Macrophages are by far the most prominent immune cells in the testis. They are found in the testes of all mammalian species so far examined, but are particularly numerous in human, rat, and mouse testes (Frungieri et al., 2002, Hume et al., 1984, Mendis-Handagama et al., 1987, Miller et al., 1984, Vergouwen et al., 1993, Wang et al., 1994). Typically, macrophages arise from myeloid precursors in the bone marrow and circulate in the blood as monocytes. They are capable of adopting a broad range of structural and functional phenotypes that are largely dictated by the tissues in which they become resident (Mantovani et al., 2005, van Furth, 1988). Characteristically, they are robustly phagocytic and possess a potent arsenal of antimicrobial and cytotoxic activities, such as the ability to generate large amounts of ROS. They produce many immunoregulatory products, including cytokines, eicosanoids (prostaglandins, thromboxanes, leukotrienes, and lipoxins), and nitric oxide, all of which are important in establishing and controlling the inflammatory process, but also have important roles in tissue remodeling, healing, and normal homeostasis. All macrophage lineages can express MHC class II antigens and have the capacity to present to CD4⁺ helper or T_reg cells (Unanue 1984). Thus, macrophages are the cells that link innate and adaptive immunity, and dendritic cells are a highly specialized macrophage subset that is particularly effective in this role (Heath and Carbone 2001).

Testicular macrophages have been most extensively studied in the rat, with a few studies also in the mouse and human. Studies in the rat indicate that they are heterogenous in their morphology and functions (Bryniarski et al., 2004, Gerdprasert et al., 2002a, Itoh et al., 1995b, Wang et al., 1994). Their absolute numbers are primarily under Leydig cell control, but evidence suggests that FSH acting via the Sertoli cell has a role in regulating their functional activity (Duckett et al., 1997a, Duckett et al., 1997b, Gaytan et al., 1994c, Wang et al., 1994). Macrophages accumulate within the interstitial tissue at the time of puberty, and play critical roles in the normal development of the testis, particularly by promoting the proliferation and maturation of the Leydig cell population (Cohen et al., 1997, Gaytan et al., 1994a, Gaytan et al., 1994b, Hardy et al., 1989, Itoh et al., 1999, Raburn et al., 1993, Vergouwen et al., 1993). Through their ability to produce a broad array of growth factors, prostanoids, and vasoactive regulators, other developmental and homeostatic roles within the testis may be predicted as well.

Data concerning the immunological functions of testicular macrophages are based on a relatively small number of studies in either the rat or the mouse, and are patchy at best. Indeed, there is some evidence for distinct differences even between these two closely related species. In general, testicular macrophages have intact phagocytic, cytotoxic, and antimicrobial functions (Miller et al., 1984, Wei et al., 1988). Rat testis macrophages express MHC class II antigens (Head and Billingham, 1985, Hedger and Eddy, 1987, Wang et al., 1994), but display a reduced capacity for lymphocyte activation in vitro (Kern and Maddocks 1995). Testicular macrophages in the human and mouse also express MHC class II, although expression in the mouse appears to be more restricted (Haas et al., 1988, Itoh et al., 1995b, Pöllänen and Niemi, 1987, Tung et al., 1987). A lack of expression of the essential costimulatory molecules, B7-1 (CD80) and B7-2 (CD86), has been reported in the mouse testis as well (Sainio-Pöllänen et al. 1996). Studies also indicate that macrophages from the rat testis have a significantly reduced ability to produce proinflammatory cytokines, such as IL1β and TNFα, following activation (Gerdprasert et al., 2002a, Hayes et al., 1996, Kern and Maddocks, 1995, Kern et al., 1995, Meinhardt et al., 1996), and stimulated mouse testicular macrophages produce the anti-inflammatory/immunosuppressive cytokines IL10 and TGFβ₁ in vitro (Bryniarski et al., 2004, Bryniarski et al., 2005). While the data suggest that testicular macrophages may have a predominantly anti-inflammatory and/or immunosuppressive phenotype, this has yet to be formally established. Furthermore, dendritic cells, which are a highly specialized macrophage subset also found within the testicular interstitium (Fijak et al., 2005, Hoek et al., 1997, Itoh et al., 1995b, Rival et al., 2006a), presumably play a more important role in controlling immune responses to antigens within the testis (Fijak et al., 2005, Rival et al., 2006a, Rival et al., 2007).

In the rat testis, macrophages can be discriminated by their differential expression of two molecular markers recognized by the antibodies ED1 and ED2. Antibody ED1 recognizes the lysosomal protein CD68, which is associated with antigen processing and presentation (Groisman et al. 2002), while ED2 recognizes the scavenger receptor CD163 (Fabriek et al. 2005). CD163 is involved in the clearance of hemoglobin:haptoglobin complexes and the resolution of inflammation (Philippidis et al. 2004). About 80–85% of rat testicular macrophages express this scavenger receptor and about half of these cells also lack expression of CD68 (Gerdprasert et al., 2002a, Meinhardt et al., 1998, Wang et al., 1994). In the human testis, most macrophages appear to be CD68⁺, but expression of CD163 is a feature of a subset of these cells as well (Frungieri et al. 2002). The heterogeneous expression of these two markers points to the existence of multiple populations of testicular macrophages, and evidence suggests that these phenotypic differences also correspond to functional differences. In the normal and inflamed adult rat testis, CD163⁺ cells lack expression of the key inflammatory mediators IL1β and inducible nitric oxide synthase (iNOS), as measured by double-label immunohistochemistry, but these molecules are detectable in CD68⁺ cells that do not express CD163 (Gerdprasert et al., 2002a, O’Bryan et al., 2005). Indeed, purification of CD163⁺ macrophages from the rat testis using immunomagnetic beads confirms that these cells have reduced capacity to produce IL1β, IL6, or NO upon stimulation, although they retain the ability to produce prostaglandins (Gerdprasert et al. 2002a; Hedger, unpublished data). This suggests that the CD163⁺ macrophages have reduced proinflammatory activities, while the CD163-negative testicular macrophage population retains normal proinflammatory functions. This observation is important because macrophages that express CD163 and produce IL10, but display reduced expression of CD80, CD86, iNOS, TNFα, and IL1β, are polarized toward the immunoregulatory M2 subset (Mantovani et al., 2005, Sica et al., 2008). The M2 macrophage subset is analogous to the Th2 subset among T cells, and is associated with late-stage inflammation and its resolution, as well as tumors and other immunologically privileged tissue sites. It would appear that a significant proportion of the resident macrophages of the testis possess an immunoregulatory phenotype that is consistent with immune privilege. The functional roles of this macrophage population, as well as those of the macrophages that appear to retain normal inflammatory (i.e., M1 subset) properties, demand particular attention if we are to understand the unique immunological environment of the testis.

