Background

A large variation exists in the individual response to a Chlamydia (C.) trachomatis infection. Some women clear a C. trachomatis infection adequately without developing tissue damage, whereas others get a persistent infection which may ascend to the upper genital tract, increasing the risk of tubal damage and subfertility. The susceptibility, course and outcome of infectious diseases are determined by environmental factors, virulence factors of the pathogen and host factors.

Immunogenetic studies evaluate the role of genetic variations in immunologically important host genes as determinants of the susceptibility, course and outcome of infectious diseases. Among these variations are single nucleotide polymorphisms (SNPs), in which one nucleotide has been substituted, inserted or deleted. This may lead to synthesis of a potentially aberrant protein, or to up- or downregulation of the normal protein, and subsequently to an aberrant immune response, increasing the risk of late sequelae of infectious diseases (e.g. tubal pathology following a C. trachomatis infection).

In the present study, we have evaluated SNPs in genes encoding for pattern recognition receptors (PRRs). PRRs are present on or in circulating cells of the innate immune system (e.g. macrophages) and local cells (e.g. epithelial cells of the upper genital tract). PRRs are involved in the bacterial sensing pathways of the innate immune system by recognizing the so-called pathogen-associated molecular patterns (PAMPs), which are pathogen-specific cell wall components or intracellular components. Since different PRRs recognize different PAMPs, pathogen recognition and initiation of the immune response is a complex and flexible system.

Carrying a SNP in a single PRR may not result in a large effect on disease severity, since other PRRs may compensate for the partial loss of function in a specific pathogen recognition route. Subsequently, SNPs in only one PRR may not play a significant role as risk factors for the development of C. trachomatis-associated tubal pathology, as shown for the PRR toll-like receptor (TLR) 4 [1] and its co-receptor cluster of differentiation (CD) 14 [2]. However, carrying multiple SNPs in one gene or in multiple genes (in so-called carrier traits) may be associated with an increased risk of tubal pathology. Smirnova et al. (2003) [3] have found that combinations of TLR4 variants are markedly more common in patients with meningococcal infections, whereas single variants are not over-represented in those patients. In studies on gastrointestinal malignancies, it has been concluded that carrying multiple pro-inflammatory polymorphisms is associated with an increased risk of gastric cancer [4, 5]. Furthermore, studies on the relationship between caspase recruitment domain (CARD) 15/nucleotide oligomerisation domain (NOD) 2 genetic variants, of which SNP8, SNP12 and SNP13 are most studied, and Crohn's disease have shown that compound heterozygous subjects (carriers of two different genetic variants, e.g. SNP12 genotype 1.2 and SNP13 genotype 1.2) have a higher risk of Crohn's disease as compared to homozygous subjects (carriers of the same genetic variant on both chromosomes, e.g. SNP12 genotype 2.2) [6, 7].

Analogous to these findings, we hypothesized that carrying multiple genetic variations in multiple PRRs (in a so-called carrier trait) may increase the risk of C. trachomatis-associated tubal pathology in subfertile women. According to their biological function (recognition of C. trachomatis PAMPs: see Table 1), four PRRs were selected: TLR9, TLR4, CD14 and CARD15/NOD2. Five relatively common SNPs, which are assumed to influence the receptor function, in these four PRR genes were studied in this carrier trait analysis (see Table 1).

Table 1 The pattern recognition receptors (PRRs), which recognize C. trachomatis pathogen-associated molecular patterns (PAMPs), and the single nucleotide polymorphisms (SNPs) studied
Full size table

Methods

Study population

The study was performed in women who visited the Academic Hospital Maastricht between December 1990 and November 2000 because of subfertility. In all patients blood was drawn at their initial visit for a Chlamydia IgG antibody test (CAT). All spare sera were cryopreserved. Only patients who had undergone a laparoscopy and tubal testing as part of their fertility work-up were included in the present study. Since the prevalence of SNPs may depend on ethnical background, only Dutch Caucasian women were included. Patients who had undergone previous pelvic surgery (except for an uneventful appendectomy or Caesarean section) were excluded. In the Netherlands, for retrospective analysis of anonymized patient data and stored sera no ethical committee approval is required. In the fertility clinic of the Academic Hospital Maastricht, all couples are informed at intake about possible use of their anonymized data and stored sera for research purposes, and a "no objection procedure" is followed. Only patients having not objected participated in the present study.

Two independent investigators, who were unaware of the CAT results, scored 259 successive laparoscopy reports to assess the grade of tubal pathology. Tubal pathology was defined as extensive peri-adnexal adhesions and/or distal occlusion of at least one tube [8]. In case of disagreement, consensus was reached by consultation.

Of the 259 women who underwent a laparoscopy, 43 (17%) had tubal pathology (according to the above-mentioned definition) and 184 (71%) had no tubal pathology (no peri-adnexal adhesions and patent tubes), and these 227 women participated in the present study. Thirty-two women (12%) had minor or non-C. trachomatis-related abnormalities (any peri-adnexal adhesions and/or proximal occlusion of at least one tube) and were excluded.

