| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Biological Sciences, Fordham University, Bronx, New York 10458
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
gametogenesis, Sertoli cells, signal transduction, spermatogenesis, testis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Paracrine and endocrine communication pathways play major roles in the seminiferous epithelium. Gap junctions between the different cell types may also establish additional communication pathways that are, in turn, subject to regulation by paracrine and endocrine factors. The occurrence of gap junctions in the seminiferous epithelium has been firmly established by morphological, immunocytochemical, and functional assays (discussed in [5, 6]). Risley et al. [7] and Tan et al. [8] have shown that the predominant connexin (Cx) channel proteins detected in the rat seminiferous epithelium by immunocytochemistry are Cx33 and Cx43 between Sertoli cells. The importance of Cx43 gap junctions to gametogenesis is indicated by the severe depletion of germ cells in prenatal male and female mice lacking the Cx43 gene [9]. Postnatal proliferation of spermatogonia is also impaired in Cx43 null mutants [10]. Insertion of Cx32 or Cx40 coding regions into the Cx43 coding region of Cx43-/- mice restored oogenesis and other deficiencies caused by Cx43 deletion, but spermatogonial amplification and spermatogenesis remained defective [11]. Thus, Cx43 is an essential component of communication pathways supporting early phases of spermatogenesis. The importance of Cx43 gap junctions to regulation of spermatogenesis in adults is suggested by the fact that Sertoli cell Cx43 assembly varies during the maturation and cycle of the seminiferous epithelium [7, 8, 12–14], and Cx43 immunoreactivity is reduced in spermatogenesis-deficient mutants [13, 14]. Cx33 assembly also varies with the maturation and cycle of the epithelium [8], but this connexin does not form channels in heterologous cell assays and may act as an inhibitor of channel assembly and function [15].
Other connexin proteins have been detected by immunocytochemistry in the Sertoli cell and germ cell membranes and intracellular compartments [7, 16, 17]. Risley [6] recently reported that at least 12 connexin genes are transcribed and translated in the rat seminiferous epithelium. Thus, a highly complex array of connexin proteins is produced in the seminiferous epithelium. Gap junction channel permeability and regulation is connexin dependent, and structurally and functionally diverse channel types may be assembled depending on connexin expression and compatibility [18]. Dye coupling assays have revealed complex asymmetric and dye-dependent communication pathways in brain [19] and retinal [20, 21] cells. These observations led us to hypothesize that structurally and functionally diverse gap junctions are assembled in the seminiferous epithelium to establish unique communication pathways between the different cell types [6–8].
In situ electrophysiological measurements [22, 23] and dye coupling [5, 7, 13] studies with lucifer yellow have shown that rodent Sertoli cells are coupled by gap junctions, and the coupling may be stage dependent [5] and regulated by second messengers [23]. The functional status of Sertoli-germ cell junctions is much less clear. Although Sertoli-germ cell dye coupling has been detected using lucifer yellow [5, 13], Enders [5] reported that it is not common. Technical problems may have prevented sensitive detection of the lucifer yellow tracer in germ cells, or Sertoli-germ cell junctions may be functionally different from Sertoli-Sertoli junctions, with different permeability properties. We analyzed gap junction coupling in the seminiferous epithelium using a variety of tracers that differ in their physicochemical properties to determine whether there are unique coupling pathways and to better characterize the functional states of these gap junctions.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Sprague-Dawley rats (Zivic-Miller, Pittsburgh, PA) were maintained on a 12L:12D cycle. Animal care and use was approved by the Institutional Animal Care and Use Committee and was consistent with the Guide for Care and Use of Laboratory Animals. Rats were killed by CO2 asphyxiation, and testes were immediately dissected, decapsulated, and processed for dye coupling or to prepare cell cultures.