While the functions and regulation of lymphocytes within the testis have yet to be properly characterized, by their presence alone these cells will have significant effects on testicular events. It may also be assumed that, in contrast to the largely resident population of testicular macrophages, lymphocytes circulate through the testis. This may be because they possess specificity for testicular antigens, or because they have been activated during earlier intratesticular inflammatory or infectious events (Czerkinsky et al., 1999, Springer, 1994). Quantitative studies in the rat indicate that, at any given time, lymphocytes are approximately 10% as numerous as the macrophages in the normal testis (Hedger and Meinhardt, 2000, Hedger et al., 1998b, Wang et al., 1994). In the rat and mouse, intratesticular lymphocytes express T cell and NK cell markers, and comprise distinct T cell, NK cell, and NKT cell subsets (Tompkins et al. 1998; Hedger, unpublished data). Both T cell and NK cell markers are expressed by lymphocytes in the human testis as well (Hedger, unpublished data). B cells are not usually observed in normal tissues, and this includes the testis. In the rat, many T cells possess a memory phenotype, as would be expected (Hedger et al., 1998b, Tompkins et al., 1998). Reanalysis of the data for individual animals across several studies suggests that their numbers tend to be proportional to the numbers of macrophages, but gradually increase with the age of the animal (Hedger et al., 1998b, Hedger and Meinhardt, 2000, Wang et al., 1994). Among the T cells of the testis, there is a strong bias toward the CD8⁺ (i.e., MHC class I restricted, largely cytotoxic) subset (Hedger et al., 1998b, Pöllänen and Niemi, 1987, Ritchie et al., 1984, Tompkins et al., 1998, Wang et al., 1994). The prominence of NK cell subsets in the testis is also intriguing given the reduced MHC class I antigen expression in the seminiferous compartment, which would tend to make these cells more susceptible to NK cells (Lanier 2001). Consistent with the anti-inflammatory/immunosuppressive phenotype of the major testicular macrophage subset, T cells and NKT cells in the rat testis constitutively produce IL10 (Hedger, unpublished data). While much remains to be discovered, the existing data concerning the types and activity of testicular lymphocytes are consistent with maintaining immune privilege (i.e., the presence of immunoregulatory lymphocytes) and increased innate immunity (i.e., cytotoxic cell activity).

Evidence that PMNs are important in testicular function comes from the fact that mast cells and eosinophils are significant elements of the interstitial tissue in many species. In the rat, mouse, dog, cat, bull, and deer, both cell types are largely absent from the testicular parenchyma, but are associated with blood vessels in the testicular capsule (Anton et al. 1998). Mast cells are found throughout the interstitial tissue in equine and human testes (Anton et al., 1998, Nistal et al., 1984, Yamanaka et al., 2000), and both mast cells and eosinophils are present in porcine species (Anton et al. 1998). The role of these cells in vascular control and innate immunity can be predicted, but they may also play immunoregulatory roles, as mast cells do in other tissues (Galli et al. 2005).

In humans, testicular mast cell numbers are increased in infertility (Meineke et al., 2000, Nagai et al., 1992, Yamanaka et al., 2000), but decline with advancing age (Nistal et al. 1984). The possibility that they accumulate due to past immune events needs to be considered, but they also appear to be actively regulated. In the adult rat testis, mast cell proliferation is under the control of the Leydig cells, suggesting that these cells produce an inhibitor of mast cell activity (Gaytan et al., 1990, Wang et al., 1994). Neonatal estrogen treatment increases mast cell numbers in the rat testis (Gaytan et al., 1989, Gaytan et al., 1990), and a relationship with elevated Leydig cell aromatase activity is suggested by the prevalence of these cells in boar and equine testes (Eisenhauer et al., 1994, Raeside and Renaud, 1983). Neutrophils are found in the testis only under conditions of testicular inflammation or damage (Gerdprasert et al., 2002a, Kohno et al., 1983a, O’Bryan et al., 2000b).

The relatively recent discovery of the TLRs has radically expanded our understanding of innate immunity and promises to do the same for testicular immunology. In addition to their predictable expression on testicular macrophages and dendritic cells, TLR mRNA, protein, and/or ligand responses have been described in several other testicular cells. Rat and mouse Sertoli cells express TLR2, 3, 4, and 5 (Bhushan et al., 2008, Riccioli et al., 2006, Starace et al., 2008, Wu et al., 2008). TLR1, 10, and 11 mRNA expression has been observed in rat Sertoli cells (Bhushan et al. 2008), and low levels of TLR6, 7, and 13 mRNA have been detected in mouse Sertoli cells (Riccioli et al., 2006, Wu et al., 2008). In the rat, germ cells also produce mRNA for TLR2, 3, and 4 in a stage-specific manner, and the peritubular cells express TLR3 and 11, along with low levels of TLR2, 4, and 6 (Bhushan et al. 2008). Rat Leydig cells express TLR2 mRNA and low levels of TLR10 (Bhushan et al. 2008). However, the cytoplasmic receptors TLR8 and 9 have yet to be observed in any nonmyeloid testicular cell type. Of the other pattern-recognition receptors, Nod1 mRNA has been detected in rat Sertoli cells, peritubular cells, and spermatogonia, while Nod2 was detected in spermatogonia only (Bhushan et al. 2008).

Expression of TLRs and other pattern-recognition receptors in the testis is obviously important for responses toward intratesticular pathogens. Infections and inflammation, both systemic and localized, have negative effects on testicular function, resulting in reduced androgen production, lowered sperm counts, and temporary loss of fertility (Adamopoulos et al., 1978, Baker, 1998, Carlsen et al., 2003). Presumably, TLRs expressed by the Sertoli cells and germ cells are necessary to detect pathogens that ascend the reproductive tract and, as a consequence, lie behind the blood–testis barrier. This expression within the seminiferous epithelium also suggests a mechanism whereby inflammation can directly inhibit spermatogenesis. Activation of Sertoli cells by TLR ligands induces production of inflammatory mediators, such as IL1α, IL6, and NO, which have direct effects on germ cell development and Leydig cell steroidogenesis (Gérard et al., 1992, Stéphan et al., 1995, Stéphan et al., 1997, O’Bryan and Hedger, 2008, Riccioli et al., 2006, Wu et al., 2008). Moreover, while inhibition of Leydig cell steroidogenesis during inflammation may involve indirect effects at the level of the pituitary, or the actions of inflammatory mediators produced by the testicular macrophages and Sertoli cells, there is evidence that these cells can also respond directly to TLR ligands (Bhushan et al., 2008, Xiong and Hales, 1993a).