C. trachomatisIgG antibody testing

IgG antibodies to C. trachomatis were detected using the species-specific Chlamydia pneumoniae IgG micro-immunofluorescence (MIF) test (AniLabsystems, Finland), as described previously [9]. This species-specific test is able to detect IgG antibodies to both C. pneumoniae and C. trachomatis (using an antigen derived from a C. trachomatis LGV strain, serovar L2). We have previously studied the test performances of five commercially available C. trachomatis IgG tests, including the C. trachomatis IgG spot in the C. pneumoniae MIF (AniLabsystems) [10]. In our hands, the C. trachomatis IgG titre obtained by the C. pneumoniae MIF (AniLabsystems) had the best predictive value for tubal factor subfertility [10]. Therefore, we have used this test in the present study. The cut-off titre used for a positive test was 32.

Immunogenetic analysis

For the immunogenetic analyses, genomic DNA was extracted from the cryopreserved serum samples using either the MagNaPure LC isolator according to the manufacturers' instructions (Roche Molecular Biochemicals, Germany) or the High Pure PCR Template Preparation (HPPTP) Kit according to the manufacturers' instructions (Roche Molecular Biochemicals, Germany). Both techniques provide enough DNA for reproducible genetic analyses. Genotyping was performed using a polymerase chain reaction (PCR)-based restriction fragment length polymorphism (RFLP)-assay or TaqMan-assay as described previously [2, 11–13]. The SNPs studied are summarized in Table 1. Chromosomal locations and further information on the genes studied are: TLR9 chromosomal location 3p21.3, TLR9 -1237 T>C rs5743836 and TLR9 +2848 G>A rs352140; TLR4 chromosomal location 9q32-q33, TLR4 +896 A>G rs4986790; CD14 chromosomal location 5q31.1, CD14 -260 C>T rs25691909; CARD15/NOD2 chromosomal location 16q21, CARD15/NOD2 Leu1007fsinsC (SNP13) rs2066847. For ethnically-matched background genotyping, genomic DNA was extracted from whole blood of 97 healthy Dutch Caucasian employees of the VU University Medical Center. They gave written informed consent for use of their anonymized sera to serve as control sera for genetic research purposes.

Statistical methods

The genotype distribution was tested for Hardy-Weinberg equilibrium to assess Mendelian inheritance. Fisher's exact or χ2 tests were used to compare the single genotypes between C. trachomatis IgG-positive and IgG-negative subfertile women with and without tubal pathology and the healthy control group. Subsequently, the single genotypes were used to define carrier traits. The carrier traits were tested in χ2 and trend analyses. P < 0.05 was considered statistically significant.

Results

Of all 227 subfertile women participating in the present study, 43 (19%) had tubal pathology, whereas 184 (81%) did not have tubal pathology. C. trachomatis IgG antibodies were present in 39 women, of whom 26 (67%) had tubal pathology and 13 (33%) did not have tubal pathology. C. trachomatis IgG antibodies were absent in 188 women, of whom 17 (9%) had tubal pathology and 171 (91%) did not have tubal pathology.

For all genes studied, the genotype distribution was in Hardy-Weinberg equilibrium in the subfertile women and the ethnically-matched control group.

The genotype distribution did not differ between subfertile women, the C. trachomatis IgG-positive subgroup of subfertile women and the healthy control group (Table 2), indicating that the subfertile women participating in the study reflect an average Dutch Caucasian population regarding the genotype distribution.

Table 2 The risk of tubal pathology (TP) in relation to the genotype of the single genes studied.
Full size table

Single gene analysis

The risk of tubal pathology was assessed in all subfertile women and in the C. trachomatis IgG-positive subgroup in relation to the genotype of TLR9, TLR4, CD14 and CARD15/NOD2 (Table 2; Figure 1). An increasing risk of tubal pathology was observed across the genotypes in all genes except CD14. Carrying SNPs in these genes increased the risk of tubal pathology (on average almost 20%). These differences did not reach statistical significance. These single genotypes were used to define carrier traits.

Figure 1
figure 1

The risk of tubal pathology (TP) in C. trachomatis IgG-positive subfertile women in relation to the genotype of the single pattern recognition receptor genes. a Adapted from Morré et al., 2003 [1]. b Adapted from Ouburg et al., 2005 [2].

Full size image

Carrier trait analysis

The SNPs in the single genes were combined in carrier traits. The risk of tubal pathology was assessed in C. trachomatis IgG-positive and IgG-negative subfertile women in relation to the number of SNPs. Carrying two or more SNPs did not influence the risk of tubal pathology in C. trachomatis IgG-negative women as compared to C. trachomatis IgG-negative women carrying less than two SNPs (9% vs. 8% risk respectively; Figure 2). However, carrying multiple SNPs doubled the risk of tubal pathology in C. trachomatis IgG-positive women as compared to C. trachomatis IgG-positive women with less than two SNPs (73% vs. 33% risk respectively; Figure 2). These differences did not reach statistical significance (P = 0.15) but a clear trend was observed.

