An Altered Immunity Hypothesis for the Development of Symptomatic Bacterial Vaginosis
- 1Department of Obstetrics and Gynecology, Weill Medical College of Cornell University, New York, New York
- 2Department of Gynecology and Obstetrics, University of Sao Paulo Medical School, Sao Paulo
- 3Department of Gynecology and Obstetrics, State University of Campinas, Brazil
- Reprints or correspondence: Dr. Steven S. Witkin, Dept. of Obstetrics and Gynecology, Weill Medical College of Cornell University, 525 E. 68th St., Box 35, New York, NY 10021 (switkin{at}med.cornell.edu).
Abstract
The hypothesis is advanced that the transition from a Lactobacillus-dominated vaginal microflora to a microflora characteristic of bacterial vaginosis (BV), as well as development of the adverse consequences of BV in some women but not in others, are due to alterations in innate immunity. A microbial-induced inhibition of Toll-like receptor expression and/or activity may block induction of proinflammatory immunity and lead to the proliferation of atypical vaginal bacteria. A lack of 70-kDa heat-shock protein production and release in response to abnormal flora would compound this failure to activate antimicrobial immune responses. A deficit in vaginal mannose-binding lectin concentrations would further decrease the capacity for microbial killing and increase the likelihood of bacterial migration from the vagina to the upper genital tract.
Bacterial vaginosis (BV) is the clinical term describing an alteration in the composition of the vaginal bacterial flora of reproductive age women from flora dominated by Lactobacillus species to flora severely depleted of lactobacilli and replaced by a predominance of other bacteria, including Gardnerella vaginalis, gram-positive anaerobic rods (Prevotella, Bacteroides, and Porphyromonas species), Mycoplasma hominis, Mobiluncus species, and Atopobium vaginae [1, 2]. This change in vaginal composition is typically accompanied by an amine odor upon addition of KOH to a drop of secretion, the presence of clue cells on microscopic examination, and occasionally by the presence of a thin discharge. Clue cells are defined as vaginal epithelial cells whose external surfaces are blurred as a result of the attachment of a multitude of bacteria. The rate of transition between a Lactobacillus-dominated flora to a flora characteristic of BV fluctuates in individual women; in some women, 1 type of flora will always predominate [3]. Many nonpregnant and pregnant women with BV have no clinical symptoms and only receive the diagnosis following a microscopic examination of a wet mounted or Gram-stained specimen. In other women, a BV-dominated vaginal flora is accompanied by symptoms such as an unpleasant odor and irregular discharge; these women may be at risk for serious adverse consequences. Although definitive evidence is still lacking, investigations have reported associations between BV in some women and adverse pregnancy outcomes [4], pelvic infection [5], postsurgical infection [6], and acquisition of sexually transmitted microorganisms, including HIV [7]. Why BV is without apparent adverse consequences in some women and is a potentially serious health problem in others has only begun to be addressed [8] but is still largely undetermined. Similarly, the factors that trigger the transition from a normal to a BV-dominated vaginal flora remain unknown, although associations have been observed between acquisition of BV and variables such as sexual intercourse with specific male partners, vaginal douching, and chronic stress [3, 9]. It is also unclear whether the disappearance of vaginal lactobacilli is a preceding cause or a secondary consequence of BV.
Recent advances in the field of innate immunity, also called nonspecific immunity, lead us to now propose immune-mediated mechanisms for development of BV and adverse sequela. The innate immune system is the most primitive and evolutionarily conserved arm of the immune system. It recognizes pathogen-associated molecular patterns that are nonvariant components of many microbial invaders, rather than specific antigens characteristic only of 1 microorganism. Recognition of microorganisms by the innate immune system is the initial trigger for a successful antimicrobial immune defense. Transmembrane proteins called Toll-like receptors (TLRs) are components of this system and have been shown to be major determinants for the recognition of microbial pathogens and the initiation of immune system activation [10, 11]. Eleven TLRs have been identified thus far in humans. TLR2 combines with either TLR1 or TLR6 in the recognition of lipoprotein and peptidoglycan membrane components of gram-positive bacteria, as well as yeast cell wall mannan and lipids of Trypanosoma cruzi. TLR3 binds double-stranded viral RNA, TLR4 recognizes lipopolysaccharide from gram-negative bacteria, TLR5 binds bacterial flagellen, TLR8 interacts with single-stranded viral RNA, and TLR9 recognizes a unique DNA sequence, unmethylated cytosine-guanosine dinucleotides (CpG), present only in bacterial DNA. TLR11 recognizes urinary tract pathogens.