In Situ Dye Coupling in Sertoli and Germ Cells
Fluorescent tracers of different charges and relative masses were used to study differential gap junction permeabilties. Dextran (0 charge, 10 000 Mr) coupled to fluorescein isothiocyanate (FITC) or tetrarhodamine isothiocyanate (TRITC) was used to identify bulk-loaded cells and nonjunctional dye transfer. Gap junctional permeability was studied with lucifer yellow (-2 charge, 457 Mr), neurobiotin (+1, 287 Mr; Vector Laboratories, Burlingame, CA), biotin cadaverine (+1, 328 Mr), and biotin-X-cadaverine (+1, 556 Mr). Unless otherwise noted, all tracers were purchased from Molecular Probes (Eugene, OR). Dextran and lucifer yellow were detected directly by fluorescence microscopy, but biotin tracers required indirect detection.
A modification of the chop-loading procedure [5] was used to examine coupling in seminiferous tubules, as previously described [7]. Adult rat testes were decapsulated in Hanks balanced salt solution (HBSS) at 32–33°C, blotted with absorbent paper, and cut into three fragments. Each testis fragment was added to 200 µl of PBS containing fluorescent dyes and quickly cut with scissors four times. The fragments were immediately transferred to 10 ml warm PBS for 10 min followed by two or three washes (10 ml, 5 min each) with ice cold PBS to remove excess dye. Testis fragments were fixed immediately with 2–5 ml cold 4% paraformaldehyde (PFA) in PBS for 2 h. Alternatively, rinsed, dye-loaded fragments were incubated in a shaker bath with 10 ml 0.05% collagenase (type 1A; Sigma Chemical Co., St. Louis, MO) in PBS for 5–10 min at 32–33°C, with gentle mixing to partially release separate tubules. The tubules were settled, rinsed with cold PBS, and then fixed with cold 4% PFA/PBS. This procedure facilitated microscopic analyses of individual tubules and indirect detection of biotin conjugates.
Tubules loaded with lucifer yellow and dextran only were observed after rinsing with PBS. Fixed fragments or tubules loaded with biotin tracers were washed twice (5 min each) with 10 ml ice cold PBS containing 0.05% Tween-20 and permeabilized by incubation for 30 min in 10 ml ice cold PBS containing 0.5% Tween-20. Tubules were then incubated 1 h in blocking solution (PBS containing 0.05% Tween-20, 1% biotin-free BSA; Vector Laboratories). This solution was then replaced with blocking solution containing 1:1500 to 1:3000 dilutions of Extravidin-TRITC (Sigma), and tubules were incubated for 2 h. Tubules were then rinsed twice with cold PBS containing 0.05% Tween-20 and resuspended in cold PBS for microscopy. Nuclei in tubules were also stained with 5 µg/ml Hoechst 33258 or 4',6'-diamidino-2-phenylindole to assist in cell identification.
Dissociated tubules or tubules teased away from fixed fragments were placed on glass slides. To minimize distortion, tubules were surrounded with a small amount of vaseline prior to adding coverslips. An Optiphot microscope (Nikon, Tokyo, Japan) equipped with fluorescence and differential interference contrast optics was employed to identify dye-loaded cell types. Dye coupling was indicated by the presence of cells containing only a low molecular weight tracer adjacent to cells containing both the dextran tracer and the low molecular weight tracer. All data were recorded as instances of particular types of coupling (e.g., Sertoli-Sertoli or Sertoli-germ cell). Total numbers of coupled cells were not recorded. Selective dye coupling was studied in tubules coloaded with lucifer yellow and a biotin tracer and was identified as the occurrence of cells with only one of the tracers adjacent to cells with both tracers. The cell types were characterized as type A spermatogonia, type B spermatogonia, early spermatocytes, adluminal spermatocytes, round spermatids, or elongate spermatids based predominantly on their basal or adluminal location, arrangement as single cells, pairs, chains, or clusters, and cell size. Higher resolution identifications were not attempted. All data are presented as means of the different types of coupling observed in each of multiple experiments. The significance of the difference between means was evaluated by t-test (Sigma Stat; Jandel Scientific, Sausalito, CA).
Photomicrographs were made using Ektachrome color slide film (Kodak, Rochester, NY). The slides were scanned into Adobe Photoshop (Adobe Systems, San Jose, CA) with a Microtek Scanmaker 5 (Hsinchu, Taiwan) and processed to create image overlays and plates.
