| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Minireview |
a Population Council, Center for Biomedical Research, New York, New York 10021
b Department of Zoology, The University of Hong Kong, Hong Kong, China
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONFUNCTIONS OF THE TJ...STRUCTURE AND MOLECULAR...MODELS FOR STUDYING TJ...REGULATION OF TJ DYNAMICSCONCLUSIONS AND FUTURE...REFERENCES |
|---|
Sertoli cells, signal transduction, spermatogenesis, testis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONFUNCTIONS OF THE TJ...STRUCTURE AND MOLECULAR...MODELS FOR STUDYING TJ...REGULATION OF TJ DYNAMICSCONCLUSIONS AND FUTURE...REFERENCES |
|---|
|
Based on the results of studies conducted with other epithelia over the past decade, such as Madin-Darby Canine Kidney (MDCK) cells and keratinocytes, it is increasingly clear that junction dynamics are regulated largely by TJ-integral membrane proteins, and by their associated signaling molecules at the junction via protein phosphorylation/dephosphorylation (for reviews, see [8, 9]). Although TJ structures in the seminiferous epithelium possess some features unique to the testis, they are similar to those found in other epithelia (for reviews, see [1, 5–7]). As such, it has been postulated that junction dynamics in the seminiferous epithelium are regulated by mechanisms similar to those in other epithelia [4, 5]. It is ironic that the study of junction dynamics in the testis is a formidable task, largely because of the complexity of spermatogenesis, which is intriguingly regulated by different molecules and pathways at the cellular, biochemical, and molecular levels. Also, these regulatory events likely occur concurrently in the seminiferous epithelium at each stage of the cycle.
Although the structure and function of the Sertoli cell TJ that constitutes the BTB in the testis have been reviewed (see [4–7]), the molecules and pathways that regulate TJ dynamics in the testis have remained largely obscure until recently. This minireview presents the results of recent findings in this area, and in particular, the current status of the research. Furthermore, we attempted to highlight specific research areas that deserve attention in future studies. The minireview is not intended to be exhaustive, rather, it is meant to serve as a guide for future studies. Readers are strongly encouraged to read earlier reviews (see [4–7]) to gain a more comprehensive view of this field.
|
FUNCTIONS OF THE TJ AND THE CURRENT THEORY OF TJ REGULATION IN THE TESTIS |
|---|
|
TOPABSTRACTINTRODUCTIONFUNCTIONS OF THE TJ...STRUCTURE AND MOLECULAR...MODELS FOR STUDYING TJ...REGULATION OF TJ DYNAMICSCONCLUSIONS AND FUTURE...REFERENCES |
|---|
Tight junctions are the only known occluding junction types in mammalian epithelia and endothelia. They are located at the most apical region of epithelia and endothelia, they seal intercellular spaces, and contribute to the permeability barrier across an epithelium or endothelium [10, 11]. Behind TJs are cell-cell, actin-based adherens junctions (AJs) and desmosomes (cell-cell, intermediate, filament-based anchoring junctions). Altogether, these structures are referred to as the junctional complex in epithelia (for a review, see [12]). Yet in the testis, the relative location of TJs, AJs, and desmosomes is in sharp contrast to their location found in other epithelia. For instance, Sertoli cell TJs are adjacent to the basement membrane (a modified form of extracellular matrix; ECM) [2] and are present alongside AJs and desmosomes (for reviews, see [1, 4, 6]). Such morphological intimacy between TJs and the ECM has prompted us to postulate that the ECM may affect Sertoli cell TJ dynamics. Indeed, recently completed studies in this laboratory have illustrated the pivotal role of ECM proteins, such as collagen, in regulating TJ dynamics [13]. For instance, the use of an anticollagen antibody that perturbed the function of collagen can reversibly disrupt the Sertoli cell TJ barrier in vitro [13]. TJs also segregate a cell into two physiological compartments conferring to cell polarity, and they create a barrier that selectively permits the transepithelial flux of molecules and ions across the intercellular spaces between epithelial cells via the paracellular pathway (for reviews, see [14–17]). For instance, Sertoli cell TJs that create the BTB divide the seminiferous epithelium into the basal and the adluminal compartments (see Fig. 1) (for a review, see [1]). Moreover, TJs also act as a major site and platform for vesicle trafficking and signal transduction [18].
