Association of tubal factor infertility with elevated antibodies to Chlamydia trachomatis caseinolytic protease P

Abstract

OBJECTIVE

The objective of the study was to assess antibodies against Chlamydia trachomatis heat shock proteins (HSP) in patients with tubal factor infertility (TFI), infertility controls (IFC), and fertile controls (FC). HSPs assist organisms in surviving caustic environments such as heat.

STUDY DESIGN

Twenty-one TFI, 15 IFC, and 29 FC patients were enrolled after laparoscopic tubal assessment. The titers of antibodies against C trachomatis organisms and 14 chlamydial HSPs were compared among the 3 groups.

RESULTS

TFI patients developed significantly higher levels of antibodies against C trachomatis and specifically recognizing chlamydial HSP60 and caseinolytic protease (Clp) P, a subunit of the ATP-dependent Clp protease complex involved in the degradation of abnormal proteins.

CONCLUSION

In addition to confirming high titers of antibodies against C trachomatis organisms and HSP60 in TFI patients, we identified a novel link of TFI with anti-ClpP antibodies. These findings may provide useful information for developing a noninvasive screening test for TFI and constructing subunit anti-C trachomatis vaccines.

Chlamydia trachomatis is the most common reported agent of sexually transmitted infections worldwide.¹ The rate of C trachomatis infection in the United States has increased significantly over the last 2 decades.² Infection with C trachomatis poses serious health risks, including long-term reproductive tract sequelae such as infertility, chronic pelvic pain, ectopic pregnancy,^3-6 and development of cervical cancer.^7,8

The linkage of tubal factor infertility (TFI) to C trachomatis infection has been extensively studied. C trachomatis organisms can be isolated from a large portion of women with TFI³ and elevated anti–C trachomatis antibodies can be detected in more than 70% of women with tubal occlusion.⁹ Women with prior C trachomatis infection usually maintain high titers of C trachomatis antibodies.¹⁰ Although urogenital tract infections with C trachomatis is common and has been recognized as a significant cause of tubal infertility, the pathogenic mechanisms of C trachomatis–induced tubal damage remain unknown and no effective vaccines are available.

It has been hypothesized that host responses triggered by chlamydial infection contribute to both protective immunity and pathogenesis. Antibodies against the chlamydial major outer membrane protein (MOMP) are associated with protective host immune responses, which is consistent with the recent findings that immunization with a native MOMP-induced protection.^11,12 In contrast, antibodies to chlamydial heat shock protein (HSP) 60 are associated with pathologies,^4,13-15 which may provide a partial explanation for the half-century-old observation that whole chlamydial organism-based vaccines designed for preventing trachoma in children actually exacerbated pathologies.^16-18 HSPs assist organisms in surviving stressful environments such as acidity or heat.

Our objective was to test whether human antibodies against other C trachomatis HSPs are also associated with chlamydia-induced tubal pathologies by comparing all 14 chlamydial HSPs for their reactivity with antibodies in patients with TFI, infertility controls (IFC), and fertile controls (FC).

Materials and Methods

Human antisera

Following approval by the institutional review board at the University of Texas Health Science Center at San Antonio, 21 TFI, 15 IFC, and 29 FC patients were enrolled. All recruited women underwent diagnostic laparoscopy with chromotubation. Diagnosis of tubal infertility was based on 1 of the following findings: hydrosalpinx, fimbrial phimosis, or peritubal adhesions. Women with prior tubal ligation or a history of pelvic infection or inflammation other than pelvic inflammatory disease such as appendicitis were excluded. IFC patients were women with normal pelvic findings and tubal patency at laparoscopy. FC patients had no history of infertility with at least 1 live birth and normal pelvic findings at time of tubal ligation. All participants underwent a single blood draw. Serum samples were stored at −20°C until analyzed.

