Macroautophagy, commonly referred to as autophagy, is the name given to an intracellular process whereby long-lived defective macromolecules, dysfunctional organelles, such as aged mitochondria, as well as intracellular microorganisms become sequestered in a double-membrane structure known as the autophagosome. Fully formed autophagosomes then fuse with lysosomes to form autolysosomes, and the membrane-bound sequestered components are degraded into their constituent peptides and amino acids by lysosomal enzymes, and are returned to the cytoplasm. Thus autophagy is a major mechanism to rid the cell of unneeded and/or destructive components, and in addition provides a means for the recycling of degraded macromolecules into components that can be reused to provide nutrients for essential metabolic processes. Autophagy is a highly conserved process, being remarkably similar in yeasts, invertebrates, and mammals. It has also been identified in plants. In addition to its direct destructive effect of engulfing and eliminating intracellular microbial pathogens, autophagy is a critical component of both innate and acquired immunity. The binding of highly conserved molecules expressed on a wide range of microorganisms, called pathogen-associated molecular patterns (PAMPs), to pattern recognition receptors (PRRs) present either on the surface or in the cytoplasm of mammalian cells initiates an antimicrobial immune response. This PAMP–PRR interaction, the hallmark of antimicrobial innate immunity, also induces autophagy. Induction of the autophagic process in antigen-presenting cells such as monocytes, macrophages, and dendritic cells results in the presentation of antigenic microbial peptides, in conjunction with major histocompatibility complex class-II molecules, on the cell surface. Subsequent interaction with activated T lymphocytes results in the generation of antigen-specific acquired cell-mediated immunity, a major component of immune defence, as well as immunological memory, against microbial pathogens.
Autophagy formation is a complex process involving about 35 different proteins.[6, 7] Under normal conditions – an adequate nutrient supply, with physiological homeostasis – autophagy induction is blocked by the action of a kinase known as mammalian inhibitor of rapamycin (mTOR) that has the ability to sense nutrient and energy availability. Under conditions of amino acid deprivation or the intracellular accumulation of molecules associated with non-physiological conditions, mTOR activity is inhibited. This induces the sequential synthesis and activation of proteins called autophagy gene (ATG) proteins, and results in the assembly of the autophagosome. This process is composed of three stages: initiation, elongation, and maturation. The initiation stage involves the formation of a membranous structure called the phagophore. Three of the key proteins involved in phagophore formation are Ulk1, Beclin-1, and ATG14. Phagophore elongation and closure to form the mature autophagosome involves ATG5 and ATG12. Concomitantly, the inactive cytosolic form of a protein called light-chain 3-I (LC3-I) becomes conjugated to phosphatidylethanolamine to form LC3-II, and further promotes autophagosome formation. The mature autophagosome then fuses with lysosomes to form an autolysosome, the sequestered autophagosomal contents are degraded by lysosomal acid hydrolases, and its components are transported back into the cytoplasm through channels in the autolysosomal membrane. The entire process is illustrated in Figure 1.
Bacterial and viral pathogens are major causes of premature delivery, and neonatal and maternal morbidity and mortality. In non-pregnant women sexually transmitted infections caused by protozoa, bacteria, and viruses, as well as other pathogens that are not sexually transmitted, continue to plague large numbers of women worldwide. The development of improved protocols to prevent or treat these infections depends on gaining a more complete understanding of: (1) the pathogens' mechanisms of survival and pathogenesis within the human host; (2) the activation and regulation of the host's defences against specific microbial invaders; and (3) an appreciation of the interactions between different microorganisms and their influence on pathogenicity and host defences. Recent investigations have identified autophagy as a major contributor to the control of a multitude of infections, as well as a mechanism that can be sequestered by microorganisms to enhance their pathogenicity. Hardly any autophagy studies related to infectious diseases affecting the female genital tract have appeared in the obstetrics and gynaecology literature. The aim of this review is to familiarise clinicians and researchers with what is currently known about autophagy in relation to microorganisms that infect the female genital tract or negatively impact on pregnancy. Based on these studies, we highlight potentially productive areas for future research.
