Role of Acid/Base Transporters in the Male Reproductive Tract and Potential Consequences of Their Malfunction

Nuria Pastor-Soler, Christine Piétrement, Sylvie Breton


Acid/base transporters play a key role in establishing an acidic luminal environment for sperm maturation and storage in the male reproductive tract. Impairment of the acidification capacity of the epididymis, via either genetic mutations or exposure to environmental factors, may have profound consequences on male fertility.

Of the large number of couples that are affected by infertility, ~30% of the problems result from a male fertility defect. Between 30 and 50% of these male infertility cases are idiopathic. Although a significant number of these cases are likely to have a genetic basis (90), environmental factors are also significant contributors (118). Several studies have shown a progressive decrease in sperm counts and semen quality over the last few decades, although this issue is the subject of considerable current debate (76, 90). In addition to an inability to produce an adequate number of spermatozoa, a major cause of male infertility is the production of sperm with reduced function, including low motility and poor interaction with the oocyte (41).

The generation of competent sperm is a complex process that includes the production of a large number of spermatozoa by the testis, followed by several maturation steps that occur in the excurrent duct. The excurrent duct of the male reproductive tract is composed of highly heterogeneous tissues, including the efferent ducts, epididymis, and vas deferens. Spermatozoa acquire their ability to become motile and to fertilize an egg in the lumen of the tubules that form the epididymis (68, 98, 137). Therefore, failure of the epididymis to provide an adequate environment for sperm maturation may have profound effects on male fertility. One critical feature of the epididymal luminal fluid is that it is maintained acidic (84, 85). Low pH and HCO3 concentration help to keep spermatozoa in a quiescent, immotile state while they mature and are stored in the epididymis (1, 32, 57, 68). In this review, the expression and regulation of key players in acid/base transport in the epididymis will be described.

The proximal epididymis is connected to the testis via the efferent ducts (FIGURE 1). In the rat, the epididymis is composed of one long convoluted tubule that is divided into several regions, including the initial segments and caput, located in the proximal epididymis (FIGURE 1), the corpus, and the cauda, which connects into the vas deferens. At least three epithelial cell types exist in the epididymis: principal cells are located along the entire length of the epididymal tubule, the so-called narrow and clear cells are intercalated between principal cells, and basal cells are located at the base of the epithelium and are not in contact with the luminal side. Narrow cells are relatively low in number and are located in the initial segments. Clear cells, which are more numerous than narrow cells, are located in the caput, corpus, and cauda epididymidis. Narrow, clear, and principal cells play a vital role in establishing the optimal environment for the maturation and storage of spermatozoa (57, 63, 108, 109, 133).


Part of the male reproductive tract

Most of our knowledge of the composition of the luminal fluid of the epididymis originated from the early pioneering work of several groups (84, 85, 91, 127, 132). The luminal fluid in the male reproductive tract undergoes considerable changes in composition after it leaves the seminiferous tubules of the testis and moves through the epididymis and into the vas deferens. These changes include net water, Na+, Cl , and HCO3 reabsorption, K+ secretion, and luminal acidification (4, 84, 85, 127). Several lines of evidence indicate the occurrence of significant transepithelial acid/base transport by the epididymal epithelium. The luminal concentration of HCO3 becomes significantly lower than that of blood as the efferent duct fluid transits through the proximal regions of the epididymis (85). Luminal HCO3 concentration then remains low in the more distal parts of the epididymis and the vas deferens (85). This is accompanied by a significant acidification of the luminal fluid (30, 84, 85, 136).