Current Theory of TJ Regulation in the Testis
Two theories are currently found in the literature that attempt to explain the regulation of TJ dynamics in the testis, and which are based largely on results of morphological studies. Because both theories were recently reviewed by Pelletier [5, 7], only a brief summary is presented here. The "zipper" theory proposes that TJs at the basal domain of Sertoli cells break down to accommodate the passage of preleptotene and leptotene spermatocytes across the BTB, while new TJ fibrils are formed under the migrating preleptotene and leptotene spermatocyes [5, 7]. The "repetitive removal of membrane segments" theory proposed by Pelletier, Byers, and colleagues [4] suggests that the upward movement of preleptotene and leptotene spermatocytes creates stress against Sertoli cell TJs, resulting in changes in the orientation, disintegration, and proliferation of TJ fibrils. Yet both postulates do not take into account recent biochemical and molecular findings on TJ-integral membrane proteins, such as claudins, occludins, and junctional adhesion molecules (JAMs), and TJ-associated signaling molecules such as cytokines and protein kinases. Needless to say, these theories will be updated when we learn the precise role in the regulation of Sertoli cell TJs of other players such as cytokines, phosphatases, and kinases.
|
STRUCTURE AND MOLECULAR COMPOSITION OF TJs |
|---|
|
TOPABSTRACTINTRODUCTIONFUNCTIONS OF THE TJ...STRUCTURE AND MOLECULAR...MODELS FOR STUDYING TJ...REGULATION OF TJ DYNAMICSCONCLUSIONS AND FUTURE...REFERENCES |
|---|
|
Occludins
Occludin, a 65 kDa protein, was the first TJ-integral membrane protein identified in multiple epithelia [27, 28], including the rat testis [29]. Until now, no other occludin-related gene has been identified in any species, except for a splicing variant of occludin, designated occludin 1B. Occludin 1B is found in most tissues, including the testis, but it is not found in the spleen [30]. Each occludin molecule consists of four transmembrane domains with a long carboxyl-terminus and a short amino-terminus in the cytoplasm, one intracellular loop, and two extracellular loops (for reviews, see [31, 32]) (Fig. 1). The first extracellular domain is involved in cell-cell coupling, which confers the cell adhesion function [28]. The second extracellular loop apparently is needed for the assembly and sealing of TJs, thereby conferring the TJ functionality [33, 34].
Several studies have demonstrated the unique function of occludin in TJs. For instance, overexpression of chicken occludin in MDCK cells can elevate the transepithelial electrical resistance (TER) across the cell epithelium, making the TJ barrier tighter [35]. Addition of a synthetic peptide corresponding to the second extracellular loop of occludin can down-regulate TER in cultured Xenopus epithelial cells [33] and rat Sertoli cells [34]. Furthermore, intratesticular administration of a 22-amino acid synthetic peptide corresponding to the outermost region of this loop can reversibly arrest spermatogenesis, thereby perturbing BTB function, which in turn disrupts the underlying AJs and desmosomes, leading to germ cell loss from the seminiferous epithelium in vivo [34].
Recent findings, however, have shown that the structure and function of TJs cannot be maintained by occludin alone. For instance, occludin is found in TJ strands in most epithelial cells, yet Sertoli cells in the testes of humans and guinea pigs [29], and endothelial cells in non-neuronal tissues [36] do not have occludin. Furthermore, in occludin-/- mice, TJs do not appear to be morphologically affected, and the barrier function of the intestinal epithelium appears normal when examined electrophysiologically [37]. Although the testes and the seminiferous tubules of occludin-deficient mice appear to be normal with developing germ cells observed in early postnatal development, tubules show typical atrophy and are devoid of germ cells in adult mice [37]. These results seemingly suggest that occludin is critical to the BTB function, which is consistent with an earlier report illustrating that an intratesticular injection of a 22-amino acid peptide corresponding to the second extracellular loop of occludin could reversibly disrupt the BTB as demonstrated by micropuncture analysis [34]. For instance, the BTB became leaky to [125I]-BSA when a tracer was administered i.p. via the jugular vein within 2 wk after a single intratesticular administration of occludin peptide, which persisted for 6 wk [34]. Also, this BTB damage was accompanied by a gradual loss of germ cells from the seminiferous epithelium, and the tubules were virtually devoid of spermatids by day 27 after peptide treatment [34]. And by 68 days after occludin peptide treatment, normal spermatogenesis was found in all tubules and the BTB was no longer leaky [34]. Furthermore, the assembly of Sertoli cell TJ-barrier and its reassembly in vitro after CdCl2-induced disruption was shown to associate with a significant induction in occludin [38]. Taken collectively, these data illustrate the pivotal role of occludin in BTB function and spermatogenesis.
Although the mechanism or mechanisms that mediate the peptide-induced germ cell loss are not known, these results tempt us to speculate that damage to the BTB by perturbing the occludin function can lead to a loss of the cell adhesion function. Such a postulate is not without precedence. For instance, it is known that cross-talk exists between TJs and the cadherin/catenin-based AJ structures, and that a disruption of AJs can indeed alter TJ functionality [39]. Furthermore, other histological abnormalities are also found in occludin-/- mice [37]. These include hyperplasia of the gastric epithelium, lack of cytoplasmic granules in striated duct cells of the salivary glands, accumulation of calcium deposits in the brain, and thinning of the compact bone [37]. These phenotypic changes in occludin-/- mice suggest that occludin may have other yet-to-be-identified functions in addition to its structural role in TJs.
