Abstract
Our knowledge of the immune response to genital tract infection has progressed appreciably in recent years. This review focuses on the innate immune system, in particular the role of Toll-like receptors (TLRs), in controlling genital tract infection. Research into the role of TLRs in recognizing ‘pathogen-associated molecular patterns’ (PAMPS) has provided an important insight into the host's early immune response. TLRs are activated following binding of microbial components leading to cytokine production, which, in turn, stimulate phagocytic and natural killer cells and mobilize T and B lymphocytes of the antigen-specific acquired immune system. The therapeutic use of TLR agonists as topical agents or for improving CD4+ and CD8+ T-cell responses to microbial vaccines is an important area of ongoing research, particularly with respect to genital mucosal infection.
INTRODUCTION
Our understanding of the immune response to genital tract infection has advanced in recent years following the discovery of novel interactions between pathogens and the host cell. The purpose of this review is to highlight some of these advances with respect to the innate immune system, in particular the role of Toll-like receptors (TLRs).
The innate and adaptive immune systems are sequentially activated by infection and work together to eradicate the microbial agent. The innate immune system is the first line of defence against invading microbial pathogens, while the adaptive immune system becomes active 4–7 days after infection and includes a specific and long-lasting immunity involving T and B lymphocytes.
The ‘innate immune system’ is a primitive and evolutionary conserved defence system and relies on a large family of pattern recognition receptors (PRRs), which recognize non-specific pathogen-associated molecular patterns (PAMPS), rather than specific antigens on individual organisms. Host cell soluble proteins and cellular elements can rapidly recognize these PAMPs resulting in an early defence, mostly by phagocytosis. 1 Although innate immunity is unable to specifically recognize pathogens and provide specific immunity, it can instruct the adaptive immune response. For example, macrophages and dendritic cells (DCs) degrade ingested microbes and present the microbial antigens to T-cells in lymph nodes. This leads to a clonal expansion of antigen-specific T-cells with the resulting production of cytokines, which in turn stimulate B-cell proliferation and antibody production. Natural killer (NK) cells have an important role in innate immune responses and recent work has shown an important interaction (so-called ‘NK–DC cross-talk’). 2 As this review will focus predominantly on the role of TLRs in genital tract immunity, other sources are recommended for further information on DCs, NK cells and other aspects of innate immunity as related to the genital tract. 3,4
METHODS
Relevant articles were identified by searching the National Library of Medicine Medline database, accessed through PubMed, using the following keywords in addition to the organism or disease: innate immunity, Toll-like receptor. Key references were also selected from reviewed papers, if not already identified from the original search.
Toll-like receptors
One of the major advances in our understanding of the innate immune system was the discovery of a group of highly conserved receptors, the so-called ‘TLRs’. The first member of the family, Toll, was discovered in the fruit fly, Drosophila malanogaster, where it was shown to be responsible for embryonic development and essential for protection against fungal infection in adult flies. Homologues to Toll have been identified in humans and are considered to play a major role in recognizing PAMPs and hence initiating immune system activation. 5–7 TLRs recognize a wide range of lipid, carbohydrate, peptide and nucleic acid structures that are broadly expressed by different groups of microorganisms. Each TLR consists of an external leucine-rich repeat (LRR) area, which is involved with ligand binding, a transmembrane region and a cytoplasmic Toll/interleukin-1 receptor (TIR). The types of PAMP that each TLR recognizes are very diverse (Box 1).
Toll-like receptor (TLR) ligand recognition
TLR1: May play a minor role in pneumococcal infections, although there is generally little information available on the role of TLR1 in vivo.
TLR2: Has been regarded as the primary Gram-positive TLR that responds to a variety of ligands, including bacterial lipoproteins, lipoarabinomannan, teichoic acid, porins (e.g. Neisserial species) and possibly peptidoglycan; TLR2 appears to interact and form heterodimers with other members of the TLR family, in particular TLR1 and TLR6, allowing discrimination of subtle changes in ligand structure. The TLR1/TLR2 and TLR2/TLR6 heterodimers recognize triacylated and diacylated lipoproteins, respectively.