Although the TLRs have evolved to recognize pathogen-derived molecules, such as LPS (TLR4) and bacterial lipoproteins (TLR2), recent data show that they can also be activated by certain endogenous molecules. The ability of preparations of some endogenous molecules, such as heat-shock proteins, to activate TLR signaling is now attributed to contamination of these preparations by bacterial products (Tsan and Gao 2004), but there is convincing evidence that mammalian mRNA is a ligand for TLR3 (Karikó et al. 2004) and mammalian CpG DNA can activate TLR9 (Yasuda et al. 2005). The high mobility group box chromosomal protein 1 (HMGB1), produced by both spermatogonia and Sertoli cells, is a TLR2 and 4 ligand (Yu et al., 2006, Zetterström et al., 2006). Expression of TLRs by the Sertoli cells under normal conditions, therefore, suggests a potential role for these receptors in responding to these endogenous ligands.

IL1 is the best studied of all the cytokines normally expressed in the testis. It is produced in two forms, α and β, which are encoded by separate genes and share approximately 25% sequence homology. Both forms act through the same receptor complex (IL1R1), and exert a similar range of proinflammatory and physiological effects (Dinarello, 1996, O’Neill and Dinarello, 2000). Most IL1 signaling occurs through activation of the MyD88/TRAF and Jnk/p38 MAP kinase pathways, thereby regulating many proinflammatory genes through stimulation of the transcription factors NFκB and activated protein 1 (AP1) (Medzhitov et al. 1998). The IL1s are synthesized as precursor proteins, which are cleaved to produce 17 kDa active proteins. In the case of IL1α, the precursor possesses low-level biological activity, but the IL1β precursor protein is inactive (Black et al., 1988, Watanabe and Kobayashi, 1994). The precursor of IL1β is cleaved by the IL1 converting enzyme (ICE, caspase 1) during the process of secretion, whereas IL1α is cleaved by the calcium-dependent membrane-associated cysteine protease (calpain) or by extracellular proteases (Thornberry and Molineaux, 1995, Watanabe and Kobayashi, 1994). IL1β is produced by activated monocyte/macrophages and is the main secreted form during inflammation, whereas IL1α is more commonly found within the cell, where it may act as an autocrine or paracrine growth factor involved in direct cell-to-cell communication.

In addition to the prototypical IL1α and IL1β, the IL1 family comprises a number of structurally related proteins, which appear to have arisen by gene duplication (Dunn et al. 2001). The closest structurally to IL1α/β is IL18, which is processed by caspase 1 activity, but acts via a separate receptor rather than IL1R1 (Gracie et al. 2003). Another member of the family, which does bind to IL1R1 but lacks the ability to transduce a signal, blocks IL1α/β action and is called IL1 receptor antagonist (IL1ra) (Arend 1991).

Production of IL1α is constitutively high in the normal testis, as IL1α is produced by the Sertoli cells in a cyclical manner within the mature seminiferous epithelium (Jonsson et al., 1999, Söder et al., 1991). Its production is driven by the developing germ cells and by phagocytosis of the residual cytoplasm cast off by the spermatids at the time of their release into the tubule lumen (Gérard et al., 1992, Syed et al., 1995). Sertoli cell IL1α production is also stimulated by LPS, but not by FSH (Stéphan et al. 1997). Production of IL1β by the normal testis is low, but is upregulated in a subset of testicular macrophages and the Leydig cells during inflammation (Gérard et al., 1991, Jonsson et al., 1999, O’Bryan et al., 2005). Curiously, the majority of this secreted testicular IL1β exists as the precursor (O’Bryan et al. 2005).

The receptor IL1R1 has been localized to germ cells and Sertoli cells (Gomez et al. 1997). IL1α stimulates DNA synthesis in spermatogonia (i.e., mitosis) and early spermatocytes (i.e., meiosis), proliferation of immature Sertoli cells, and a number of activities of the mature Sertoli cell that support spermatogenesis, such as the production of lactate and transferrin (Hakovirta et al., 1993b, Hoeben et al., 1996, Nehar et al., 1998, Parvinen et al., 1991, Pöllänen et al., 1989, Söder et al., 1991). It also regulates the ability of the Sertoli cell to maintain contact with other Sertoli cells and the developing germ cells, by altering the Sertoli cell cytoskeleton (Sarkar et al. 2008). In the interstitium, both IL1α and IL1β suppress LH-stimulated androgen production by adult Leydig cells through inhibition of steroidogenic enzyme activity (Hales, 1992, Xiong and Hales, 1994, Xiong and Hales, 1997). IL1ra is produced by the Sertoli cell, and is stimulated by FSH, IL1, and LPS (Zeyse et al. 2000). Presumably, its role is to regulate the activity of IL1α/β within the testis.

Both IL18 and its receptor have been found in the seminiferous epithelium, with IL18 mRNA and protein localized to spermatocytes and round spermatids (Strand et al. 2005). As is the case for IL1β, the majority of IL18 produced within the testis is in the precursor form, indicating that the processing of these cytokines by caspase 1 in the testis is an area requiring more investigation. IL18 stimulates spermatogonial DNA synthesis in cultures of rat seminiferous tubules, without influencing germ cell apoptosis (Strand et al. 2005). Another member of the family localized to the testis is IL1F8, although its function there is entirely unknown (Kumar et al. 2000).

TNFα and FasL are transmembrane and secreted cytokines with typically trimeric structures and they interact with specific receptors to mediate a cell death signal (Ju et al., 1995, Schütze et al., 2008). They are implicated in both immunoregulation and the control of normal spermatogenic function in the testis.

TNFα is a 17 kDa glycosylated polypeptide and is a major early product of activated monocytes and macrophages. It binds to either of two TNF receptor subsets (TNFR1 and TNFR2) and plays a central role in initiating the inflammatory response by stimulating the production of IL1 and IL6 (Basak and Hoffmann, 2008, Bradley, 2008). Whether TNFα exerts proinflammatory or cytotoxic effects depends on the receptor subtype engaged and the expression of specific adaptor proteins within the target cell (Basak and Hoffmann, 2008, Chung et al., 2002, Hsu et al., 1996, Mak and Yeh, 2002). Consistent with its name, TNFα exerts a cell death signal via TNFR1, through interaction with the TNFR-associated death domain protein (TRADD) or the Fas-associated death domain protein (FADD), and activation of the caspase-dependent apoptotic pathway (Schütze et al. 2008). However, interaction with TRADD can also result in recruitment of other adaptor proteins leading to the binding of TRAF2 and activation of the NFκB pathway and/or the MAP kinases Jnk or p38 (Chung et al. 2002). Moreover, TNFR2 and its related receptors, which include CD40, do not contain a death domain, and instead associate with the TRAFs leading to activation of cell signaling events (Bradley, 2008, Mak and Yeh, 2002).

