Circulating antibodies to a conserved epitope of the Chlamydia ...

Human Reproduction vol.13 no.5 pp.1175–1179, 1998

Circulating antibodies to a conserved epitope of the Chlamydia trachomatis 60 kDa heat shock protein (hsp60) in infertile couples and its relationship to antibodies to C.trachomatis surface antigens and the Escherichia coli and human HSP60 Steven S.Witkin1,4, Myriam Askienazy-Elbhar2, Jeanine Henry-Suchet3, Joelle Belaisch-Allart3, Joelle Tort-Grumbach3 and Kamal Sarjdine3 1Division

of Immunology and Infectious Diseases, Department of Obstetrics and Gynecology, Cornell University Medical College, 515 East 71st Street, New York, New York 10021, USA, 2Laboratoire de Biologie Medicale Magenta, Paris, France and 3C.H.Jean Rostand, Unit of Obstetrics and Gynecology and Reproductive Medicine, Se`vres, France 4To

whom correspondence should be addressed

To evaluate the relationship between immunity to specific regions of the Chlamydia trachomatis 60 kDa heat shock protein (hsp60), autoimmunity to human HSP60 and infertility, sera from 50 women and 45 men seen for an infertility evaluation were tested. Humoral immunity to human HSP60 was detected in 18% of women and 8.9% of men while antibodies to the Escherichia coli hsp60 were detected in 12% of women and 4.4% of men. These differences were not statistically significant. In contrast, antibodies to a synthetic peptide epitope of the chlamydial hsp60, encompassing amino acids 260–271 (chsp 260–271), were present in sera from 16 (32%) of the women but in only six (13.3%) of the men (P J 0.03). Antibodies to chsp 260– 271 were present in 11 out of 17 (64.7%) individuals with high titre (>1:160) immunoglobulin (Ig)G antibody to C.trachomatis surface antigens as opposed to only two out of 15 (13.3%) with low titre antibody and two out of of 17 (11.8%) with undetectable chlamydial antibody (P < 0.004). Antibodies to chsp 260–271 were also associated with humoral immunity to human HSP60. 50% of sera with, as opposed to only 18.6% of sera without, anti-human HSP60 IgG were positive for antibodies to chsp 260–271 (P J 0.03). In contrast, there was no relationship found between immunity to the E.coli hsp60 and antibodies to human HSP60. Antibodies to chsp 260–271 were more prevalent in women with at least two spontaneous abortions (eight out of 13, 61.5%) than in women with other infertility diagnoses (six out of 35, 17.1%) (P J 0.004). Thus, immunity to chsp 260–271 is more prevalent in women than in men, associated with autoimmunity to human HSP60 and may be an immunological marker for spontaneous abortion. Key words: Chlamydia trachomatis/60 kDa heat shock protein/ infertility/spontaneous abortion

Introduction Chlamydia trachomatis infections are a major cause of infertility due to occlusion of the Fallopian tubes (Cates and © European Society for Human Reproduction and Embryology

Wasserheit, 1991). Several recent studies have demonstrated an association between tubal damage and immunity to the chlamydial 60 kDa heat shock protein (hsp60) (Brunham et al., 1985; Wager et al., 1990; Toye et al., 1993; Witkin et al., 1993; Arno et al., 1995). Hsp60 is present in organisms ranging from bacteria to man and is a highly conserved protein. The chlamydial hsp60 shares almost a 50% amino acid sequence homology with the heat shock proteins of other bacteria as well as with the human HSP60 (Morrison et al., 1992). It has been postulated that C.trachomatis-associated early stage pregnancy loss may occur via an immune mechanism involving the reactivation, by human HSP60 expressed during pregnancy, of lymphocytes that were previously sensitized to conserved microbial hsp60 epitopes (Witkin et al., 1994, 1996). Hsp60 expression has been identified in the glandular epithelial cells in human decidua (Mincheva-Nilsson et al., 1994), in follicular fluids of women with failed in-vitro fertilization (IVF) cycles (Neuer et al., 1997) and in the early stage mammalian embryo (Bensuade and Morange, 1983). To further define the relationship between immunity to the chlamydial hsp60 and the human HSP60 and infertility we evaluated sera from 50 women and 45 men being seen for an infertility evaluation for antibodies to C.trachomatis surface components, two synthetic peptides corresponding to conserved and Chlamydia-specific epitopes of the chlamydial hsp60, the Escherichia coli hsp60 and human HSP60. The results demonstrate that humoral immunity to the conserved chlamydial hsp60 epitope is more prevalent in women than in men and is associated with sensitization to the human HSP60 and with a history of two or more spontaneous abortions. Materials and methods Subjects Peripheral blood was obtained from 50 women and 45 men being seen for an infertility evaluation at the Hoˆpital de Se`vres, Se`vres, France. Sera were obtained by centrifugation and stored at –20°C until utilized. Clinical evaluation All female patients had a complete clinical evaluation including history of past pregnancies, previous fetal losses, ectopic pregnancies, gynaecological surgery, treatment for pelvic inflammatory disease (PID) or genital infections. The genital tract flora was evaluated by cervico–urethral sampling. All subjects had routine hormonal evaluations as well as uterotubal evaluation by hysterosalpingography. All but 12 women with patent tubes also underwent a laparoscopic examination. The male subjects evaluated were partners of the female subjects. Each underwent a clinical examination, semen evaluation including