A recent report has also shown that the expression of occludin by Sertoli cells cultured in vitro can be stimulated by testosterone [38], and the presence of testosterone can even protect the Sertoli cell TJ-barrier from the disruptive effects of CdCl2 [38]. Recent studies in our laboratory have demonstrated that transforming growth factor ß3 (TGFß3) could effectively suppress the expression of occludin in Sertoli cell cultures, thus perturbing the Sertoli cell-TJ permeability barrier [40]. Furthermore, this TGF-ß-induced disruption of the TJ barrier is mediated via the p38 mitogen-activated protein (MAP) kinase pathway [41]. These results collectively suggest that the dynamics of Sertoli cell TJs in the testis are regulated at least in part by cytokines in order to permit preleptotene and leptotene spermatocytes to translocate across the blood-testis barrier at stages VIII–IX [3] of the epithelial cycle during spermatogenesis. It is likely that germ cells are the major source of cytokines that regulate the timely opening and closing of the BTB.
Claudins
Claudin-1 and claudin-2, two members of the claudin protein family, were initially identified in chicken liver [20]. Similar to occludins, claudins are also TJ integral membrane proteins (
23 kDa) found at the sites of TJs in both epithelia and endothelia. However, claudins do not show any sequence similarity to occludin. Each claudin molecule consists of four transmembrane domains with both NH2 and COOH termini present in the cytoplasm (Fig. 1). The amino acid sequences of the first and fourth transmembrane segments and the extracellular loops are highly conserved among different claudins. Claudins are composed of a large gene family with more than 20 members [42–47], the sequence identities of which range from 12.5% to 69.7% (for a review, see [32]). Overexpression of claudin-1 and claudin-2 in fibroblasts lacking TJs can induce the assembly of TJ strands similar to other epithelial TJ strands (for a review, see [48]), illustrating their significance in TJ assembly. Claudins interact directly with TJ membrane-associated guanylate kinase homologues, ZO-1, ZO-2, ZO-3, and multi-PDZ domain protein 1 (MUPP1), and they interact indirectly with AF-6 (also known as s-afadin) and cingulin [49, 50]. As such, claudins are important structural and functional components of TJs (for a review, see [47]).
Each mammalian tissue can express a unique combination of different claudin members, but not all [20, 42, 48]. For example, claudin-3 mRNA is detected largely in the lung and liver, whereas claudin-1 and claudin-2 are expressed predominantly in liver and kidney. However, some cell types express their own claudins. For example, claudin-5 is restricted to the endothelial cells of blood vessels [44]. The testis expresses several claudins (e.g., claudins 1, 3, 4, 5, 7, 8, and 11; for a review, see [51]). Claudin-11, or oligodendrocyte-specific protein, is highly expressed in the brain and testis, but it can also be found in the kidney [43]. Claudin-11 is an important TJ building block that constitutes TJ strands between the lamellae of myelin sheaths of oligodendrocytes in the brain and between Sertoli cells in the testis [43, 48]. Among all claudins, claudin-11 is one of the best studied TJ molecules. In claudin-11-/- mice, TJ strands were absent in the myelin sheaths of oligodendrocytes and Sertoli cells [52]. Male claudin-11 null mice are sterile, and their seminiferous tubules are filled with aggregates of nucleated cells and Sertoli cells without normal spermatogenesis [52], illustrating the significance of claudin-11 in spermatogenesis. Recent studies have shown that cytokines such as TGFß3 can down-regulate the expression of claudin-11 in Sertoli cell cultures at the time of TJ assembly, which in turn perturbs the TJ permeability barrier [40]. These results thus suggest claudin-11 plays a crucial role in the formation and maintenance of TJ barrier in the testis. Much work remains to be done to address the significance of claudins in TJ dynamics.
Junctional Adhesion Molecules
Another family of TJ integral membrane proteins is the JAM family. JAMs are localized at the sites of TJs in endothelial and epithelial cells [19]. To date, three members of JAM (JAM 1, 2, and 3) have been identified [19, 53]. Studies with Northern blot analysis have identified the mRNAs encoding JAM-1 and JAM-2 in the testis [53], but it is not known whether JAM-3 is present in the testis. JAMs are members of the immunoglobulin superfamily. Each molecule is composed of an extracellular domain comprised of two immunoglobulin V loops, a single-span transmembrane domain, and a short cytoplasmic tail [19, 54] (Fig. 1). JAMs also associate intracellularly with other TJ-associated proteins such as ZO-1, AF-6, MUPP1, ASIP/PAR-3, and cingulin, suggesting that JAM is a critical component of the multiprotein complex of TJs [19, 50, 55–57]. JAMs are also known to modulate the transmembrane cell migration of neutrophils and monocytes in the endothelium [19, 53]. In view of the physiological significance of JAMs in cell movement, it is ironic that JAMs should be vigorously studied in the testis for the purpose of investigating their role in the migration of preleptotene/leptotene spermatocytes across the BTB during late stage VIII and early stage IX of the epithelial cycle [3].