TLR3: Expressed intracellularly (endosome/lysosome compartment) and recognizes viral double-stranded RNA and the synthetic analogue polyinosine–polycytidylic acid (poly(I:C)).
TLR4: Has been identified as the principle signal transducer in the recognition of lipopolysaccharide (LPS) of Gram-negative bacteria. It also recognizes yeast mannan, viral heat-shock proteins and fibrinogen and streptococcal cytotoxin.
TLR5: Recognizes bacterial flagellin monomers located in a highly conserved region within the flagellum.
TLR6: Recognizes mycobacterial lipoarabinomannan and yeast zymosan, although little information is available of its role in vivo.
TLR7 and TLR8: Intracellular TLRs that recognizes viral single-stranded RNA and imidazoquinolones (e.g. imiquimod). To date, no bacterial ligands have been found.
TLR9: Expressed primarily intracellularly and recognizes bacterial cytosine-phosphate-guanosine (CpG) motifs in Gram-positive and Gram-negative bacterial DNA, including mycobacteria.
TLR10: No ligand has been identified yet for human TLR10.
To date, TLRs 1–10 have been identified in humans and TLRs 1–9 and 11 in mice. Most of the TLRs are expressed at the cell surface; however, TLRs 7–9 appear to function inside the cell and require endosomal maturation in the signalling process. Once the TLR binds to an appropriate ligand, a series of downstream signalling events takes place. Two distinct pathways have been identified in this signalling cascade, one myeloid differentiation factor 88 (MyD88)-dependent and one MyD88-independent, and these, in turn, are selectively activated by four potential adaptor proteins. All TLRs except TLR3 trigger the MyD88-dependent pathway, which results in the translocation of nuclear factor-kappa B (NF-kappaB) into the nucleus with the subsequent expression of proinflammatory and inflammatory cytokines and chemokines and increased expression of costimulatory molecules such as CD40, CD80 and CD86. 8 These cytokines stimulate phagocytic and NK cells and mobilize T and B lymphocytes of the antigen-specific acquired immune system. Human TLRs have been identified with DCs, neutrophils, monocytes/macrophages, fibroblasts and epithelial cells (ECs), including those in the female genital tract. However, the distribution of TLRs is complex in terms of which cells express which TLRs. Recent reports suggest that distinct TLRs can discriminate between different PAMPs from the same organism and that the various TLRs activated by the same microorganism can activate different host immune pathways. This implies differences between the intracellular signal pathways triggered by the TLRs.
In addition to TLRs, there are two groups of structurally related cytosolic receptors that recognize microbial components gaining cell entry. These are the NLR family (nucleotide-binding oligomerization domain [NOD]-like receptor family) and RLR family (RIG [retinoic acid-inducible gene 1]-like receptor family). There is growing evidence of cooperation between different TLRs and between TLRs and other pattern recognition receptors, such as the NLRs and RLRs. 9
Specific genital infections
Chlamydia trachomatis infection
With respect to the innate immune response, Chlamydiae express a variety of ligands that could serve as potential TLR ligands. Lipopolysaccharide (LPS) (previously known as endotoxin) is the main component of the outer leaflet of the outer membrane of Gram-negative bacteria and acts as a potent inducer of inflammation. Interestingly, chlamydial LPS has been reported to be of low endotoxic activity, although as with enteric LPS, TLR4 appears to be utilized for signalling. 10,11 Another important chlamydial antigen, heat-shock protein (HSP60), has been shown to activate TLR2 and/or TLR4. 12 The chlamydia genome also contains a number of known and hypothetical lipoproteins that would be potential TLR2 ligands. The role of specific TLRs during productive infection remains less clear. Using a mouse model to study genital tract inflammatory pathology, TLR2 has been identified as a predominant receptor. 13 Similarly, using an immortalized human cervical EC line, the expression of MyD88 and TLR2 was required for cellular activation following infection with C. trachomatis, whereas TLR4 had a minor effect. 10 TLR2 was closely associated with the organism during the intracellular phase and is thought to be responsible for the initiation of signal transduction events during infection with C. trachomatis.