1175

S.S.Witkin et al.

antisperm antibody testing and urethral cultures. History of treatments for previous genital infections was noted. Clinical diagnosis Tubal factor infertility was diagnosed by laparoscopy or hysterosalpingography as an incomplete or complete stenosis, hydrosalpinx or peritubo–ovarian adhesions. PID was diagnosed by laparoscopy or by the presence of at least three of the following clinical signs: pelvic pain, fever, hyperleukocytosis, modified erythrocyte sedimentation rate, elevated C-reactive protein. Endometriosis was diagnosed by laparoscopy and confirmed by histology. Spontaneous abortion was defined as a pregnancy loss prior to the 13th week of gestation. Microbiology Ureaplasma urealyticum and Mycoplasma hominis were identified by ammonia production after urea or arginine hydrolysis (Mycofast Screening Evolution; International Microbio, Signes, France). C.trachomatis was detected in cervical and female urethral swabs by Amplicor PCR (Produits Roche, Neuilly sur Seine, France). Anaerobic bacteria were identified by transporting cervical swabs in Shaedler medium and plating on Shaedler agar (BioMe´rieux, Marcy-L’Etoile, France). After 2 and 7 days in an anaerobic chamber at 37°C colonies not growing under aerobic or microaerobic conditions were isolated and identified (API-ANA; BioMe´rieux). The samples positive for anaerobic bacteria contained one or more of the following: grampositive cocci (peptostreptococci), gram-negative bacilli (Bacteroides, Prevotella, Mobiluncus, Fusobacterium) and gram-positive bacilli (Propionibacterium, Eubacterium, Catenebacterium). Bacterial vaginosis was diagnosed when at least three of the four Amsel criteria (Amsel et al., 1983) were present, a gram stain yielded a Nugent score (Nugent et al., 1991) of ù6 and at least one anaerobic organism, and/or U.urealyticum or M.hominis were identified by culture. Antibody assays Sera were assayed for immunoglobulin (Ig)G antibodies to C.trachomatis surface components by the commercially available SeroELISA (Savyon Diagnostics, Ashdod, Israel) and the titres determined in reference to a microimmunofluorescence antibody scale in an arithmetic progression. Antibodies to recombinant human and E.coli hsp60 (StressGen, Victoria, BC, Canada) were detected by enzyme-linked immunosorbent assay (ELISA). The hsp60 were bound to wells of a microtitre plate (1 µg/well) by overnight incubation in carbonate buffer (Witkin et al., 1996). Sera to be tested were diluted 1:200 in phosphatebuffered saline–Tween 20 detergent buffer (PBS–Tween) and 0.1 ml aliquots were added to duplicate wells and incubated by floating the plate in a 37°C water bath for 60 min. The wells were then washed with PBS–Tween and a 1:200 dilution of alkaline phosphatase (AP)-conjugated antibody to human IgG (Kirkegaard and Perry, Gaithersburg, MD, USA) was added to the wells. After an additional 60 min incubation at 37°C the wells were again washed and the colourless AP substrate, p-nitrophenylphosphate in 10% diethanolamine buffer, was added. After a 30–60 min incubation at room temperature, the intensity of the yellow colour that formed was measured at 405 nm. Known positive and negative sera were always assayed in parallel to the test samples. Duplicates differing by .10% were retested. Antibodies to two synthetic biotinylated peptides corresponding to epitopes of the C.trachomatis hsp60 were detected by ELISA as previously described (Witkin et al., 1996). The peptides corresponded to amino acids 151–162 (chsp 151–162) and 260–271 (chsp 260– 271) of chlamydial hsp60. Antibodies in human sera to chsp 260–