HCO3 Reabsorption in the Proximal Epididymis

Previous studies have measured significant water reabsorption, coupled to Na+ and HCO3 reabsorption, in the efferent ducts and in the initial segments (28, 49, 85, 140). These tissues are related embryologically to the kidney, and they share similar transport mechanisms (58). In the kidney, the Na+/H+ exchanger NHE3 plays a major role in acid-base balance and Na-fluid homeostasis (113). It is highly expressed in the apical membrane of kidney proximal tubules, where it is involved in NaCl and HCO3 reabsorption (62, 96, 135). In the male reproductive tract, NHE3 is highly expressed in the apical membrane of nonciliated cells of the efferent ducts and principal cells of the epididymis, and it is absent from narrow and clear cells (6, 72, 106) (FIGURE 2, A AND B). Interestingly, the level of expression of NHE3 varies in different regions of the rat epididymis (6). In the proximal parts, NHE3 is most abundant in the initial segment and in the proximal caput epididymidis. In the distal regions of the epididymis, NHE3 is expressed in the proximal cauda only, being absent from the distal cauda and vas deferens. In addition, another Na+/H+ exchanger, NHE2, has been described in the apical membrane of principal cells from the caput, corpus, and cauda epididymidis but was shown to be absent from the initial segments (36). Functional studies using epididymal tubules perfused in vitro and loaded with the intracellular pH indicator BCECF showed that EIPA-dependent Na+/H+ exchange accounts for ~50% of apical acid extrusion in the nominal absence of extracellular HCO3 (6). Because the initial segments were reported to be negative for NHE2 (36), NHE3 was the most likely isoform to be involved in this region. However, HOE694, used at a concentration that was expected to inhibit most NHE2 activity but not NHE3 (37), inhibited all of the EIPA-sensitive intracellular pH recovery. This unexpected result indicates that either 1) NHE2 is present in the initial segments but has so far remained undetected; 2) an NHE3 splice variant, sensitive to HOE694, is expressed in the epididymis; or 3) epididymal posttranslational modification confers HOE694 sensitivity to NHE3. The two latter possibilities were proposed to explain a similar unanticipated inhibition of NHE3 by HOE694 in the main pancreatic duct (82). It is, therefore, conceivable that, as in the pancreatic duct, an HOE694-sensitive NHE3 is expressed in the initial segments of the epididymis. Additional experiments using NHE2, NHE3, and double-knockout mice will be required to answer these questions.


Immunofluorescence localization of the Na+/H+ exchanger NHE3, the Na+-HCO3 cotransporter NBCe1-A, and the Cl/HCO3 exchanger AE2 in the epididymis
A: principal cells in the distal initial segment show a strong apical staining for NHE3 (green). Narrow cells are identified by their positive immunoreactivity for the E subunit of the vacuolar H+-ATPase (V-ATPase; yellow). The long stereociliae of principal cells stained for NHE3 are masking the apical pole of narrow cells in this conventional microscopy image. B: confocal image shows the absence of NHE3 from a narrow cell stained for the E subunit of the V-ATPase (red) and its presence in principal cell stereociliae (green). C and D: all epithelial cells of the initial segment show intense basolateral staining for NBCe1-A (C) and AE2 (D).

In addition, immunofluorescence studies showed that the electrogenic Na+-HCO3 cotransporter NBCe1-A (originally called NBC) and the Cl /HCO3 exchanger AE2 are expressed in the basolateral membrane of all epididymal cells, whereas AE1 is not present in this tissue (65, 66) (FIGURE 2, C AND D). The levels of expression of NBCe1-A and AE2 were higher in the proximal regions of the epididymis, where a low luminal concentration of HCO3 is established (84, 85), indicating their potential role in HCO3 reabsorption. Epididymal principal cells also show strong cytosolic carbonic anhydrase (CA) activity, especially in the proximal regions (40) and CA-IV is expressed in their apical and basolateral membranes (73, 101) (see also FIGURE 3). More recently, additional CA isoforms have been identified in kidney and epididymal epithelial cells. In the kidney, CA-XII is present in the basolateral membrane of epithelial cells of the thick ascending limb of Henle, distal convoluted tubule, and collecting duct (78, 102), and membrane-bound CA-XIV message has been found in proximal tubules (74, 92). In human epididymis, CA-XII colocalizes with CA-II in apical mitochondria-rich cells (which correspond to rat narrow and clear cells) (71). More recently, a complex pattern of CA expression was reported in the rat epididymis (54). CA-XII was located in the basolateral membrane of narrow cells of initial segments and in principal cells of the corpus and cauda epididymidis, and CA-XIV was located apically and basolaterally in principal cells.


Immunofluorescence localization of carbonic anhydrase-IV in the epididymis
Strong apical staining and weaker basolateral staining is detected in principal cells of the caput epididymidis. The subepithelial smooth muscle cells are also stained for carbonic anhydrase (CA)-IV.

In summary, principal cells of the proximal regions of the epididymis are fully equipped for net HCO3 reabsorption and use transport mechanisms that are very similar to those used by the kidney proximal tubule (48). In the initial segments of the epididymis, the apical Na+/H+ exchanger NHE3 or an “NHE3-like” exchanger secretes protons into the lumen (FIGURE 4). These H+ ions combine with luminal HCO3 to form CO2 and H2O under the enzymatic activity of CA-IV and CA-XIV. Newly formed CO2 then diffuses into the cell through the apical membrane and is hydrated by cytosolic CA-II to form H+ and HCO3 . The protons recycle back into the lumen via NHE3, whereas HCO3 would diffuse passively across the basolateral membrane via the Na+-HCO3 cotransporter NBCe1-A or the anion exchanger AE2. The result of these processes is net HCO3 reabsorption with no net proton secretion. Basolateral CAs were proposed to facilitate Na+-HCO3 cotransport by preventing the development of an alkaline disequilibrium pH in the interstitium (116, 126).


