immunolocalization of ca ii and h v-atpase in … · ap indicates apical region; g, golgi aparatus;...

376

Journal of Andrology, Vol. 21, No. 3, May/June 2000Copyright � American Society of Andrology

Immunolocalization of CA II and H�V-ATPase in EpithelialCells of the Mouse and Rat Epididymis

LOUIS HERMO, HUZAIFA I. ADAMALI, AND SERO ANDONIAN

From the Department of Anatomy and Cell Biology, McGill University, Montreal, Quebec, Canada.

ABSTRACT: Acidification of the epididymal lumen has been sug-gested to play an important role in sperm functions; however, thecell types, pumps, and mechanisms involved have not been fullyaddressed. In this study, carbonic anhydrase II (CA II) and a 67-kdsubunit of Neurospora crassa vacuolar proton adenosinetriphospha-tase (H�V-ATPase) pump were immunolocalized using light micros-copy and electron microscopy (EM) in the epididymis of rats andmice. In both animals, narrow cells, identified in the initial segmentand intermediate zone of the epididymis, contained numerous smallvesicles in their apical region, often cup-shaped in appearance. Inthe mouse but not rat, these cells also possessed numerous cister-nae of smooth endoplasmic reticulum, suggesting steroid synthesis;and cytoplasmic blebs of their apical cell surface, which appearedto detach, suggesting apocrine secretion. Anti-CA II antibody wasimmunocytochemically localized in the light microscope within nar-row cells but not over any other cell types of the entire epididymis.Anti-H�V-ATPase antibody was also localized in narrow cells of theinitial segment and intermediate zone; as well as clear cells of thecaput, corpus, and cauda regions. Using EM, gold particles for anti-CA II and H�V-ATPase antibodies were noted in the apical regionof narrow cells in relation to the numerous, small, cup-shaped ves-

icles. Although CA II was mainly located in the cytosol near thesevesicles, H�V-ATPase appeared on their delimiting membrane andon the apical plasma membrane of these cells. A similar distributionwas noted for H�V-ATPase in clear cells. The nature of the smallvesicles of the apical region of narrow cells was examined with elec-tron-dense fluid phase tracers that were introduced into the epidid-ymal lumen. The tracers appeared within these vesicles and a fewendosomes 1 hour after injection, suggesting that they contact theapical plasma membrane. Since these vesicles are also related toCA II and H�V-ATPase, the data suggests that, as the site of protonproduction, the vesicles recycle to and from the apical cell surface,and in this way, deliver protons to the epididymal lumen for acidifi-cation. Clear cells and their expression of H�V-ATPase may alsoserve in this function. In summary, both narrow and clear cells ap-pear to be involved in luminal acidification, an activity that may beessential for sperm as they traverse and are stored in the epididy-mis.

Key words: Luminal acidification, proton secretion, sperm matu-ration.

J Androl 2000;21:376–391

Narrow cells have been identified in the epididymis ofnumerous species, including humans (Hamilton,

1975; Sun and Flickinger, 1980; Abou-Haıla and Fain-Maurel, 1984; Robaire and Hermo, 1988; Palacios et al,1991). In rats, they are present only in the initial segmentand intermediate zone. They are characterized as slendercells with an apically located elongated nucleus, andshow, in appropriate sections, a thin stalk of cytoplasmthat extends to the basement membrane (Adamali andHermo, 1996). Using electron microscopy (EM), numer-ous small vesicles, often cup-shaped in appearance, canbe seen occupying the apical region of narrow cells, buttheir significance has remained obscure (Sun and Flick-inger, 1980; Abou-Haıla and Fain-Maurel, 1984; Adamaliand Hermo, 1996). Apical cells have also been noted inthe same regions of the rat epididymis (Adamali and Her-

Supported by a grant from the Medical Research Council of Canada.Correspondence to: Dr Louis Hermo, Department of Anatomy and Cell

Biology, McGill University, 3640 University St, Montreal, Quebec, H3A2B2 Canada (e-mail: [email protected]).

Received for publication March 31, 1999; accepted for publication Sep-tember 30, 1999.

mo, 1996). In contrast to narrow cells, apical cells havebeen characterized as goblet-shaped, they do not contactthe basement membrane, and they contain few small api-cal vesicles. Apical cells represent a distinct cell type inthe initial segment and intermediate zone of the rat epi-didymis. They show differences in expression of gluta-thione S transferases and lysosomal enzymes, comparedwith principal, basal, and narrow cells (Adamali and Her-mo, 1996). In the rat epididymis, another cell type withan apically located nucleus has also been noted in theinitial segment. The cells were originally termed apicalcells by Reid and Cleland (1957); however, they struc-turally resemble principal cells, even though they do notappear to contact the basement membrane (Sun and Flick-inger, 1980; Robaire and Hermo, 1988; Adamali and Her-mo, 1996).

In the mouse epididymis, two types of cells with api-cally located nuclei have been identified; each is distin-guished from the other by various histochemical proper-ties (Abou-Haıla and Fain-Maurel, 1984). One type, re-ferred to as apical cells, are numerous in segment 1, butfewer are found in proximal epididymal regions. Theypresent the same morphological and histochemical fea-

377Hermo et al · CA II and H�V-ATPase in Adult Epididymis

tures as the adjacent principal cells and, in appropriatesections, contacted the basement membrane. The secondcell type consists of narrow cells that contain numerousmitochondria, small apical vesicles, and few endoplasmicreticulum (ER) elements (Abou-Haıla and Fain-Maurel,1984).