In the mouse testis, TNFα is expressed by round spermatids, pachytene spermatocytes, and testicular macrophages, and mRNA for TNFR1 has been located on both Sertoli and Leydig cells (De et al. 1993). In porcine Sertoli cells, TNFα receptor subunit protein expression is stimulated by FSH (Mauduit et al. 1996). There is no evidence that TNFα is produced by the Sertoli cell, but expression of TNFα by the germ cells within the seminiferous epithelium, like that of IL1α and IL6, is cyclical. Within the seminiferous epithelium, TNFα produced by the germ cells appears to play a complex role in the control of both Sertoli cell function and spermatogenesis. TNFα reduces spontaneous germ cell degeneration in cultured human and rat seminiferous tubules, suggesting a germ cell survival effect presumably mediated through the Sertoli cell (Pentikäinen et al., 2001, Suominen et al., 2004). Conversely, TNFα disrupts Sertoli cell tight junction assembly by inhibiting the production of junction proteins and by regulating matrix metalloprotease and protease inhibitor activity (Siu et al. 2003). TNFα also stimulates plasminogen activator inhibitor expression in rat testicular peritubular cells (Le Magueresse-Battistoni et al. 1997). Similar to IL1, TNFα stimulates basal lactate production by cultured Sertoli cells, but TNFα generally antagonizes the actions of FSH on Sertoli cell function, including the stimulation of aromatase activity and lactate production (Mauduit et al., 1993, Nehar et al., 1997, Riera et al., 2001). However, Delfino et al. (2003) have shown that TNFα stimulates androgen receptor expression in Sertoli cells via upregulation of NFκB, which binds to several enhancer motifs in the androgen receptor promoter. TNFα also stimulates the expression of inflammatory cytokines, chemokines, and leukocyte adhesion molecules in both Sertoli cells and peritubular cells (Aubry et al., 2000, De Cesaris et al., 1998, Stéphan et al., 1997). In testicular pathology, TNFα has been implicated as a major causative agent in the development of experimental autoimmune orchitis (EAO) (Yule and Tung 1993). There is a significant increase in the number of TNFα-positive testicular macrophages and the number of TNFR1-positive germ cells undergoing apoptosis during EAO in rats (Suescun et al. 2003). Within the interstitium, TNFα acts as an effective regulator of Leydig cell steroidogenesis, by inhibiting LH receptor binding and steroidogenic gene expression (Li et al., 1995, Mauduit et al., 1991b, Mauduit et al., 1998, Xiong and Hales, 1993b, Xiong and Hales, 1994).

No intratesticular cytokine has excited more controversy than the cell death signaling ligand FasL. Interactions between FasL and its receptor on activated T cells are important in the regulation of the immune response (Ju et al. 1995). The death domain in the cytoplasmic region of the Fas receptor recruits the FADD adaptor protein and induces T cell death via caspase-dependent apoptosis (Schütze et al. 2008). There is evidence that FasL expression on epithelial cells in immune-privileged tissues, including Sertoli cells, may be responsible for deleting activated T cells within the tissue, thereby suppressing adaptive immunity (Bellgrau et al., 1995, Griffith et al., 1996, Saas et al., 1997, Stuart et al., 1997). However, this hypothesis has been challenged by the observation that increased FasL expression on cells such as tumor cell lines and pancreatic islet cells does not increase protection in transplantation studies, but actually provokes a massive inflammatory reaction and rejection (Allison et al., 1997, Kang et al., 1997). Moreover, FasL appears to be abundantly expressed in the epithelia of several human tissues that lack any evidence for immunological privilege (Xerri et al. 1997), while absence or inhibition of Fas or FasL does not automatically cause failure of immune privilege (Rogers et al., 1998, Suarez-Pinzon et al., 2000, Wahlsten et al., 2000). It seems unlikely that FasL expression on its own is a fundamental explanation for immune privilege, although it may contribute through interaction with other immunoregulatory processes (Chen et al. 1998).

Localization of FasL and Fas in the testis under normal conditions also has proven controversial, which may be attributed to differences in detection methods, limitations of some of the reagents that have been used, and the fact that these molecules are inducible (D’Alessio et al., 2001, Restifo, 2000). Studies have shown FasL to be present in rat, mouse, porcine, and human Sertoli cells and absent in most germ cells (Bellgrau et al., 1995, Brændstrup et al., 1999, D’Abrizio et al., 2004, Lee et al., 1997), but other studies have reported that FasL expression in the rat seminiferous epithelium is confined to the germ cells (Celik-Ozenci et al., 2006, D’Alessio et al., 2001). The receptor Fas has been found on isolated mouse Sertoli cells (Riccioli et al. 2000), but in intact testes has been localized to spermatogonia and spermatocytes from the pubertal period onward (Celik-Ozenci et al., 2006, Lee et al., 1997, Lizama et al., 2007, Ogi et al., 1998). In germ cells, Fas expression is associated with cells that are undergoing apoptosis (Lee et al., 1997, Lizama et al., 2007), and Fas is induced by TNFα and IFNγ in the Sertoli cell (Riccioli et al. 2000). Curiously, in one study using fragment cultures of human seminiferous tubules, TNFα downregulated FasL expression and inhibited apoptosis of germ cells (Pentikäinen et al. 2001). Less controversially, Fas and FasL expression is upregulated in various models of seminiferous epithelium damage, indicating that this mechanism is important in regulating germ cell apoptosis in cases of physical and toxicological insult (Lee et al., 1997, Ogi et al., 1998).

Members of the IL6 family of cytokines exert their actions via binding to specific receptors that associate with a common membrane signal transducer gp130, leading to the activation of the Janus kinase/signal transducers and activators of transcription (Jak/Stat) and MAP kinase cascades (Heinrich et al. 2003). In addition to their functions in inflammation and the immune response, these cytokines play crucial roles in hematopoiesis, liver and neuronal regeneration, embryonic development, and fertility. Members of this family that have been detected in the testis are leukemia inhibitory factor, oncostatin M, ciliary neurotropic factor, IL11, and IL6 itself (de Miguel et al., 1996, de Miguel et al., 1997, Du et al., 1996, Jenab and Morris, 1998, Okuda et al., 1994, Stéphan et al., 1997). The principal roles of most of these cytokines in the testis appear to lie in regulating Leydig cell development and the onset of spermatogenesis. A role in testicular immunology is most likely for IL6, which possesses both proinflammatory and anti-inflammatory properties, and regulates the acute-phase response as well as several aspects of immune cell development and activity (Kopf et al., 1994, Tilg et al., 1997).