Model for HCO3 reabsorption in the initial segments of the epididymis
Protons are transported out of the cell via apical Na+/H+ exchanger NHE3. The driving force for proton extrusion is provided by basolateral Na+-K+-ATPase, which maintains a low intracellular Na+ concentration. K+ is recycled back through the basolateral membrane via potassium channels. Luminal protons are combined with luminal HCO3 under the activity of CA-IV and CA-XIV to form H2O and CO2. Luminal CO2 diffuses into the cell through the apical membrane and is hydrated to form H+ and HCO3 , a reaction that is catalyzed by cytosolic CA-II. Newly formed HCO3 is transported through the basolateral membrane via the Na+-HCO3 transporter NBCe1-A or the Cl /HCO3 exchanger AE2. Basolateral CA-IV is proposed to facilitate HCO3 transport through the basolateral membrane by facilitating HCO3 dissipation in the interstitium.

Net Proton Secretion in the Distal Epididymis: Role of the Vacuolar H+-ATPase

The luminal fluid of the distal epididymis is maintained at the acidic pH of 6.8 (84, 85), indicating that active proton secretion takes place in this segment. Previous studies have shown that: 1) narrow and clear cells express high levels of the vacuolar H+-ATPase (V-ATPase) in their luminal plasma membrane, as well as in intracellular vesicles (14, 22, 25, 50, 53, 104, 120); 2) CA-II is highly expressed in narrow and clear cells in the epididymis and vas deferens (11, 14, 15) (FIGURE 5); and 3) the bulk of proton secretion in the vas deferens is inhibited by the specific V-ATPase inhibitors bafilomycin and concanamycin (14). Coexpression of the V-ATPase and CA-II in the same subpopulation of cells indicated the participation of this CA in net proton secretion. It has to be noted that an absence of CA-II in clear cells has been reported previously (54). Whether this apparent discrepancy is attributed to different fixation protocols [a fixation solution containing paraformaldehyde, lysine, and periodate (PLP) gave positive results, whereas Bouin’s or St-Marie’s fixatives gave negative results] or different primary antibodies (an antibody provided by William S. Sly, St. Louis University Medical Center and School of Medicine, gave positive results, whereas a commercial antibody gave negative results) still remains to be elucidated. Au and Wong have shown a marked reduction in the rate of luminal acidification by the CA inhibitor acetazolamide in the cauda epididymidis perfused in vivo (4). In addition, our laboratory reported that acetazolamide strongly reduced bafilomycin-dependent proton secretion in the isolated vas deferens (11). It appears, therefore, that CA-II plays a significant role in luminal acidification of the distal segments of the male reproductive tract (cauda epididymidis and vas deferens).


Immunofluorescence localization of V-ATPase (E subunit) and CA-II in a cryostat section of the cauda epididymidis
Clear cells show strong apical staining for V-ATPase (yellow) and intense cytoplasmic staining for CA-II (red). No CA-II or V-ATPase staining is detected in adjacent principal cells.

These results indicate that proton secretion is accompanied by HCO3 reabsorption. A basolateral Cl /HCO3 exchange mechanism has been described in primary cultures of cauda epididymidis (83). In the isolated vas deferens, the rate of proton secretion is independent of Cl but is strongly inhibited by SITS under normal and Cl -free conditions (11). Low expression of AE2 and NBCe1-A was detected in the basolateral membrane of epithelial cells lining the cauda epididymidis and the vas deferens (65, 66). However, because net proton secretion is independent of Cl in this segment (11), AE2 does not appear to be involved, and NBCe1-A is the most likely candidate for basolateral HCO3 transport.