Cells with apically located nuclei in the initial segmenthave been shown by histochemistry to contain carbonicanhydrase (CA; Cohen et al, 1976). In the rat epididymisthese cells have been confirmed through immunocyto-chemical analysis to be narrow cells and to express CAII (Kaunisto et al, 1995; Adamali and Hermo, 1996);however, no reaction was associated with apical cells ofthis region (Adamali and Hermo, 1996). In more distalepididymal sites, conflicting results exist on the locali-zation of CA II, with both principal and clear cells beingreactive (Brown et al, 1992; Kaunisto et al, 1995; Bretonet al, 1999).

In the rat epididymis, apical and narrow cells, locatedin the caput region of the epididymis, were reported toexpress H�V-ATPase, as did clear cells of the cauda ep-ididymidis. With EM, a 56-kd subunit of anti-H�V-ATPase was localized on the apical plasma membrane andapically located intracellular vesicles of these cells(Brown et al, 1992, 1997; Breton et al, 1999).

Although several studies have been done on the struc-ture and function of narrow cells of the rat, there are nodetailed reports on these parameters in the mouse epidid-ymis. We report new findings on the ultrastructural ap-pearance of narrow cells, including the presence of nu-merous, small, apical cup-shaped vesicles; widely dis-persed smooth endoplasmic reticulum (sER) elements attimes forming distinct piles; and apical blebs of cyto-plasm; and propose a possible functional role for each. Inaddition, we report the cellular and subcellular localiza-tion of CA II and H�V-ATPase in distinct epithelial cellsof the different regions of the epididymis in both rats andmice using both LM and EM immunocytochemistry.Combining tracer studies and the subcellular localizationof CA II and H�V-ATPase, we propose a role for thesmall, apical, cup-shaped vesicles in releasing protonsinto the epididymal lumen as they recycle to and fromthe apical cell surface.

Materials and Methods

Routine Light and Electron MicroscopyFour adult male Sprague-Dawley rats (weighing 350–450 g) and4 adult male mice (weighing 30–40 g) were obtained fromCharles River Laboratories (St Constant, Quebec). All animalswere anesthetized with sodium pentobarbital (Somnitol, MTCPharmaceuticals, Hamilton, Ontario). The epididymides of eachmouse were fixed via cardiac perfusion; those of the rat were

fixed by perfusion through the abdominal aorta with 2.5% glu-taraldehyde buffered in sodium cacodylate buffer (0.1M) con-taining 0.05% CaCl2 at pH 7.4. After perfusion for 10 minutes,the tissue was removed and dissected to obtain parts of the initialsegment and intermediate zone. The tissue from each region wascut into 1-mm3 pieces and placed in the same fixative for anadditional 2 hours at 4�C. After an overnight wash in buffer, thetissue was postfixed in potassium ferrocyanide–reduced osmiumtetroxide (Karnovsky, 1971) for 1 hour to enhance the membranestaining. The tissue was then rinsed several times in sodiumcacodylate buffer, dehydrated in ethanol and propylene oxide,and embedded in Epon. Sections 0.5–1 �m in size were cut,stained with toluidine blue, and examined under a light micro-scope. Thin sections of selected areas were cut with a diamondknife, placed on copper grids, counterstained with uranyl acetateand lead citrate, and examined with a Philips 400 electron mi-croscope.

Tracer StudiesIn order to determine if the small, apical, cup-shaped vesicleswere endocytic in nature, native ferritin, a fluid phase tracer, wasused at a dose of 0.3 mg in 0.1 mL of physiological saline. Afteranesthesia, the tracer was injected into the lumen of the retetestis of 3 rats. One hour later, the animals were sacrificed, whichallowed ample time for the tracers to enter the initial segmentas demonstrated by English and Dym (1982). The tissue wasfixed and embedded as described earlier.

Light Microscope ImmunocytochemistryFour additional adult male Sprague-Dawley rats weighing 350–450 g each and 4 adult mice obtained from Charles River Lab-oratories (St Constant, Quebec) were anesthetized with an intra-peritoneal injection of sodium pentobarbital (Somnitol, MTCPharmaceuticals, Hamilton, Ontario). The epididymides of themice were fixed through the heart; those of rats were fixed for10 minutes by perfusion through the abdominal aorta withBouin’s fixative. After fixation, the epididymides were removedand immersed in Bouin’s fixative for another 24 hours. Beforeimmersion, the tissue was cut along its long axis so that sectionswould include 1 or more of the major regions of the epididymis(ie, initial segment; intermediate zone; and caput, corpus, andcauda epididymidis). After fixation, the tissue was dehydratedand eventually embedded in paraffin.

Immunoperoxidase StainingImmunoperoxidase staining of the epididymis was carried outaccording to the procedure described by Oko and Clermont(1989). Polyclonal anti-H�V-ATPase (67-kd) and anti-CA II an-tibodies were used at a dilution of 1:100 in 20 mM of tris-buffered saline (TBS), pH 7.4, with 0.1% bovine serum albumin.The anti-rabbit H�V-ATPase antibody was obtained from BarryJ. Bowman (University of California, Santa Cruz; characterizedin Bowman et al, 1986, 1988), while the anti-rat CA II antibodywas obtained from Nick Carter (University of London, UnitedKingdom; described in Carter et al, 1981; and Jeffery et al,1986).

Paraffin sections (5 �m) were deparaffinized in xylene andhydrated in a series of graded ethanol solutions. During hydra-

378 Journal of Andrology · May/June 2000

andr 21 304 Mp 378File # 04em

→

Figure 1. Electron micrograph montage of a narrow cell of the initial segment of the mouse epididymis. Ap indicates apical region; G, Golgi aparatus;L, lysosomes; m, mitochondria; Lu, lumen; n, narrow cell nucleus; N, principal cell nucleus; P, principal cell; and arrowheads, small vesicles (7000�).Upper inset: Light micrograph of the initial segment of the mouse epididymis immunostained with anti-carbonic anhydrase II antibody revealing intensereaction (arrow) over narrow cell (350�). Lower inset: Light micrograph of the initial segment of the mouse epididymis (glutaraldehyde-fixed, Epon-embedded) showing a narrow cell in the epithelium. Note apical position of its elongated nucleus and slight apical bulge (arrow) projecting into thelumen. Adjacent tall columnar principal cells show spherical, basally located nuclei (arrowhead) 570�.