Sertoli cells, Leydig cells, and peritubular cells have been shown to produce IL6 in vitro (Cudicini et al., 1997b, Okuda et al., 1994, Syed et al., 1993, Syed et al., 1995). In rat Sertoli cells, IL6 production is stimulated by FSH, testosterone, phagocytosis, and other inflammatory stimuli, including IL1α, IL1β, TNFα, and LPS (Cudicini et al., 1997a, Cudicini et al., 1997b, Okuda et al., 1994, Okuda et al., 1995, Stéphan et al., 1997, Syed et al., 1995). Production of IL6 by cultured mouse Sertoli cells is inhibited by INFγ (Riccioli et al., 2000, Stéphan et al., 1997). Within the seminiferous epithelium, endogenous production of IL6 follows a cyclical pattern that corresponds with the changes in the stages of the spermatogenic cycle, most probably under the influence of IL1α and FSH (Hakovirta et al., 1995, Syed et al., 1993, Syed et al., 1995). Both the IL6-specific receptor subunit (IL6R) and the gp130 mRNA are expressed in rat Sertoli cells, and are stimulated by IL1 and IL6, but only the IL6R subunit is stimulated by FSH (Fujisawa et al. 2002). IL6 increases basal and FSH-induced transferrin and cyclic GMP secretion by the Sertoli cell, and inhibits meiotic DNA synthesis in preleptotene spermatocytes (Boockfor and Schwarz, 1991, Hakovirta et al., 1995, Hoeben et al., 1997). In models of EAO, IL6 appears to play an ameliorative or protective role within the seminiferous epithelium (Li et al., 2002, Rival et al., 2006b). Leydig cells are also active producers of IL6 following stimulation by LH, LPS, or IL1β in vitro, and evidence suggests that these cells actually may be the major testicular source (Boockfor et al., 1994, Cudicini et al., 1997b, Okuda et al., 1994, Okuda et al., 1995).

Several studies have implicated IL10 as a key player in testicular immunology. Testicular IL10 production is preferentially induced by systemic LPS treatment in the adult rat testis (O’Bryan et al. 2005), and macrophages and T cells isolated from the rat testis constitutively produce IL10, which may contribute to testicular immune privilege (Hedger, unpublished data). In a mouse model of EAO, elevation of endogenous IL10 levels by adenoviral-mediated transfection of human IL10 was found to significantly reduce the incidence of immune activation, orchitis, and spermatogenic damage (Watanabe et al. 2005). Although male mice deficient in IL10 do not show evidence of male fertility problems, intratesticular immune responses have yet to be examined in these animals (Kühn et al. 1993).

Production of ROS, particularly by macrophages, is an important component of the early response to infection (Takemura and Werb 1984). Significant ROS include the superoxide anion (O2−), hydrogen peroxide (H₂O₂), the hydroxyl radical (HO·), nitric oxide (NO·), and the peroxynitrite anion (ONOO⁻). Most are products of normal cellular metabolism, but their production is stimulated during inflammation through the activity of the enzymes NADPH oxidase (nicotinamide adenine dinucleotide phosphate-oxidase) and iNOS (Forman and Torres, 2002, Nathan, 2006). These small molecules are cytotoxic to many microorganisms, although some play a more complex role in cell signaling processes. They also interact with crucial intracellular macromolecules, such as proteins, lipids, and DNA, causing oxidative damage. Although there are endogenous repair systems to correct oxidative damage, excessive production of ROS contributes to the pathology of many diseases.

Among the ROS, nitric oxide represents a special case. At high levels, NO is highly cytotoxic, especially as a precursor of the peroxynitrite anion, but at low levels NO acts as an intracellular and extracellular signaling molecule (Bogdan, 2001, Schmidt and Walter, 1994). It is produced by one of three related enzymes, neuronal NOS (nNOS or NOS1), inducible NOS (iNOS or NOS2), and endothelial NOS (eNOS or NOS3), which catalyze the conversion of l-arginine to l-citrulline and NO (Alderton et al. 2001). The NOS enzymes are homodimeric proteins encoded by three separate genes (Table 2). Both nNOS and eNOS are constitutively expressed enzymes whose activity is regulated through a calcium–calmodulin-mediated mechanism, whereas iNOS is a constitutively activated enzyme that is regulated at the transcriptional and translational levels by various inflammatory mediators, including LPS, IL1, and TNFα (Wolkow 1998). Table 2

Major inflammatory enzymes – properties and distribution in the testis

In studies in various species, NOS has been found in Sertoli cells, Leydig cells, peritubular cells, germ cells, testicular macrophages, and vascular endothelial cells (Kim et al., 2007, Lee and Cheng, 2004). The nNOS gene also produces a testis-specific isoform, TnNOS, which has been localized to Leydig cells (Wang et al. 1997). In the normal rat seminiferous epithelium, iNOS is expressed by elongating spermatids and pachytene spermatocytes, particularly during the stages immediately following sperm release, with relatively lower levels of expression in Sertoli cells and peritubular cells throughout the entire cycle (O’Bryan et al. 2000a). In the interstitium, macrophage expression of iNOS appears to be largely confined to the minority of CD163-negative macrophages of the testis, and is not detectable in the majority of resident macrophages even during inflammation (O’Bryan et al., 2000a, Gerdprasert et al., 2002a). Testicular cell iNOS expression and NO production are increased by inflammatory events induced by LPS, testicular torsion, or testicular heating (Lue et al., 2003, O’Bryan et al., 2000a, Shiraishi et al., 2001).

