Association of Chlamydia trachomatis Heat-Shock Protein 60 Antibody and HLA Class II DQ Alleles
- 1Molecular Biology Laboratory, Puget Sound Blood Center, Seattle, Washington
- 2Laboratory Center for Disease Control
- 3Departments of Community Health Sciences
- 4Medical Microbiology, University of Manitoba, Winnipeg, Canada
- 5Department of Microbiology, University of Nairobi, Nairobi, Kenya
- Reprints or correspondence: Dr. Robert C. Brunham, Dept. of Medical Microbiology, University of Manitoba, 543-730 William Ave., Winnipeg, Manitoba, R3E 0W3 Canada (Robert_Brunham{at}UManitoba.CA).
Abstract
A total of 113 female commercial sex workers had individual alleles for HLA class II genes determined by using labeled sequence-specific oligonucleotide probes to hybridize to polymerase chain reaction products of amplified DNA. Women also had microimmunofluorescent (MIF) antibody titers to Chlamydia trachomatis elementary bodies and ELISA antibody to recombinant chlamydial heat-shock protein 60 (Chsp60) determined. Women were prospectively followed at monthly intervals over 2 years for incident C. trachomatis infection and acute pelvic inflammatory disease (PID). HLA DQA1*0401 and DQB1*0402 alleles were statistically associated with increased prevalence and amount of antibody to Chsp60 but not MIF antibody. However, these alleles did not alter the risk for chlamydial PID. The potential role that HLA DQ may play in chlamydial disease pathogenesis requires further study.
Antibody to the Chlamydia trachomatis heat-shock protein 60 (Chsp60) has been correlated with the late tissue-damaging sequelae of C. trachomatis infection [1]. The mechanism(s) through which such a correlation might operate is unknown. One intriguing observation is that, in mice, antibody responses to Chsp60 are genetically determined, in part, by genes located within the major histocompatibility complex (MHC) locus [2]. Within the MHC locus, the gene(s) controlling Chsp60 antibody response was mapped in or near the IA genes, the human orthologs of which are HLA-DQ α and β genes. Recently, we reported that commercial sex workers with preexisting antibody to Chsp60 were at greater risk for acute pelvic inflammatory disease (PID) when infected with C. trachomatis than were women without preexisting Chsp60 antibody [3]. On the basis of these observations, we were interested in determining whether HLA class II genes among these women correlate with antibody responses to the Chsp60 and with chlamydial PID susceptibility.
Methods
Study population, clinical evaluation, and C. trachomatis antigen detection
The study population, clinical evaluation, and laboratory methodology have all been described [4–6]. PID was clinically diagnosed when a woman complained of the recent onset of lower abdominal pain and was found to have pelvic tenderness on bimanual vaginal examination [7].
Serology
Serum IgG antibody titers (1:8, 1:32, and 1:128) to formalin-fixed C. trachomatis elementary bodies of individual serovars were measured by the indirect microimmunofluorescent (MIF) assay of Wang and Grayston [8]. Sera were tested for antibody to recombinant fusion protein between glutathione-S-transferase (GST) and Chsp60 in an ELISA, as previously reported [9]. Sera were evaluated at a dilution of 1:500, and the optical density (OD) was recorded after subtracting the background value obtained with the GST fragment alone.
HLA class II genotyping
DNA was extracted from frozen buffy coats. Briefly, 1 mL of lysing solution (Becton Dickinson, San Jose, CA) was added to the buffy coat and incubated for 10 min at room temperature. The nucleated cell pellet was resuspended in Hanks' balanced salt solution, and cells were counted and pelleted again. Pelleted cells were then resuspended in a measured volume of polymerase chain reaction (PCR) lysis buffer (standard PCR buffer with 10% Nonidet-40 and 10% Tween 20) to give a concentration of 5 million cells per milliliter and incubated at 56°C for 1 h in the presence of proteinase K. Approximately 250 ng of DNA from each sample was amplified for DQA, DQB [10], and DRB loci in the presence of locus-specific primers [11] and digoxigenin -labeled dUTP. The digoxigenin-labeled dUTP incorporates into the DNA during the amplification cycle. Oligonucleotides of varying length that discriminate between allelic variants of DQA, DQB, and DRB loci were blotted on to nylon membranes [12], and the labeled PCR products were hybridized to the corresponding oligoblot. Excess unhybridized PCR product was washed out in two stringent washes with buffer, and an immunologic detection step was carried out. Positive reactions were visualized as a colored precipitate.