Figure 2. Electron micrograph of an oblique section through the apical region of a narrow cell of the mouse epididymis. N indicates nucleus; L,lysosomes; Lu, lumen; m, mitochondria; arrowheads, small vesicles; E, large endosomes; and e, small endosomes (15 750�).

Figure 3. Electron micrograph of the apical region of a narrow cell of the mouse epididymal initial segment. Lu indicates lumen; m, mitochondria; sER,smooth endoplasmic reticulum; small arrows, elongated fenestrated sER elements (side view); large arrows, loose anastomotic sER (face view); andcurved arrows, piles of sER (15 750�)

Figure 4. Electron micrograph of the apical region of a narrow cell of the mouse epididymal initial segment. Lu indicates lumen; E, large endosomes;e, small endosomes; m, mitochondria; �, bleb of cytoplasm protruding from apical surface; �, bleb of detached cytoplasm; small and large arrows,smooth endoplasmic reticulum; open arrows, membranous remnants; and arrowheads, small vesicles (15 400�).

Figure 5. High-power electron micrograph of the apical region of a narrow cell of the mouse epididymal initial segment. The apical region is filled withsmall vesicles, many of which are cup-shaped and appear as double-walled structures in section planes (arrowheads). Lu indicates lumen; m,mitochondria; and E, endosome (33 440�).

tion, residual picric acid was neutralized in 70% ethanol con-taining 1% lithium carbonate, and endogenous peroxidase activ-ity was abolished in 70% ethanol containing 1% (vol/vol) H2O2.Once hydrated, the tissue sections were washed in distilled watercontaining 300 mM glycine to block free aldehyde groups.

Before immunostaining, the sections were blocked for 30 min-utes with 10% goat serum in TBS. This and subsequent treat-ments were accomplished by placing 100 �L of this solutiononto a coverslip and overturning the tissue face of the slide ontothe drop, thus ensuring that the entire tissue was treated withminimal fluid (Oko and Clermont, 1989). Coverslips were re-moved by dipping the slides in TBS containing 1% Tween-20(TWBS). Sections were then incubated in a 37�C humidified in-cubation chamber for 1 hour with the appropriately diluted pri-mary antibody. After 4 5-minute washes in TWBS, sections wereonce again blocked with 10% goat serum in TBS. They werethen incubated for 30 minutes at 37�C with goat anti-rabbit im-munoglobulin G (IgG) conjugated to peroxidase (Sigma Chem-ical Company, St Louis, Mo) diluted 1:250 in TBS. This wasfollowed by 4 5-minute washes in TWBS.

The final reaction product was achieved by incubating the sec-tions for 10 minutes in 500 mL of TBS containing 0.03% H2O2,0.1 M imidazole, and 0.05% diaminobenzidine tetrahydrochlor-ide (Sigma Chemical Company, St Louis, Mo), pH 7.4. Slideswere then washed in distilled water and counterstained with0.1% methylene blue for 1 minute. The tissue was dehydratedby passing slides through a graded ethanol series, after whichthe sections were immersed in xylene and mounted with Per-mount. Specificity of the immunostaining was confirmed by in-cubating tissue sections without the primary antibody or use ofnormal rabbit serum.

Electron Microscope ImmunocytochemistryTissue Preparation—Four additional adult male Sprague-

Dawley rats each weighing 350–450 g and 4 adult mice obtainedfrom Charles River Laboratories (St Constant, Quebec) wereanesthetized with an intraperitoneal injection of sodium pento-barbital (Somnitol, MTC Pharmaceuticals, Hamilton, Ontario).

The epididymides of the mice were fixed through the heart; thoseof rats were fixed by perfusion through the abdominal aorta for10 minutes with 0.5–1% glutaraldehye and 4% paraformalde-hyde in 0.1 M phosphate buffer containing 50 mM lysine at pH7.4. After fixation, the epididymides were removed, cut intosmall pieces, immersed for 2 hours in the same fixative at 4�C,washed with 0.15 M phosphate-buffered saline (PBS) containing4% sucrose (pH 7.4) at 4�C and then treated with PBS containing4% sucrose and 50 mM NH4Cl for 1 hour at 4�C. The tissue wasthen washed, dehydrated in graded methanol up to 90%, andembedded in Lowicryl K4M as described by Oko (1988).

Ultrathin sections of selected epididymal regions were mount-ed on 300-mesh formvar-coated nickel grids. Each section wasfloated for 15 minutes on a drop of 20 mM tris-HCl bufferedsaline containing 0.1% bovine TBS and 10% goat serum andthen incubated for 1 hour on 15-�L drops of a polyclonal anti-CA II or H�V-ATPase antibody diluted 1/20 in TBS. Sectionswere washed 4 times for 5 minutes each in TWBS, transferredfor 15 minutes to drops of TBS containing 5% goat serum, andincubated for 30 minutes on 20-�L drops of goat anti-rabbit IgGantibodies conjugated to 10-nm colloidal gold particles. The sec-tions were subjected to 3- to 5-minute washes in TWBS followedby 5-minute washes in distilled water. Sections were counter-stained with uranyl acetate in 30% ethanol and lead citrate. Pho-tographs were taken on a Philips 400 electron microscope. Nor-mal rabbit serum at a dilution of 1:20 served as a control.