Chlamydia trachomatis infection
Chlamydia are gram-negative obligate intracellular bacteria that consist of several different species. Chlamydia trachomatis is the most common bacterial sexually transmitted infection, and is the leading cause of infertility and ectopic pregnancy as a result of fallopian tube occlusion. It is also the causative agent of trachoma, one of the leading causes of preventable blindness worldwide. The developmental cycle of C. trachomatis consists of an extracellular form, the elementary body (EB), and an intracellular form, the reticulate body (RB). The chlamydial EB binds to a heparin-containing surface component on epithelial cells, and is internalised into the host cell cytoplasm in an intracellular vacuole called an inclusion body. Within this structure the EBs differentiate into RBs. Following several cycles of multiplication the RBs convert back to EBs, are released from the host cell by lysis, and the EBs bind to nearby epithelial cells and begin a new cycle of infection. The C. trachomatis-containing inclusion body persists in the cytoplasm and does not fuse with the host's lysosomes. This strongly suggests that the inclusion is resistant to autophagy.
There is a paucity of experimental evidence on autophagy and Chlamydia infection. A single study suggests that ATG proteins do not interact with the chlamydial inclusion, and that therefore there is no fusion with autophagosomes and lysosomal destruction. When autophagy was induced in HeLa cells prior to their infection in vitro with the L2 strain of C. trachomatis, there was no apparent effect on chlamydial replication. Similarly, in vitro chlamydial L2 infection and replication within murine epithelial fibroblasts occurred equally well, regardless of whether or not the autophagosome protein, ATG5, was deleted. However, the introduction of interferon gamma was shown to result in the fusion of C. trachomatis L2 inclusions with autophagosomes that were subsequently eliminated through the autophagy pathway. Thus, the elicitation of an effective antichlamydial interferon gamma-mediated host immune response may be required to facilitate chlamydial destruction by autophagy. Recent evidence suggests that interferon gamma induces guanylate binding proteins that re-route the chlamydial inclusions into autophagosomes. One must remain skeptical, however, of the relevance of in vitro studies using cell lines to the in vivo situation, where chlamydial serovars infect cervical and fallopian tube epithelial cells in the presence of immune mediators.
The preferential use of host nutrients by the invading Chlamydia for RB growth and replication may reduce amino acid availability for host cell functions. This would be expected to result in the inhibition of mTOR and the induction of autophagy. That this does not occur suggests that autophagy is actively inhibited in Chlamydia-infected cells. In symptomatic chlamydial cervical infections as well as in clinically unapparent infections of the fallopian tubes, chlamydial inclusions clearly persist for prolonged periods of time within the host cell cytoplasm and produce new infectious EBs. The inhibition of apoptosis in Chlamydia-infected cells to foster survival and prolonged bacterial replication has been demonstrated.[16, 17] It is reasonable to expect that autophagy is similarly inhibited. Studies are needed to determine the mechanism(s) whereby Chlamydia inhibits autophagy induction and intracellular chlamydial inclusions avoid being sequestered into autophagosomes. This would be the first step towards designing protocols to block this inhibition. The identification of new, effective, and specific autophagy-inducing agents in addition to interferon gamma, and their use in women infected with this pathogen, could provide a novel method for reducing the deleterious consequences of chlamydial infection of the female genital tract.
Listeria monocytogenes is a gram-positive bacillus that causes gastroenteritis after ingestion of contaminated food. Listeriosis is usually self-limiting, but can be fatal in immunocompromised patients and pregnant women. Infection with L. monocytogenes is common during pregnancy. One-quarter to one-fifth of L. monocytogenes infections occur in pregnant women, most commonly during the third trimester. Listeriosis in pregnancy has been associated with an increased rate of spontaneous abortions, stillbirth, and preterm labour. Women who are infected at earlier gestational ages have a worse prognosis.[18, 19] Listeria monocytogenes can be transmitted from mother to fetus either by vertical transmission (during the delivery or by ascending vaginal colonisation) or by haematogenous spread. Listeria monocytogenes is a common cause of neonatal meningitis, and has a mortality rate of 20–60%. It is thus important to prevent listeriosis and its consequences during pregnancy.
Listeria monocytogenes is an intracellular pathogen that must escape both phagocytosis and autophagy to proliferate in the host cell cytoplasm. Following entry into non-phagocytic cells by inducing uptake into an endocytic-like vesicle, listeriolysin O (LLO) and two C-type phospholipases, phosphatidylinositol-specific (PI-PLC) and broad-range phosphatidylcholine (PC-PLC), facilitate bacterial escape from the vesicle and entry into the cytoplasm.[21, 22] A number of in vitro and in vivo studies have detailed the interaction of L. monocytogenes with the host's autophagy machinery. Co-localisation of the autophagy-inducing protein LC3 with L. monocytogenes in macrophages results in either bacterial clearance by autophagy or the formation of large vacuoles called spacious Listeria-containing phagosomes (SLAPs). Within SLAPs L. monocytogenes proliferates and chronically persists within its host. ActA is a Listeria protein that promotes actin polymerization, and is necessary for the spread of L. monocytogenes from cell to cell. It also participates in vacuolar disruption and the escape of L. monocytogenes from SLAPs. In addition, ActA prevents the initiation of the autophagy pathway.[20, 26] Two in vivo studies have demonstrated that interference with autophagy in mice and Drosophila resulted in markedly increased susceptibility to L. monocytogenes infection.[27, 28] Protocols specifically designed to prevent L. monocytogenes from subverting the host's autophagy machinery, perhaps by inhibiting or inactivating ActA, may be beneficial in short-circuiting its pathogenicity. Studies are needed to identify the mechanism leading to ActA production, and to identify inhibitors of this process.
