Regulation of blood-testis barrier dynamics by focal adhesion kinase (FAK): An unexpected turn of events

Abstract

The blood-testis barrier (BTB) is conferred by co-existing tight junctions (TJs), basal ectoplasmic specializations (basal ES), desmosome-like junctions and gap junctions (GJs) between adjacent Sertoli cells near the basement membrane in the seminiferous epithelium. While the concept of the BTB has been known for more than a century and its significance to spermatogenesis discerned for more than five decades, its regulation has remained largely unknown. Recent studies, however, have demonstrated that focal adhesion kinase (FAK), a modulator of the integrin-based signaling that plays a crucial role in cell movement, apoptosis, cell survival and gene expression at the focal adhesion complex (FAC, also known as focal contact, a cell-matrix anchoring junction type), is an integrated component of the BTB, associated with the TJ-integral membrane protein occludin and its adaptor zonula occludens-1 (ZO-1). Herein, we summarize recent findings in the field regarding the significance of FAK in conferring BTB integrity based on some unexpected observations. We also critically discuss the role of FAK in regulating the timely “opening” and “closing” of the BTB to facilitate the transit of primary preleptotene spermatocytes across the BTB at stage VIII of the seminiferous epithelial cycle of spermatogenesis. Lastly, we describe a working model, which can be used to design future functional experiments to explore the involvement of FAK in BTB dynamics by investigators in the field.

Introduction

Focal adhesion kinase (FAK), as the name implies, is a non-receptor protein tyrosine kinase (PTK) restricted to the focal adhesion complex (FAC, also known as the focal contact which is an actin-based, cell-matrix type of anchoring junction) at the interface of the cell and extracellular matrix (ECM). While FAK is ubiquitously expressed in mammalian cells and tissues, it is highly expressed in the testis and brain, as well as in osteoclasts in the bone.^1,2 FAK is a 125 kDa protein composed of four functional domains: (1) a band 4.1, ezrin, radixin, moesin (FERM) domain near the N-terminus, (2) a central catalytic kinase domain, (3) three proline-rich regions, PRI, PRII, PRIII, and (4) a focal adhesion targeting (FAT) domain near the C-terminus (Fig. 1).^3,4 FAK is known most notably for its ability to mediate integrin-based signaling,⁵ but FAK activation can also regulate multiple cellular processes including cell growth, cell survival, apoptosis, cell movement, gene expression and junction restructuring.^4–9 FAK is different from other non-receptor PTKs in two important aspects. First, FAK lacks a myristylation site which is needed to anchor PTKs to the plasma membrane.^2,6 Second, while FAK has multiple binding partners, it lacks putative Src homology 2 (SH2) and SH3 domains, which are required for protein-protein interactions. Instead, proline (Pro)-rich regions such as PRI, PRII and PRIII provide the binding sites for SH3 domain-containing proteins such as p130Cas, Graf [GTPase-activating protein (GAP) for Rho associated with FAK], PSGAP (PH and SH3 domain containing Rho GAP) and ASAP1 [ArfGAP (ADP ribosylation factor/GTPase-activating protein) with SH3 domain, ANK (ankyrin) repeats and PH domain 1] (Fig. 1).^3,4

Figure 1.

A schematic drawing illustrating the different functional domains of FAK. FAK is a known moderator of integrin-based signaling. Upon coupling of an integrin receptor with one of its various ligands (e.g., collagens, laminins and/or one of their fragments), FAK mediates various cellular functions in multiple epithelia and endothelia, including cell movement, apoptosis, gene expression, cell survival and junction dynamics. FAK is composed of a FerM (band 4.1, ezrin, radixin, moesin homology) domain near its N-terminus, followed by the catalytic kinase domain, three pro-rich regions (pri, prii and priii) and a FAt (focal adhesion targeting) domain near the C-terminus. FAK also has six putative phosphorylation sties at tyr-397, -407, -576, -577, -861 and -925. Different regulatory proteins (e.g., c-src) and adaptors (e.g., vinculin, talin) can bind to FAK upon its activation. thus, FAK also serves as a platform for signal transduction by recruiting many regulatory and adaptor proteins to regulate diverse cellular processes. This figure was prepared based on the following reviews.³–⁵