Zonula Occludens
Several cytosolic proteins were found to associate with the intracellular domain of TJ integral membrane proteins (Fig. 1). The first TJ peripheral protein that was subjected to extensive investigation was ZO-1 (Mr 220 kDa) [58]. Subsequent immunoprecipitation studies had identified two ZO-1 associated proteins of Mr 160 kDa and 130 kDa in MDCK cells, and they were designated ZO-2 and ZO-3, respectively [59–61]. These proteins also belong to the membrane-associated guanylate kinase protein family that share extensive sequence homology at their NH-terminal region. They all contain three PDZ (PSD-95-Discs Large-ZO-1) domains; namely, PDZ1, PDZ2, and PDZ3, an SH3 domain, and one guanylyl kinase-like domain [62]. Among these domains, PDZ domains bind directly to the COOH-terminus of various proteins such as occludin, claudins, and JAMs [19, 49, 61, 63–65]. In the testis, only ZO-1 and ZO-2 have been positively identified to date [21, 22].
|
MODELS FOR STUDYING TJ DYNAMICS IN THE TESTIS |
|---|
|
TOPABSTRACTINTRODUCTIONFUNCTIONS OF THE TJ...STRUCTURE AND MOLECULAR...MODELS FOR STUDYING TJ...REGULATION OF TJ DYNAMICSCONCLUSIONS AND FUTURE...REFERENCES |
|---|
In Vitro Model
Sertoli cells cultured in vitro on Matrigel-coated bicameral units have been used for almost two decades, when this technique was first developed [66–68] for studying the biology and regulation of the TJ permeability barrier. This model was also used in our laboratory to identify several target genes that are implicated in the regulation of Sertoli cell TJ dynamics, such as occludin, claudin-11, ZO-1, 
2-macroglobulin, and others [34, 38, 40, 69–75]. Furthermore, it was shown that TGF-ß3 perturbed the Sertoli cell-TJ permeability barrier in vitro, possibly mediating its effects via its action on the expression of occludin, ZO-1, and claudin-11 [40], using the p38-MAP kinase signaling pathway for downstream signaling [41]. The novelty of this model is as follows. First, Sertoli cells are cultured in chemically defined serum-free medium (e.g., F12/Dulbecco modified Eagle medium) on Matrigel to permit these cells to establish TJs, and to form an epithelium with cellular behavior and morphology similar to those found in vivo, such as polarized secretion of various Sertoli cell products [66–68, 73].
As such, this in vitro culture system can be used to identify the one or more molecules that play a crucial physiological role in regulating TJ dynamics. Indeed, using this model, it was shown that testosterone, cAMP [38, 76, 77], and protease inhibitors [78] are important regulators of Sertoli cell-TJ dynamics. Second, this model can quantitatively assess the dynamics of the TJ barrier when coupled with the use of other techniques such as restricted diffusion of fluorescein isothiocyanate-dextran, [3H]-inulin or [125I]-BSA, polarized secretion of Sertoli cell proteins [66–68, 73, 74], along with or in place of the measurement of TER across the Sertoli cell epithelium [34, 40, 75, 77, 78]. Indeed, this model was used in conjunction with CdCl2 to study Sertoli cell TJ dynamics [38, 77]. The system has also shown that testosterone is one of the crucial regulators of Sertoli cell TJ dynamics [38].
In this regard, it is of interest to note that whereas the inter-Sertoli-TJ barrier is one of the tightest in the mammalian body (for reviews, see [4, 79, 80]), studies with TER using Sertoli cells cultured in vitro have consistently yielded a reading of <100 ohm·cm2 [38, 40, 71, 75, 78, 81] versus >1000 ohm·cm2 in MDCK cells and keratinocytes [82–84]. These data taken collectively seemingly suggest that the TJs found between Sertoli cells are less tight than those present in the kidney and the epidermis. That androgens [38, 77] and protease inhibitors [78] can boost the Sertoli cell-TJ barrier clearly illustrates that many biological factors present in vivo indeed contribute to the integrity and maintenance of the BTB in the testis, making this in vitro model exceedingly useful for studying TJ dynamics and their regulation. For instance, it was shown that the disruption of Sertoli cell-TJ barrier in vitro induced by CdCl2 was associated with a transient reduction in occludin expression, which was accompanied by a surge in urokinase-type plasminogen activator (uPA; a serine protease known to be produced by Sertoli cells), suggesting that the cleavage of TJs require the involvement of proteases [38]. These studies also reveal many targets (e.g. protease and protease inhibitors) that can be tackled to compromise the BTB function and to perturb spermatogenesis, forming the basis of novel and new approaches for male contraceptive development.
In Vivo Models
Several in vivo models are currently available for the study of BTB dynamics in the testis, many of which were not intended for studying junction dynamics, such as the aging and the cryptorchidism/orchitis models. However, subsequent studies have demonstrated that these conditions are associated with changes in the BTB function. These models should aid future studies in understanding the biology and regulation of the BTB.