Candida albicans infection
Candida albicans is a dimorphic fungus that can exist either as a yeast or as an invasive filamentous form. In many women Candida is a harmless commensal organism in the vagina and/or the intestinal tract; however, in response to a variety of conditions that either alter the normal composition of the vaginal flora (antibiotic usage) or induce a transient localised immune suppression (vaginal allergic response, steroid usage), this organism becomes invasive and elicits a symptomatic episode of vaginitis. In other women the predisposing factors leading to the transition of C. albicans to a pathogen remains unidentified.
A very recent study has elegantly described how C. albicans subverts the host's immune response to aid in its survival and the involvement of autophagy in this process. In the vagina, C. albicans concentrations are largely controlled by the activity of phagocytic neutrophils and macrophages. The newly recognised Th17 class of T lymphocytes is now thought to be the major inducer of phagocyte recruitment and activation. Interleukin 17A (IL-17A), one of the primary cytokines produced by Th17 cells, activates phagocytic cells and, in addition, induces epithelial cells to produce and release peptides with antimicrobial properties. Candida albicans has now been shown to bind IL-17A, and as a consequence to induce a condition of nutrient starvation within the yeast cell. This leads to the inhibition of candidal mTOR and induction of autophagy. A consequence of this activation is the induction of germ tube formation and the formation of Candida biofilms. These processes enhance the resistance of the microbe to the immune system of the host and foster its persistence. Thus, activation of autophagy in C. albicans in response to the production by its host of an immune mediator meant to combat its proliferation, conversely is used by Candida to ensure its survival under adverse conditions. In contrast to the beneficial effects of autophagy induction in Candida, its induction in infected host cells contributes to fungal clearance. When macrophages were co-incubated with C. albicans in vitro, autophagosomes containing the yeast as well as the autophagosome marker, LC3, were identified. When autophagy was inhibited by deleting the gene coding for ATG5, a significant decrease in phagocytosis of C. albicans was observed. Thus, the clearance of C. albicans by macrophages involves the activation of the autophagy pathway.
Many women suffer from recurrent vulvovaginal candidiasis (RVVC), a disorder defined by the occurrence of at least four distinct culture-proven episodes in 1 year of a symptomatic vulvovaginal C. albicans infection. The mechanism responsible for RVVC remains undetermined and treatment is at best only temporarily effective. Symptoms typically recur following the cessation of antibiotic administration. Therefore, new protocols to determine why some women are susceptible to develop RVVC as well as novel methods to prevent the recurrences are urgently needed. Based on the data presented above, an attractive and completely untested possibility would be to determine whether women with RVVC differ from other women in the production of IL-17A in response to a yeast infection, and/or are deficient in yeast-induced autophagy induction in phagocytes. The identification of either variation could then lead to the development of beneficial autophagy-related adjuvants to conventional antibiotic treatment, and thereby increase the potency and longevity of antifungal therapies.
Trichomonas vaginalis is an anaerobic flagellated protozoan and the most prevalent nonviral sexually transmitted disease agent worldwide. Female genital tract infections by this organism can cause vaginitis, urethritis, cervicitis, and a multitude of pregnancy complications. In many women the infection is persistent and asymptomatic. Whereas functional homologues of autophagy proteins have been identified in single-cell protozoa, this has not been studied as yet in T. vaginalis. A very recent investigation has illustrated that the majority of T. vaginalis strains are infected with an endogenous virus, and that it is the release of this virus during the course of the infection or following antibiotic treatment that significantly amplifies the host's proinflammatory immune response and increases pathogenicity. There is a good probability that the virus evades destruction within T. vaginalis by interfering with autophagy. Studies to examine autophagy induction in T. vaginalis, and its effect on genital tract persistence and endogenous viral replication, should be encouraged. The discovery of autophagy-related mechanisms in this pathogenic protozoan may lead to novel means to combat this prevalent infection.