Furthermore, FAK contains six putative phosphorylation sites at Tyr-397, -407, -576, -577, -861 and -925.² Amongst these, Tyr-397 is the only site that is autophosphorylated. Studies have shown that the interaction of FAK with integrin receptors such as β1-integrin near the N-terminus of FAK (Fig. 1) induces conformational changes, exposing the Tyr-397 site.^2,10 This, in turn, creates a high affinity-binding site for different regulatory proteins such as Src family kinases (e.g., c-Src), PI-3K (phosphatidylinositol 3-kinase), PLC-γ (phospholipase C-γ), Grb7 (growth factor-receptor bound protein 7), PTEN (phosphatase and tensin homolog on chromosome 10) and Nck-2 (SH2/SH3 adaptor protein-2)^3,4 so that the SH-2 domain of these proteins can adhere to FAK (Fig. 1). Thus, even though integrin does not possess intrinsic catalytic activity, its interaction with FAK following the coupling of integrin (e.g., β1-integrin) with one of its ligands (e.g., laminins, collagens and/or fragments of laminins and collagens) can induce downstream signaling to affect cellular function via the recruitment of signaling proteins and adaptors to the site. Indeed, it was shown that p-FAK-Tyr³⁹⁷ structurally interacted with β1-integrin, vinculin and c-Src in Sertoli-germ cell cocultures¹¹ with established anchoring junctions (e.g., desmosome-like junction, basal ES) that mimicked those found in vivo.¹² The five remaining putative phosphorylation sites other than Tyr-397 are putative targets of the non-receptor Src kinase protein family, and the activation of FAK via phosphorylation at any one of these sites can induce either intrinsic kinase activity and/or recruitment of regulatory or signaling proteins to FAK to affect cell function.^3,4 For instance, it is known that phosphorylation of FAK at Tyr-576 and Tyr-577 found within the kinase domain enhances its intrinsic kinase catalytic activity,¹³ whereas phosphorylation of FAK at Tyr-925 generates a binding site for Grb2 (growth factor-receptor bound protein 2) (Fig. 1), which serves as an adaptor to recruit Sos [a guanine nucleotide exchange factor (GEF) for Ras GTPase] to associate with FAK. This, in turn, leads to activation of downstream ERK1/2 (extracellular regulated kinase 1/2) and JNK (c-Jun N-terminal kinase, also known as stress-activated protein kinase) MAPK (mitogen activated protein kinase) signaling events,¹⁴ contributing to FAK-mediated cell growth and cell survival.^5,15,16

Finally, the FAT domain in FAK can associate with adaptors paxillin and talin, as well as with the signaling protein Stat-1 (signal transducer and activator of transcription 1) (Fig. 1). The FAT domain can also link FAK to integrins mediated via adaptors, such as paxillin (Fig. 1). In short, FAK is a crucial regulatory protein for diverse cellular events in multiple epithelia. Interestingly, while the expression level of FAK in the testis is at least 20- to 100-fold higher than in non-gonadal tissues,^1,2 its physiological significance in the testis has remained unexplored for almost two decades since its discovery in the late 1980s for three reasons. First, FAC is absent in the seminiferous epithelium of mammalian testes. Thus, there was little interest to study an FAC-associated protein, such as FAK, in the testis. Second, germ cells are not motile cells per se. Third, FAK is intimately related to cell movement (e.g., fibroblasts, macrophages and tumor cells) at the cell-matrix interface in other epithelial cells. Although Sertoli cells are highly motile cells when cultured in vitro, they are relatively static in the seminiferous epithelium in vivo. Yet, they provide nutritional and structural support to developing germ cells and maintain BTB function near the basal region of the seminiferous epithelium. As such, interest on the physiological role of FAK in spermatogenesis has remained relatively low. Herein, we summarize recent findings and critically discuss the unexpected turn of events regarding the role of FAK in regulating BTB dynamics.