Statistics
Associations between HLA class II alleles and categories of Chsp60 and MIF antibody responses were tested by χ2 analysis. Mean ODs for antibody among different alleles were first compared by using analysis of variance. For alleles with statistical differences in antibody levels, this was followed by t tests comparing results for those with and without the specific allele. Significance levels were set at α = .05, reflecting the investigative rather than inferential nature of the study.
Results
Study population
Of 280 commercial sex workers whose data previously demonstrated a correlation between Chsp60 antibody and risk of C. trachomatis PID, 113 had HLA class II genotyping performed. In general, women who were or were not HLA class-typed were similar in terms of age, prostitution duration, number of clients per day, chlamydial PID rate, and Chsp60 and MIF seropositivity (data not shown). The group of women who were HLA class II—genotyped differed from the group who were not genotyped by a greater number of clinical visits, longer follow-up, lower human immunodeficiency virus infection rate, higher CD4 cell counts, more chlamydial antigen tests performed, and a greater number with one or more chlamydial infections (data not shown).
HLA class alleles
Overall, 25 different DRB1 alleles, 10 DQA1 and 15 different DQB1, were identified. Certain DRB1 alleles, as well as those associated with the DRB3 locus, were predominant in this population; the most common DRB1 allele was DRB1*1101, found in 30% of the women; the most common DRB3 allele was DRB3*02, found in 55% of the women. Similarly, certain DQA1 and DQB1 alleles were also predominant; for example, DQA1*0102 was found in 51% and DQB1*0301 in 43% of the women. A complete list of DRB1, DQA1, and DQB1 alleles identified among the 113 women is available on request from the authors.
Association of selected HLA class II alleles with Chsp60 and MIF antibody
Alleles that were identified in ⩾10% of the study participants were evaluated for their statistical association with Chsp60 antibody. This was compared in two ways: by the mean OD units among allele-positive and -negative women or the proportion of women with Chsp60 OD units ⩾0.5, a cutoff previously validated as correlated with an increased risk for C. trachomatis PID in this group of women [3]. To determine whether an HLA class II allele association with Chsp60 antibody was specific, the relationship between class II genes and MIF antibody was also determined.
Overall, the strongest statistical association between Chsp60 antibody and HLA class II genes was found with the HLA DQA1*0401 allele (table 1). The mean ± SD OD unit for allelepositive women was 0.74 ± 0.77, compared with 0.33 ± 0.45 among allele-negative women (P = .02). Furthermore, the proportion of women with Chsp60 antibody ⩾0.5 OD units was higher among allele-positive than allele-negative women (10/24 allele-positive women vs. 16/89 allele-negative women; odds ratio [OR] = 3.26, 95% confidence interval [CI] = 1.23–8.64, P = .01). Women with this allele were as likely to have MIF antibody at ⩾1:8 (20/24) as were allele-negative women (73/89, P = .88). A weaker statistical association with Chsp60 antibody was found for the HLA DQB1*0402 allele. Allele-positive women had a higher mean OD unit value (0.72 ± 0.8) than allele-negative women (0.35 ± 0.46, P = .05) and were slightly more likely to have Chsp60 antibody at ⩾0.5 OD units (8/21) than allele-negative women (18/92; OR = 2.53, 95% CI = 0.91–7.02, P = .07). Women with this allele were as likely to have MIF antibody at ⩾1:8 (17/21) as women without the allele (76/92, P = .86).
Table 1
Correlation of selected HLA class II DQ and DR alleles with Chlamydia trachomatis heat-shock ptotein 60 antibody.
Two HLA DRB1 alleles were associated with reduced Chsp60 OD values. Women with the HLA DRB1*0102 allele or the HLA DRB1*0301 allele had significantly lower mean ± SD OD units (0.21 ± 0.22 vs. 0.45 ± 0.57, P = .007; 0.21 ± 0.34 vs. 0.45 ± 0.57, P = .03) compared with women who lacked either of these alleles. The proportion of women with Chsp60 antibody at ⩾0.5 OD units was not significantly different for allele-positive versus allele-negative women. These alleles were not associated with an altered response in the MIF antibody assay. No other DQ or DR alleles were associated with an altered Chsp60 antibody response.