Thus net luminal acidification in the cauda epididymidis and vas deferens is achieved by the V-ATPase, which works in conjunction with basolateral HCO3 transporters to couple proton secretion with HCO3 reabsorption. The V-ATPase is a complex enzyme that is composed of several subunits (20, 45, 93, 95, 119, 129). It is divided into two distinct sectors, the V0 and V1 sectors. The V0 sector is composed of transmembrane subunits (subunits a and c) as well as subunit d, which is closely associated with the membrane. The V0 sector is responsible for proton translocation. The V1 sector forms a cytosolic complex of eight subunits (subunits A–H) and is associated with the membrane via its interaction with the V0 sector. Several subunits of the V0 and V1 sectors exist in more than one isoform. The exact function of each of these isoforms is the subject of intensive research. These isoforms are expressed by distinct cell types and are located in different compartments of the cells, and it has been proposed that the assembly of a particular set of subunit isoforms may determine the intracellular targeting of the V-ATPase (intracellular organelles vs. plasma membrane) and may control its function.

Regulation of Proton Secretion via Recycling of V-ATPase

Similarly to renal type A intercalated cells, α-CA-rich cells of the turtle bladder, and osteoclasts, proton secretion in clear cells is regulated via recycling of V-ATPase-containing vesicles to and from the apical membrane (8, 13, 103). This active recycling is reflected by a very high rate of endocytosis (2, 18, 19, 24, 26, 55, 89, 91, 114, 121). In these cells, an increase in V-ATPase surface expression closely correlates with an increase in proton secretion (8, 18, 81, 89, 103, 121). The molecular mechanisms responsible for the regulation of V-ATPase recycling are still poorly characterized. In renal intercalated cells, V-ATPase-containing vesicles possess an extensive cytoplasmic coat consisting largely of V-ATPase subunits (19). Previous studies have shown that these vesicles are devoid of clathrin (21, 23) and of caveolin-1 (12). Epididymal epithelial cells are also negative for caveolin-1 (S. Breton and D. Brown, unpublished data). Thus V-ATPase recycling in intercalated cells as well as epididymal clear cells may rely on different and possibly unique clathrin- and caveolin-independent mechanisms for the regulation of proton secretion. The V-ATPase itself may be involved in these trafficking processes.

Role of SNARE proteins

A family of specialized proteins, the so-called soluble N-ethyl-maleimide-sensitive fusion protein attachment protein (SNAP) receptor (SNARE) proteins are major players in the docking and subsequent fusion of a variety of vesicles, and they have been implicated in the regulation of most intracellular membrane-trafficking events (61). It is interesting to note that some subunits of the V-ATPase are homologous to synaptic vesicle-associated proteins. Subunits a and c of the V-ATPase associate with synaptobrevin and synaptophysin on synaptic vesicles (46). In addition, the B1 subunit of the V-ATPase is present in aquaporin 2 (AQP2)-containing endosomes isolated from kidney principal cells (111). Because these endosomes do not contain other subunits of the V-ATPase and do not acidify their lumen, it is possible that the B1 subunit has a function independent of proton-pumping activity. Therefore, some subunits of the V-ATPase may be involved, in a novel way, in the recycling and targeting machinery underlying the delivery of the V-ATPase itself, as well as some other membrane proteins, to their target membrane. Previous reports have shown interaction between the V-ATPase and members of the SNARE family, including syntaxin 1A and SNAP23 in inner medullary collecting duct cultured cells (7, 86, 94). In the epididymis, we have shown that cellubrevin, a vesicle-associated SNARE protein [v-SNARE, also known as R-SNARE (35)], is highly expressed in clear cells of the epididymis and vas deferens and that cleavage of cellubrevin by tetanus toxin markedly inhibited bafilomycin-dependent proton secretion in isolated vas deferens (13). Because tetanus toxin did not directly inhibit the V-ATPase pumping activity in endo-somes isolated from rat kidney cortex, we proposed that inhibition of proton secretion by tetanus toxin resulted from a decrease in the number of membrane-inserted V-ATPase molecules, due to an impairment of the exocytotic process. SNAP23 is also expressed in epididymal clear cells, where it partially colocalizes with the V-ATPase (39). However, we did not detect syntaxin 1A mRNA or protein in epithelial cells of the epididymis, and the target membrane T-SNARE that would interact with the V-ATPase in the male reproductive tract is still unknown.