The CdCl2 model Cadmium toxicity and cadmium-induced damage to the BTB in the testis and TJs in vascular endothelium have been known since the early 1900s [85]. Subsequent studies have shown that cadmium causes damage to Sertoli TJs that constitute the BTB in vivo [79, 86] and to the Sertoli cell-TJ barrier in vitro [38, 77]. At low doses, cadmium can cause failure of spermiation in rats [87]. It remains to be investigated whether this relates to a disruption of the dynamics of the actin filament, the intermediate filament, or the microtubule network, such as their reorganization and polymerization. The drawback of this model is that the cadmium-induced damage to the BTB is irreversible [86], making it difficult to assess changes in target genes during BTB recovery. Immunofluorescence confocal microscopy studies have shown that within 24 h after an administration of CdCl2 at 1 mg/kg of body weight, disorganization of the TJ-associated microfilaments in Sertoli cells occurs primarily at stages VIII through II–III tubules [86]. Yet the Sertoli cell microfilament bundles in several stages prior to VIII, and peritubular myoid cells are not affected by CdCl2 treatment [86]. These results thus suggest that the microfilament bundles in Sertoli cells are the primary target of CdCl2, but only at stage VIII–III tubules. Collectively, these results clearly show that this is a useful model for studying BTB dynamics, and in particular, the role of TJ-associated microfilaments in TJ function [38, 74, 86]. Because E-cadherin (a putative AJ integral membrane protein) is the apparent primary target of CdCl2 in other epithelial cells in vitro (for a review, see [111]), the CdCl2-induced damage to the BTB may be secondary to the disruption of AJs. As such, this model is extremely helpful for investigating the intriguing physiological and functional relationship between TJs and AJs in the testis, because in particular, cross-talk exists between TJs and AJs [39].
The glycerol model When glycerol is applied intratesticularly, it causes long-term cessation of spermatogenesis in rats, rabbits, and squirrel monkeys without having any apparent effects on Leydig cell steroidogenesis, on levels of serum FSH, LH, and testosterone; and on secondary sexual characteristics [88–92]. This effect is mediated by the glycerol-induced structural damage to the BTB [88]. Indeed, recent studies with fluorescence and confocal microscopy have shown that the networks of occludin, actin filaments, and microtubules in the seminiferous epithelium in glycerol-treated rats are damaged, suggesting that these molecules may be the putative targets of glycerol [93]. Yet the signaling pathway or pathways by which glycerol affects the occludin-based TJ fibrils at the site of the BTB is not known. This information, if it is known, will be crucial in understanding TJ dynamics in the testis. Again, glycerol-induced BTB damage is irreversible [88], making it difficult to assess changes during BTB reassembly.
The occludin-peptide model When a single dose of a synthetic occludin peptide, NH2-GSQIYTICSQFYTPGGTGLYVD corresponding to residues 209–230 in the second extracellular loop of rat occludin, was administered to adult rats intratesticularly at 1.5–10 mg/testis, it caused reversible germ cell depletion from the seminiferous epithelium [34]. Morphological analysis of the treated testis revealed that more advanced germ cells, such as elongated spermatids, began to deplete from the epithelium between 8 and 16 days after the occludin peptide treatment [34]. Massive depletion of germ cells from the epithelium occurred in virtually all the tubules by 27 days after an intratesticular occludin peptide injection. In addition, the seminiferous tubules shrank significantly, with the tubular diameter reduced by as much as 20% to 30% compared with control rat testes receiving vehicle or a myotubularin synthetic peptide alone [34]. Germ cells began to repopulate the epithelium 27 days after the occludin peptide treatment. By 47 days, spermatocytes were clearly visible in all the tubules examined, and the morphology of the seminiferous epithelium appeared indistinguishable from control rats by 68 days after occludin peptide treatment, thereby showing full recovery.
That the testes recovered almost fully within 40 days suggests that spermatogonia were not destroyed by this occludin peptide treatment. This occludin peptide-induced germ cell loss was also accompanied by a reversible disruption of the BTB when assessed by micropuncture technique quantifying [125I]-BSA in rete testis fluid and seminiferous tubular fluid following i.v. administration of [125I]-BSA through the jugular vein [34]. The recovery of the BTB coincided with the reappearance of spermatogenesis in the seminiferous epithelium when examined via histological analysis [34]. Also, this event of TJ reassembly (but not TJ disruption) was accompanied by a tumbling in the testicular TGFß3 protein level (unpublished observations), which was consistent with earlier in vitro studies that the assembly of Sertoli cell TJs is associated with a transient plummeting of TGFß3 mRNA levels. This peptide apparently exerts its effects by interfering with the homotypic interactions of two occludin molecules between adjacent Sertoli cells at the sites of TJs, thereby disrupting the BTB. The BTB resealed once the peptide was metabolized, cleared, or both. These results clearly illustrate that this occludin peptide-induced reversible damage of the BTB [34] is a useful model for studying BTB dynamics. For example, it can be used to assess changes in other target genes, such as TGFß3 and occludin, during the disruption and reassembly of the BTB. This model is particularly useful for identifying molecules and signal pathways that are crucial to the event of BTB reassembly.