Results

Light and Electron Microscopic Appearance of NarrowCells of the Mouse and Rat EpididymisIn the mouse epididymis, as noted in LM, narrow cellswere present in the initial segment and intermediate zone.They were characterized by an apically located nucleus,which was usually elongated, and an apical cytoplasm,which often bulged into the lumen (Figure 1 lower inset;



Figure 6. Electron micrograph of the apical region of an initial segment narrow cell of a rat sacrified 1 hour after injection with native ferritin into thelumen of the rete testis. The electron-dense tracers are located in several small vesicles (arrowheads) and in an endosome, which indicates anendocytic function. E indicates endosome; m, mitochondria; g, glycogen; and arrows, smooth endoplasmic reticulum; (42 500�).

Figures 7b and 8b). Excluding the apical bulges, narrowcells of the rat were similar to those of the mouse epi-didymis (Figures 7a and 8a) and located in the same re-gions. In appropriate sections, narrow cells appeared at-tenuated as they contacted the basement membrane (Fig-ures 1, upper inset; 7a).

In EM, narrow cells of the mouse showed a slight api-cal bulge of cytoplasm protruding into the lumen and aslender process extending toward the basement membrane(Figure 1). The nucleus was located in the upper third ofthe cell cytoplasm (Figure 1). Subjacent to the apical plas-ma membrane were numerous small vesicles, which often


Figure 7. Light micrographs of initial segment of rat (a) and mouse (b) epididymides immunostained with anti-carbonic anhydrase II antibody. Intensereaction is localized over narrow cells in both. P indicates principal cell; Lu, lumen; IT, intertubular space; and arrows, narrow cells (350�).



Figure 8. Light micrographs of the rat (a) and mouse (b) intermediate zone and rat corpus regions (c,d) of the epididymis immunostained with anti-H�V-ATPase antibody. In (a) and (b), narrow cells are intensely reactive and in (b) their apices protrude into the lumen. In (c) and (d), clear cellsshow an intense apical band of reaction product. P indicates principal cells; Lu, lumen; IT, intertubular space; arrows, clear cells; and arrowheads,narrow cells. Magnification: (a) 300�; (b) 400�; (c) 350�; (d) 400�.


Figure 9. Light micrographs of the initial segment (a) and corpus (b)regions of the rat epididymis immunostained with normal rabbit serum.E indicates epithelium; Lu, lumen; IT, intertubular space; and arrow-heads, clear cells. Magnification: (a) 162�; (b) 202�.

appeared as cup-shaped or double-walled structures in ap-propriate planes of sections (Figures 2 through 5). Thesevesicles were observed near and removed from the apicalcell surface, and as such, formed a conspicuous layer ofthe apical cell cytoplasm (Figures 2 through 5). Inter-spersed among these cup-shaped vesicles were a fewsmall and large endosomes, designated as early and lateendosomes, respectively; whereas supranuclearly, the cy-toplasm was filled with mitochondria (Figures 1 through5). Lysosomes were noted in narrow cells both supranu-clearly and infranuclearly, whereas the Golgi apparatuswas located juxtanuclearly (Figures 1 and 2). Widely dis-persed among these organelles were elements of sER.They were seen as flattened, fenestrated profiles in sideview or as loose, anastomotic networks in face view (Fig-ure 3). At times, sER elements were layered on top ofanother, which formed distinct piles (Figure 3). Some nar-row cells displayed prominent blebs of cytoplasm pro-truding from their apical surface into the lumen and, only

occasionally, contained an sER element (Figure 4). A ho-mogeneous granular material constituted the matrix ofthese so-called apical blebs (Figure 4). Although someblebs were continuous with the apical plasma membrane,others appeared to be detached (Figure 4). The lumenadjacent to the blebs often contained membranous profiles(Figure 4). Unlike narrow cells of the mouse epididymis,those of the rat did not display the apical blebs, nor weresER elements prominent; in other respects, they displayedfeatures that were similar to those of the mouse (Adamaliand Hermo, 1996).

Tracer StudiesIn order to determine if the small, apical vesicles of nar-row cells were endocytic in nature, native ferritin wasinjected into the lumen of the rete testis. The animalswere sacrificed 1 hour later, at which interval electron–dense tracers were noted in many of the small, apicalvesicles of these cells as well as within their endosomes(Figure 6).

Light Microscope ImmunocytochemistryStaining with anti-CA II antibody revealed an intense re-action in narrow cells of the initial segment and inter-mediate zone of both rats and mice (Figures 7a and b,respectively), but not in any other cell type of the entireepididymis. Anti-H�V-ATPase antibody was also local-ized over narrow cells of the initial segment and inter-mediate zone of rats (Figure 8a) and mice (Figure 8b)alike. In addition, staining was also noted over the apicalregion of clear cells of the caput, corpus, and cauda ep-ididymidis of both rats and mice (Figures 8c and d). Useof normal rabbit serum or omitting the primary antibodyfrom sections demonstrated no reaction over the epithe-lium, lumen, or intertubular space of the entire epididymisof both rats and mice (Figures 9a and b).

Electron Microscope ImmunocytochemistryImmunogold labeling for anti-CA II antibody was scat-tered throughout the apical region of narrow ceells of therat and mouse. Gold particles appeared mainly in the cy-tosol either near or slightly removed from the numerous,small apical vesicles (Figure 10). There was no labelingover other epididymal cell types. Gold particles for anti-H�V-ATPase antibody were also scattered throughout theapical region of narrow cells in both rats and mice andwere seen near the apical, small vesicles where they wereoften distributed along their delimiting membrane (Figure11). A similar distribution of labeling was also noted withanti-H�V-ATPase antibody in the clear cells (Figure 12).Only a few gold particles, which were consistent withbackground levels of labeling, were noted when sectionswere immunostained with normal rabbit serum.