Dye Coupling in Sertoli-Germ Cell Cocultures
Seminiferous tubules were obtained from decapsulated testes of 20- to 30-day-old rats, as described previously [6]. Testes were dissociated for 10–20 min at 32–33°C (5–10 ml/testis) in HBSS lacking calcium and magnesium (HBSS-CMF) and containing 0.1% collagenase type 1A. Free tubules were settled, supernatants was discarded, and tubules were poured into a Petri dish. A dissection microscope was used to identify clumps of undissociated interstitium, which were manually removed. The tubules were settled and further dissociated in a shaker bath at 32–33°C using HBSS-CMF containing 0.1% BSA, 0.05% collagenase, 0.05% hyaluronidase (Sigma), and 50 µg/ml DNase I (Sigma). Tubules were settled and resuspended in this solution with trituration two or three times at 10-min intervals to reduce tubules to Sertoli-germ cell aggregates and remove peritubular cells (as assessed by repeated microscopic analyses). The final pool of aggregates (with approximately 5–50 cells/aggregate) was settled and resuspended in Dulbecco modified Eagle medium (DMEM)-F12 supplemented with 15 mM Hepes and 5% heat-inactivated, dialyzed fetal calf serum for 10 min. After a final settling, the aggregates were suspended in Hepes-supplemented DMEM-F12 containing 0.01% BSA, 5 µg/ml insulin, 5 µg/ml transferrin, 10-7 M 
-tocopherol, 50 ng/ml ovine FSH (NHPP; Dr. A.F. Parlow, NIDDK), and an antibiotic/antimycotic (Gibco BRL, Grand Island, NY). Culture media and supplements were obtained from Sigma unless otherwise noted.
Cell aggregates were cultured in 12- and 24-well plates coated with Matrigel (35 µl/22-mm well; Collaborative Biomedical Products, Bedford, MA). Medium was added to each well (2 ml for 12-well plates and 1 ml for 24-well plates), and a concentrated suspension of aggregates in medium was added to cover approximately half of the surface of each well (as evaluated with an inverted microscope). The cultures were maintained at 3% CO2 and 33°C in a humidified incubator. The medium was changed on Days 3 and 5.
Dyes were bulk loaded into Sertoli and germ cells by a modification of the scrape-loading technique [24]. On occasion, 20 µM 18--glycyrrhetinic acid (Sigma), an inhibitor of gap junction coupling [25], was added to wells from a 20 mM stock in dimethyl sulfoxide for 15 min prior to scrape loading. Medium was removed from a well, the well was gently and rapidly rinsed with 1 ml warm PBS, and dye in 33°C PBS (100 µl/22-mm well) was added. Fragments (5 mm) broken from the cutting edge of size 10 scalpels were lightly and carefully touched to the cell monolayer in multiple places to bulk load dyes. The dye was removed, the well was rinsed twice with warm HBSS, and the medium was added back to each well for 15 min. Each well was then rinsed two or three times with ice cold HBSS and fixed for 30 min with ice cold 2% PFA in PBS and then 2 h with 4% PFA in PBS. Fixative was removed, wells were rinsed with cold PBS, and cells were permeabilized by incubation with cold 0.5% Tween-20 in PBS for 10 min. Cells were then incubated with blocking solution (0.5 ml/well) and then with blocking solution containing either extravidin-TRITC or streptavidin-TRITC (2–40 µg/ml; Molecular Probes) for 2–3 h. After staining, wells were rinsed two or three times with cold PBS containing 0.05% Tween-20. Rinsing was monitored by microscope to assure that dyes were removed from the Matrigel layer. Solutions were changed with a minimum of turbulence to avoid detachment of cells from cut areas. All wells were examined and photographed with a Nikon Diaphot inverted microscope equipped with fluorescence and Hoffman modulation optics. Slides were processed as described above.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
LY coupling was most often seen between Sertoli cells (Fig. 1, top row). About 59% of clusters of bulk-loaded Sertoli cells exhibited LY coupling to neighboring cells (Fig. 2). Thus, about 41% of patches of bulk-loaded Sertoli cells lacked evidence of LY coupling. Diffusion of LY was rarely detected beyond one or two Sertoli cells attached in series to a bulk-loaded Sertoli cell. Of all clusters of bulk-loaded Sertoli cells, 35.7% exhibited Sertoli-Sertoli coupling (Fig. 1, row 1; Fig. 2, S>S), whereas only 20.4% showed Sertoli-germ cell coupling (Fig. 1, row 2; Fig. 2, S>GC). It was uncommon (<4%) to encounter Sertoli cells coupled to both germ cells and other Sertoli cells. Most of the germ cells coupled to Sertoli cells were late spermatogonia or basal spermatocytes. Sertoli coupling to adluminal spermatocytes was rare, and coupling to spermatids was not detected. LY in bulk-loaded germ cells was never observed to diffuse into Sertoli cells (Fig. 2, GC>S).