The aging model—changes in the BTB functionality during aging Studies using an intercellular tracer such as lanthanum nitrate to assess the integrity of the BTB in rats have shown that lanthanum nitrate can permeate the Sertoli cell-TJ barrier, entering to the tubular lumen upon aging [94]. This model illustrates a breakdown of the BTB during aging, which is associated with fewer Sertoli cell TJ fibrils. This in turn renders a lower efficiency of spermatogenesis in aging rats [94]. Similar to rats, the Sertoli cell population in human testes also tumbles with age and studies in the literature have reported a reduction in Sertoli cell number that can in turn affect the integrity of the BTB, thereby reducing the sperm production rate [95–97]. The drawback of this model is that age-related TJ disruption may also be conferred by other factors, such as an alteration of the Leydig cell steroidogenic function, during aging. Also, the loss of BTB function during aging apparently is irreversible.
The cryptorchidism and orchitis model Some cases of male infertility are caused by cryptorchidism and orchitis. Cryptorchidism is a development abnormality caused by the failure of the testes to descend into the scrotum. Orchitis is caused by an inflammation of the testis, which is associated with pain and swelling. Cryptorchidism leads to impaired spermatogenesis with defects in the proliferation and differentiation of Sertoli and Leydig cells in cryptorchid boars, mice, and humans [98–101]. Because of a reduction in the number of differentiated Sertoli cells in the testis, affected males display defective BTB development with fewer TJ fibrils in the seminiferous epithelium, and abnormal spermatogenesis [102–105]. The seminiferous epithelium of cryptorchid males also display unusual thickening of the basement membrane [106], and seemingly suggest that changes in the BTB function may possibly be secondary to the damage in the basement membrane, which is a modified ECM [2]. Aspermatogenesis can also be induced when mice, guinea pigs, or both are immunized with testicular homogenates or spermatozoa emulsified in Freunds adjuvant to induce experimental orchitis [107–109]. This orchitis-induced aspermatogenesis is also accompanied by a disruption of the BTB when assessed by using lanthanum nitrate. Lanthanum was found to deposit between Sertoli cells and inflammatory cells in the seminiferous tubules [108]. The damaged BTB induced by orchitis or cryptorchidism apparently is irreversible [108, 110]. Also, the molecular mechanisms that lead to the BTB damage are not known. For example, it is not known whether cytokines are activated or inactivated during BTB damage, and whether there are any alterations on the functionality of one of the three classes of TJ integral membrane proteins.
Future Perspectives
The usefulness of these models in understanding the regulation and biology of TJ dynamics is obvious. For example, studies conducted using the in vitro model have unequivocally demonstrated that the assembly of Sertoli cell TJ barrier is not only constituted by a series of morphological changes; instead, multiple signaling molecules, proteases, and protease inhibitors are likely to be involved in each step of this event. Using this in vitro model to assess changes in the expression of target genes during Sertoli cell TJ assembly, it has been shown that the assembly of the TJ barrier coincides with a transient plummeting of the levels of TGFß3 protein and mRNA [40]. And the inclusion of recombinant TGFß3 in Sertoli cell cultures during TJ assembly can indeed perturb the assembly of the TJ barrier in a dose-dependent manner, which also inhibits the transient induction of occludin and ZO-1 by Sertoli cells [40].
Also, addition of recombinant TGFß3 to Sertoli cell epithelium with an intact TJ barrier can indeed perturb the barrier (unpublished observations). TGFß3 apparently affects the Sertoli cell TJ barrier in vitro via the p38-MAP kinase pathway [40, 41], and the use of an inhibitor specific to p38-MAP kinase, such as SB202190 [4-(4-fluorophenyl)-2-(4-hydroxyphenyl)-5-(4-pyridyl)IH-imidazole], can indeed abolish the TGFß3-induced disruptive effect on Sertoli cell TJ in vitro [41]. In addition, the Sertoli cell TJ barrier in vitro is regulated at least in part by protein kinases A and C [26, 70].
Obviously, one can argue the in vivo physiological relevance of these in vitro findings. That a 22-amino acid synthetic occludin peptide that can perturb the Sertoli cell TJ assembly in vitro can indeed be reproduced in vivo by reversibly disrupting the BTB [34] seemingly suggests the in vivo relevance of this model for studying TJ dynamics. Furthermore, recently completed studies in this laboratory have shown that the CdCl2-induced damage to the BTB is associated with a surge in TGFß3 in the testis both at the mRNA and protein levels (unpublished observations). Also, the reassembly of the BTB after occludin peptide-induced damage is associated with a plummeting of testicular TGFß3 mRNA level (unpublished observations). These latter studies thus support the physiological significance and the in vivo relevance of the in vitro studies that have clearly illustrated TGFß3 is a crucial regulator of Sertoli cell TJ dynamics [40]. Also, occludin [93] and E-cadherin [111] are the likely primary targets of glycerol- and CdCl2-induced BTB damage in vivo, respectively. And either the damage of BTB in vivo, or the disruption of the Sertoli cell TJ barrier in vitro, induced by glycerol, CdCl2, or both was also shown to be associated with subtle changes in the expression of multiple target genes such as occludin, proteases, and/or protease inhibitors [34, 38, 93].