Current Concept of the BTB

The concept of the blood-testis barrier (BTB) and blood-brain barrier (BBB) is based on observations from more than a century ago in which dyes administered intravenously into rodents failed to ‘stain’ the seminiferous tubules in the testis, as well as the brain, respectively, because of the presence of a blood-tissue barrier in these organs.^17,18 Subsequent studies illustrated that unlike other blood-tissue barriers (e.g., the BBB and blood-retina barrier) which are created exclusively by tight junctions (TJs) between endothelial cells in microvessels, the BTB is conferred by adjacent Sertoli cells in the seminiferous epithelium near the basement membrane (Fig. 2).^18,19 In fact, microvessels in the interstitium of the testis contribute relatively little to barrier function. Instead, the BTB in mammalian testes is constituted by co-existing TJs, basal ectoplasmic specializations (basal ES, a testis-specific atypical adherens junction type^20,21), desmosome-like junctions and gap junctions (GJs) present between adjacent Sertoli cells, and these junctions are known to function collectively in the maintenance of barrier integrity^17,22 (Fig. 2). As such, the BTB is one of the tightest blood-tissue barriers in the mammalian body. The BTB also divides the seminiferous epithelium into a basal and an apical (i.e., adluminal) compartment and segregates the events of post-meiotic germ cell development from the systemic circulation by conferring an immunological barrier so that the host body does not develop antibodies against specific antigens residing on spermatids during spermiogenesis. The BTB also confers Sertoli cell polarity and creates a “fence” by restricting paracellular transport of substances (e.g., water, ions and electrolytes) into the apical compartment. While the BTB is a highly dynamic structure since it must “open” at stage VIII of the seminiferous epithelial cycle to allow the transit of primary preleptotene spermatocytes, yet the BTB cannot be compromised at this time, even transiently, to avoid the production of anti-sperm antibodies. Thus, it is believed that a novel mechanism is in place in the testis to confer barrier function while at the same time permitting restructuring of the BTB in order to facilitate the transit of primary spermatocytes at stage VIII of the epithelial cycle of spermatogenesis.

Figure 2.

A model illustrating the mechanism by which FAK regulates BtB function by causing its “opening” and/or “closing” to facilitate the transit of primary preleptotene spermatocytes at stage viii of the seminiferous epithelial cycle of spermatogenesis via its effects on the phosphorylation status of integral membrane proteins at the BTB, such as occludin. This model was prepared based on recent findings in the field as detailed in the text. The panel on the left represents a schematic drawing illustrating the relative location of the BtB in the seminiferous epithelium of adult mammalian testes. the two panels on the right illustrate the status of the BtB (either “closed” or “opened”) which is regulated by FAK via its effects on the phosphorylation status of integral membrane proteins (e.g., occludin) at the BtB. the upstream external stimuli are likely to be cytokines (e.g., tGFβ3, tNFα) and testosterone, which have opposing effects on the BtB integrity by either perturbing or promoting the integrity of the junctional complexes at the BtB, respectively. the combined effects of these external stimuli, in turn, affect the levels of integral membrane proteins at the cell-cell interface via a balance between protein endocytosis, endosome-mediated protein degradation, recycling and/or transcytosis. the overall effect of these cellular processes facilitates the transit of primary preleptotene spermatocytes across the BtB during spermatogenesis.

Regulation of the BTB: Current Status of Research

In epithelia and endothelia (e.g., microvessels), only about one-fifth of cell junctions are TJs that are restricted to the apical region of adjacent cells.^23,24 TJs are known to contribute almost exclusively to the integrity of the permeability barrier in most blood-tissue barriers, such as BBB and blood-retina barrier. Behind TJ-fibrils are adherens junctions (AJs), which form the adhesion belt, and desmosomes, which are cell-cell intermediate filament-based anchoring junctions that are found immediately behind the adhesion belt. Collectively, TJs, AJs and desmosomes form the junctional complex with gap junctions (GJs) lying outside of this specialized ultrastructure. In most cases, but not all, a disruption of the TJ can affect the integrity of the AJ and vice versa, illustrating the intimate relationship between TJs and AJs in epithelia and endothelia. The Sertoli cell BTB, in contrast, is composed of co-existing TJs, basal ES, desmosome-like junctions, and gap junctions. Throughout spermatogenesis, particularly during spermiogenesis, post-meiotic spermatids transform from step 1 round to step 19 elongated spermatids that must traverse the seminiferous epithelium. This involves extensive restructuring of anchoring junctions present between spermatids and Sertoli cells known as desmosome-like junctions and apical ES. Thus, a mechanism must be in place to avoid disruption of TJs at the BTB during restructuring of anchoring junctions. A recent study in the testis has shown that multi-protein complexes at the TJ (e.g., occludin-ZO-1) and at the basal ES (e.g., N-cadherin-β-catenin) associate structurally via an interaction between ZO-1 and β-catenin which reinforce barrier function when in an “engaged” state.²⁵ However, during anchoring junction restructuring throughout spermatogenesis to facilitate the transit of germ cells across the epithelium, as well as after treatment of rats with adjudin, a chemical known to induce anchoring junction disruption in the testis,²⁶ interactions between these TJ- and AJ-based protein complexes become “disengaged,”²⁵ such that a disruption of AJs fails to perturb TJ function. In this way, the function of the immunological barrier in the seminiferous epithelium can be maintained at all times during spermatogenesis. Moreover, co-existing desmosome-like junctions and GJs have also been recently shown to facilitate cross-talk between different junction types at the BTB, as well as to contribute to the maintenance of the TJ-permeability barrier.²⁷ For instance, it was shown that connexin43 (Cx43) and plakophilin-2 (PKP-2) form a functional protein complex in the testis.²⁷ While knockdown of either one of these proteins in vitro failed to affect the Sertoli cell TJ barrier, simultaneous knockdown of both proteins by RNA interference (RNAi) impeded TJ barrier function by perturbing the distribution of the TJ protein occludin and reducing the steady-state level of basal ES proteins N-cadherin and coxsackie and adenovirus receptor (CAR).²⁷ Additional research is needed to investigate the mechanism(s) by which desmosome-like junctions and GJs affect TJ and basal ES function at the BTB, as these cell junctions contribute collectively to BTB integrity and remodeling.