|
Biotin cadaverine (BC) dye transfer relative to D-FITC was studied next. The BC tracer was chosen because it has a positive charge compared with the -2 charge on LY and is smaller than LY. Thus, BC may pass through gap junctions that LY does not pass through. Similar to LY, Sertoli cells were most often bulk-loaded with D and BC (Fig. 1, row 3). The frequencies of total BC coupling (74%) and Sertoli-Sertoli coupling (35.6%) in bulk-loaded clusters of Sertoli cells were not significantly different from those for LY coupling (Fig. 2). In contrast, BC coupling between Sertoli and germ cells occurred in 64.3% of patches of BC-loaded Sertoli cells, a frequency about 3-fold higher than that for LY and nearly 2-fold greater than the BC S>S coupling frequency. The frequency of diffusion of BC from Sertoli to Sertoli and germ cells concurrently (S>S and GC) was almost 9-fold greater than that for LY. Sertoli cells were coupled to chains of spermatogonia (see Fig. 1, BC) and to basal spermatocytes. Adluminal spermatocyte and spermatid coupling to Sertoli cells was not seen. Also, unlike LY, BC showed weak coupling (faint fluorescence) from about 40% of chains or clusters of bulk-loaded spermatogonia to neighboring Sertoli cells (Fig. 2). Spermatogonia that received BC from bulk-loaded Sertoli cells, however, were not observed to pass the BC on to other Sertoli cells (Fig. 1, row 3, arrow), which suggests that the diffusion rate for BC from Sertoli cells to spermatogonia is greater than it is in the opposite direction.
Neurobiotin (NB) coupling relative to D-FITC was also examined to determine whether this relatively small tracer exhibited greater Sertoli to germ cell coupling than did LY or BC. Again, bulk-loaded cells were mostly Sertoli cells (Fig. 1, row 4). NB diffused extensively from bulk-loaded Sertoli cells to both Sertoli and germ cells, making it difficult to characterize and quantify. Nevertheless, the extent of coupling was clearly greater with NB than with LY or BC.
These data suggested that gap junctions between Sertoli cells and spermatogonia were more permeable to biotinylated tracers than to LY. To directly examine this possibility, LY was bulk loaded with a biotin tracer to determine whether dye-selective pathways of coupling were present in the seminiferous epithelium. Figure 3 (LY and BC) clearly shows that LY and BC bulk loaded into Sertoli cells diffuse to different cell types. The LY image shows predominantly Sertoli cells with their long basal-apical axis oriented oblique to the transverse axis of the tubule. The BC and merge images show BC in most of the Sertoli cells with LY but also in neighboring chains of type A spermatogonia that lack detectable LY. Selective transfer of any dye was detected in 58.8% of bulk-loaded cell clusters, with LY and BC each accounting for about half of this value (Fig. 4). However, when Sertoli cells adjacent to bulk-loaded cells contained only one of the two tracers, it was LY in 84.5% of these instances (Fig. 4, S>S). When spermatogonia next to bulk-loaded Sertoli cells contained only one of the two tracers, it was BC in 78.2% of such occurrences (Fig. 4, S>GC). The relatively high rate of BC selectivity in Sertoli-germ cell coupling was consistent with the relative frequencies of Sertoli-germ cell coupling observed for the two dyes independently (Fig. 2). LY was also bulk loaded together with the largest tracer studied, biotin-X-cadaverine (BXC). The LY image shows fluorescent Sertoli cells observed from their basal surfaces (Fig. 3, LY). Similar to the smaller BC tracer, BXC appeared in LY-positive Sertoli cells and in adjacent LY-negative spermatogonia. Thus, the charge and possibly the shape of the tracer appears to be more important to selectivity than the size.