Production of NO by germ cells, in particular, is implicated in the control of the formation and disassembly of the Sertoli cell junctions that constitute the blood–testis barrier, as well as the junctional complexes involved in Sertoli–germ cell adhesion (Lee and Cheng, 2003, Lee et al., 2005). Moreover, pachytene and round spermatid apoptosis is significantly reduced in iNOS null mice, leading to an increase in daily sperm output and indicating a key role for iNOS in limiting germ cell survival and/or the carrying capacity of the Sertoli cells (Lue et al. 2003). NO inhibits Leydig cell steroidogenesis directly, an action that is probably mediated through oxidative damage, and treatment with NOS inhibitors counteracts the normal decrease in testosterone associated with sepsis or stress (Del Punta et al. 1996; Sharma et al. 1998; Welch et al. 1995). Similarly, production of other ROS has been implicated in the loss of steroidogenic function in various inflammatory models and in Leydig cell cultures (Georgiou et al., 1987, Quinn and Payne, 1984). Last but not least, NO is a potent vasodilator, and plays a role in the endogenous control of testicular blood flow and formation of the interstitial fluid (Lissbrant et al., 1997, O’Bryan et al., 2000a). Production of NO is an important mediator of germ cell death in testicular torsion (Moon et al., 2005, Shiraishi et al., 2001), and may be involved in the testicular damage associated with varicocele (Santoro et al., 2001, Türker Köksal et al., 2004).

The IFNs are functionally related cytokines with antiviral activity and they comprise three main groups (α, β, and γ), based on their structural relationships and major cellular sources (Hertzog et al., 2003, Malmgaard, 2004). The type I IFNs (IFNα and β) are produced by a variety of cell types in addition to leukocytes, and exert primarily antiviral and antiproliferative effects via the IFNα receptor (IFNAR). The type II IFNγ is produced by NK and NKT cells, activated T cells and, under certain conditions, by macrophages and dendritic cells (Schoenborn and Wilson, 2007, Varma et al., 2002). It acts through a separate receptor, and regulates the activity of APCs as part of the adaptive immune response in addition to its antiviral functions. Hence, IFNγ is a type II IFN, but a type 1 cytokine.

Type I IFN production is stimulated through TLR3, which detects double-stranded viral RNA, and TLR4, the bacterial LPS receptor (Hertzog et al. 2003). These two TLRs, which are expressed in the testis (Bhushan et al., 2008, Riccioli et al., 2006, Starace et al., 2008, Wu et al., 2008), are able to induce the IFN transcription factor IRF3, via engagement of the TRIF adapter protein (Akira and Takeda, 2004, Hertzog et al., 2003, O’Neill and Bowie, 2007). The regulation of IFNγ is cell specific and complex, but involves Jak/Stat signaling and various transcription factors, including NFκB and AP1 (Schoenborn and Wilson, 2007, Varma et al., 2002). Its production is stimulated by type I IFNs and by IL18, among other cytokines. Viral infections stimulate type 1 IFN and, somewhat more surprisingly, IFNγ production by Sertoli cells, peritubular cells, Leydig cells, and testicular macrophages, leading to typical antiviral responses in the testis (Dejucq et al., 1995, Dejucq et al., 1998a, Dejucq et al., 1998b, Hu et al., 1998, Starace et al., 2008). A role for IFNγ in maintaining testicular immune privilege has been indicated by the observation that mouse Sertoli cells upregulate their expression of the negative costimulatory ligand B7-H1 in response to IFNγ in vitro, but remain devoid of positive costimulatory molecules (Dal Secco et al. 2008). On the other hand, in vivo studies have identified IFNγ as a causative factor in EAO in mice (Itoh et al. 1998b). Moreover, IFNγ stimulates Sertoli cell production of Fas and caspase 1, and is implicated as the mediator of Sertoli cell and germ cell apoptosis under various conditions (Kanzaki and Morris, 1998, Riccioli et al., 2000). These disparate observations are indicative of a complex relationship between IFNγ, the local immune system, and spermatogenesis.

IFN production also affects testicular steroidogenesis. Normal healthy men treated with human IFNα had significantly decreased serum testosterone levels in conjunction with normal serum gonadotropins, and both IFNα and IFNγ inhibit testosterone production in primary cultures of Leydig cells (Orava et al. 1986). Studies using porcine Leydig cells indicate that IFNγ exerts its inhibitory effect on testosterone production at the level of cholesterol transport into the mitochondria, and inhibits expression of the steroidogenic acute regulatory (StAR) protein and the P-450 steroidogenic enzymes, cholesterol side-chain cleavage complex (P450scc) and 17α-hydroxylase/C17–C20 lyase (P450c17) (Orava et al., 1985, Orava et al., 1989). These data indicate that IFNs may contribute to the decline in steroidogenic function of patients with viral infections, in particular. However, Dejucq et al., 1995, Dejucq et al., 1998a have shown that an increase in IFNα and IFNγ expression in Sertoli and Leydig cells of rats infected with Sendai virus was actually associated with an increase in testosterone production. These results suggest that indirect or secondary stimulatory effects on testicular steroidogenesis may be involved, providing further evidence that IFNs play a complex role in testicular function.

The three TGFβ family members that are expressed in mammals (TGFβ₁, TGFβ₂, and TGFβ₃) are the archetypes of a much larger superfamily of mostly homo- and heterodimeric proteins, which includes the activins, the bone morphogenetic proteins (BMPs), inhibin, and anti-Müllerian hormone (Lin et al. 2006). The TGFβs themselves are multifunctional growth and differentiation factors involved in many aspects of tissue remodeling and repair as well as regulation of the immune system (Ashcroft, 1999, Chang et al., 2001, Licona-Limón and Soldevila, 2007). Nearly all cells synthesize a form of TGFβ and possess functional receptors for these cytokines. Most TGFβ family ligands act via dual transmembrane serine/threonine kinase receptors, called type I and type II, which interact upon ligand binding (Cárcamo et al., 1994, Ebner et al., 1993, Massagué et al., 1992). Signals are transduced from the membrane to the nucleus through phosphorylation of the regulatory Smad signaling proteins, with Smads 2, 3, and 4 responsible for mediating TGFβ and activin signaling (Datto et al., 1999, de Caestecker et al., 1998).

All three mammalian TGFβ forms are differentially expressed by Sertoli cells, peritubular cells, and Leydig cells in the fetal and immature testis, but production declines considerably during sexual maturation (Avallet et al., 1994, Gautier et al., 1994, Mullaney and Skinner, 1993, Olaso et al., 1997). In the postpubertal testis, they have also been localized to the germ cells (Caussanel et al., 1997, Teerds and Dorrington, 1993), and TGFβ receptors are found in both somatic and germ cells (Caussanel et al., 1997, Goddard et al., 2000, Le Magueresse-Battistoni et al., 1995). The TGFβs appear to be involved in testicular development by controlling, among other things, apoptosis of undifferentiated spermatogonia (gonocytes) (Olaso et al. 1998), seminiferous tubule formation, and Leydig cell differentiation (Cupp et al., 1999, Dickson et al., 2002, Khan et al., 1992a, Konrad et al., 2000). In the adult testis, TGFβ₂ and TGFβ₃ regulate Sertoli cell tight junction dynamics, indicating a role in regulating the permeability of the blood–testis barrier to facilitate the passage of spermatocytes across the barrier during spermatogenesis (Lui et al., 2003, Xia et al., 2006). However, the TGFβs are also potent inhibitors of the activity of immune cells, especially T cells, B cells, and NK cells (Ahmad et al., 1997, Ashcroft, 1999, Cousins et al., 1991, Licona-Limón and Soldevila, 2007, Wahl et al., 2006), suggesting a prominent role in creating testicular immune privilege. Accordingly, TGFβ₁ contributes to the lymphocyte inhibitory activity of testis extracts (Pöllänen et al. 1993) and to the immunoprotective properties of Sertoli cells in cotransplantation studies (Suarez-Pinzon et al. 2000).

