- Top of page
Although previous studies employing efferent duct ligation in intact rat testes (Setchell, 1967) and micropuncture and microcannulation of the rat seminiferous tubules in vivo (Tuck et al. 1970; Levinè & Marsh, 1971) and in vitro (Cheung et al. 1977) have provided evidence that the seminiferous tubules elaborate a watery secretion into the tubular lumen, the mechanism of secretion and its regulation have not been fully explored. This study represents a first attempt to unravel secretory mechanisms by measuring electrogenic ion transport across cultured monolayers of rat Sertoli cells. Immature rat Sertoli cells grown on filters coated with basement membrane matrices have been shown to form confluent cell monolayers which assume an epithelial-like appearance reminiscent of Sertoli cells in vivo (Byers, 1986; Dym, 1994). These epithelia have been used to study protein biosynthesis and secretion (Onoda & Djakiew, 1990; Grima et al. 1992), FSH-induced cAMP production (Dym et al. 1991) and Sertoli cell differentiation (Hadley et al. 1985), etc. However, the use of cultured Sertoli cell epithelia to study ion transport has not been attempted.
When clamped in the Ussing chamber, cultured immature rat Sertoli cell epithelia did not exhibit a measurable transepithelial potential. The low transpithelial resistance of 60 Ω cm2 probably reflects a leaky intercellular pathway. Measurement of transtubular potential difference in intact rat seminiferous tubules in vivo revealed a value of 4 mV, lumen negative (Levinè & Marsh, 1971). However, direct comparison between the two sets of data is difficult since the seminiferous tubule fluid contains a high concentration of K+ ions (Tuck et al. 1970; Levinè & Marsh, 1971) which would have depolarized the luminal membrane of the tubular cells in vivo. The short-circuit current (Isc) is virtually zero at rest but could be increased by thapsigargin and forskolin, which are known to increase intracellular Ca2+ and cAMP, respectively. It transpires that, as in other secretory epithelia, transepithelial ion transport in the Sertoli cells is under the control of a dual intracellular signalling pathway involving Ca2+ and cAMP. The magnitude of the response to thapsigargin was greater than that to forskolin and cpt-cAMP (Figs 1 and 8), indicating that the Ca2+ pathway predominates in the control of transepithelial ion secretion by Sertoli cells. Although forskolin and cpt-cAMP produced Isc responses of similar magnitude, the time courses were very different (Fig. 1). Forskolin, being more lipid soluble than cpt-cAMP, would elicit a faster response. In rat Sertoli cells, cAMP was found to raise intracellular Ca2+ (Gorczynska et al. 1994). A possibility therefore exists that the stimulation of Isc by forskolin and cpt-cAMP could be mediated by a rise in intracellular Ca2+.
ATP (100 μm) added to the apical side caused a relatively large and sustained increase in Isc. The inward current could be wholly inhibited by apically applied 1 mm DPC or 0.25 mm DIDS (Fig. 2), suggesting that the current could be attributed to chloride secretion. The lack of effect of apical amiloride (0.01 mm) argues against Na+ reabsorption being responsible for the inwardly flowing current (Table 1). Ion substitution experiments also supported anion secretion as the basis of the Isc response to ATP. Removal of chloride from the external medium greatly attenuated the Isc, suggesting that the current is likely to be due to chloride secretion. However, removal of HCO3− also reduced the Isc by 40 %. This reduction in current after HCO3− removal could be due to loss of HCO3− secretion per se or HCO3−-dependent Cl− secretion. The former seems unlikely for the following reasons. First, acetazolamide and 6-ethoxyzolamide, inhibitors of carbonic anhydrase (CA) did not affect the Isc response to ATP (Table 1). Rete testis fluid collection following efferent duct ligation has shown no acute effect of acetazolamide on fluid production by the rat testis (Setchell, 1967). Using a histochemical method to detect CA, Cohen et al. (1976) and Ekstedt & Ridderstrale (1992) localized CA in the interstitial cells but not in the seminiferous tubules of the rat and rabbit testes. We too were unable to detect CA in the primary cultures of immature rat Sertoli cells (authors' unpublished observation). This negative staining for CA in the Sertoli cells contrasts sharply with the abundant CA expressed by the rat epididymal cells which have been shown to secrete HCO3− copiously (Chan et al. 1996). Secondly, amiloride (1 mm), bafilomycin A1 (0.8 μm) and H2DIDS (300 μm), inhibitors of the Na+-H+ exchanger, H+-ATPase and Na+-HCO3− cotransporter, respectively, added basolaterally did not affect the Isc response to ATP (Table 1). This indicates that these transporters and pump could not have contributed to the Isc, and by inference HCO3− secretion, on account of the fact that the Na+-H+ exchanger, H+-ATPase, and Na+-HCO3− cotransporter have been implicated in HCO3− secretion in other epithelia (see Gleeson, 1992; Raeder, 1992). Thirdly, contrary to the earlier work of Tuck et al. (1970), a more recent micropuncture study by Caflisch & DuBose (1990) showed that the rat seminiferous tubular fluid contains a very low concentration of HCO3−. These results led these authors to argue that active HCO3− secretion is not an important factor in the formation of the seminiferous tubular fluid.