Herpes virus infections
Genital herpes simplex virus (HSV), a double-stranded DNA virus, is becoming increasingly common among women of childbearing age. Although HSV-2 is the most common cause of genital herpes, the incidence of genital herpes caused by HSV-1 is increasing in frequency. Prior infection with HSV-1 also increases the incidence of asymptomatic HSV-2 infection. Genital herpes occurring during pregnancy is a major concern for neonatal mortality and morbidity. The seroprevalence of HSV in pregnant women has been found to be 63% for HSV-1 and 22% for HSV-2. Transmission to the fetus occurs most commonly through direct contact with vaginal secretions. Transplacental and ascending infection can also occur, but is rare.
There is growing evidence that herpes viruses can manipulate autophagy to evade the host's immune response. The host's defence against intracellular viral infection involves the activation of cellular double-stranded RNA-dependent protein kinase R (PKR), which interferes with the translation of viral RNA into protein, and thereby inhibits viral replication; however, HSV-1 and HSV-2 encode a protein called infected cell protein 34.5 (ICP34.5) that prevents PKR-dependent antiviral activity. This viral protein also binds to and inhibits the activation of the autophagy gene, Beclin-1, thereby preventing the formation of autophagosomes in response to viral invasion. In mice possessing an altered ICP34.5 protein that lacks the ability to bind to the Beclin-1 gene, HSV-1 neurovirulence is greatly reduced.[41, 42] Autophagy has also been shown to influence vaginal HSV infection in a mouse model. Depletion of the autophagy gene, ATG5, in dendritic cells led to increased pathology and morbidity in vaginally infected mice.
Human cytomegalovirus (CMV) is a member of the beta herpes virus family. Congenital infection by CMV during pregnancy is the leading infectious cause of neonatal neurologic impairment and hearing loss. The virus synthesizes a protein called TRS1, a functional homologue of ICP34.5, that also binds to Beclin-1 and effectively inhibits the induction of autophagy after viral invasion.
The development of protocols to manipulate autophagy induction in women infected with herpes virus family members may provide a novel mechanism to subvert or shorten viral replication cycles, as well as the length of clinical symptoms.
Human immunodeficiency virus
The human immunodeficiency virus type-1 (HIV-1) is a member of the retrovirus family that replicates by converting its RNA genome into DNA using the enzyme reverse transcriptase. The DNA provirus is then transcribed back into RNA that is used to synthesize HIV proteins. HIV-1 primarily infects CD4+T lymphocytes as well as monocytes/macrophages, and is the causative agent of the pandemic known as acquired immune deficiency syndrome (AIDS).
Human immunodeficiency virus-1 is remarkable in its ability to commandeer the host's autophagy machinery for its own advantage. No fewer than four HIV proteins have been shown to interact with ATG proteins. Infection of macrophages results in increased autophagosome formation, and the inhibition of autophagy is associated with diminished HIV production. Thus, it appears that autophagy is required for HIV replication in macrophages. The HIV Gag protein interacts with LC3, whereas the HIV Nef protein binds to Beclin-1 in the macrophage cytoplasm. It has been postulated that Gag promotes the autophagy-dependent assembly of HIV components, whereas Nef inhibits the later proteolytic stage of autophagy and thereby prevents the destruction of viral intermediates. A third retroviral protein, Tat, has also been shown to block the ability of interferon gamma to induce autophagy in macrophages.
Human immunodeficiency virus has also been shown to block the initiation stage of autophagy in CD4+ T cells. Quite remarkably, in the presence of an HIV infection autophagy is induced in non-infected bystander CD4+ T cells. This has been suggested to result from the ability of the HIV Env protein that is released from infected cells to bind to these cells and to reduce mTOR levels. A consequence of Env-directed autophagy in uninfected CD4+ T lymphocytes is a profound increase in apoptosis in these cells. This may provide the major mechanism responsible for depletion of uninfected CD4+ T cells in AIDS patients.
Studies to develop effective pharmacological interventions to modulate the distinct autophagic responses to HIV in infected T cells and macrophages, as well as in bystander T cells, may help to modulate or prevent the negative clinical picture associated with HIV infection.