Activins are homodimers or heterodimers of homologous subunits, designated β_A–β_E, which are structurally related to the TGFβs (Chang et al., 2001, de Kretser et al., 2002). Homodimers of β_A form activin A, which is widely expressed and has been extensively studied. Relatively less is known about the other activin forms, which appear to be both less abundant and less widely distributed. Activin A is a multifunctional growth factor and inflammatory regulator (Phillips et al. 2001), although activins were originally named for the ability of activin A and B, in particular, to stimulate pituitary FSH secretion (Corrigan et al., 1991, Ling et al., 1986). Heterodimers of either β_A or β_B subunits with a homologous α subunit, are called inhibin A and inhibin B, respectively, and act as feedback inhibitors of FSH secretion from the pituitary (Robertson et al. 1986). Unlike the activins, inhibins are not widely produced, and their chief source is the Sertoli cell, the cellular target of FSH action (de Kretser et al., 2004, Meunier et al., 1988).

Activin functions are highly regulated. Like the TGFβs, they act via specific type I and type II transmembrane serine/threonine kinase receptors, linked to the Smad signaling pathway (Ethier and Findlay, 2001, Vale et al., 2004). The inhibins are competitive inhibitors of activin receptor binding, but activin homodimers and heterodimers comprising the β_B and β_C subunits also appear to act as weak competitive agonists of activin A (Brown et al., 2000, Makanji et al., 2008, Mellor et al., 2003). These interactions are modulated and/or facilitated by specific inhibitory coreceptor proteins, such as betaglycan and the BMP and activin membrane-bound inhibitor (BAMBI) (Lewis et al., 2000, Makanji et al., 2008, Onichtchouk et al., 1999). Moreover, activin bioactivity can be effectively neutralized in the circulation by the high-affinity activin binding protein, follistatin (Nakamura et al. 1991).

Activin A is abundantly produced in the testis, particularly during early testicular development (Barakat et al., 2008, Buzzard et al., 2004). Even in the adult rat, intratesticular fluid levels of activin A are 5–10 times higher than normal circulating concentrations (O’Bryan et al. 2005). The Sertoli cells appear to be the main source in the adult testis, although the βA subunit is expressed in most germ cells (Buzzard et al., 2004, Kaipia et al., 1992), and activin A protein is present in the resident macrophages and mast cells (Okuma et al., 2005b, Okuma et al., 2006). Activin A is expressed at low levels throughout the cycle of the seminiferous epithelium, but there is a distinct peak of production immediately following spermiation, which is driven by the surge of ILα produced by the Sertoli cell at this time (Kaipia et al., 1992, Okuma et al., 2006). Activin receptors are expressed by most, if not all, of the somatic and germ cells in the testis (Cameron et al., 1994, de Winter et al., 1992, Feng et al., 1993, Kaipia et al., 1993). Activin A exerts a complex regulation of germ cell, Sertoli cell, and Leydig cell proliferation and/or differentiation during development and in the adult (Barakat et al., 2008, Buzzard et al., 2003, Mather et al., 1990, Mauduit et al., 1991a, Meehan et al., 2000, Meinhardt et al., 2000), and disorders of activin signaling are implicated in the onset of testicular cancer (Dias et al. 2008).

Activin A is produced by activated monocytes, macrophages, dendritic cells, bone marrow stromal cells, and some lymphocytes (Erämaa et al., 1992, Ogawa et al., 2006, Robson et al., 2008, Yamashita et al., 1992), and plays a facilitating role in early inflammation responses (Jones et al. 2007). Consistent with its homology to the TGFβ proteins, however, activin A also possesses a repertoire of anti-inflammatory and immunoregulatory activities, and displays characteristics of a type 2 cytokine (Hedger et al., 1989, Ogawa et al., 2006, Wang et al., 2008, Yu et al., 1998). In various experimental systems, activin A has been found to antagonize the production and actions of IL1 and IL6, to inhibit critical T cell and B cell activation responses, and to induce macrophages to polarize toward the M2 phenotype (Brosh et al., 1995, Gribi et al., 2001, Ogawa et al., 2006, Russell et al., 1999). Both Sertoli cells and testicular macrophages in culture respond to LPS by increasing activin A production (Okuma et al. 2005a) (Hedger, unpublished data), and IL1 is a very potent stimulus for activin A production by the Sertoli cell (Okuma et al., 2005b, Okuma et al., 2006). The actual role of activin A in testicular inflammation remains unclear, since the high intratesticular levels of activin A are not affected by LPS treatment in vivo (O’Bryan et al. 2005), but a role for activin A in modulating testicular immune responses can be predicted.

Androgens produced by the Leydig cells have immunosuppressive properties that contribute to differences in immunity between the sexes (Cutolo et al., 2004, Miller and Hunt, 1996). The majority of these effects are believed to be exerted at the level of the immune tissues, rather than by direct effects on circulating leukocytes, which lack classical androgen receptors (Grossman et al., 1979, Sasson and Mayer, 1981). Recently, however, it has been discovered that steroids can interact with membrane-bound G-protein-coupled receptors to trigger nongenomic responses (Braun and Thomas, 2004, Rahman and Christian, 2007). Studies have shown that androgens alter [Ca] fluxes in lymphocytes and macrophages via such membrane-mediated interactions, affecting gene expression and function in the target cells (Benten et al., 2004, Walker, 2003, Wunderlich et al., 2002). Although studies of the role of androgens in suppressing immune responses, such as graft rejection, in the testis have been somewhat equivocal (Cameron et al., 1990, Selawry and Whittington, 1988, Whitmore and Gittes, 1978), the observation that immune cells can respond directly to androgens has resurrected the intriguing possibility that androgens, and possibly other testicular steroids, may directly regulate immune cell activity within the testis and adjacent draining lymph nodes.