HCO3 -regulated soluble adenylyl cyclase modulates V-ATPase recycling

Soluble adenylyl cyclase (sAC) is a chemosensor that mediates HCO3 -dependent elevation of cAMP (34). It is distinct from transmembrane adenylyl cyclases and is directly stimulated by HCO3 ions. We have shown that sAC is highly expressed in clear cells of the epididymis (103). V-ATPase recycling in epididymal clear cells is strongly dependent on luminal pH or luminal HCO3 concentration. In vas deferens perfused in vivo with a physiological solution containing HCO3 and adjusted to pH 7.1, clear cells exhibited numerous, well-developed V-ATPase-labeled microvilli (FIGURE 6A). Under these conditions, very little endocytosis of the V-ATPase was detected. In contrast, when vas deferens was perfused in a HCO3 -free solution kept at the same pH value, a larger portion of V-ATPase molecules were present in subapical endosomes and shorter and fewer V-ATPase-labeled microvilli were detected (FIGURE 6B). A similar response was observed when the vas deferens was perfused at very alkaline pH values (pH 7.8) in the absence of HCO3 (103). The sAC inhibitor 2-hydroxyestradiol prevented the alkaline pH- and HCO3 -induced apical accumulation of the V-ATPase, indicative of the participation of sAC in this process. Addition of the permeant analog cpt-cAMP mimicked the effect of luminal HCO3 or alkaline pH. Interestingly, acetazolamide induced a complete internalization of the V-ATPase and a complete retraction of apical microvilli. This last result correlates with the inhibition of proton secretion that we had previously observed in isolated vas deferens exposed to acetazolamide (11) and indicates that intracellular production of HCO3 is a key process for the targeting of the V-ATPase to the apical membrane. Together, these results suggest that intracellular HCO3 elevation activates sAC and triggers cAMP production, which in turn leads to the accumulation of the V-ATPase in the apical membrane.


Effect of luminal HCO3 on V-ATPase localization in clear cells
Confocal images showing clear cells from cauda epididymidis perfused luminally in vivo with modified Hanks buffer containing 12 mM HCO3 (pH 7.1, 5% CO2) (A) or PBS adjusted to pH 7.1 (B). Endocytosis was detected by adding horseradish peroxidase (HRP) into the luminal solutions. Double staining for V-ATPase (green) and HRP (red) was performed. A: In the presence of luminal HCO3 , the V-ATPase is mainly located in well-developed apical microvilli. B: In the absence of HCO3 , the V-ATPase is distributed between apical microvilli and subapical vesicles. The arrows indicate the frontier between apical microvilli and the subapical region of the cell.

Primary cultures of epididymal and vas deferens principal cells have the ability to secrete HCO3 in response to a variety of extracellular stimulation (115, 131). HCO3 secretion by epididymal principal cells is achieved through cAMP-dependent CFTR located in the apical membrane (130) that works in conjunction with basolateral Na+/H+ exchanger NHE1 (36). Luminal elevation in HCO3 concentration was proposed to occur during sexual arousal to “prime” spermatozoa before ejaculation (31). However, HCO3 secretion by principal cells would lead to an increase in luminal pH. We propose that clear cells respond to a rise in luminal HCO3 concentration by increasing their rate of proton secretion following HCO3 -induced sAC activation and intracellular cAMP elevation (FIGURE 7). This would reestablish the luminal pH to its resting acidic value. A potential apical transport route for HCO3 entry through the apical membrane might be the electroneutral Na+-HCO3 cotransporter NBC3 (also known as NBCn1), which is expressed in the apical membrane of clear cells (106).


Model for HCO3-induced apical V-ATPase accumulation in clear cells
Increase in luminal HCO3 concentration leads to an increase in intracellular HCO3. The Na+-HCO3 cotransporter NBC3 represents a potential route for apical entry of HCO3. Intracellular HCO3 activates soluble adenylyl cyclase (sAC), which results in the production of cAMP. An increase in intracellular cAMP leads to an increase in V-ATPase exocytosis and/or a decrease in V-ATPase endocytosis, resulting in the accumulation of V-ATPase in well-developed microvilli and an increase in net proton secretion. This would reestablish the luminal pH to its resting acidic value.

A variety of proton-secreting epithelia, including kidney thick ascending limbs, distal tubules, collecting ducts, and the choroid plexus also express sAC (34, 103). sAC may, therefore, be a common sensor that allows acidifying cells to sense and modulate the pH of their environment.

Modulation of the actin cytoskeleton via gelsolin participates in the regulation of V-ATPase recycling