1176

Table I. Antibodies to Chlamydia trachomatis and to heat shock proteins in sera from men and women seeking a consultation for infertility (sera from 50 women and 45 men were tested for immunoglobulin G antibodies by enzyme-linked immunosorbent assay) Antibody

C.trachomatis Escherichia coli hsp60 Human HSP60 C.trachomatis hsp60: chsp 151–162 chsp 260–271

No. positive (%) Women

Men

32 (64.0) 6 (12.0) 9 (18.0)

23 (51.1) 2 (4.4) 4 (8.9)

2 (4.0) 16 (32.0)a

2 (4.4) 6 (13.3)

aP 5 0.03 compared with men. hsp60 5 60 kDa heat shock protein.

271, but not antibodies to chsp151–162, were previously shown to react with the corresponding epitope in human HSP60 (Yi et al., 1993). For each ELISA, a sample was defined as positive for IgG antibodies if it yielded an absorbance value .1.0, which in each case was .2 SD above the mean value obtained with a panel of negative control sera. Statistical analysis Correlations between discrete variables were evaluated using Fisher’s exact test. P , 0.05 was considered to be statistically significant.

Results Circulating antibodies to C.trachomatis and heat shock proteins Results of IgG antibody testing on the 50 female and 45 male subjects are summarized in Table I. Individual results of the testing for hsp60 antibodies are shown in Figure 1. Antibodies to C.trachomatis surface components were detected in 32 (64%) of the women and 23 (51.1%) of the men. Six women (12%) and 2 men (4.4%) were positive for antibodies to the E.coli hsp60 while 9 women (18%) and 4 men (8.9%) had antibodies to the human HSP60. None of these differences between women and men was statistically significant. Utilizing the two synthetic peptide epitopes corresponding to regions of the chlamydial hsp60, 16 (32%) of the women had antibodies to chsp 260–271. This was greater than the prevalence in men (six out of 45, 13.3%) (P 5 0.03). Only two women (4%) and two men (4.4%) were positive for antibodies to chsp 151–162. The presence of antibodies to chsp 260–271 in women was associated with a high antibody titre to chlamydial surface antigens (Table II). Antibodies to chsp 260–271 were detected in 11 of 17 subjects (64.7%) with a C.trachomatis antibody titre of .160, in two out of 15 (13.3%) with a titre of 20– 160 (P 5 0.004) and in two out of 17 with a titre of ,20 (P 5 0.001). Antibodies to the human HSP60 were also associated with humoral immunity to chsp 260–271 (Table III). Antibodies to this chlamydial hsp60 epitope were present in 50% of subjects with antibodies to the human HSP60 as opposed to only 18.6% of subjects negative for antibodies to human HSP60

Antibodies to chlamydial hsp60

(P 5 0.03). In contrast, only two subjects with antibodies to human HSP60 were also positive for E.coli hsp60 antibodies. Antibodies to C.trachomatis surface antigens and hsp60 and pregnancy outcome The relationship between infertility diagnosis and antibodies to chsp 260–271, E.coli hsp60 and human HSP60 in women is shown in Table IV. Antibodies to this chlamydial hsp60 epitope were more prevalent (P 5 0.004) in women with a history of at least two spontaneous abortions (eight out of 13, 61.5%) than in women with other known diagnoses (6/ 35, 17.1%). In contrast, antibodies to C.trachomatis surface antigens were identified with a similar prevalence in women with (69.2%) and without (62.2%) a history of at least two spontaneous abortions.

Figure 1. Antibodies to the 60 kDa heat shock protein (hsp60) of Escherichia coli, Chlamydia trachomatis and man. Sera from 50 females (F) and 45 males (M) were tested by enzyme-linked immunosorbent assay for immunoglobulin (Ig)G antibodies to recombinant E.coli and human HSP60 and to two synthetic peptides corresponding to epitopes of the C.trachomatis hsp60. The horizontal line indicates the lower boundary of a positive antibody response.