Cell culture and chlamydial infection

HeLa cells (American Type Culture Collection, Manassas, VA) were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Gibco PRL, Rockville, MD) with 10% fetal calf serum (FCS; Gibco BRL) at 37°C with 5% carbon dioxide (CO₂) as previously described.¹⁹ C trachomatis serovar D or C pneumoniae AR39 organisms were grown, purified, and titrated as previously described.^20,21

After titration, organisms were stored at −80°C. For immunofluorescence assay, chlamydial organisms were used to infect HeLa cells grown on glass coverslips in 24-well plates. The subconfluent HeLa cells were treated with DMEM containing 30 μg/mL of diethylaminoethyl (DEAE)-dextran (Sigma, St Louis, MO) for 10 minutes. After removal of DEAE-dextran solution, chlamydial organisms were added to the wells for 2 hours at 37°C. The infected cells were continuously cultured in DMEM with 10% FCS and 2 μg/mL of cycloheximide (Sigma).

For preparing whole-cell lysates, infection was carried out in tissue culture flasks. Infected cultures were processed or harvested 48 hours after infection or as indicated in individual experiments.

Immunofluorescence assay

Antichlamydial organism antibodies in human sera were titrated using an immunofluorescence assay as previously described.^22,23 Briefly, HeLa cells grown on coverslips were infected with C trachomatis or C pneumoniae organisms, fixed 48 hours after infection for C trachomatis and 72 hours for C pneumoniae with 2% paraformaldehyde, and permeabilized with 2% saponin. After blocking, human antisera were added to the chlamydia-infected cell samples. Goat antihuman immunoglobulin (Ig) G conjugated with Cy2 (green; Jackson ImmunoResearch Laboratories, Inc, West Grove, PA) was used to visualize human antibody binding and a Hoechst deoxyribonucleic acid (DNA) dye (blue; Sigma) to visualize HeLa and chlamydial DNA. The highest dilution of a serum that still gave a positive reactivity was defined as the titer of the given serum sample.

All human serum samples were serially diluted, and the appropriate dilutions were repeated multiple times based on the results obtained from prior dilutions to obtain a more accurate titer for each serum. For the time-course study, the C trachomatis–infected HeLa cells were processed as described above at various time points after infection as indicated in the data figure.

The processed samples were coimmunostained with a mouse anti-HSP60 (unpublished data) or anti-caseinolytic protease (Clp) P (unpublished data) plus rabbit anti–C trachomatis serovar D organisms. The primary antibody binding was visualized with a goat antimouse IgG conjugated with Cy3 (red) and a goat antirabbit IgG conjugated with Cy2 (green; both from Jackson ImmunoResearch Laboratories), respectively, and DNA by a Hoechst DNA dye.

Images were acquired with an Olympus AX70 fluorescence microscope equipped with multiple filter sets (Olympus, Melville, NY) as previously described.²³ All microscopic images were processed using an Adobe Photoshop program (Adobe Systems, San Jose, CA).

Chlamydial fusion protein–arrayed microplate enzyme-linked immunosorbent assay (ELISA)

The glutathione S-transferase (GST) fusion protein ELISA for detecting human antibody recognition of chlamydial proteins was carried as previously described.²³ The bacterial lysates containing individual chlamydial GST fusion proteins were added to 96-well microplates precoated with glutathione (Pierce, Rockford, IL).

The GST fusion protein lysates included all 14 chlamydial HSP family members: GST-CT110 (GroEL, HSP60); GST-CT111 (GroES, HSP10); GST-CT113 (ClpB, ClpB-related ATP-dependent protease); GST-CT286 (ClpC, Clp protease ATP-binding subunit); GST-CT341 (DnaJ protein); GST-CT395 (GrpE, HSP70 cofactor); GST-CT396 (DnaK, HSP70); GST-CT407 (DksA, probable DnaK suppressor); GST-CT431 (ClpP, ATP-dependent ClpP endopeptidase); GST-CT604 (GroEL, HSP60); GST-CT705 (ClpX, ATP-dependent ClpX-related protease; GST-CT706 (ClpP, ATP-dependent ClpP endopeptidase subunit); GST-CT709 (MreB, Rod shape determining protein MreB/HSP70 sugar kinase); and GST-CT755 (HSP60).