←

Figure 10. Electron micrograph of a narrow cell of the rat initial segment immunostained with anti-carbonic anhydrase II antibody. Gold particles(arrowheads) are numerous throughout the apical region, appearing in their cytosol near or at a distance from the small apical vesicles. E indicateslate endosome; and m, mitochondria (46 800�).

Figure 11. Electron micrograph of the apical region of a narrow cell of the rat initial segment immunostained with anti-H�V-ATPase antibody. Goldparticles (arrowheads) are scattered throughout; many are related to the membrane of the small vesicles (46 800�).

Discussion

EM Appearance of Narrow Cells of the MouseEpididymisNarrow cells were noted only in the initial segment andintermediate zone of the mouse epididymis, correspond-ing to segment 1 (Abou-Haıla and Fain-Maurel, 1984),and as demonstrated in the rat epididymis (Adamali andHermo, 1996). As also noted by Abou-Haıla and Fain-Maurel (1984), narrow cells were characterized by aslight bulge of their apical cytoplasm and an apically lo-cated, often elongated nucleus; however, details of theirultrastructure were not assessed. In our study, narrowcells showed an apical region filled with small vesicles,often cup-shaped in appearance, which in appropriate sec-tions, appeared as double-walled vesicles (Figure 13).Narrow cells of the mouse, unlike those of the rat, oftencontained numerous, widely dispersed elements of sER,which at times, formed distinct piles. They were charac-terized as sER elements based on their similar morpho-logical appearance and topographical arrangement tothose present in Leydig cells of the testis and principalcells of the vas deferens, where steroid synthesis has beendocumented (Flickinger, 1973; Christensen, 1975; Ham-ilton, 1975; Andonian and Hermo, 1999). We propose,therefore, that narrow cells of the mouse epididymis maybe involved in steroid synthesis, but further experimen-tation is required.

In the mouse, narrow cells showed blebs of cytoplasm,called apical blebs, protruding into the lumen (Figure 13).They contained few organelles other than an occasionalsER element. Although many blebs appeared to be at-tached to the narrow cell apical surface, others appearedto be free in the lumen, which suggests that they detachedfrom the cell surface. Such apical blebs have been notedin epithelial cells of different parts of the male reproduc-tive system of various species (Niemi, 1965; Flickinger,1973; Ilio and Hess, 1994; Renneberg et al, 1995), wherethey have been reported to be involved in apocrine se-cretion, but their presence in narrow cells has not beenpreviously reported. In principal cells of the mouse vasdeferens, apical blebs have been implicated in the secre-tion of a major protein lacking an ER signal peptide se-quence, and to a glycosylation site, which limits the pro-tein to the cytosol of the blebs (Manin et al, 1995). In therat vas deferens, such blebs were noted in principal cells,

and Andonian and Hermo (1999) suggested that they mayalso represent a means of steroid secretion by these cells.While the functional role of blebs in narrow cells of themouse epididymis awaits documentation, they may playa role in apocrine secretion involving proteins, steroids,or both.

While the use of tracers has documented the endocyticrole of apical vesicles in clear cells of the caput epidi-dymidis (Moore and Bedford, 1979), the role of the nu-merous, small apical vesicles, often cup-shaped in ap-pearance, in narrow cells of the initial segment and in-termediate zone, has not been examined, and yet, theirabundance suggests they play an important role in thesecells. In our study, tracers introduced into the lumen ofthe epididymal duct revealed that these vesicles becomelabeled, indicating their role in endocytosis of substancesfrom the lumen. In many cell types, small endocytic ves-icles fuse with each other to form early endosomes (He-lenius et al, 1983). However, based on the abundance ofthe small apical vesicles, they would not all appear to beinvolved in formation of endosomes because endosomesare not a prominent feature of narrow cells. We propose,therefore, that in narrow cells, some of these vesiclesform endosomes, but others may recycle to and from theapical cell surface to secrete H� into the epididymal lu-men.

Immunolocalization of Anti-CA II and H�V-ATPaseAntibodiesCA has been histochemically localized in narrow cells ofthe initial segment of the adult rat epididymis with vari-able reaction being observed in other epithelial cell types,at times including the entire epithelium, in more distalepididymal regions (Cohen et al, 1976). Brown et al(1992) and Breton et al (1999) reported the immunocy-tochemical presence of CA II in narrow cells, apical cells,or both of the caput epididymidis, while Kaunisto et al(1995) noted staining of narrow cells of the initial seg-ment. In our study, CA II was expressed in narrow cellsof the initial segment and intermediate zone of both therat and mouse epididymis, as already reported for the ratepididymis by Adamali and Hermo (1996), but not in thecaput epididymidis. In fact, narrow and apical cells donot exist in the caput region per se, and are replaced byclear cells, as noted by differences in the appearance ofthese cell types (Hamilton, 1975; Abou-Haıla and Fain-



Figure 12. Electron micrograph of a clear cell of the rat corpus epididymidis immunolabeled with anti-H�V-ATPase antibody. Lu indicates lumen; E,endosomes; m, mitochondria; arrowheads, gold particles; and arrows, microvilli (51 680�).

Maurel, 1984; Robaire and Hermo, 1988; Hermo et al,1994). Although Brown et al (1992) and Breton et al(1999) do not appear to have discriminated between thecaput region and the initial segment, the existence of theinitial segment as a separate and distinct region in the ratand mouse epididymis has long been recognized fromearly studies (Hoffer et al, 1973; Flickinger, 1979). Fur-thermore, in the rat epididymis, an intermediate zone hasbeen described as situated between the initial segment andcaput epididymidis, and showing distinct morphologicalfeatures of their principal cells (Hermo, 1995). This isalso the case for the mouse epididymis (Adamali et al,1999).