Gonorrhoea
The first stage of gonococcal infection involves the attachment of the organism to ECs via gonococcal pili. The subsequent contact of the bacteria with the host cell surface is mediated by the gonococcal opacity (Opa) outer membrane proteins, which is followed by Opa-dependent cellular invasion. Neisserial porins, PorA and PorB, comprise more than 60% of the outer membrane protein content. Studies examining the role of TLRs suggest that gonococcal lipoprotein induces the release of EC inflammatory mediators in a TLR2-dependent manner. 14 Both porin-induced B-cell and DC activation appears to be dependent on the expression of TLR2 and MyD88 (an essential adaptor protein in the TLR signalling pathway). 15,16 Considering the role of other TLRs, the lack of TLR4 in the cervicovaginal epithelium suggests that the genital tract is capable of responding to Gram-negative pathogens without LPS recognition, at least via TLR4. 17 Thus, while TLR4 may be essential for the response to Gram-negative bacteria during sepsis, other TLRs may be more important for the recognition of these bacteria at specific sites, such as the genital tract mucosa.
Bacterial vaginosis
The pathogenesis of bacterial vaginosis (BV) remains poorly understood. Considering the possible role of innate immunity, Witkin and co-workers suggested that microbial products, such as proteases produced by BV-associated bacteria, directly inactivate EC TLRs. 18 However, a study using cervicovaginal lavage samples found that women with BV induce up to 60-fold increases in TLR4 mRNA expression compared with women without BV, partly related to increased tumour necrosis factor-alpha secretion. 19 Using live bacteria capable of protein synthesis in direct contact with vaginal ECs, Atopobium vaginae has been found to induce increased levels of interleukin-6, interleukin-8 and beta-defensin 4 (an antimicrobial peptide) transcripts. 20 This is mediated by TLR2, requiring the adaptor protein MyD88, and involves the activation of the NF-kappaB signalling pathway.
Considering a possible genetic predisposition to BV, an association has been reported between the development of BV in pregnancy and polymorphism in the gene coding for TLR4, resulting in markedly reduced TLR activity. 21 Although a more recent study found no association between gene polymorphisms and BV, some degree of genetic susceptibility involving pathogen recognition may occur with the key BV organism, A. vaginae. 22
Mycoplasmas and ureaplasmas
Mycoplasma genitalium is now recognized as an important cause of genital tract disease. A study using human embryonic keratinocyte cells expressing human TLR has suggested that M. genitalium may activate NF-kappaB via TLR2/6. 23 Interestingly, in this model vaginal ECs were less responsive than cervical ECs. A further in vitro study from Japan provides the first report of a virulence factor of M. genitalium that can induce an inflammatory response. 24 A triacylated lipoprotein of M. genitalium, MG149, was found to activate NF-kappaB through TLR1 and TLR2. Using a similar in vitro system, the same group has reported that lipoproteins from Ureaplasma parvum, interestingly not recognized as a genital tract pathogen, were found to activate NF-kappaB through TLR1, TLR2 and TLR6. 25
Trichomoniasis
Considering the role of TLRs in Trichomonas vaginalis infection, studies of the molecular pathways involved in the immune response in humans suggest an upregulation of TLR2, TLR4 and TLR9 gene expression. 26,27
Candidiasis
Cell-mediated immunity is considered the predominant host defence mechanism against mucosal candidal infection, there being less information on innate immune system involvement. The switch from blastoconidia to hyphal forms is a virulence trait of Candida albicans and different mechanisms, such as inhibition of phagocytosis and differential induction of pro- or antiinflammatory cytokines have been suggested as key factors in determining the increased invasiveness of hyphae. Studies suggest that blastoconidia stimulate both TLR2 and TLR4, the latter receptor being responsible for interferon (IFN)-gamma production. In contrast, hyphae appear not be recognized by TLR4 and although they induce larger amounts of IL-10, through TLR2, they are unable to stimulate the release of IFN-gamma. 28 Differences in cell wall composition of blastoconidia and hyphae could explain the differences in signalling pathways and cytokine release.