|
|
We also attempted to determine whether NB showed selective coupling to germ cells by studying coupling in Sertoli-germ cell cocultures where microscopic analysis would be less complex than in intact tubules. Cell cultures were established from Sertoli-germ cell aggregates to minimize disruption of the normal cell-cell contacts. Six-day cultures from rats 20 days old were scrape loaded with LY and NB. Both LY and NB spread from the cut cells through the monolayer of Sertoli cells well removed from the cut (Fig. 5). LY concentrated in nuclei, but NB was spread throughout the Sertoli cells. LY and NB did not diffuse equally to germ cells, however. Away from the cut edge, several chains of spermatogonia containing predominantly NB were routinely detected. LY and NB were both readily detected in bulk-loaded spermatogonia (Fig. 5, upper arrow). Pachytene spermatocytes present as single cells in the cultures only occasionally showed evidence of NB coupling (not shown).
|
Incubations of cell cultures with 18--glycyrrhetinic acid, an inhibitor of gap junction coupling [25], completely blocked diffusion of both LY and NB from bulk-loaded cells (Fig. 6) and eliminated the occurrence of chains of germ cells with NB only. In separate experiments, where NB or LY were coloaded with D, D remained restricted to the cells along the cut line of the monolayer, and streptavidin fluorescence in cells was always strictly dependent upon the presence of NB during cut loading (data not shown). These results indicate that the selective diffusion of NB from Sertoli cells to spermatogonia was mediated by gap junctions.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In the rat, Sertoli-Sertoli gap junctions consist primarily of Cx33 and Cx43 [8] and a very low level of Cx32 detectable only after Triton X-100 extraction [17]. The prepubertal period is accompanied by a large increase in Cx33 expression and assembly in Sertoli cells and meiotic cells [6, 8]. Chang et al. [15] suggested that Cx33 may function as an inhibitor of coupling in the seminiferous epithelium. Cx33 may be responsible for the reduced junctional permeability of adult Sertoli cells relative to immature Sertoli cells [22]. Cx33 and Cx43 immunocytochemistry results indicate that Sertoli-Sertoli junctions vary in abundance during the cycle of the rat seminiferous epithelium [8]. Dye coupling experiments by Enders [5] suggested that Sertoli-Sertoli junctional communication in the mouse is also cycle dependent. We noted Sertoli cell coupling in most tubules from the rat, suggesting that coupling occurs at most epithelial stages. In our procedure, tubule cells were dye loaded immediately after decapsulation of the testes, whereas Enders [5] dye loaded tubules after they were collagenase dissociated. Dissociated tubules were also used by Eubesi et al. [22] for electrical coupling studies. However, uncoupling may result from artifactual channel closure during the chopping and during the dissociation procedure, and the sensitivity to this artifact may be stage dependent.
Other types of gap junctions occur between Sertoli cells and germ cells. Junctions between some late spermatogonial or basal spermatocyte stages (uncharacterized) and Sertoli cells were permeable to both LY and BC, similar to Sertoli-Sertoli junctions. However, there was also a category of Sertoli-spermatogonial junctions that showed significantly higher permeability to the positively charged biotin tracers in situ and in cell cultures. This conclusion is based on the fact that BC coupling between Sertoli and all germ cell types occurred at three times the frequency as did negatively charged LY coupling, and coupling from Sertoli cells to both Sertoli and germ cells concurrently was about 9-fold higher with BC than with LY. When LY was coloaded with either larger (BXC) or smaller (BC, NB) biotin tracers, the biotin tracers were found alone in spermatogonia adjacent to Sertoli cells with both LY and a biotin tracer. Although not quantified, NB diffused from Sertoli cells to many more germ cells and Sertoli cells than did the other biotin tracers. Thus, Sertoli-early germ cell coupling may be relatively extensive with low molecular weight tracers (<287 Mr), whereas subsets of Sertoli-spermatogonial junctions are cation selective for larger tracers (>328 Mr).