The role of the actin cytoskeleton in regulating membrane protein trafficking is complex and depends on the cell type and protein being examined (3, 59). The actin cytoskeleton is a very dynamic structure that is regulated by a large number of proteins, including the calcium-activated, actin-capping and -severing protein gelsolin. We recently showed that clear cells express a very high level of gelsolin (8). When the potent actin-polymerizing agent jasplakinolide was used to overcome the severing action of gelsolin, a complete inhibition of the alkaline pH- and cAMP-induced apical membrane accumulation of V-ATPase was observed, indicating that dynamic remodeling of the actin cytoskeleton is required for the recycling of V-ATPase. A corresponding inhibition of bafilomycin-sensitive net proton secretion was detected under these conditions using a self-referencing proton-selective electrode. Conversely, when gelsolin-mediated actin-filament elongation was inhibited using a 10-residue peptide (PBP10) derived from the phosphatidylinositol 4,5-bisphosphate-binding domain (PBD2) of gelsolin (38), a significant accumulation of V-ATPase in the apical membrane was induced at luminal acidic pH. PBP10 inhibits actin assembly by preventing the uncapping of gelsolin from the growing end of the actin filaments and induces major disruption of the actin cytoskeleton (38). A similar disruption of the actin cytoskeleton by peptides containing the PBD domains of gelsolin was observed in osteoclasts (10). Thus apical accumulation of V-ATPase generated by the PBP10 peptide at acidic pH was probably secondary to the induction of a less-polymerized actin cytoskeleton. No additional effect of PBP10 was observed on V-ATPase distribution at alkaline luminal pH, suggesting that the actin cytoskeleton was already maintained in a more depolymerized state via gelsolin under these conditions. These results indicate a major role for gelsolin in modulating the intracellular and plasma membrane distribution of V-ATPase. The severing activity of gelsolin is strongly dependent on intracellular calcium (64, 87, 138). In agreement with the role of gelsolin in modulating V-ATPase recycling, we have shown the absolute requirement of calcium for the alkaline pH-induced accumulation of V-ATPase (8).

Interestingly, subunits B and C of the V-ATPase interact directly or indirectly with actin (16, 60, 128). The V-ATPase binds to F-actin but not G-actin, and in osteoclasts its internalization is correlated with increased interaction with actin (33). We are currently investigating the possibility that the interaction between V-ATPase and actin can also be modulated in clear cells.

High expression of gelsolin is a common feature of cell types that express the V-ATPase in their plasma membrane and recycling vesicles, including kidney intercalated cells, osteoclasts, and spermatozoa (10, 27, 88). Therefore, modulation of the actin cytoskeleton by this severing and capping protein may represent a common regulatory mechanism for proton secretion in these cells.


The mechanisms responsible for the apical membrane accumulation of V-ATPase in response to various stimuli are still not fully characterized. Whereas some recycling membrane proteins, such as AQP2 and CFTR, are directly phosphorylated by PKA (9, 17), none of the subunits of the V-ATPase appears to be phosphorylated by PKA (122, 129). This indicates that an indirect mechanism might be involved in the cAMP-induced apical accumulation of the pump. Modulation of the actin cytoskeleton by cAMP is a key step in the regulation of several membrane transporters (105, 107, 117, 123, 124, 134). It is, therefore, conceivable that depolymerization of the actin cytoskeleton by a cAMP-dependent mechanism, leading to inhibition of V-ATPase endocytosis, would induce the apical accumulation of V-ATPase in epididymal clear cells. We have also shown that cAMP can induce V-ATPase apical membrane accumulation after the pump had been completely internalized by acetazolamide, and we proposed that cAMP also increased the rate of V-ATPase exocytosis (103). Thus cAMP might exert its action via both an increase in V-ATPase exocytosis and a decrease in V-ATPase endocytosis. Whether cAMP acts via modulation of the actin cytoskeleton and/or via PKA phosphorylation of an intermediate protein involved in the trafficking of the V-ATPase still remains to be elucidated.

Impairment of Acidification and Consequences for Male Fertility

Mutations of V-ATPase subunits

As mentioned above, the V-ATPase is a very complex enzyme that is composed of many distinct subunits. Two of its subunits, B and a, are of particular interest because mutations of one of their isoforms in humans lead to several disease states. The B subunit is part of the V0 sector and has two isoforms, B1 and B2. The a subunit is part of the V0 sector and has four isoforms, a1, a2, a3, and a4. Mutations of the Atp6v1b1 gene coding for the B1 subunit and of the Atp6v0a4 coding for the a4 subunit cause recessive distal renal tubular acidosis, due to impairment of proton secretion by collecting duct intercalated cells (129). The exact mechanisms by which these mutations cause a dysfunction of the V-ATPase remain to be elucidated. For example, it is not known whether the mutated B1 and a4 subunits can assemble in the V-ATPase complex and, if so, whether they may affect its targeting to the apical membrane. Interestingly, whereas the B1 subunit is expressed in kidney intercalated cells and is absent from proximal tubules, the a4 subunit is present in both cell types. However, no apparent defect in proximal tubule acid/base transport has been detected in patients harboring mutations of the a4 subunit. Similarly, bilateral sensorineural hearing loss was diagnosed in the majority of the patients harboring the B1 subunit mutations but not in the patients harboring mutations of the a4 subunit. This raised the possibility that other a subunits (a1, a2, or a3) might compensate for the lack of functional a4 subunit in the proximal tubule or in the inner ear.