The binding of an invariant microbial component to a specific TLR triggers a sequence of events leading to the selective activation of genes coding for proinflammatory cytokines [11]. These cytokines, in turn, stimulate phagocytic and natural killer cells to attack the intruding pathogen, as well as to mobilize the antigen-specific acquired immune system (i.e., T and B lymphocytes), to initiate a microbial pathogen-specific immune response. TLRs have been identified on a variety of cell types, including antigen presenting cells (dendritic cells, monocytes/macrophages, and neutrophils), fibroblasts, and epithelial cells [12]. Of particular interest to BV is the localization of TLRs on epithelial cells in the female genital tract [13, 14].
The inducible 70-kDa heat shock protein (hsp70) is a newly recognized component of innate immunity. It is among the most highly conserved proteins in evolution and is essential for survival of all unicellular and multicellular organisms. Hsp70 synthesis is greatly up-regulated under nonphysiological conditions, such as infection, inflammation, ischemia, and exposure to toxic chemicals. Intracellularly, hsp70 binds to other proteins, preventing their denaturation and degradation; cell death is also inhibited by the sequestration of components of the apoptosis cascade by hsp70 [15]. Hsp70 is also released from cells into the extracellular milieu in response to stress. This extracellular hsp70 functions as an early warning signal. By binding to TLRs, hsp70 initiates a nonspecific proinflammatory immune response to combat any microbial pathogens that might be present [16]. In addition, extracellular hsp70 induces the production of nitric oxide [17], a compound with potent antimicrobial activity against a wide range of microorganisms.
Mannose-binding lectin (MBL) is another component of the innate immune system and possesses antimicrobial activity against fungi, bacteria, and viruses. It is a plasma protein and a member of the collectin protein family. The active MBL complex is composed of 6 subunits, each containing 3 polypeptide chains. The 6 subunits together form a multimeric complex that resembles a bouquet of tulips, with different carbohydrate binding sites. This presence of multiple carbohydrate-binding sites within a single MBL molecule results in high affinity binding of this protein to carbohydrate structures. MBL is defined as a pattern-recognition molecule and a component of the innate immune defense system, because it recognizes and binds to specific mannose-rich, N-acetyl-glucosamine-rich, or fucose-rich carbohydrate patterns that are present on the surface of many microorganisms [18]. Importantly, MBL does not bind to carbohydrate moieties present in human glycoproteins. Subsequent to MBL binding to a microbial surface, enzymes called “serine proteases,” which are physically associated with MBL, are activated and initiate complement system activation [19]. This leads to deposition of complement components on the surface of the affected microorganism and renders it susceptible to opsonization by phagocytic cells that contain complement receptors on their surface. Additionally, complement activation via the MBL pathway can lead directly to microbe killing. The terminal components of the complement cascade create holes in the microbial cell wall resulting in cell membrane distention and lysis. Opsonization of MBL-bound microbes can also occur by a complement-independent mechanism (i.e., by binding to specific collectin receptors present on the surface of macrophages). In addition to being present in serum, MBL has also been identified in vaginal secretions [20], suggesting a role for this innate immune system component in antimicrobial defense at this site.
We propose that BV develops as a result of inhibition of TLR activation and that the negative consequences of BV are facilitated by a decrease in hsp70 production and release and/or inadequate MBL function. We suggest that the bacteria present in the vaginas of women without BV (i.e., the normal endogenous flora), maintain vaginal epithelial cell TLR activation at a steady level, resulting in sufficient cytokine production to inhibit the proliferation of abnormal BV-associated bacteria. Examination of vaginal secretions of women with lactobacilli-dominated flora has identified the proinflammatory cytokines IL-1β, TNF-α, and IL-6 in low concentrations [21]. However, under conditions in which vaginal TLR expression is inhibited or inactivated, this capacity to prevent replication of abnormal flora would be diminished or absent, and the probability of developing BV would be increased. Bacteria can inactivate TLRs by the induction of immunosuppressive anti-inflammatory cytokines, such as IL-10, and also by the direct inhibition of pathogen-associated molecular pattern—TLR interaction [22].