The types of coupling encountered with LY and BC suggest the existence of alternative coupling pathways in addition to selective coupling pathways. Coupling from Sertoli to Sertoli only, Sertoli to germ cell only, or absence of coupling occurred with both BC and LY. These findings suggests alternative states of regulated channel gating. Independent closure of Sertoli-Sertoli junctions would favor molecular diffusion through the less abundant channels between Sertoli and germ cells. Sertoli-Sertoli coupling in vitro can be regulated by cyclic nucleotides [23], FSH [26, 27], and androgens [27, 28]. However, artifactual induction of channel closing cannot be ruled out as a contributing factor to the appearance of alternative functional states of gap junctions, but it is difficult to envision how artifactual closure would have different impacts on different junction types.
In addition to selective permeability, the Sertoli-germ cell junctions appeared to have asymmetric permeability, a phenomenon first observed between heterotypic cell pairs in culture [29]. LY in germ cells was never observed to pass to Sertoli cells (also see [13]). BC only weakly passed from spermatogonia to Sertoli cells but passed readily from Sertoli cells to spermatogonia. This phenomenon is similar to the asymmetric coupling of LY and NB in the retina [20, 21]. In the rat retina, junctions between astrocytes are permeable to LY and NB [21]. Astrocyte-Müller cell junctions are permeable to NB but not LY, and the permeability is unidirectional; neither LY nor NB could pass from Müller cells to astrocytes. The asymmetric permeability of junctions between Sertoli and germ cells may ensure that physiologically important molecules diffusing through Sertoli cell junctions can attain a relatively high concentration in neighboring spermatogonia by minimizing return diffusion rates and serial Sertoli germ cell-Sertoli coupling over long distances.
The selective permeability properties of Sertoli-spermatogonial junctions suggests that the connexins in these junctions may differ from those in Sertoli-Sertoli junctions. The connexins in the spermatogonial hemichannels may be different from those in the Sertoli cell hemichannel to which it is paired. The permeability properties described could result if connexins in the spermatogonial hemichannels restricted passage of negatively charged tracers in either direction. No one has definitively characterized the connexin composition of germ cell junctions. Mok et al. [16] immunolocalized Cx31 in rat spermatogonia, but Cx31 is incompatible with other connexins and does not show LY-NB selectivity [30]. Cx45 is expressed at high levels in rat Sertoli and germ cells [6], it is compatible with Cx43, and it has greater permeability to the cationic NB than to the anionic LY [30]. Cx57 is expressed in rat germ cells [6]. Cx57 is permeable to NB but not LY, and Cx57 is also compatible with Cx43 but not Cx32 or Cx40 in heterotypic junctions [31]. Assembly of Cx57 in spermatogonial hemichannels paired with Cx43 Sertoli cell hemichannels could give the permeability properties we have described here, and Cx32 and Cx40 would not be able to substitute for Cx43 [11]. It will be important to determine which connexins allow for selective communication between Sertoli cells and spermatogonia.
Data suggesting selective dye coupling should be interpreted with caution because two different tracers may not be detected equivalently. When LY and BC were coloaded in tubules, LY was found selectively in some Sertoli cells and germ cells. This finding was probably an artifact because of inadequate exposure of the cells to the secondary biotin-detection reagents. However, it is highly unlikely that artifacts were responsible for the occurrence of cells containing biotin but not LY. LY does not require detection by secondary reagents, and it has a very bright fluorescence because of its high molar extinction coefficient [32]. LY was readily detected together with biotin tracers in the bulk-loaded cells, so biotin detection does not interfere with LY detection. Bright LY fluorescence also occurred in Sertoli cells immediately adjacent to germ cells with only the biotin tracer. Germ cell-Sertoli cell coupling and selective permeability to the NB tracer were eliminated by 18--glycyrrhetinic acid, an inhibitor of junctional coupling [25].