Phospholipomannan (PLM) is a glycolipid expressed at the surface of the C. albicans cell wall and is recognized as a member of the PAMPs family. 29 A study using cultured human primary keratinocytes has shown that PLM upregulates the mRNA and protein levels of TLR2, whereas the mRNA level of TLR4 is not altered. 30 Secreted aspartic proteinases, an important candidal virulence factor, may be a further potential stimulator of TLRs. 31
Syphilis
Treponema, generally known for their lack of antigenicity, induce a moderate innate immune response leading to chronic rather than acute, often systemic, infections.
The mechanisms by which Treponema pallidum causes insidious ongoing inflammation are poorly understood. The organism lacks LPS, the proinflammatory constituent in the outer membranes of Gram-negative bacteria, but contains an abundance of lipoproteins. There is extensive in vitro evidence that spirochoetal lipoproteins and synthetic lipoprotein analogues (lipopeptides) are potent activators of innate immune cells and that these PAMPs trigger cellular activation by binding to the pattern recognition receptors CD14 and TLR1 and TLR2 on the surfaces of monocytes/macrophages and DCs. 32,33 The interaction between T. pallidum and DCs is a key step in the immune response, with recent studies suggesting that DCs engulf intact T. pallidum, thereby liberating lipoproteins resulting in cell activation via the participation of one or more TLRs at the level of the phagosome. 34
Human papillomavirus infection
The control of both human papillomavirus (HPV) infection and HPV-associated neoplasia is dependent predominantly on cell-mediated immune responses, regulated by cytokines secreted by T helper cells. Considering the innate immune response to HPV infection with respect to the role of TLRs, TLR5 and TLR9 expression have been examined in normal and pathological cervical tissue using immunohistochemistry (IHC) and real-time polymerase chain reaction (RT-PCR). 35,36 IHC revealed undetectable or weak expression in normal cervical squamous epithelial tissues with expression gradually increasing from low-grade cervical intraepithelial neoplasia (CIN), CIN3 through to squamous cell carcinoma. RT-PCR revealed no difference in TLR5 expression between normal cervical tissue and tumours, contrary to the IHC result, whereas TLR9 expression was significantly enhanced in tumours compared with normal cervical tissues. The authors suggest that both TLR5 and TLR9 may play a significant role in tumour progression of cervical neoplasia and may represent a useful marker for malignant transformation of cervical squamous cells. Interestingly, and partly contrary to this, HPV type 16 has been reported to interfere with innate immunity by affecting the expression of TLRs. 37 Infection of human primary keratinocytes with HPV16 E6 and E7 recombinant retroviruses has been found to inhibit TLR9 transcription resulting in a functional loss of TLR9-regulated pathways. Similar findings were achieved in HPV16-positive cancer-derived cell lines and primary cervical cancers, demonstrating that this event also occurs in vivo. HPV6 E6 and E7 were unable to downregulate the TLR9 promoter. In addition, E6 and E7 from HPV18, which persists less competently in the host than HPV16, showed reduced efficiency compared with HPV16 in inhibiting TLR9 transcription. This study reveals a novel mechanism used by HPV16 to suppress the host immune response by deregulating the TLR9 transcript, possibly as early as 8 hours postinfection. Abolishing innate responses may be a crucial step in the carcinogenic events mediated by HPVs.
Imiquimod, a TLR7 agonist, belongs to the group of imidazoquinoline compounds and is used as a topical treatment for genital warts and dysplastic HPV-associated anogenital disease. Considering a possible role for other imidazoquinolines in treating cervical HPV-related disease, a recently published study has shown that Langerhans cells exposed to HPV16 virus-like particles and subsequently treated with the TLR agonist, 83M-002, or resiquimod (TLR7/8 agonist), highly upregulate surface activation markers, secrete proinflammatory cytokines and chemokines and initiate an HPV16-specific CD8(+) T-cell response. 38 These data indicate that 83M-002 and resiquimod are promising therapeutic agents for the treatment of HPV infections and HPV-induced cervical lesions.