Both B1 and a4 subunits have been described in the male reproductive tract, where they are highly expressed in the apical plasma membrane of epididymal narrow and clear cells (14, 22, 120). Because luminal acidification is achieved in part by the V-ATPase, it is possible that fertility might be altered in patients harboring B1 and a4 subunit mutations. Several such patients were diagnosed recently, and they were all younger than 6 years old, with the majority presenting before the first year of age. Long-term clinical follow-up of these young patients will determine whether or not their fertility will be impaired.

V-ATPase B1 subunit-deficient mice

To better characterize the role of the B1 isoform in the function of the V-ATPase, mice deficient in the murine V-ATPase B1 subunit homolog (Atp6v1b1−/− mice) were generated (44). Interestingly, in contrast to human patients with B1 mutations, Atp6v1b1−/− mice appeared normal, as long as they were fed a normal diet (44). However, their urine was significantly more alkaline, and when challenged with an oral acid load, they developed a more pronounced metabolic acidosis compared with their wild-type littermates. These findings indicate that although some V-ATPase function is retained in the absence of B1, the pump cannot increase its activity after an acid challenge. Expression of the B2 subunit, together with the B1 subunit, has recently been described in renal intercalated cells (104). Under baseline conditions, the B2 isoform was detected mainly in intracellular vesicles, but under some conditions, including CA inhibition, it was also detected in the apical membrane of proton-secreting type A intercalated cells. These results indicate that the B2 isoform can compensate, at least partially, for the lack of B1 in the function of the V-ATPase. Interestingly, Atp6v1b1−/− mice were reported to be fertile. In the male reproductive tract, B2 is also expressed in clear cells together with B1, but in normal mice and rats, it is located mainly in subapical vesicles and is absent from apical microvilli (104), where other subunits of the V-ATPase predominate. However, we have recently detected a significant redistribution of the B2 subunit from intracellular vesicles to the apical plasma membrane in clear cells of Atp6v1b1−/− mice compared with wild-type mice, indicating that the B2 subunit could partially compensate for the absence of B1 and help maintain the epididymal luminal pH within the acid range compatible with fertility (Da Silva N, Paunescu TG, Brown D, and Breton S, unpublished observation). Thus the B2 isoform may serve as a potential backup for the active role played by the B1 subunit in luminal acidification of the epididymis and the kidney.

It will be interesting to determine whether the presence of mutated B1 in human patients might prevent the assembly of B2 in the V-ATPase holoenzyme, which would result in a more complete inhibition of V-ATPase function. In Atp6v1b1−/− mice, the absence of B1 (or the presence of a significantly truncated and dysfunctional B1 protein) might allow for the incorporation of the B2 isoform into the holoenzyme, which might preserve a better V-ATPase function.

Cadmium inhibits the V-ATPase

Several studies conducted in different populations of the world have suggested a significant decrease in male fertility over the past 50 years (76, 118). Although such a decline is still a matter of debate and might be attributed to methodological factors and regional differences, there is some evidence that environmental substances, including estrogenic and antiandrogenic agents, fungicides, pesticides, and heavy metals, have a detrimental effect on male fertility (5, 56, 67, 69, 75, 125). Exposure to various environmental pollutants, including cadmium, known to affect luminal acidification in the epididymis (29, 52) also induce a reduction in male fertility (100, 125). The heavy metal cadmium is of particular interest in male fertility because it is present in tobacco smoke and food, and it accumulates in the environment due to human contamination by mining, smelting, and use in industry. Cadmium exposure induces a reduction in the size of testis, epididymis, and seminal vesicles (29, 79, 80), a decrease in sperm concentration (79, 80, 99, 112), and a decrease in plasma testosterone concentration (42, 43). Interestingly, a significant alkalinization of the luminal fluid of the epididymis was observed in adult male rats treated with cadmium (29). In addition, treatment of rats with cadmium for 2 wk induced a regression of the morphology of the epididymis to a prepubertal or castrated phenotype and a marked redistribution of the V-ATPase from the apical pole of clear cells into intracellular vesicles scattered throughout the cytoplasm (52). Cadmium also directly inhibited bafilomycin-sensitive ATPase activity in preparations of epididymal plasma membranes and inhibited bafilomycin-sensitive proton secretion in isolated, cut-open vas deferens, indicating a direct inhibition of the V-ATPase by this heavy metal (52). Thus cadmium may act as an hormonal disruptor similarly to several environmental and occupational pollutants that have been implicated in the reduction of fertility in men and animals. The cadmium-induced internalization of the V-ATPase in clear cells may be at least partially responsible for the luminal alkalinization that occurs after cadmium intoxication (29). In kidney proximal tubules, cadmium-metallothionein induces a depolymerization of microtubules, leading to the internalization of a variety of apical membrane proteins, including the V-ATPase (110). Interestingly, cadmium exposure reduces cAMP levels in rat testis (47). In addition, cadmium can directly inhibit the V-ATPase, which may also contribute to the cadmium-induced alkalinization. A similar inhibition of the V-ATPase was proposed to be responsible for the detrimental effect of cadmium on kidney function (51).

