, is a common feature noted with many other proteins expressed by these cells, which appears to be due to the abundance of endocytic organelles in these cells and the resulting amplification of the diaminobenzidine reaction product. For example, in the case of lysosomal enzymes expressed by these cells and derived from the Golgi apparatus, a cytoplasmic reaction is noted, which is due to the abundance of lysosomes in these cells and which, as confirmed in electron microscope immunocytochemistry, does not represent their localization in the cytoplasm but in the numerous lysosomes of their cytoplasm (Hermo et al, 1992a, 1997; Igdoura et al, 1994). Thus, in the present study, it is proposed that in addition to being present on the plasma membranes of the clear cells serving to transport water from the lumen into the intertubular space, the plethora of early and late endosomes known to occupy the cytoplasm of the clear cells may express AQP-9 to remove water as these organelles concentrate the material contained therein to evolve into the numerous lysosomes (Robaire and Hermo, 1988), a hypothesis that would need to be confirmed in future electron microscope immuno-cytochemical analysis.
In the epididymis, as in the efferent ducts, AQP-1 was also expressed over endothelial cells of vascular channels of the intertubular space of the entire epididymis (Figure 20). In the absence of specific markers, these channels may represent lymphatics or large venous channels. Thus, in the epididymis, water may be transported from the lumen via AQP-9 expressed on the principal and clear cells and be removed in the intertubular space via AQP-1 expressed on vascular channels. The removal of water from the lumen of the initial segment and caput regions would concentrate sperm for enhancing their capacity for maturation, while in the cauda region, it may allow for more efficient space for immobilin, a protein secreted by the principal cells to immobilize sperm while they are stored in this region (Hermo et al, 1992b).
Regulation of AQP Expression in the Efferent Ducts and
Epididymis
It is well established that many epididymal functions and the expression of
different proteins are regulated by testicular factors
(Cornwall and Hann, 1995;
Robaire and Viger, 1995;
Orgebin-Crist, 1996; Kirchoff, 1999). These factors
include androgens synthesized by Leydig cells and entering the epididymis via
the circulation and lumen, as well as other luminal factors, such as sperm and
Sertoli-derived substances entering from the seminiferous tubules via the
efferent ducts. In the present study, at all time points examined after
orchidectomy or efferent duct ligation, there was no noticeable change in the
expression of AQP-1 or -9 over microvilli or cilia of the efferent ducts. This
included reaction over epithelial cells as well as vascular channels. Thus,
these 2 AQPs do not appear to be regulated by testicular factors. In several
studies, estrogen or, more specifically, a functional estrogen receptor,
(ER)-, has been suggested to regulate fluid transport in the male
reproductive tract (Lubahn et al,
1993; Eddy et al,
1996; Hess et al,
1997a,b;
Fischer et al, 1998), with ER- having been found in highest
concentrations in the efferent ducts of several species (Fischer et al, 1997;
Goyal et al, 1997; Kwon et al, 1997). While the
expression of AQP-1 appears to be modulated by estrogen exposure during
neonatal development (Fischer et al, 1998), AQP-1 expression was not affected
in the distal segments of the efferent ducts of adult ERKO mice, mice
lacking the estrogen receptor
(Zhou et al, 2001). Taken
together, these data as well as those of the present study suggest that AQP-1
expression over microvilli or cilia of the efferent ducts is not regulated
directly by testosterone or estrogens. While it is still not known whether or
not AQP-9 is regulated by estrogens, the present data would suggest that it is
not, as orchidectomy had no effect on AQP-9 expression.
In comparison, in the epididymis, AQP-9 expression was affected by both efferent duct ligation and orchidectomy but in a cell- and region-specific manner (Tables 3 and 4; Figure 20). Three days after orchidectomy and at all later time points, the initial segment displayed no reaction over microvilli of the principal cells. Since testosterone replacement at high levels did not restore the expression to the control levels, it is suggested that androgens are not the only factor regulating the expression of AQP-9 in this region. Data from efferent duct ligated animals, which also showed an absence of reaction under these conditions, suggest that a testicular factor derived from the lumen is also required for AQP-9 expression, but its identity at present is unknown. However, AQP-9 expression in the principal cells of the caput, corpus, and cauda regions was unaffected by all experimental treatments, indicating that AQP-9 in the principal cells of these regions is not regulated by testicular factors (Table 3).
|
|
On the other hand, the clear cells of the cauda epididymidis at 14 and 21 days after orchidectomy or efferent duct ligation showed an absence of AQP-9 expression (Table 4). Testosterone administration to orchidectomized animals did not restore the expression to the control levels. Thus, AQP-9 expression in the clear cells of the cauda region appears to be regulated by a testicular luminal factor, as in the case of the principal cells of the initial segment. It is not known if the same factor regulates these 2 cell types.