Correlation of HLA DQ heterodimers with Chsp60 and MIF antibody
To more completely analyze HLA DQ associations with Chsp60 antibodies, we evaluated the serologic relationships for the five most commonly observed DQ heterodimers (table 2). As suggested by the analysis of individual DQ alleles, the HLA DQA1*0401/DQB1*0402 heterodimer was associated with increased Chsp60 antibodies. Thus, of 20 women with this DQ heterodimer, the mean ± SD Chsp60 OD unit value was 0.75 ± 0.80, compared with 0.35 ± 0.46 among women who did not have the heterodimer (P = .04). Furthermore, these women more often had Chsp60 antibody at ⩾0.5 OD units (8/20) than did women who did not have this heterodimer (18/93, OR = 2.78, 95% CI = 0.99–7.80, P = .05). Women with the DQA1*0401/B1*0402 heterodimer had MIF antibody at ⩾1:8 (16/20) as often as those who lacked this heterodimer (77/93, P = .77). None of the other common DQ heterodimers correlated with Chsp60 or MIF antibody.
Table 2
Association of selected HLA DQ heterodimers with Chlamydia trachomatis heat-shock ptotein 60 antibody.
Correlation of selected HLA class II alleles with risk of C. trachomatis PID
To determine whether the class II genes that were correlated with Chsp60 antibody responses were also correlated with risk of C. trachomatis PID, we compared the proportion of women with C. trachomatis cervical infection who had PID at that visit for each class II gene. Overall, no statistically significant association was found for any of the individual DR or DQ alleles. In particular, 2/8 Chlamydia-infected women with DQA1*0401 and DQB1*402 had PID, compared with 17/60 Chlamydia-infected women without these alleles (P = .84, OR = 0.84, 95% CI = 0.15–4.60).
Discussion
The mechanism by which antibody to Chsp60 contributes to the immunopathogenesis of C. trachomatis disease remains enigmatic. A potential clue to the mechanism derived from the observation that antibody responses to Chsp60 are genetically controlled in mice [2]. Because Chsp60 antibody responses in humans are heterogenous [3], it seemed plausible that human serologic responses might also be genetically determined. The mouse data suggested that genes in or near the IA locus could be involved. The orthologous locus in humans is the DQ region. Thus, it is significant that two DQ alleles were identified as associated with a “high-responder” trait to Chsp60. The two DQ alleles identified as associated with Chsp60 antibody are known to form an authentic DQ α and β heterodimer, and women with this heterodimer were also observed to have significantly higher Chsp60 antibody responses than were women without the heterodimer. As well, two DRB1 alleles were also observed to be correlated with reduced antibody responses to Chsp60, although the consistency of these associations appeared to be less stringent than that observed for the two DQ alleles.
We reasoned that if the Chsp60 antibody response predisposition to chlamydial immunopathogenesis operated directly through HLA class II genes, it should be possible to determine whether women with these genes have an altered risk for C. trachomatis PID when infected with C. trachomatis. However, none of the alleles associated with altered Chsp60 antibody responses was associated with a significant difference in risk for C. trachomatis PID. The robustness of this conclusion needs to be viewed in light of the small number of PID cases studied. The immunopathogenesis of chlamydial PID is clearly complex. Further study to examine Chsp60 antibody, HLA DQA1*0401/B1*0402 and other class II genes, as well as cellular immune responses to chlamydial antigens, may help to clarify the steps in the causal pathway.
Footnotes
-
Financial support: National Institutes of Health (AI-34616); Medical Research Council of Canada (GR13301). F. P. is supported as a scientist by the Medical Research Council of Canada.
-
Informed consent was obtained from patients in accordance with human experimentation guidelines of the University of Nairobi and the University of Manitoba.
- Received October 14, 1998.
- Revision received March 2, 1999.
- © 1999 by the Infectious Diseases Society of America
References
This Article
- AbstractFree
- » Full Text (HTML)Free
- Full Text (PDF)Free
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: 6.288
5-Yr impact factor: 6.103
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