Importantly, a significant increase in blood cadmium was observed in smokers (70), a condition known to significantly reduce male fertility (77, 139). The reduction in male fertility in smokers was associated in part with a reduction in sperm quality and motility. Whereas the exact component of the cigarette smoke responsible for this adverse effect is still not known, it is tempting to propose that cadmium, by altering the luminal pH of the epididymis, may contribute at least partially to this defect. In this respect, it is important to note that compared with several parts of the reproductive tract, the epididymis, together with the kidney, are among the organs that accumulate cadmium most efficiently (97).


In summary, transepithelial acid/base transport in the epididymis is achieved by distinct sets of transport proteins located in principal cells and clear cells. In the proximal epididymis, the Na+/H+ exchanger NHE3 or an “NHE3-like” transporter appears to be involved in HCO3 reabsorption. NHE3 is abundantly expressed in the apical membrane of principal cells in the initial segments, where few V-ATPase-rich narrow cells are present. In contrast, NHE3 is not detectable in the distal cauda epididymidis, where V-ATPase-rich clear cells are numerous. In the distal cauda epididymidis, where spermatozoa are stored, clear cells are poised to play a central role in the final steps of luminal acidification. Another Na+/H+ exchanger, NHE2, is also present in the apical membrane of principal cells and may participate in luminal acidification (36). These apical proton-secretory proteins work in conjunction with cytosolic CA-II, membrane-bound CA-IV and CA- XIV, and basolateral HCO3 transporters, including the electrogenic Na+-HCO3 cotransporter NBCe1-A and the Cl/HCO3 exchanger AE2, to secrete protons and reabsorb HCO3.

Net proton secretion by clear cells is modulated via active endocytosis and exocytosis of V-ATPase-containing vesicles. This recycling mechanism represents a possibly unique clathrin- and caveolin-independent process. V-ATPase recycling depends on an intact microtubule network, is highly regulated via modulation of the actin cytoskeleton, and requires the participation of members of the SNARE protein family. In addition, accumulation of V-ATPase in the apical membrane is induced by either an increase in luminal pH or HCO3 concentration, following a sAC-dependent elevation of cAMP. Ongoing studies in our laboratory are aimed at examining the downstream effectors of cAMP in regulating V-ATPase recycling, as well as other potential extracellular stimuli that may regulate proton secretion in the epididymis, a process that is central to establishing a suitable acidic luminal environment in which sperm mature and are stored in a quiescent state.


We would like to thank William S. Sly (St. Louis University Medical Center and School of Medicine) for providing us with his anti-CA-IV and anti-CA-II antibodies, Walter F. Boron (Yale University School of Medicine) for his anti-NBCe1-A antibody (originally labeled NBC), and Daniel Biemesderfer (Yale University School of Medicine) for his anti-NHE3 antibody. Our special thanks go to Dennis Brown for his critical reading of the manuscript and constructive comments.

This work was supported by National Institutes of Health (NIH) Grants HD-40793 and DK-38452 to S. Breton, NIH Grant K08-HD-45524 to N. Pastor-Soler, and grants from the Committee of American Memorial Hospital of Reims, France, the Conseil Régional de Champagne-Ardenne, France, and the Ministère des Affaires Etrangères (Concours Lavoisier), France to C. Piétrement.

Present address for Nuria Pastor-Soler: Department of Medicine, Renal-Electrolyte Division, University of Pittsburgh School of Medicine, Pittsburgh, PA 15261

Present address for Christine Piétrement: Départment de Pédiatrie, Hôpital américain, Reims Cedex, France 51092.


View Abstract