Figure 2
figure 2

The risk of tubal pathology (TP) in C. trachomatis IgG-positive (CT+) and IgG-negative (CT-) subfertile women in relation to carrying five single nucleotide polymorphisms (SNPs) in four pattern recognition receptor genes.

Full size image

Discussion

Over the last decade, immunogenetic studies have provided insight in the pathogenesis of and susceptibility to infectious diseases. So far, the role of SNPs in immunologically relevant genes has been established in numerous diseases, e.g. sexually transmitted infections [14–16] and inflammatory bowel diseases (Crohn's disease and ulcerative colitis) [17, 18]. In this study, we analyzed the role of single SNPs and multiple SNPs in multiple genes (in a so-called carrier trait) as risk factors of C. trachomatis-related tubal pathology and we confirmed our hypothesis that a carrier trait based on genes in bacterial sensing pathways had a stronger association with the risk of tubal pathology than a single gene analysis.

Recent studies have shown the value of genetic traits in complex diseases. Carrying multiple SNPs in the same gene, or multiple SNPs in multiple genes, has been associated with an increased risk of infectious diseases and malignancies [3, 4, 6, 7]. Analogous to these findings, we hypothesized that the disregulation of the immune response by the presence of multiple SNPs may lead to an even higher risk of tubal pathology following a C. trachomatis infection as compared to carrying a single SNP.

In this study, we investigated the role of five SNPs in four genes which are assumed to play a role in the recognition of C. trachomatis (see Table 1). An adequate recognition of C. trachomatis by PRRs is the first step in the immune response. Recent studies have shown that TLR1-9 are expressed in the human female genital tract. TLR4 and its co-receptor CD14 are predominantly expressed in the fallopian tubes, where they may play an important role in the innate host defence mechanism against ascending C. trachomatis infections [19–21]. Regarding CARD15/NOD2, it is not clear whether it plays a role in the C. trachomatis recognition in the genital tract, although NOD proteins have been shown to be involved in the intracellular sensing of C. pneumoniae in endothelial cells [22].

Our data show a doubling of the risk of tubal pathology in C. trachomatis IgG-positive women carrying two or more SNPs as compared to C. trachomatis IgG-positive women carrying less than two SNPs (73% vs. 33% risk). The differences did not reach statistical significance (P = 0.15) due to the small sample size (227 subfertile women in total, including 39 C. trachomatis IgG-positive subfertile women, of whom 67% has tubal pathology). If the association found in this pilot study would be confirmed in a larger cohort, a 50% increase in sample size (to 341 subfertile women) would result in a significant difference (P = 0.047) between C. trachomatis IgG-positive subfertile women carrying two or more SNPs versus C. trachomatis IgG-positive subfertile women carrying less than two SNPs regarding the risk of tubal pathology. A 100% increase in sample size (to 454 subfertile women) would result in a strong association (P = 0.016; OR: 5.3) between carrying two or more SNPs and an increased risk of tubal pathology. Increasing the sample size twofold would not be possible however in our setting in a reasonable time frame. Although the present study was performed in a large fertility clinic and patients were included during a ten-year period, the number of affected women remained small, due to the low prevalence of IgG-positivity in combination with both carrying multiple SNPs and tubal pathology.

From our results it can be concluded that an adequate recognition of the pathogen at the site of infection seems to be a relevant step in the immune response, and that carrying multiple SNPs in multiple C. trachomatis PRR genes tends to increase the risk of an aberrant immune response and tubal pathology. To draw significant conclusions, our hypothesis should be retested in further studies using a larger cohort.

As expected, a difference in risk of tubal pathology between C. trachomatis IgG-positive women and C. trachomatis IgG-negative women was observed. In previous studies, the presence of C. trachomatis IgG antibodies, an indicator of a previous C. trachomatis infection, has been associated with tubal pathology [23]. Our data show that carrying multiple SNPs in bacterial sensing pathways and a previous C. trachomatis infection synergistically enhance the risk of tubal pathology, while carrying these SNPs does not influence the risk of tubal pathology in the absence of a previous C. trachomatis infection (Figure 2).

Conclusion

We hypothesized that carrier traits (i.e. carrying multiple SNPs in multiple genes) that likely result in an aberrant immune response are associated with tubal pathology following a C. trachomatis infection. In 227 subfertile women, we studied five variations in four genes encoding for pattern recognition receptors, which recognize several pathogen-associated molecular patterns of C. trachomatis. The presence of two or more SNPs tends to correlate with an increased risk of tubal pathology following a C. trachomatis infection as compared to a lower number of SNPs. Further studies in a larger cohort are needed to confirm our preliminary findings. An adequate recognition of C. trachomatis by receptors in the genital tract seems to be a relevant step in the immune response, and may play a role in protecting the host against the development of late sequelae following a C. trachomatis infection.