Secretory leukocyte protease inhibitor is a protein secreted by vaginal epithelia that inhibits neutrophil elastase activity and down-regulates the proinflammatory immune responses of monocytes and macrophages. It also specifically inhibits TLR activation [23]. Microbial products, such as proteases and other degradative products produced by BV-associated bacteria, may directly react with and enzymatically inactivate epithelial cell TLRs. In addition, an accumulation of unsaturated fatty acids in the vagina by the action of BV-associated bacteria and/or the modification of host fatty acids by these organisms would also result in the blockage of local TLR2 and TLR4 activation and further hinder development of an effective antimicrobial immune response [24]. In the absence of an effective immune response, atypical microorganisms present in the vagina in low numbers would be free to multiply. We have shown that a polymorphism in the gene coding for TLR4, which results in markedly reduced TLR activity, was associated with an altered vaginal immune response and development of BV in pregnant women [25].
Hsp70 has been identified as a component of extracellular vaginal fluids in women with recurrent vulvovaginal candidiasis and in a subset of women with BV [26, 27]. Additional studies revealed that hsp70 was associated with elevated vaginal levels of nitric oxide [28]. Thus, by activating TLRs and stimulating nitric acid production, hsp70-producing women may be more efficient at counteracting the presence of abnormal and potentially pathogenic vaginal flora; therefore, these women have a decreased risk for encountering the negative consequences of BV. The variables associated with hsp70 release into vaginal secretions in some women—but not in others—remain to be determined.
We propose that MBL, in addition to functioning as an antimicrobial agent in the vagina, is a major factor preventing the migration of bacteria to the uterus and fallopian tubes in women with BV. MBL binding to microbial surfaces renders the organisms amenable to attachment to collectin receptors present on epithelial cells and on antigen-presenting cells [29]. This sequestration of bacteria prevents their transport from the vagina to the upper genital tract. MBL binds to clue cells [30], demonstrating the interaction of MBL with BV-related bacteria. Circumstances favoring a decrease in MBL bioactivity would, therefore, be expected to increase the possibility of bacterial proliferation and migration within the genital tract. MBL degradation could be brought about by the elaboration of bacterial proteases. In addition, the gene coding for MBL (gene symbol mbl2) is polymorphic at several loci. Carriage of the variant alleles at these polymorphic loci is associated with production of an unstable MBL protein that is rapidly degraded [31]. Greatly reduced concentrations of MBL have been measured in both serum and vaginal secretions from women positive for MBL variant alleles [20, 32].
Several readily testable predictions can be examined from this model. We predict that the cell-free vaginal supernatants from some women with BV, but not from women with lactobacilli-dominated vaginal flora, will inhibit the in vitro activation of TLRs. Furthermore, the extent of this inhibition should be directly proportional to the concentration of 1 or several distinct bacterial species. In addition, women with BV, who are also positive for a functional polymorphism in the gene coding for MBL and/or are deficient in hsp70 production, would be at increased risk for BV-associated adverse sequela. If pregnant, these women would be susceptible to the induction of the sequence of events leading to preterm birth. If not pregnant, they would be at an elevated risk for development of pelvic infection. If the present hypothesis is confirmed, then the analysis of women with BV for their capacity for hsp70 induction and for carriage of MBL gene polymorphism will influence subsequent treatment options. It will eliminate unnecessary prophylactic treatment of women not at risk for BV-related adverse outcomes. More importantly, it will allow a heightened focus on the subset of patients for whom intervention could reduce the rate of BV-associated pathology.
Acknowledgments
Potential conflicts of interest. All authors: no conflicts.
- Received September 14, 2006.
- Accepted November 8, 2006.
- © 2007 by the Infectious Diseases Society of America
References
This Article
- AbstractFree
- » Full Text (HTML)Free
- Full Text (PDF)Free
- A correction has been published
Classifications
Services
Citing Articles
Google Scholar
PubMed
Related Content
Share
- Email this article
-
CiteULike
Connotea
Delicious
Facebook
Google+
Mendeley
Twitter
Published on behalf of
IDSA Members: For your free access to this journal, log in via the IDSA members area.
The Journal
Impact Factor: 8.186
5-Yr impact factor: 7.898
For The Media
For Authors
- Services for authors
- Self-archiving Policy
- IDSA Conflict of Interest policy
- Instructions for Authors
- ICMJE Form
- Submit Now!
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy
Corporate Services
Alerting Services
Related IDSA Publications