In the work by Brown et al (1992), Brown and Breton(1996), and Breton et al (1999) the researchers did notdistinguish between narrow and apical cells, even thoughthe 2 cell types differ in their morphological appearance

in both the rat and mouse epididymis (Abou-Haıla andFain-Maurel, 1984; Adamali and Hermo, 1996). In therat, apical cells do not contact the basement membrane,they are especially prominent in the proximal initial seg-ment, and they differ structurally and functionally fromnarrow and principal cells (Adamali and Hermo, 1996).In the mouse, cells referred to as apical cells are espe-cially numerous in segment 1. However, such cells struc-turally resemble principal cells except for the more apicalposition of their nucleus, and, although clearly differentstructurally and functionally from narrow cells, apicalcells do not differ functionally from principal cells (Abou-Haıla and Fain-Maurel, 1984). As judged from our study,a comparable cell type to the one described for apicalcells in the rat does not appear to exist in large numbersin the mouse epididymis. Nevertheless, in our study, asrevealed by their shape from LM analysis and distinct


Figure 13. Diagrammatic representation of a narrow cell of the mouseinitial segment. Cytoplasmic blebs protrude from the apical region intothe lumen and contain few smooth endoplasmic reticulum elements. Nu-merous mitochondria and small vesicles fill the apical region of the cell.Inset: vesicles recycling to and from the apical plasma membrane. Nindicates nucleus; L, lysosome; M, mitochondrion; P, principal cell; V,cup-shaped vesicles.

morphological features from EM observations, only nar-row cells in both the rat and mouse epididymis expressCA II.

Furthermore, while some investigators have reportedthe presence of CA II in clear cells of the epididymis(Membre and Bedford, 1985; Brown et al, 1992; Bretonet al, 1999); principal cells of the caput, corpus, and caudaregions (Cohen et al, 1976; Kaunisto et al, 1995) or both,we did not note staining of any cell types of the caput,corpus, or cauda regions. Clearly, a discrepancy existsbetween the different laboratories, one that may involvetissue preservation and immunocytochemical techniques.Nevertheless, in addition to CA II, other isoenzymes ofCA have been identified and localized in different epithe-lial cell types of the epididymis of various species, in-cluding humans, which suggests that these enzymes playan important role in this tissue (Kaunisto et al, 1990,1995; Asari et al, 1996).

It has been reported that H�V-ATPase is localized toapical cells, narrow cells, or both, of the caput region ofthe adult rat epididymis (Brown et al, 1992; Breton et al,1999). In our study, H�V-ATPase was expressed only innarrow cells of the rat and mouse initial segment and

intermediate zone, and not in the caput epididymidis orin apical cells, as judged by their distinctive appearancesas outlined earlier. As noted in the present study, H�V-ATPase has also been shown to be expressed in clear cellsof the adult rat epididymis (Brown et al, 1992, 1997;Breton et al, 1996, 1999; Brown and Breton, 1996).

Through EM, CA II was immunolocalized in narrowcells with gold particles appearing in their cytosol nearor at some distance from the numerous, small apical ves-icles. Gold particles were also abundant, representinganti-H�V-ATPase antibodies in the apical region of apicaland narrow cells (Brown et al, 1992), where they wereclosely associated with the membrane delimiting the nu-merous small apical vesicles of this region. As describedearlier, these vesicles come into contact with the apicalplasma membrane as demonstrated by their ability to in-corporate tracers introduced into the lumen of the duct.Finding both CA II and H�V-ATPase in these small apicalvesicles suggests that some are the site of H� production.Because these vesicles are extremely numerous andwould not all be used in the formation of endosomes,which are not abundant in narrow cells, they may recycleto and from the apical cell surface, and in this way, de-liver protons into the epididymal lumen (Figure 13). Inaddition to the apical plasma membrane, where H�V-ATPase has been located and where H� can be secreteddirectly into the epididymal lumen (Brown et al, 1992),the small apical vesicles would greatly increase the over-all surface area for proton production and eventual secre-tion. The distinct, cup-shaped appearance of many ofthese small vesicles points to their candidacy as the re-cycling elements because such vesicles are unlike thespherical endocytic vesicles of principal cells, which areconsistently CA II and H�V-ATPase negative. In inter-calated cells of the kidney collecting duct, H�V-ATPasemolecules are recycled from the plasma membrane by ahighly specialized population of vesicles (Brown et al,1987b, 1997). These vesicles are coated with an electron-dense material on their cytoplasmic side, composed of thecytoplasmic subunits of a H�V-ATPase, and referred toas ‘‘studs’’ (Brown et al, 1987a). The studs of the H�V-ATPase-rich vesicles have been suggested to contain pro-teins that are involved in regulating and defining theirintracellular targeting. Whereas it has been proposed thatsuch mechanisms exist in narrow cells of the epididymis,experimental data were lacking, but our tracer and im-munocytochemical studies now provide some support forthis proposal.

Clear cells also contain numerous small apical vesicles,some with a cup-shaped appearance, which have beendemonstrated to internalize tracers that are introduced intothe lumen of the caput (Moore and Bedford, 1979) andcauda (Hermo et al, 1988) epididymidis. In our study,H�V-ATPase was immunolocalized to these cells and to



the vesicles of their apical region. Although we did notfind expression of CA II in clear cells, other isoenzymesof CA may be present in these cells. Thus, some of thevesicles in these cells may have a similar function to thosesuggested for narrow cells in our study and be the site ofproton production and secretion.