Taken collectively, results derived from studies using these models have unequivocally implicated the significance of occludin, E-cadherin, proteases, and protease inhibitors in TJ dynamics. Needless to say, some models may turn out to be less useful than others for delineating the regulatory pathways of TJ dynamics in the testis. Still, mechanistic studies generated by using these models should shed new light on the development of novel male contraceptives targeted to disrupt the BTB function, and thereby arresting spermatogenesis.
|
REGULATION OF TJ DYNAMICS |
|---|
|
TOPABSTRACTINTRODUCTIONFUNCTIONS OF THE TJ...STRUCTURE AND MOLECULAR...MODELS FOR STUDYING TJ...REGULATION OF TJ DYNAMICSCONCLUSIONS AND FUTURE...REFERENCES |
|---|
Cytokines
A transient plummeting in TGF-ß3 mRNA and protein has been detected at the time of Sertoli cell TJ assembly. Addition of TGF-ß3 into Sertoli cell cultures, however, can effectively suppress the expression of occludin, ZO-1, and claudin-11, thus perturbing the Sertoli cell TJ barrier [40]. Also, addition of recombinant TGFß3 can perturb a well-formed Sertoli cell TJ barrier in vitro (unpublished observations). These results taken collectively suggest that TGFß3 is an important cytokine in the regulation of Sertoli TJ dynamics. Recent studies have shown that TGFß3 apparently perturbs the Sertoli cell TJ barrier via p38 MAPK pathway. SB202190, a specific p38 MAPK inhibitor, can effectively abolish the TGFß3-induced Sertoli cell TJ disruption in vitro [41]. These observations thus clearly illustrate that if TGFß3 can indeed perturb the BTB in vivo, a disruption of the cytokine function in the testis can effectively block the TJ barrier function. Because TGFß receptors are found in extragonadal tissues, the use of SB202190 or even a new analogue based on this core structure may not necessarily be the best candidate because this will block other TGFß-mediated biological effects. The potential of this implication thus largely relies on the ability to develop a novel delivery agent, such as an FSH mutant having the ability to bind to its receptor on Sertoli cells without any hormonal activity, which can in turn conjugate to SB202190 or a more potent analogue, to block the TGFß-induced disruption of the Sertoli cell TJ barrier to permit cell movement across the BTB.
Protein Phosphorylation
Tyrosine phosphorylation of TJ-associated proteins is known to play a crucial role in modulating TJ dynamics [115–117]. For instance, both occludin and ZO-1 can be tyrosine phosphorylated and are putative substrates of tyrosine kinases [117]. Also, occludin found at the site of TJs is highly phosphorylated [118, 119]. In MDCK cells, an increase in tyrosine phosphorylation of occludin, ZO-2, and ZO-3 was detected during TJ barrier assembly, which could be inhibited by genistein and PP-2 (a protein tyrosine kinase inhibitor) [119, 120]. When MDCK cells were cultured in media with low calcium, the TJ barrier was disrupted, coinciding with a reduced serine/threonine phosphorylated occludin level [118]. However, other studies using MDCK cells have shown that increased tyrosine phosphorylation by using inhibitors of protein tyrosine phosphatase (PTPi) can cause redistribution of TJ proteins, which in turn perturbs the TJ barrier [121]. In the mouse testis, there is a gradual increase in the level of phosphorylated occludin coinciding with the development of the BTB, indicating the level of phosphorylated occludin correlates with the event of Sertoli cell TJ assembly [29]. Indeed, the Sertoli cell TJ barrier can be perturbed reversibly by using inhibitors of protein tyrosine phosphatase (PTP), such as sodium orthovanadate, to increase the intracellular phosphoprotein content [26]. Furthermore, both Sertoli and germ cells in rats are known to express rat myotubularin (rMTM) and to produce the rMTM protein [70, 122], a putative PTP. A tumbling in rMTM expression was also detected in the testis during glycerol-induced BTB damage in the rat [70]. These data thus strongly support the notion that protein phosphorylation plays a crucial role in TJ dynamics. Yet the molecules or pathways that trigger changes in protein phosphorylation are not known even though protein kinases A and C are implicated in this event [26]. Furthermore, the subsequent downstream molecular and cellular events by which phosphoproteins can affect TJ dynamics are not known.