Glucocorticoids play an important role in modulating both testicular immunity and steroidogenesis via a complex extratesticular regulatory loop. During inflammation, proinflammatory cytokines activate the hypothalamic–pituitary–adrenal axis leading to increased secretion of glucocorticoids, which act through the ubiquitously expressed glucocorticoid receptors (Buckingham et al., 1996, Sapolsky et al., 2000). Glucocorticoids suppress innate and acquired immunity at multiple levels and are an essential control in the inflammatory/immune response. They inhibit the inflammatory response primarily by repression of NFκB, suppressing the production and actions of the proinflammatory cytokines, reducing adhesion molecule expression, and stimulating the production of anti-inflammatory cytokines (Auphan et al., 1995, Brack et al., 1997, Kapcala et al., 1995). Glucocorticoids also exert inhibitory effects at all levels of the hypothalamic–pituitary–Leydig cell axis (Bambino and Hsueh, 1981, Hales and Payne, 1989, Hardy et al., 2005, Monder et al., 1994), thereby contributing to the inhibition of androgen production that accompanies local and systemic inflammatory events.

Prostaglandins (PGD, PGE, and PGF), prostacyclins (PGI), and thromboxanes (Tx), collectively called the prostanoids, are fundamental to many physiological processes, including inflammation and immunity. Prostanoids arise from hydrolysis of membrane glycerophospholipids to release the free fatty acid arachidonic acid through the action of one of the more than 20 enzymes with phospholipase A₂ (PLA₂) activity (Kudo and Murakami, 2002, Schaloske and Dennis, 2006). Arachidonic acid is in turn directed down the prostanoid synthetic pathway by the action of the cyclooxygenase (COX; also called prostaglandin-endoperoxide synthase) enzymes (Figure 3). There are two COX enzymes, which are the products of different genes: the constitutively expressed COX1 and an inflammation-induced form, COX2 (Table 2) (Smith et al., 2000, Tanabe and Tohnai, 2002). While conversion of arachidonic acid to the prostanoid precursor PGH₂ by COX is the rate-limiting step, production of specific prostanoids depends on the activity of the prostaglandin synthases, that is, PGE synthase (PGES), PGIS, PGDS, PGFS, and thromboxane A synthase (TxAS) (Helliwell et al., 2004, Ueno et al., 2005). Each prostanoid acts via one or more specific receptors to regulate cell growth, vascular smooth muscle constriction or relaxation, vascular permeability, and immune cell activity (Bos et al., 2004, Hata et al., 1990). Several prostanoids exert both proinflammatory and immunosuppressive actions through different receptor interactions, or through metabolism to other active immunoregulatory metabolites, such as 15-deoxy-PGJ₂, a ligand for the peroxisome proliferator-activated receptor γ (PPARγ) (Jiang et al. 1998). An alternative synthetic pathway, catalyzed by the lipoxygenases, converts arachidonic acid to the related leukotrienes and lipoxins, which also possess diverse metabolic, vascular, and inflammation-regulating actions (Samuelsson et al., 1987, Takano et al., 1997). Figure 3

Prostanoid and lysophosphatidylcholine synthesis and targets. Synthesis of the prostanoids commences through the action of phospholipase A₂ (PLA₂) on arachidonyl-containing phospholipids (usually a 1-acyl-2-arachidonyl-glycerophospholipid) to produce the free fatty acid arachidonic acid and the corresponding lysophospholipid. Arachidonic acid is converted into the prostanoid precursor prostaglandin H₂ (PGH₂) by the action of cyclooxygenases and this is further converted into the various prostanoid subtypes, prostaglandins (PGD, PGE, and PGF), prostacyclins (PGI), and thromboxanes (Tx), by the action of specific prostaglandin and thromboxane synthases. Synthesis of lipoxins and leukotrienes from arachidonic acid occurs via the activity of a separate lipoxygenase enzyme (not shown). The prostanoids interact with their various receptor subtypes: there are two PGD receptors (DP1–2) and four PGE receptors (EP1–4). It is the cellular expression of these receptor subtypes that ultimately determines the eventual biological responses; for example, different prostanoid and receptor combinations may produce either proinflammatory or anti-inflammatory responses, or may have opposing vasodilatory and vasoconstrictive effects. When the phospholipid is a phosphatidylcholine, a lysophosphatidylcholine (LPC) is formed, which may interact with cells via G-protein-coupled receptor-mediated or direct membrane interactions, or may undergo further enzymatic modification to form other bioactive lipids.

Inhibition of complement activation is an important mechanism for reducing transplant rejection, particularly in the early phase. Membrane complement inhibitors, such as CD46, CD55, CD59, and the complement receptor-related protein (Crry), are expressed in several immune-privileged tissues and have been implicated in suppression of inflammation in the eye and placenta (Bora et al., 1993, Bulla et al., 2005). CD46 and CD55 appear to be confined to spermatids in the adult rat testis, Crry is expressed on early spermatocytes and interstitial cells, but CD59 is widely expressed throughout the seminiferous epithelium and interstitium (Mizuno et al. 2006). A role for CD59 expression by the Sertoli cells in prolonging the survival of these cells in transplantation studies has been suggested (Lee et al. 2007). Moreover, complement regulators both in seminal plasma and on the surface of the spermatozoa (CD46, CD55, and CD59) may contribute to complement evasion by the spermatozoa in the female tract (Harris et al. 2006).

Classical MHC class I molecules (HLA-A, HLA-B, and HLA-C) are expressed on nearly all cell types, and facilitate antigen presentation by these cells and activation of cytotoxic T cells (Cresswell et al., 1999, Janeway, 1992, Sant et al., 1999). By contrast, the nonclassical MHC class Ib molecules (HLA-E, -F, and -G) are much less widely expressed, and are associated with suppression of the adaptive immune response. Membrane-bound and soluble variants of these MHC class Ib molecules have been implicated in the apoptosis of alloreactive cytotoxic T cells, inhibition of NK cell cytolytic activity, suppression of T cell proliferation, and deviation of type 1 to type 2 responses, most notably in pregnancy (Bainbridge et al., 2000, Fournel et al., 2000, Ishitani et al., 2006, Kapasi et al., 2000, Kanai et al., 2001, Riteau et al., 1999). HLA-G has been found in the rhesus monkey testis (Ryan et al., 2002, Slukvin et al., 1999), and HLA-E is expressed on germ cells in the human testis (Fiszer et al. 1997), indicating that these molecules may contribute to immune privilege in the testis.