Postnatal Expression of AQP-1, -8, and -9
In the present study, AQP-1 expression was noted in the epithelium of the
efferent ducts by postnatal day 7 and as early as fetal day 20.5 by others
(Fischer et al, 1998; Table 5).
Thus, AQP-1 expression depends neither on the presence of Sertoli-derived
factors that would appear at about day 18, when the seminiferous tubular lumen
develops, nor on high androgen levels, which appear by day 39
(Tindall et al, 1975;
Scheer and Robaire, 1980;
Robaire and Hermo, 1988). This
is also verified by our orchidectomy and ligation experiments, which
demonstrated that AQP-1 expression over microvilli and cilia was not regulated
by testicular factors in the efferent ducts, including the presence of water,
which would be absent in both of these experimental conditions.
|
AQP-9 expression took on an adultlike staining pattern only by postnatal day 21 in the efferent ducts. These data suggest that the expression of AQP-1 and -9 in nonciliated cells of the efferent ducts may be regulated by different factors. By day 21, AQP-9 expression in the principal cells of the caput, corpus, and cauda regions of the epididymis was comparable to that in the control adult animals but was comparable only by postnatal day 29 in the case of the principal cells of the initial segment and clear cells of the cauda epididymidis (Table 5). Thus, different factors appear to regulate AQP-9 expression in the case of these cell types and regions, an observation identical to that noted for experimentally treated animals. In addition, data regarding AQP-9 expression in the epididymis appearing between days 21 and 29 suggest that it is regulated by factors other than androgens and sperm, which appear by days 39 and 49, respectively (Robaire and Hermo, 1988).
In summary, a cell-, subcellular-, region-, and tissue-specific distribution was observed for AQP-1, -8, and -9 in the testis, efferent ducts, and epididymis of normal adult animals. The expression of AQP-1 and -9 over microvilli and cilia of cells of the efferent ducts was not regulated by testicular factors, as was the case for AQP-9 expression in the principal cells of the caput, corpus, and cauda regions. However, AQP-9 expression in the principal cells of the initial segment and in the clear cells of the cauda region appears to be regulated by luminal factors derived from the testis other than androgens and sperm.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Barham SS, Berlin JD, Brackeen RD. The fine structural localization of testicular phosphatases in man: the control testis. Cell Tissue Res. 1976;166: 497 -510.[Medline]
Borgnia M, Nielsen S, Engel A, Agre P. Cellular and molecular biology of the aquaporin water channels. Annu Rev Biochem. 1999;68: 425 -458.[Medline]
Brawer JR, Schipper H, Robaire B. Effect of long term androgen and estradiol exposure on the hypothalamus. Endocrinology. 1983; 112: 194 -199.[Medline]
Brown D, Katsura T, Gustafson CE. Cellular mechanisms of aquaporin trafficking. Am J Physiol Renal Physiol. 1998; 44: F328 -F331.
Brown D, Katsura T, Kawashima M, Verkman AS, Sabolic I. Cellular distribution of the aquaporins: a family of water channel proteins. Histochem Cell Biol. 1995; 104: 1 -9.[Medline]
Brown D, Verbavatz JM, Valenti G, Lui B, Sabolic I. Localization of the CHIP28 water channel in reabsorptive segments of the rat male reproductive tract. Eur J Cell Biol. 1993; 61: 264 -273.[Medline]
Byers S, Graham R. Distribution of sodium-potassium ATPase in the rat testis and epididymis. Am J Anat. 1990; 188: 31 -43.[Medline]
Calamita G, Mazzone A, Cho YS, Valenti G, Svelto M. Expression and
localization of the aquaporin-8 water channel in rat testis. Biol
Reprod. 2001;64:
1660
-1666.