Lysates containing GST alone, as negative, and GST-chlamydial protease-like activity factor, as positive controls, were also included. After blocking, human antisera preabsorbed with a bacterial lysates containing GST alone were reacted with the plate-immobilized fusion proteins. The human antibody reactivity was detected with a goat antihuman-IgG conjugated with horseradish peroxidase (HRP; Jackson ImmunoResearch Laboratories) plus the substrate 2,2′-azinobi(2-ethylbenzothiazoline-6-sulforic acid) diammonium salt (ABTS; Sigma). The optical density (OD) was measured at 405 nm using a microplate reader (Molecular Devices Corp, Sunnyvale, CA).

To confirm the antibody-binding specificity, all antisera were further absorbed with lysates made from either HeLa cells alone or C trachomatis serovar D–infected HeLa cells prior to reacting with the fusion protein-coated plates. The antibody binding that remained positive after HeLa-alone lysate absorption but significantly reduced by chlamydia-HeLa lysate absorption was considered true positive.

Western blot

Western blot with GST fusion proteins as antigens was carried out as previously described.²⁰ GST fusion proteins (GST-HSP60, GST-HSP10, GST-ClpP) were purified from the corresponding bacterial lysates using glutathione agarose beads as previously described.²⁴ The purified fusion proteins were resolved on a sodium dodecyl sulfate-polyacrylamide gel and transferred to a nitrocellulose membrane. Membrane-immobilized proteins were reacted with human sera pooled from each patient group and preabsorbed with bacterial lysates containing GST alone. Human antibody binding was detected with a goat antihuman IgG-HRP secondary antibody and visualized with an enhanced chemiluminescence kit (Santa Cruz Biotechnology, Inc, Santa Cruz, CA).

Data analyses

Data were analyzed using SPSS version 15.0 software (IBM, Chicago, IL). As a preliminary step, titer values were log transformed to produce a normal distribution and analyses were performed on transformed values. Analysis of variance was used to assess anti-C trachomatis and anti-C pneumoniae antibodies to evaluate overall mean differences among the 3 groups of patients.

The Student t test was utilized to compare differences between groups. Because the antibody titers had large variations within a given group, the serum titers were evaluated by ranges of less than 1:10 (negative), 1:10 to 1:10,000 (low), and greater than 1:10,000 (high). The χ² and Fisher’s exact tests were used to compare TFI, IFC, and FC overall antibodies to C trachomatis and antibodies to C pneumoniae.

Finally, we evaluated pairwise differences between TFI vs IFC, TFI vs FC, and FC vs IFC in C trachomatis using logistic regression. ELISA results were analyzed also using χ² and Fisher’s exact tests as appropriate.

Results

When C trachomatis–infected cells were used as antigens to titrate the patient serum antibodies, the TFI group had higher titers than the IFC and FC groups (Table). The titers of anti-C trachomatis antibodies were significantly greater in patients with TFI. Because the antibody titers had large variations within a given group, the serum titers were evaluated by ranges of less than 1:10 (negative), 1:10 to 1:10,000 (low), and greater than 1:10,000 (high).

TABLE 1.

Titers of human antibodies against C trachomatis and C pneumoniae

	Antibodies to C trachomatis			Antibodies to C pneumoniae
Variable	TFI (n = 21)	IFC (n = 15)	FC (n = 29)	TFI (n = 21)	IFC (n = 15)	FC (n = 29)
Mean	90,199	4488	36,994	56,010	32,027	56,429
SD	123,700	10,110	83,570	75,580	41,930	65,140
ANOVA	P = .018			P = .45
Student t test	TFI vs IFC, P = .012
	TFI vs FC, P = .075
	IFC vs FC, P = .142
Categorization of serum samples into negative, low, and high titer groups
Negative titers (<1:10)	1 (5%)	2 (13%)	1 (3%)	3 (14%)	4 (27%)	0 (0%)
Low titers (1:10-1:10,000)	6 (29%)	12 (80%)	15 (52%)	3 (14%)	1 (7%)	5 (17%)
High titers (>1:10,000)	14 (67%)	1 (7%)	13 (45%)	15 (71%)	10 (67%)	24 (83%)
χ2 test	P = .009			P = .09
Logistic regression	TFI vs IFC	High vs negative	P = .04
	TFI vs IFC	High vs low	P = .004
	TFI vs FC	High vs low	P = .03
	FC vs IFC	High vs negative	P = .04
	FC vs IFC	High vs low	P = .03