The last category of gap junctions consists of the small gap junctions between Sertoli cells and adluminal spermatocytes, and these may be functionally different from the Sertoli-Sertoli and the Sertoli-spermatogonial junctions. Gap junctions are more abundant on pachytene spermatocytes than spermatogonia [33, 34], and rat pachytene spermatocytes contain at least 12 different connexin mRNAs [6]. However, in this study, dye coupling between Sertoli cells and adluminal germ cells was uncommon, consistent with Ender's observations [5]. We do not believe that this lack of observable dye coupling was due solely to the failure of conventional microscopic techniques to be sensitive to fluorescence from sources in whole tissue, because dyes were readily detected in Sertoli cell extensions surrounding adluminal germ cells and in bulk-loaded adluminal germ cells. We were also unable to detect dye transfer from adluminal germ cell clones to Sertoli cells. Cx33 expressed in meiotic cells, but not spermatogonia [6], may act with Sertoli cell Cx33 to restrict coupling between Sertoli cells and adluminal germ cells, as suggested previously [15]. However, adluminal germ cell clones consist of many relatively large cells connected by cytoplasmic bridges. Dyes entering the large cytoplasmic volumes of adluminal germ cell clones from Sertoli cells will be relatively dilute and more difficult to detect.
Although we were unable to determine whether adluminal germ cell gap junctions restrict dye passage or whether dye diffusion within an adluminal germ cell clone is so rapid and extensive that tracers (and perhaps physiologically important molecules) fail to concentrate sufficiently for detection, either alternative may be biologically significant. Thus, even if adluminal germ cell gap junctions were equally permeable to basal germ cell gap junctions (unlikely), a burst in Sertoli cell production of molecules (including second messengers) would result in an effective concentration of such molecules in the spermatogonia prior to the adluminal spermatocyes.
We demonstrated the occurrence in the rat seminiferous epithelium of multiple gap junction communication pathways, each with different permeability properties and possibly different regulatory properties. Because most second messengers have short half-lives, even small permeability differences could be physiologically significant [30] and could be important in regulating gene expression in spermatogenic cells and Sertoli cells. Gap junctional communication should be considered an important component in the network of cell-cell communications supporting and regulating spermatogenesis.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 This material is based upon work supported by the National Science Foundation under grant IBN-9722987. 
2 Correspondence. FAX: 718 817 3645; risley{at}fordham.edu
3 Current address: Department of Biology, New York University, 1009 Main Bldg., 100 Washington Square East, New York, NY 10003
4 Current address: Department of Pharmacology, Weill-Cornell University Graduate School of Medical Sciences, New York, NY 10021
Accepted: April 10, 2002.
Received: September 18, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Brehm, M. Zeiler, C. Ruttinger, K. Herde, M. Kibschull, E. Winterhager, K. Willecke, F. Guillou, C. Lecureuil, K. Steger, et al. A Sertoli Cell-Specific Knockout of Connexin43 Prevents Initiation of Spermatogenesis Am. J. Pathol., July 1, 2007; 171(1): 19 - 31. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. de Montgolfier, J. Dufresne, M. Letourneau, J. J. Nagler, A. Fournier, C. Audet, and D. G. Cyr The Expression of Multiple Connexins Throughout Spermatogenesis in the Rainbow Trout Testis Suggests a Role for Complex Intercellular Communication Biol Reprod, January 1, 2007; 76(1): 2 - 8. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. N Kahiri, M W. Khalil, F. Tekpetey, and G. M Kidder Leydig cell function in mice lacking connexin43. Reproduction, October 1, 2006; 132(4): 607 - 616. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Fiorini, X. Decrouy, N. Defamie, D. Segretain, and G. Pointis Opposite regulation of connexin33 and connexin43 by LPS and IL-1{alpha} in spermatogenesis Am J Physiol Cell Physiol, March 1, 2006; 290(3): C733 - C740. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Fischer, R. Brehm, L. Konrad, S. Hartmann, S. Kliesch, R. M. Bohle, and M. Bergmann Connexin 33: A Rodent-Specific Member of the Gap Junction Protein Family? J Androl, January 1, 2005; 26(1): 75 - 84. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. W. White Nonredundant Gap Junction Functions Physiology, June 1, 2003; 18(3): 95 - 99. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