Herpes
Cell-mediated immune mechanisms, in particular CD4 T-cells, are well established as important in controlling herpes simplex virus (HSV) replication and spread of infection. DCs, potent antigen-presenting cells, particularly at mucosal surfaces, have also been shown to play an essential role. The plasmacytoid DC (pDC), one of a number of DC subtypes, has the ability to secrete type 1 interferons (e.g. IFN-alpha, IFN-beta) in response to viral infection. One of the major pathways for DC activation involves TLRs and in humans pDCs, as compared with other DCs, express high levels of TLR9. TLR9 expression by pDCs is required for the recognition and secretion of IFN-alpha in response to HSV type 2 (HSV-2) in mice; 39 however, some workers have suggested that IFN-beta rather than IFN-alpha may play a more vital role in the innate protection against HSV-2 infection. 40 The novel finding of sequential recognition of virus by DCs via TLR2 → TLR9 has recently been demonstrated, 41 suggesting that cells expressing multiple TLRs can detect pathogens with multiple PAMPs in an orchestrated sequence, thereby linking the uptake and degradation of pathogens to microbial recognition.
To date, vaccines against HSV infection have proved unsuccessful since although a parenterally delivered vaccine may induce a strong systemic immune response, the response at mucosal sites is generally poor. Consequently, the development of a mucosal vaccine or therapy against HSV, and potentially other STIs, has been a focus of attention. Initial studies using TLR ligands as adjuvants in mucosal immunization and as locally delivered prophylactic agents against HSV-2 infection have been encouraging. 42 Double-stranded DNA (dsDNA) is produced by most viral infections at some point during replication and is a ligand recognized by TLR3. Polyinosine-polycytidylic acid (poly-I:C) is a synthetic analogue of dsDNA that has been shown to elicit protective immune responses against HSV-2 when administered intravaginally in mice. A single dose delivered one day prior to and up to 4 hours postinfection provided protection. 43 A further study has shown that mucosal, but not systemic, delivery of ligands for TLR3, but not TLR4, induced protection against intravaginal genital HSV-2 challenge in mice. 44 TLR9 is expressed intracellularly in endosomes and recognizes bacterial cytosine-phosphate-guanosine (CpG) motifs in Gram-positive and Gram-negative bacterial DNA. Synthetic CpG oligonucleotides have been found to elicit strong cellular and humoral immune responses in the genital tract. The delivery of CpG as an adjuvant in both intranasal and vaginal immunization in mice has been shown to trigger an effective adaptive immune response. 45,46 Further study in mice suggests that CpG is most effective when delivered directly into the vagina, with the timing of delivery being essential. 47,48 Considering other TLRs, some protection against HSV-2 infection in mice has been demonstrated using an intravaginally delivered peptide epitope vaccine targeting TLR2. 49 Topical use of the imidazoquinoline compound and TLR7/8 agonist resiquimod has been studied in patients with recurrent genital herpes, and although one study demonstrated reduced HSV-2 genital shedding, 50 this was not confirmed in a subsequent study. 51
DISCUSSION
Our understanding of the functioning and complexity of the innate immune system has progressed appreciably in recent years. In contrast to the adaptive immune system, innate immune system regulation, recognition of pathogens and downstream signalling may be specific for certain cell and anatomical sites. 52 TLR signalling appears to be influenced by various endogenous and exogenous factors such as hormones, cell differentiation and the mucosal microbial flora. The negative feedback loops controlling the duration and extent of innate immune responses is considered crucial for controlling the commensal flora without causing inflammation. Considering therapeutic implications, individual TLR7, TLR8 and TLR9 agonists have been used to boost CD4+ and CD8+ T-cell responses to microbial vaccines. In addition, the use of topical TLR7/8 agonists, such as the imidazoquinolines, remains a key area for further research, particularly with respect to genital mucosal infection. The concept that multiple TLR–ligand interactions are required for inducing host resistance to pathogens has important implications for future vaccine design and immunotherapy against infection. As mentioned in a recent review, ‘TLRs are among the most important receptors ever discovered and, just as they are tied to the initial phase of immune defense, they are tied to some of the basic mechanisms of disease’. 53