Serum samples from women with TFI, IFC, or FC were 2-fold serially diluted starting with 1:10 and reacted with HeLa cells infected with either C trachomatis or C pneumoniae. The highest dilution that still gave a positive reactivity was defined as the serum titer. Each serum sample was titrated 3 times, and the average from the 3 independent titrations was used as the geometric titer of a given serum sample. ANOVA was used to analyze the overall differences among the 3 groups. There is a statistically significant difference in titers of antibodies against C trachomatis (P = .018) but not C pneumoniae (P = .45) organisms. The significant difference was determined between the TFI and IFC groups by Student t test (P = .012). When the serum samples were divided into 3 categories (negative, low, and high) based on antibody titers, the χ2 test still revealed a significant difference in the number of sera in different categories among the 3 groups of patients for antibodies against C trachomatis (P=.009) but not C pneumoniae (P=.09) organisms. Further logistic regression analyses of the anti–C trachomatis antibodies revealed significant differences between the TFI and IFC, TFI and FC, and the IFC and FC groups. The number of individuals with high anti–C trachomatis antibody titers in the TFI group is significantly higher than those in either the IFC or FC groups, although there are also differences between the IFC and FC groups.

ANOVA, analysis of variance; FC, fertile controls; IFC, infertility controls; TFI, tubal factor infertility.

Further logistic regression (Table) analyses revealed significant differences between TFI and IFC in the high compared with both negative and low titers, TFI and FC in the high compared with negative titers, and IFC and FC in the high compared with both negative and low titers. The number of individuals with high anti–C trachomatis antibody titers in TFI group is significantly more than those in either the IFC or FC groups.

These results have demonstrated an association of TFI with anti–C trachomatis antibodies, which is consistent with various previous observations.²⁵ The anti–C pneumoniae antibody titers among the 3 groups were not significantly different (Table 1). The high titers of anti–C pneumoniae antibodies in most of the patients in all 3 groups did not interfere with the measurements of anti-C trachomatis antibodies because high titers of anti–C trachomatis antibodies were detected only in most TFI patients.

Serum samples from 16 TFI, 7 IFC, and 13 FC patients with high anti–C. trachomatis antibodies (≥1:1000) were further evaluated in a fusion protein ELISA (Figure 1). Antibodies against ClpP were significantly higher in the TFI group as compared with the control groups. Antibodies against the remaining 12 HSPs including HSP10 were not significantly different among the groups.

FIGURE 1. Reactivity of human antibodies with chlamydial fusion proteins arrayed to microplate wells.

The bacterial lysates containing individual chlamydial GST fusion proteins or GST alone (listed along the X-axis) were directly added to glutathione-coated microplates. Human antisera from 3 groups of patients (listed along the Y-axis) were first preabsorbed with bacterial lysates containing GST alone and then reacted with the plate-immobilized chlamydial fusion proteins. The human antibody binding was detected with a goat antihuman IgG conjugated with HRP plus the soluble substrate ABTS (Sigma, St Louis, MO) and measured in OD values at 405 nm. A reaction with an OD value of 2 SD about the mean was considered positive as indicated with horizontal bars. The number of positive individuals from different groups of patients was compared with Pearson’s χ² test. The number of sera that positively recognized HSP60 (P = .001) or ClpP (P = .03) was significantly higher in the TFI group when compared with either the IFC or the FC groups.

ClpP, caseinolytic protease P; FC, fertile controls; GST, glutathione S-transferase; HSP, heat shock proteins; IFC, infertility controls; OD, optical density; SD, standard deviation; TFI, tubal factor infertility.