On a final note, recent studies have shown that testosterone, which promotes BTB and junctional complex integrity in the testis,^28–31 is likely to induce the assembly of “new” TJ-fibrils behind a migrating primary preleptotene spermatocyte,³² possibly via de novo protein synthesis and/or transcytosis. On the other hand, cytokines (e.g., TGFβ3 and TNFα) have been reported to disrupt “old” TJ-fibrils which are situated above a spermatocyte in transit via accelerated protein endocytosis and endosome-mediated degradation.³² Thus, the combined effects of testosterone and cytokines facilitate the transit of primary spermatocytes across the BTB while maintaining the immunological barrier. However, the precise mechanism(s) underlying testosterone- and cytokine-mediated regulation of protein endocytosis, degradation, recycling and/or transcytosis, and the regulatory molecule(s) involved in these events remain unknown.

FAK is an Integrated Component of the Occludin-ZO-1 Protein Complex at the BTB

While putative protein complexes (e.g., occludin/ZO-1, claudins/ZO-1, JAM-A/ZO-1, JAM-B/ZO-1, nectins/afadins) that constitute the BTB in the testis have been known for almost two decades, the regulatory molecules that affect BTB restructuring remain largely unexplored. The first report on a possible regulator of the BTB appeared in the literature in 2003 when it was shown that FAK localized to the BTB in virtually all stages of the seminiferous epithelial cycle by immunohistochemistry and fluorescence microscopy.¹¹ However, its staining at the BTB was considerably diminished in stage VIII of the epithelial cycle during the transit of primary spermatocytes across the BTB,¹¹ illustrating FAK is a possible regulator of BTB integrity. However, the two phosphorylated forms of FAK, p-FAK-Tyr³⁹⁷ and p-FAK-Tyr⁵⁷⁶ were restricted mostly to the apical ES, co-localizing with vinculin, a putative apical ES protein.¹¹ This latter observation was subsequently confirmed by other investigators in the field,³³ and this thus illustrated that activated p-FAK-Tyr³⁹⁷ is important for apical ES function. However, these findings were not immediately pursued and expanded because FAK is a well-studied component of the FAC in other epithelia, but in the testis, the FAC is absent. Since this time, numerous studies have illustrated that the apical and basal ES are constituted by proteins usually restricted to the FAC including integrin, vinculin, integrin-linked kinase (ILK), c-Src and others.^34,35 This led us to speculate that the BTB may indeed utilize components of the FAC for its regulation to provide an effective mechanism of restructuring during spermatogenesis. As such, components of the FAC, such as FAK and c-Src, may be present at the BTB to regulate its restructuring during spermatogenesis. Indeed, by co-immunoprecipitation FAK was shown to be an integrated component of the occludin-ZO-1 multi-protein complex, and dual-labeled immunofluorescence analysis also illustrated co-localization of FAK with occludin and ZO-1 in cross-sections of testes in vivo and Sertoli cells cultured in vitro.³⁶ Equally important, when the BTB was induced to undergo restructuring following administration of cadmium which resulted in its eventual disruption, a transient but significant increase in association between occludin and ZO-1 with FAK was detected,³⁷ seemingly suggesting that an increase in occludin and ZO-1 phosphorylation induced by FAK leads to BTB disruption. To further explore the functional significance of FAK in Sertoli cell barrier function, FAK was silenced by RNAi in Sertoli cell cultures having an established TJ permeability barrier, as well as containing TJ and basal ES ultrastructures that mimicked the BTB in vivo. It was shown that knockdown of FAK led to a loss of TJ barrier function when transepithelial electrical resistance (TER) across the Sertoli cell epithelium was quantified.³⁷ This loss of barrier function was associated with changes in the distribution of occludin and JAM-A in Sertoli cells. Specifically, these proteins moved away from the cell-cell interface and into the cytosol, possibly as a result of enhanced protein endocytosis.³⁷ A study to assess the level of phosphorylated occludin in Sertoli cells following FAK silencing indeed demonstrated a significant decline in occludin phosphorylation at Tyr and Ser, but not Thr, residues,³⁷ revealing the internalized occludin to be less phosphorylated. These findings are consistent with the concept³⁸ that highly phosphorylated occludin is recruited to TJ-fibrils to confer TJ-permeability barrier function,^39,40 whereas less phosphorylated occludin is found in the basolateral region of epithelial cells and not recruited to TJ-fibrils.^39,41