Clulow J, Jones RC, Hasen LA, Man SY. Fluid and electrolyte reabsorption in the ductuli efferentes testis. J Reprod Fertil. 1998;53(suppl): 1 -14.
Connolly DL, Shanahan CM, Weissberg PL. The aquaporins. A family of water channel proteins. Int J Biochem Cell Biol. 1998; 30: 169 -172.[Medline]
Cooper TG. Role of the epididymis in mediating changes in the male gamete during maturation. In: Mukhopadhyay AK, Raizada MK, eds. Tissue Renin-Angiotensin Systems. New York: Plenum Press; 1995: 87-101.
Cornwall GA, Hann SR. Specialized gene expression in the
epididymis. J Androl.
1995; 16:
379
-383.
Crabo B. Studies on the composition of epididymal content in bulls and boars. Acta Vet Scand. 1965; 6: 8 -94.
Deen P, van Os C. Epithelial aquaporins. Curr Opin Cell Biol. 1998;10: 435 -442.[Medline]
Denker BM, Smith BL, Kuhajda FP, Agre P. Identification,
purification, and characterization of a novel Mr 28 000 integral membrane
protein from erythrocytes and renal tubules. J Biol
Chem. 1988;263:
15634
-15642.
Echevarria M, Ilundain AA. Aquaporins. J Physiol Biochem. 1998;54: 107 -118.[Medline]
Eddy EM, Washburn TF, Bunch DO, Goulding EH, Gladen BC, Lubahn DB, Korach KS. Targeted disruption of the estrogen receptor gene in male mice causes alteration of spermatogenesis and infertility. Endocrinology. 1996; 137: 4796 -4805.[Abstract]
Elkjaer M, Vajda Z, Nejsum LN, Kwon T, Jensen UB, Amiry-Moghaddam M, Frokiaer J, Nielsen S. Immunolocalization of AQP9 in liver, epididymis, testis, spleen, and brain. Biochem Biophys Res Commun. 2000; 276: 1118 -1128.[Medline]
Elkjaer ML, Nejsum LN, Gresz V, Kwon TH, Jensen UB, Frokiaer J,
Nielsen S. Immunolocalization of aquaporin-8 in rat kidney, gastrointestinal
tract, testis, and airways. Am J Physiol Renal
Physiol. 2001;281:
F1047
-1057.
Engel A, Fujiyoshi Y, Agre P. The importance of aquaporin water channel protein structures. EMBO J. 2000; 19: 800 -806.[Medline]
Fawcett DW, Neaves WB, Flores MN. Comparative observations on intertubular lymphatics and the organization of the interstitial tissue of the mammalian testis. Biol Reprod. 1973; 9: 500 -532.[Abstract]
Fisher JS, Turner KJ, Fraser HM, Saunders PTK, Brown D, Sharpe RM.
Immunoexpression of aquaporin-1 in the efferent ducts of the rat and marmoset
monkey during development, its modulation by estrogens, and its possible role
in fluid resorption. Endocrinology.
1998; 139:
3935
-3945.
Goyal HO, Bartol FF, Wiley AA, Neff CW. Immunolocalization of receptors for androgen and estrogen in male caprine reproductive tissues: unique distribution of estrogen receptors in efferent ductule epithelium. Biol Reprod. 1997; 56: 90 -101.[Abstract]
Gravis CJ, Yates RD, Chen IL. Light and electron microscopic localization of ATPase in normal and degenerating testes of Syrian hamsters. Am J Anat. 1976; 147: 419 -432.[Medline]
Hansen LA, Clulow J, Jones RC. The role of Na+—H+ exchange in fluid and solute transport in the rat efferent ducts. Exp Physiol. 1999; 84: 521 -527.[Abstract]
Hermo L, Adamali HI, Mahuran D, Gravel RA, Trasler JM. ß-Hexosaminidase gene expression and enzyme immunolocalization in the rat testis and epididymis. Mol Reprod Dev. 1997; 46: 227 -242.[Medline]
Hermo L, Barin K, Robaire B. Structural differentiation of the epithelial cells of the testicular excurrent duct system of rats during postnatal development. Anat Rec. 1992c; 233: 205 -228.[Medline]
Hermo L, Morales C. Endocytosis in nonciliated epithelial cells of the ductuli efferentes in the rat. Am J Anat. 1984; 171: 59 -74.[Medline]
Hermo L, Morales C, Oko R. Immunocytochemical localization of sulfated glycoprotein-1 (SGP-1) and identification of its transcripts in epithelial cells of the extratesticular duct system of the rat. Anat Rec. 1992a; 232: 401 -422.[Medline]
Hermo L, Oko R, Morales C. Secretion and endocytosis in the male reproductive tract: a role in sperm maturation. Int Rev Cytol. 1994;154: 105 -189.