We further confirmed the specificity of the human antibody binding to ClpP fusion proteins using an absorption approach (Figure 2). In addition to the preabsorption with bacterial lysates containing GST alone, the human sera from the TFI group were further absorbed with either C trachomatis–infected or HeLa-alone lysates prior to reacting with the fusion proteins in the ELISA assay.

FIGURE 2. Absorption of human sera with endogenous C trachomatis antigens blocks the binding of human antibodies to chlamydial fusion proteins.

The bacterial lysates containing individual chlamydial GST fusion proteins or GST alone (as listed along the left side of the figure) were allowed to bind to microplates, and the ELISA was carried out as described in the legend for Figure 1. The 4 human antisera from the TFI group as listed on top of the figure were preabsorbed with bacterial lysates containing GST alone and then further absorbed with either HeLa alone or chlamydia-infected HeLa lysates prior to reacting with the chlamydial fusion proteins on the microplate. Please note that none of the 4 sera bound to the other subunit of the ClpP complex (GST-CT706) and the binding of the 4 sera to both GST-CT431 and GST-CT858 was completely blocked by absorption with the chlamydia-infected but not HeLa alone lysates.

ClpP, caseinolytic protease P; ELISA, enzyme-linked immunosorbent assay; GST, glutathione S-transferase; TFI, tubal factor infertility.

Absorption with C trachomatis–infected HeLa lysate but not the HeLa-alone lysate completely removed ClpP-reactive antibodies from all 4 TFI antisera, demonstrating that the recognition of ClpP by the TFI antisera was specific. Binding of TFI sera to ClpP was confirmed on Western blot (data not shown).

Protein expression of ClpP and HSP60 was assessed over time in cell culture following chlamydial infection (Figure 3). HSP60 was detected as early as 12 hours after infection, whereas ClpP was expressed 24 hours after infection. Both proteins were restricted to the intracellular chlamydia inclusions and persisted throughout the infection cycle.

FIGURE 3. Expression of CT110 (HSP60) and CT431 (ClpP) during C trachomatis infection.

HeLa cells grown on coverslips were infected with C trachomatis serovar D organisms, and at various times after infection as listed on the top of the figure, the infected cultures were processed for immunofluorescence labeling with mouse antibodies against HSP60 or ClpP (red). The samples were colabeled with an anti-MOMP antibody (green) and a DNA dye (blue). Please note that HSP60 was detected as early as 12 hours (yellow = red overlapping with green), whereas ClpP was detected only by 24 hours (white arrows).

ClpP, caseinolytic protease P; HSP, heat shock proteins; MOMP, major outer membrane protein.

Comment

Heat shock proteins are stress response proteins that increase expression with stress such as temperature changes and hypoxia. HSPs are evolutionarily highly conserved and found in bacteria and humans.^26-28 Antibody responses to chlamydial HSP60 and HSP10 have been linked to chlamydia-induced pathologies.²⁹ However, it was unknown whether the antibody responses to any of the remaining 12 chlamydial HSPs are also associated with chlamydial pathogenesis.

We demonstrated that TFI patients displayed significantly higher levels of anti–C trachomatis antibodies, whereas there was no significant difference in the anti–C pneumoniae antibody titers between TFI and control patients, which is consistent with what has been previously reported.¹⁰ C pneumoniae is a ubiquitous human respiratory pathogen. Although infection with C pneumoniae has been associated with both airway allergic diseases and cardiovascular pathologies, C pneumoniae infection has not been linked to tubal factor infertility. Indeed, we found that there were no significant differences in anti–C pneumoniae antibodies among the 3 groups.

This observation has not only confirmed the lack of association of C pneumoniae infection with TFI but also suggested that coinfection with C pneumoniae did not significantly affect the detection specificity when measuring anti–C trachomatis antibodies despite the fact that C pneumoniae and C trachomatis share a very similar genome.