Human papillomavirus (HPV), a double-stranded DNA virus, is the cause of cervical carcinoma, and has also been strongly associated with the development of additional anogenital cancers and head and neck malignancies. Of the many different HPV serotypes, only several, predominantly HPV 16 and 18, induce oncogenesis. In the female genital tract HPV 16 and 18 infect terminally differentiated epithelial cells. Two HPV proteins, E6 and E7, expressed during persistent infection, are the triggers of oncogenic progression. The E6 protein binds to and inactivates p53, a host tumour suppressor protein. E7 interacts with, and induces the suppression of, the host retinoblastoma tumour suppressor protein.[49, 50]
The HPV-infected genital epithelial cells are stimulated to replicate, whereas uninfected sister cells are dormant, having already reached the stage of terminal growth arrest. Evidence has been advanced that the induction of autophagy by the infecting virus provides essential nutrients, and thereby fosters host cell survival under these atypical conditions of forced replication. Uninfected epithelial cells that have been altered to express E7 also undergo autophagy, even under favourable growth conditions.
It will be of considerable interest to determine the extent of autophagy in cells at different stages of cervical dysplasia. This will help to pinpoint when exogenous intervention would be most beneficial. Protocols to block the induction of autophagy in HPV-infected cervical epithelial cells may provide a novel mechanism to limit or prevent their survival, and thereby short-circuit the progression to malignancy.
Studies have demonstrated that several strains of lactobacilli, the predominant bacterial species found in the lower genital tract of women of reproductive age, are potent inducers of interferon gamma. As interferon gamma is an inducer of autophagy, whether lactobacilli or other bacterial phylotypes predominate will exert an influence on the likelihood of autophagy induction in the female reproductive tract in response to microbial pathogens. Perhaps an unsuspected consequence of bacterial vaginosis, the common vaginal disorder where lactobacilli are replaced by large numbers of anaerobic bacteria, is the elimination or diminution of interferon gamma-related autophagy induction. It is known that bacterial vaginosis increases susceptibility to a multitude of sexually transmitted infections, and perhaps an alteration in autophagy induction is at least partially responsible. It has also been demonstrated that a concomitant infection with herpes simplex virus type 2 promotes the persistence of Chlamydia trachomatis within infected cells.[55, 56] Although the precise mechanism remains to be elucidated, it is likely that the prolonged induction of a local proinflammatory response triggered by chlamydial persistence augments the damage to epithelial cells. Whether this dual infection increases the likelihood of fallopian tube occlusion, and thereby elevates the risk of ectopic pregnancy, is a rational possibility that to date has not been the subject of any research. The role of autophagy in fostering dual herpes virus and chlamydial infection of epithelial cells, and the pathological consequences of this interaction, deserves to be investigated.
Infection-related pregnancy outcome
The involvement of autophagy in gestation has not received much research attention. Studies have identified autophagosomes and autolysosomes in human amnion epithelial cells, where they may enhance viability under conditions of nutrient limitation. This hypothesis was reinforced by a later study in pregnant mice demonstrating that the inhibition of autophagy resulted in the induction of preterm birth. Conversely, an overly vigorous placental autophagy response has been associated with the induction of severe pre-eclampsia and intrauterine growth restriction.[59, 60] Thus, alterations from the physiological autophagy response, either an up- or a down-regulation, may negatively influence pregnancy outcome. Microbial infections may influence host cell autophagy by a multitude of mechanisms. Depending on the pregnancy stage when an infection becomes evident, and/or the properties of the specific microbial invader, and/or the host's genetic make-up, alterations in autophagy at any of several maternal or fetal sites may impinge upon the normal sequence of events critical for the successful timing of gestational processes. Although it is becoming increasingly evident that bacteria migrating from the lower genital tract into the pregnant uterus are a major cause of preterm birth, the possibility that alterations in autophagy resulting from an infection contribute to this pathological process has not yet been studied.
It has recently been shown in a pregnant mouse model that a latent genital tract infection with a murine herpes virus similar to the Epstein–Barr virus (EBV) increased the likelihood that a concomitant bacterial infection will induce preterm labour. Because EBV, as well as other viruses, can influence autophagy induction, even in uninfected bystander cells, the alteration in autophagy responses to a bacterial infection caused by the simultaneous presence of an unapparent viral infection appears to be a likely factor influencing bacterial pathogenicity. The comparative influence on pregnancy outcome of a bacterial and viral co-infection in the lower genital tract, compared with a bacterial infection alone, is a worthwhile endeavour that has not yet been explored. Initiation of research investigations to evaluate microbial-induced alterations in autophagy during gestation, and the delineation of mechanisms to negate or overcome this alteration, may lead to novel and previously uninvestigated protocols to prevent or overcome infection-induced preterm delivery.