It has been known for almost four decades that the BTB is exceedingly vulnerable to cadmium, an environmental toxicant.⁴² Studies have shown that treating rats with cadmium disrupted TJ-associated microfilaments,⁴³ however, the molecular mechanism underlying this damage has remained largely unknown until recently. Interestingly, when FAK was silenced in Sertoli cells with an established barrier that mimicked the BTB ultrastructurally and functionally in vivo, the loss of FAK rendered Sertoli cells less susceptible to cadmium-induced barrier disruption.³⁷ This finding further illustrates the functional significance of FAK at the BTB. Collectively, these observations not only illustrate that FAK is an integrated and structural component of the occludin/ZO-1 complex at the BTB, but that it also serves as a crucial regulator to affect the cellular distribution of occludin at the cell-cell interface. It is likely that FAK can affect the BTB integrity to accommodate the transit of primary preleptotene spermatocytes across the barrier at stage VIII of the epithelial cycle via its effects on the phosphorylation status of occludin.

FAK Regulates BTB Dynamics: A Novel Model

As briefly discussed above, the occludin/ZO-1 protein complex at the BTB is a putative substrate of FAK (Fig. 2). During the seminiferous epithelial cycle when the BTB must remain “closed,” FAK regulates the phosphorylation status of occludin so that phosphorylated occludin can be recruited to the BTB and assembled into TJ-fibrils in order to maintain BTB integrity (Fig. 2). However, external stimuli, such as an increase in the levels of cytokines (e.g., TGFβ3 and TNFα) or a lowering of the androgen/androgen receptor steady-state level (or both) within the microenvironment of the BTB, or an exposure of the testis to environmental toxicants (e.g., cadmium), can induce dissociation of FAK from the occludin/ZO-1 protein complex and/or a loss in FAK function. Thus, this would lead to a reduction in occludin phosphorylation at the BTB, possibly causing an acceleration in its endocytosis and leading to an increase in occludin endosome-mediated degradation (Fig. 2). This would contribute to a destabilization of the BTB, allowing the BTB to “open” and to facilitate the transit of primary spermatocytes across the BTB at stage VIII of the epithelial cycle. Figure 2 summarizes our current concept regarding the role of FAK in the regulation of the “opening” and “closing” of the BTB during the epithelial cycle of spermatogenesis.

Concluding Remarks and Future Perspectives

In short, FAK is an integrated and a regulatory component of the occludin/ZO-1 protein complex at the BTB. It is likely that FAK determines and maintains the phosphorylation status of the occludin-ZO-1 complex, thereby regulating the cellular distribution of occludin and ZO-1 at the BTB, possibly via endocytic vesicle-mediated protein endocytosis. Endocytosed proteins can be recycled back to the cell surface, undergo endosome-mediated protein degradation, or be transcytosed to another site such as below a primary spermatocyte in transit across the BTB. This would assemble “new” TJ-fibrils to maintain immunological barrier function. It is expected that many other regulatory proteins will be identified at the BTB microenvironment in the coming years, and the fact that FAK can recruit other regulatory proteins (e.g., c-Src) and adaptors (e.g., vinculin) to this site based on studies in other epithelia and endothelia suggests that much work is needed in order to determine how these proteins contribute to the “opening” and the “closing” of the BTB during the epithelial cycle of spermatogenesis. Additionally, since FAK is detected in germ cells besides Sertoli cells,¹¹ it is of interest to investigate if FAK plays any role in germ cell apoptosis during spermatogenesis.