The proper luminal environment of the epididymis,which is essential for sperm maturation, is conditioned bythe secretion and endocytosis of various substances, in-cluding proteins, by its epithelial lining cells (Hamilton,1975; Cooper, 1986; Robaire and Hermo, 1988; Hermoet al, 1994). We examined the structure and functions ofepithelial narrow cells and clear cells for H� productionand secretion. Localization of CA II and H�V-ATPase innarrow cells suggests that they are involved in protonsecretion and luminal acidification of the initial segmentand intermediate zone, whereas clear cells perform thisfunction in the caput, corpus, and cauda regions by virtueof their expression of H�V-ATPase. In the cauda, an api-cal sodium-hydrogen exchanger also appears to acidifythe lumen (Au and Wong, 1980; Wong and Huang, 1989).

The luminal fluid along much of the epididymis ismaintained at an acidic pH (Levine and Kelly, 1978). Thelow pH has been suggested to be important for spermmaturation and maintaining sperm in an immotile stateduring epididymal transit along with specific proteins,weak acids and ions, and prevention of premature acti-vation of acrosomal enzymes (Wong et al, 1981; Carr etal, 1985; Cooper, 1986; Hinton and Palladino, 1995). Nar-row and clear cells, although not noted for their role inprotein secretion (Hermo et al, 1994) would, however,appear to play important roles in H� secretion and acid-ifying the luminal environment during sperm passage andstorage in the epididymis. In addition, narrow cells of themouse also appear to be involved in apocrine secretionand steroid synthesis.

AcknowledgmentsWe appreciate the assistance of Jeannie Mui, Matilda Cheung, and P.Mukhopadhay. The antibodies were provided by Dr Nick Carter and DrBarry J. Bowman. The drawing in Figure 13 was kindly prepared by DrY. Clermont.

ReferencesAbou-Haıla A, Fain-Maurel M-A. Regional differences of the proximal

part of mouse epididymis: morphological and histochemical charac-terization. Anat Rec. 1984;209:197–208.

Adamali HI, Hermo L. Apical and narrow cells are distinct cell typesdiffering in their structure, distribution, and functions in the adult ratepididymis. J Androl. 1996;17:208–222.

Adamali HI, Somani IH, Huang J-Q, Mahuran D, Gravel RA, Trasler JM,Hermo L. I. abnormalities in cells of the testis, efferent ducts, and

epididymis in juvenile and adult mice with �-hexosaminidase A andB deficiency. J Androl. 1999;20:779–802.

Andonian S, Hermo L. Principal cells of the vas deferens are involvedin water transport and steroid synthesis in the adult rat. J Androl.1999;20:158–176.

Asari M, Sasaki K, Miura K, Ichihara N, Nishita T. Immunohistolocali-zation of the carbonic anhydrase isoenzymes (CA-I, CA-II, and CA-III) in the reproductive tract of male horses. Am J Vet Res. 1996;57:439–443.

Au CL, Wong PY. Luminal acidification by the perfused rat cauda epi-didymis. J Physiol (Lond). 1980;309:419–427.

Bowman EJ, Mandala S, Taiz L, Bowman BJ. Structural studies of thevacuolar membrane ATPase from Neurospora crassa and comparisonwith the tonoplast membrane ATPase from Zea mays. Proc Natl AcadSci USA. 1986;83:48–52.

Bowman EJ, Tenney K, Bowman BJ. Isolation of genes encoding theNeurospora vacuolar ATPase. J Biol Chem. 1988;263:13994–14001.

Breton S, Smith PJ, Lui B, Brown D. Acidification of the male repro-ductive tract by a proton pumping (H�)-ATPase. Nat Med. 1996;2:470–472.

Breton S, Tyszkowski R, Sabolic I, Brown D. Postnatal development ofH�ATPase (proton-pump)–rich cells in rat epididymis. Histochem CellBiol. 1999;111:97–105.

Brown D, Breton S. Mitochondria-rich, protein-secreting epithelial cells.J Exp Biol. 1996;199:2345–2358.

Brown D, Gluck S, Hartwig J. Structure of the novel membrane-coatingmaterial in proton-secreting epithelial cells and identification as anH�ATPase. J Cell Biol. 1987a;105:1637–1648.

Brown D, Lui B, Gluck S, Sabolic I. A plasma membrane proton ATPasein specialized cells of rat epididymis. Am J Physiol. 1992;263:C913-C916.

Brown D, Smith PJS, Breton S. Role of V-ATPase–rich cells in acidifi-cation of the male reproductive tract. J Exp Biol. 1997;200:257–262.

Brown D, Weyer P, Orci L. Non-clathrin coated vesicles are involved inendocytosis in kidney collecting duct intercalated cells. Anat Rec.1987b;218:237–242.

Carr NW, Usselman MC, Acott TS. Effects of pH, lactate and viscoelasticdrag on sperm motility: a species comparison. Biol Reprod. 1985;33:588–595.

Carter ND, Hewett-Emmett D, Jeffery S, Tashian RE. Testosterone-in-duced sulfonamide resistant carbonic anhydrase isozyme of rat liveris indistinguishable from skeletal muscle carbonic anhydrase III.FEBS Lett. 1981;128:114–118.

Christensen AK. Leydig cells. In: Greep RO, Astwood EB, eds. Hand-book of Physiology. Section 7, Vol 5. Washington, DC: AmericanPhsiological Society. 1975;14:339–349.

Cohen JP, Hoffer AP, Rosen S. Carbonic anhydrase localization in theepididymis and testis of the rat: histochemical and biochemical anal-ysis. Biol Reprod. 1976;14:339–346.

Cooper TG. Function of the epididymis and its secretory products (PartIII). In: Cooper TG, The Epididymis, Sperm Maturation and Fertil-ization. Berlin: Springer-Verlag; 1986:117–230.