Hermo L, Oko R, Robaire B. Epithelial cells of the epididymis show regional variations with respect to the secretion or endocytosis of immobilin as revealed by light and electron microscope immunocytochemistry. Anat Rec. 1992b; 232: 202 -220.[Medline]
Hermo L, Papp S. Effects of ligation, orchidectomy, and hypophysectomy on expression of the Yf subunit of GST-P in principal and basal cells of the adult rat epididymis and on basal cell shape and overall arrangement. Anat Rec. 1996; 244: 59 -69.[Medline]
Hermo L, Spier N, Nadler NJ. Role of apical tubules in endocytosis in nonciliated cells of the ductuli efferentes of the rat: a kinetic analysis. Am J Anat. 1988; 182: 107 -119.[Medline]
Hermo L, Wright J, Oko R, Morales C. Role of epithelial cells of the male excurrent duct system of the rat in the endocytosis or secretion of sulfated glycoprotein-2 (Clusterin). Biol Reprod. 1991; 44: 1113 -1131.[Abstract]
Hess RA, Bunick D, Lee KH, Bahr J, Taylor JA, Korach KS, Lubahn DB. A role for oestrogens in the male reproductive system. Nature. 1997a;390: 509 -512.[Medline]
Hess RA, Gist DH, Bunick D, Lubahn DB, Farrell A, Bahr J, Cooke PS,
Greene GL. Estrogen receptor ( and ß) expression in the excurrent
ducts of the adult male rat reproductive tract. J
Androl. 1997b;18:
602
-611.
Hinton BT, Setchell BP. Fluid secretion and movement. In: Russell L, Griswold M, eds. The Sertoli Cell. Clearwater, Fla: Cache River Press; 1993: 249 -267.
Igdoura SA, Morales CR, Hermo L. Differential expression of cathepsins B and D in testis and epididymis of adult rats. J Histochem Cytochem. 1994;43: 545 -557.
Ilio KY, Hess RA. Localization and activity of Na+, K+-ATPase in the ductuli efferentes of the rat. Anat Rec. 1992;234: 190 -200.[Medline]
Ilio KY, Hess RA. Structure and function of the ductuli efferentes. Miscrosc Res Tech. 1994; 29: 432 -467.
Ishibashi K, Kuwahara M, Gu Y, Tanaka Y, Marumo F, Sasaki S. Cloning and functional expression of a new aquaporin (AQP9) abundantly expressed in the peripheral leukocytes permeable to water and urea, but not to glycerol. Biochem Biophys Res Commun. 1998; 244: 268 -274.[Medline]
Ishibashi K, Kuwahara M, Kageyama Y, Tohsaka A, Marumo F, Sasaki S. Cloning and functional expression of a second new aquaporin abundantly expressed in testis. Biochem Biophys Res Commun. 1997; 237: 714 -718.[Medline]
Kageyama Y, Ishibashi K, Haashi T, Xia G, Sasaki S, Kihar K. Expression of aquaporins 7 and 8 in the developing rat testis. Andrologia. 2001; 33: 165 -169.[Medline]
King LS, Agre P. Pathophysiology of the aquaporin water channels. Annu Rev Physiol. 1996; 58: 619 -648.[Medline]
King LS, Yasui M, Agre P. Aquaporins in health and disease. Mol Med Today. 2000; 6: 60 -65.[Medline]
Kirchoff C. Gene expression in the epididymis. Int Rev Cytol. 1999;188: 133 -202.[Medline]
Koyama Y, Yamamoto T, Kondo D, et al. Molecular cloning of a new
aquaporin from rat pancreas and liver. J Biol Chem.
1997; 272:
30329
-30333.
Kuriyama H, Kawamoto S, Ishida N, Ohno I, Mita S, Matsuzawa Y, Matsubara K, Okubo K. Molecular cloning and expression of a novel human aquaporin from adipose tissue with glycerol