Cyclic AMP, Ca2+, Proteases, and Protease Inhibitors
It has been known for a decade that low concentrations of dibutyryl cAMP (4–20 µM) can stimulate the Sertoli cell TJ barrier assembly in vitro, whereas at high levels (100–500 µM), it can perturb the Sertoli cell TJ barrier [76], suggesting that the TJ dynamics in the testis can be regulated by the similar mechanisms and pathways via cAMP, at least in part, as in other epithelia. Yet the trigger of these changes in intracellular cAMP level is not known. Furthermore, a loss in intracellular [Ca2+] can also perturb the Sertoli cell TJ barrier in vitro, and its replacement can quickly reseal the barrier [75]. Recent studies have strongly suggested that protein kinases A and C play a critical role in regulating Sertoli cell TJ dynamics. For instance, both adenosine 3',5'-cyclic monophosphothioate (a PKA inhibitor), chelerythrine chloride (a PKC inhibitor), and D-erythro-sphingosine (a PKC inhibitor and a calmodulin-dependent kinase inhibitor) were shown to modulate the assembly and maintenance of the Sertoli cell TJ barrier in vitro [26]. These results taken collectively obviously show that intracellular phosphoprotein content and cAMP levels are two major determinants that regulate Sertoli cell TJ dynamics. Other studies have also implicated the significance of proteases and protease inhibitors in the Sertoli cell TJ function [4, 71, 123]. A wide range of proteases and protease inhibitors have been shown to be expressed and secreted by Sertoli cells, germ cells, or both in the testis [124, 125]. Some of these play a significant role in the regulation of TJ dynamics in the testis, such as protein C inhibitor (PCI), and several tissue inhibitors of metalloproteases [71, 126, 127]. For instance, chloroquine, a protease inhibitor, can facilitate the assembly of Sertoli cell TJ barrier in vitro, making the TJ barrier tighter [78]. Also, PCI-/- male mice are infertile, and this infertility is apparently due to the unchecked proteolytic activity in the testis, causing the destruction of the Sertoli cell TJ barrier, which in turn affects spermatogenesis [126]. These results taken collectively clearly illustrate the crucial role of proteases and protease inhibitors in TJ dynamics.
|
CONCLUSIONS AND FUTURE PERSPECTIVES |
|---|
|
TOPABSTRACTINTRODUCTIONFUNCTIONS OF THE TJ...STRUCTURE AND MOLECULAR...MODELS FOR STUDYING TJ...REGULATION OF TJ DYNAMICSCONCLUSIONS AND FUTURE...REFERENCES |
|---|
It is apparent that multiple regulatory molecules and pathways are operating simultaneously to modulate Sertoli cell TJ dynamics in the testis, such as the TGFß3/p38-MAP kinase pathway and protein kinases A and C [26, 40, 41, 70]. Yet it is also possible that different pathways can eventually converge to a single (or a single set of) downstream pathways, transcription factors, or both. The findings from deletion studies that transgenic mice lacking either occludin or claudins were sterile with impaired spermatogenesis also clearly illustrate the significance of the TJ barrier in spermatogenesis. Perhaps it is noteworthy that virtually no studies can be found in the literature that have investigated the role of TJ-associated signaling molecules, such as protein kinases, in TJ dynamics. The only exception to this is a small scale study using inhibitors of different kinases and phosphatases to assess their effects on the Sertoli cell TJ barrier in vitro [26, 70]. This apparently is a priority area that needs to be investigated in the near future.
Another interesting topic that deserves further investigation is whether developing germ cells migrate as a clustered clone or individually. For example, developing germ cells are cytoplasmically interconnected by intercellular bridges, presumably to permit communication between germ cells within a clone, and germ cells within each clone differentiate synchronously [128–132]. Spermatogonia that undergo apoptosis naturally or induced by irradiation are also known to degenerate in clusters, suggesting that one or more factors can diffuse through these intercellular bridges so that all cells within a clone can undergo programmed cell death simultaneously [131, 133]. If preleptotene and leptotene spermatocytes indeed migrate across the BTB as clustered clones, the risk of potential fallout of Sertoli cell TJs and the BTB is high because of the extensive TJ restructuring associated with such massive cell movement, which may also be the reason why TJ dynamics require multiple pathways for their regulation to ensure the integrity of the BTB.
In summary, a brief review of the biology and regulation of TJ dynamics in the testis was presented. Although many questions still remain unanswered, recent advances in biochemistry, and cellular and molecular biology will likely address most if not all of these questions in the coming decade. Needless to say, a thorough understanding of TJ dynamics can have significant impact on the development of novel male contraceptives. For example, a shutdown of the BTB can disrupt the timely migration of preleptotene and leptotene spermatocytes across the Sertoli cell TJ barrier. An unexpected, yet prolonged opening of the BTB can also perturb spermatogenesis. In both instances, infertility will result without affecting the hypothalamus-pituitary-testicular hormonal axis.
|
FOOTNOTES |
|---|
2 Correspondence: C. Yan Cheng, Population Council, 1230 York Avenue, New York, NY 10021. FAX: 212 327 8733; y-cheng{at}popcbr.rockefeller.edu
Received: 14 August 2002.
First decision: 30 August 2002.
Accepted: 30 October 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONFUNCTIONS OF THE TJ...STRUCTURE AND MOLECULAR...MODELS FOR STUDYING TJ...REGULATION OF TJ DYNAMICSCONCLUSIONS AND FUTURE...REFERENCES |
|---|
, gelatinase B (matrix metalloprotease-9), and tissue inhibitor of metalloproteases-1, in the basal lamina regulates Sertoli-cell tight junction dynamics in the rat testis. Endocrinology 2003; 144:371–387 