Using fusion protein ELISA, we both confirmed the association of the antichlamydial HSP60 antibodies with TFI and found a new link of TFI to human antibodies against C trachomatis ClpP. ClpP is a proteolytic subunit of the ATP-dependent Clp protease complex. The endopeptidase Clp is also called Ti endopeptidase or ATP dependent endopeptidase Ti, which is found in prokaryotes, chloroplasts, and mitochondria and plays an important role in the degradation of abnormal proteins. The remaining 12 HSPs were not associated with TFI in our patients.

The mechanisms on how the HSPs and their antibodies contribute to the tubal pathologies are still unknown. Some have proposed that the large amounts of bacterial HSPs secreted during infection can lead to an autoimmune response, resulting in tubal pathologies.^30,31 Although immune dominant B cell epitopes of chlamydia HSP60 has been mapped,^32,33 potential autoreactive epitopes have been identified³⁰ and the association between chlamydial HSP60-induced circulating autoantibodies and tubal pathologies has been established,^34-36 there is still a lack of direct demonstration for a role of the HSP60 autoreactive epitopes in chlamydial pathogenesis.

Antibody responses might just indicate the presence of chlamydial antigens in the host, and it is the chlamydial antigen-induced inflammatory^37,38 and cellular immune responses³⁹ that may be mainly responsible for causing the pathologies. Chlamydia HSP60 is a powerful inflammatory stimulus that can activate both macrophages and epithelial cells to secrete inflammatory cytokines.³⁸ HSP60 can also induce T cell responses,³⁹ which can be pathogenic, depending on the phenotype, time, and extent of the responses.^40-42

ClpP is a proteolytic subunit of the ATP-dependent Clp protease complex. The Clp proteases represent a distinctive family of energy-dependent serine proteases that are highly conserved throughout bacteria and eukaryotes.⁴³ Chlamydial ClpP share 45% amino acid sequence identity with its homolog in humans (http://blast.ncbi.nlm.nih.gov/Blast.cgi). Sequence alignment analysis led to the identification of 5 distinct regions each with more than 5 identical amino acids between chlamydial and human ClpPs. These 5 stretches of sequences may serve as potential cross-reactive linear epitopes. It is possible that some of the antichlamydial ClpP human antibodies may recognize the cross-reactive epitopes and attack human ClpP in the tubal tissues.

Although our sample size is limited, the significantly elevated anti-ClpP antibodies in TFI patients may serve as a potential marker for aiding in diagnosis of chlamydia-induced tubal damage. Diagnostic laparoscopy with chromotubation is the gold standard for evaluating tubal patency in an infertility evaluation.

Hysterosalpingogram (HSG) is less invasive in evaluating tubal patency, but HSG does have inherent risks of peritonitis or endometritis. Furthermore, a metaanalysis suggests that HSG has a sensitivity of only 65% and specificity of 83% in diagnosing tubal occlusion.⁴⁴ Thus, there is an urgent need for developing a more reliable and noninvasive marker for diagnosing tubal infertility. The observation that detection of anti–C trachomatis antibody titers can be as good as HSG in diagnosing tubal occlusion⁴⁵ suggests that chlamydial proteinspecific antibodies can be explored for predicting TFI.

Efforts have been made to use antibodies against chlamydial HSPs for screening for TFI.^45-47 In the population recruited into the current pilot study, the anti-HSP60 antibodies can be used to detect TFI with 56% sensitivity and 95% specificity (Figure 1). Inclusion of anti-ClpP antibody detection would increase the sensitivity of this screening test to 69%. The negative predictive value of using ClpP plus HSP60 is 79% and the positive predictive value is 92%. Thus, anti-ClpP antibody may prove to be a valuable marker for improving both detection sensitivity and specificity for the antibody-based diagnosis of tubal occlusion.

The discovery of a unique marker for detecting TFI using the limited number of chlamydial fusion proteins has encouraged us to expand the scope of our assay to include the entire genome. We obtained fusion protein clones covering all open reading frames encoded by C trachomatis genome and plasmid and are in the process of preparing a whole-genome scale proteome ELISA for screening the TFI patient sera as more patient sera are obtained. We hope to use the whole genome scale approach to identify additional unique markers for TFI so that we can further improve the detection specificity and sensitivity of the antibody-based diagnostic approach.