It would seem highly probable that the Isc response to apically applied ATP is due to transepithelial Cl− secretion. However, basolaterally applied bumetanide at a concentration which has been shown to abolish Cl− secretion in other Cl−-secreting epithelia produced only a small inhibition of the ATP-induced Isc in cultured Sertoli cell epithelia (Fig. 6). Bumetanide added in combination with the other inhibitors also did not affect the Isc response to ATP (Table 1). Removal of extracellular K+ also produced no noticeable effect although Na+ removal significantly reduced the Isc response to ATP. However, the dependence on Na+ may reflect a requirement of Na+ in Ca2+ influx (see below). Ouabain (2 mm), a Na+-K+-ATPase inhibitor, applied basolaterally did not affect the Isc and this may be related to the well-known phenomenon that rat tissues are rather insensitive to ouabain. The present study did not show conclusively how Cl− is taken up into the cell across the basolateral membrane. What is clearer, however, is the exit pathway at the apical membrane. The complete inhibition of the Isc by apical DPC (1 mm) and DIDS (0.25 mm) (Fig. 2) supports the notion that apical anion channels are the pathway for Cl− exit. The identification of an ATP-activated Cl− conductance in cultured rat Sertoli cells lends further support to this contention (Fig. 7).
It is likely that the Isc response of the Sertoli cells to ATP is mediated by an increase in cell Ca2+. The Isc response to ATP is mimicked by thapsigargin (Fig. 8), which increases intracellular Ca2+ (Rossato et al. 1996). Extracellular ATP has been shown to increase intracellular Ca2+ via a G-protein-coupled phosphatidyl inositol pathway in rat Sertoli cells (Filippini et al. 1994; Rudge et al. 1995; Foresta et al. 1995). The EC50 value for ATP-induced increase in Ca2+ (10–15 μm) (Filippini et al. 1994) is in close agreement with the ATP-stimulated Isc (Fig. 3). The present study has therefore ascribed a functional meaning to the ATP-induced Ca2+ response. The potency of stimulation of Isc by ATP and its analogues is in the order UTP ATP > ADP >> AMP = adenosine. This order of potency is also observed for the stimulation of inositol phosphate accumulation in prepubertal rat Sertoli cells (Filippini et al. 1994; Rudge et al. 1995). Current research in the field of P2Y receptors has led to the identification of four receptor subtypes, viz. P2Y1, P2Y2, P2Y4 and P2Y6 with different sensitivities towards the adenine and uridine nucleotides. By matching the known nucleotide selectivity of these receptor subtypes (Nicholas et al. 1996) with the potencies of the nucleotides obtained in our present study, it transpired that there could be a mixture of P2Y receptors, for instance, P2Y2 and/or P2Y4 (with selectivity to UTP and ATP but not to ADP) and P2Y6 receptors (with selectivity to UTP and ADP but not to ATP) to account for the UTP, ATP and ADP effects on anion secretion by the Sertoli cells. However, support for this notion requires more detailed pharmacological studies. As with other anion-secreting epithelia, a capacitative Ca2+ entry process (Putney, 1986) has been identified in rat Sertoli cells (Rossato et al. 1996). Gorczynska & Handelsman (1993) show that Ca2+ influx following depletion of the intracellular Ca2+ store with thapsigargin is dependent on membrane depolarization secondary to Na+ influx. This may have a bearing on the observed reduction of the Isc response to ATP, where in the absence of extracellular sodium, the plateau phase of the response was apparently affected to a greater extent than the peak response (Fig. 5). The former is thought to be associated with Ca2+ influx.
The physiological significance of the effect of apically applied ATP on transepithelial ion transport remains to be elucidated. High-affinity purine receptors that bind adenosine and ATP are identified in the rat Sertoli cells (Monaco et al. 1984). In the intact seminiferous epithelium, germ cells are held in close association with the Sertoli cells. It is conceivable that ATP released from germ cells interacts with apical purinoceptors in the Sertoli cells to regulate electrolyte and fluid secretion. In this way ATP serves as a paracrine factor that mediates ‘cross-talk’ between the developing germ cells and the Sertoli cells. Such a proposition is not untenable in view of the recent reports that germ cells in vitro activate the phosphatidyl inositol pathway (probably through ATP release) in rat Sertoli cells (Welsh & Ireland, 1992). Furthermore, the detection of cystic fibrosis transmembrane conductance (CFTR) gene expression in germ cells in vivo (Trezise et al. 1993) gives added weight to this hypothesis, for there is now a growing body of evidence that CFTR mediates ATP efflux in a number of cell systems (Rotoli et al. 1996; Cantiello et al. 1997). Alternatively, ATP may be released by the Sertoli cells in the testis and act on the same or adjacent Sertoli cells to control fluid secretion. However, such a proposed autocrine role for ATP in the Sertoli cells is only speculative at this stage.
In conclusion, the present work is the first attempt to explore electrolyte transport by Sertoli cells. It demonstrates an increase in Isc by ATP acting from the apical side of the epithelium. These effects are mediated by an increase in Ca2+. The increase in Isc is likely to be caused by anion secretion, although the process is quite insensitive to inhibitors which are known to affect Cl− and HCO3− secretion in other epithelia. The resistance of the Sertoli cell epithelium to perturbation by pharmacological agents can be construed as an advantage to the seminiferous epithelium, whose major function is to create a favourable environment for sperm production to ensure the perpetuation of the species. The stimulation by ATP of transepithelial transport may be part of the complex mechanism regulating fluid secretion by the mammalian testis.