English HF, Dym M. The time required for materials injected into therete testis to reach points in the caput epididymis of the rat and ob-servations on the absorption of cationic ferritin. Ann NY Acad Sci.1982;383:445–446.

Flickinger CJ. Regional variations in endoplasmic reticulum in the vasdeferens of normal and vasectomized rats. Anat Rec. 1973;176:205–224.

Flickinger CJ. Synthesis, transport and secretion of protein in the initialsegment of the mouse epididymis as studied by electron microscoperadioautography. Biol Reprod. 1979;20:1015–1030.

Hamilton DW. Structure and function of the epithelium lining the ductuliefferentes, ductus epididymis and ductus deferens in the rat. In: Greep


RO, Astwood EB, eds. Handbook of Physiology. Section 7, Vol 5.Washington, DC: American Physiological Society. 1975;303–317.

Helenius A, Mellman I, Wall D, Hubbard A. Endosomes. Trends BiochemSci. 1983;7:245–250.

Hermo L. Structural features and functions of principal cells of the in-termediate zone of the epididymis of adult rats. Anat Rec. 1995;242:515–530.

Hermo L, Dworkin J, Oko R. Role of epithelial clear cells of the ratepididymis in the disposal of the contents of cytoplasmic dropletsdetached from spermatozoa. Am J Anat. 1988;183:107–124.

Hermo L, Oko R, Morales CR. Secretion and endocytosis in the malereproductive tract: a role in sperm maturation. Int Rev Cytol. 1994;154:105–189.

Hinton BT, Palladino MA. Epididymal epithelium: its contribution to theformation of a luminal fluid microenvironment. Microsc Res Tech.1995;30:67–81.

Hoffer AP, Hamilton DW, Fawcett DW. The ultrastructure of principalcells and intraepithelial leucocytes in the initial segment of the ratepididymis. Anat Rec. 1973;175:169–202.

Ilio K, Hess RA. Structure and function of the ductuli efferentes: a review.Microsc Res Tech. 1994;29:432–467.

Jeffery S, Carter ND, Smith A. Immunocytochemical localization of car-bonic anhydrase isozymes I, II, and III in rat skeletal muscle. J His-tochem Cytochem. 1986;34:513–516.

Karnovsky MJ. Use of ferrocyanide-reduced osmium tetroxide in electronmicroscopy. In: Proceedings of the 11th American Society of Cell Bi-ology. New Orleans, La: American Society of Cell Biology; 1971;284:146.

Kaunisto K, Parkkila S, Parkkila A-K,Waheed A, Sly WS, Rajaniemi H.Expression of carbonic anhydrase isozymes IV and II in rat epidid-ymal duct. Biol Reprod. 1995;52:1350–1357.

Kaunisto K, Parkkila S, Tammela T, Ronnberg L, Rajaniemi H. Immu-nohistochemical localization of carbonic anhydrase isoenzymes in thehuman male reproductive tract. Histochemistry. 1990;94:381–386.

Levine N, Kelly H. Measurement of pH in the rat epididymis in vivo. JReprod Fertil. 1978;52:333–335.

Manin M, Lecher P, Martinez A, Tournade S, Jean CL. Exportation of

mouse vas deferens protein, a protein without a signal peptide, frommouse vas deferens epithelium: a model of apocrine secretion. BiolReprod. 1995;52:50–62.

Membre H, Bedford JM. The differential absorptive activity in the epi-didymis of the developing rat. Reprod Nutr Dev. 1985;25:867–872.

Moore HD, Bedford JM. The differential absorptive activity of epithelialcells of the rat epididymis before and after castration. Anat Rec. 1979;193:313–327.

Niemi M. The fine structure and histochemistry of the epithelial cells ofthe rat vas deferens. Acta Anat. 1965;60:207–219.

Oko R. Comparative analysis of proteins from the fibrous sheath andouter dense fibers of the rat spermatozoa. Biol Reprod. 1988;39:169–182.

Oko R, Clermont Y. Light microscopic immunocytochemical study offibrous sheath and outer dense fiber formation in the rat spermatid.Anat Rec. 1989;225:46–55.

Palacios J, Regadera J, Nistal M, Paniagua R. Apical mitochondria-richcells in the human epididymis: an ultrastructural, enzymohistochem-ical, and immunohistochemical study. Anat Rec. 1991;231:82–88.

Reid BL, Cleland KW. The structure and function of the epididymis:histology of the rat epididymis. Aust J Zool. 1957;5:223–252.

Renneberg H, Konrad L, Aumuller G. Immunohistochemistry of secre-tory particles (‘Vesiculosomes’) from the epithelium of bovine sem-inal vesicles and ampulla of the vas deferens. Acta Anat. 1995;153:273–281.

Robaire B, Hermo L. Efferent ducts, epididymis, and vas deferens: Struc-ture, functions and their regulation. In Knobil E, Neill J, eds. ThePhysiology of Reproduction. New York: Raven Press; 1988:999–1080.

Sun EL, Flickinger CJ. Morphological characteristics of cells with apicalnuclei in the initial segment of the adult rat epididymis. Anat Rec.1980;196:285–293.

Wong PYD, Huang SJ. Intracellular pH measurement in primary mono-layer cultures of rat epididymal cells. Pflugers Arch. 1989;413:414–421.

Wong PYD, Lee WM, Tsang AYF. The effects of extracellular sodiumon acid release and motility initiation in rat caudal epididymal sper-matozoa in vitro. Exp Cell Res. 1981;131:97–104.

immunolocalization of ca ii and h v-atpase in … · ap indicates apical region; g, golgi aparatus;...

Documents