Table II. Relationship between Chlamydia trachomatis antibody titre and antibodies to a synthetic peptide (chsp 260–271) of the chlamydial hsp60 C.trachomatis antibody titre

No. subjects

No. positive for antibodies to epitope 260–271 (%)

.160 .20 to ,160 ø20 unknown

17 15 17 1

11 (64.7)a,b 2 (13.3) 2 (11.8) 1

5 0.004 compared with .20 to ,160; bP 5 0.001 compared with ø20. hsp60 5 60 kDa heat shock protein.

aP

Table III. Relationhip between antibodies to the human 60 kDa heat shock protein (HSP60) and antibodies to chsp 260–271 and to the Escherichia coli hsp60 Pair Chlamydia trachomatis hsp 260–271 IgG in samples positive for human HSP60 IgG C.trachomatis hsp60 260–271 IgG in samples negative for human HSP60 IgG E.coli hsp60 IgG in samples positive for human HSP60 IgG E.coli hsp60 IgG in samples negative for human HSP60 IgG IgG 5 immunoglobulin G. NS 5 not significant.

No. positive (%) 5/10 (50.0)

P value 0.03

16/86 (18.6) 2/10 (20.0) 6/86 (7.0)

NS

Antibodies to C.trachomatis, E.coli and human HSP60 and lower genital tract microbiological findings The relationship between current lower genital tract microbiological findings and humoral immunity to the hsps is shown in Table V. There was no relationship between microbial findings and prevalence of antibody to chsp 260–271 or the E.coli hsp60. However, although the numbers are small, women with U.urealyticum in their lower genital tract had a higher prevalence of circulating antibody to human HSP60 (four out of six, 66.7%) than did women negative for this mycoplasma (eight out of 35, 22.9%) (P 5 0.05). Discussion Immunity to the C.trachomatis hsp60 is more typically associated with a chronic upper genital tract infection and Fallopian tube damage than with an acute infection of the lower genital tract (Brunham et al., 1985; Wager et al., 1990; Witkin et al., 1993; Toye et al., 1993; Arno et al., 1995). In addition, since hsp60 is a highly conserved protein, development of immune sensitization to conserved hsp60 epitopes probably requires prolonged exposure to an elevated concentration of hsp60. Such a situation has been postulated to occur in women as a consequence of an asymptomatic, and therefore undetected and untreated, chlamydial infection of the Fallopian tubes (Morrison et al., 1992; Witkin, 1995). In-vitro studies have demonstrated that C.trachomatis may persist in an unculturable intracellular state in which the synthesis of structural proteins is greatly reduced but hsp60 production actually increases (Beatty et al., 1993). Evidence consistent with chlamydial persistence within Fallopian tubes of infected women has also been published (Patton et al., 1994). These findings might explain the observation that, although men and women in this study had an equivalent prevalence of antibodies to C.trachomatis surface antigens, humoral immunity to a conserved peptide epitope of the C.trachomatis hsp60, chsp 260– 271, was more prevalent in women than in men. In a previous study it was demonstrated that chsp 260–271 was an immunodominant epitope in the chlamydial hsp60 for women undergoing IVF and that the presence of cervical IgA antibodies to this peptide correlated with embryo loss after a transient implantation of embryos in the uterus (Witkin et al., 1177

S.S.Witkin et al.

Table IV. Relationship between cause of infertility and antibodies to the Chlamydia trachomatis, Escherichia coli or human 60 kDa heat shock protein (HSP60) Diagnosis

PID SpAba Tubal factor Endometriosis Fertile Unknown

No. subjects

17 13 8 7 3 2

No. positive (%) chsp 260–271

E.coli

Human

2 8 2 2 0 2

3 (17.6) 1 (7.7) 2 (25.0) 0 0 0

3 3 2 1 0 0

(11.8) (61.5)b (25.0) (28.6)

(17.6) (22.1) (25.0) (14.3)

PID 5 pelvic inflammatory disease; SpAb 5 spontaneous abortion. aWomen with two or more spontaneous abortions. bP 5 0.004 compared with women with other known diagnoses.

Table V. Relationship between current lower genital tract infections and antibodies to the Chlamydia trachomatis, Escherichia coli and human 60 kDa heat shock protein (HSP60) Organism

None Bacterial vaginosis Anaerobes G.vaginalis Ureaplasma urealyticum C.trachomatis Unknown aP

No. subjects

13 10 8 7 6 3 9

chsp 260–271

E.coli

human

3 3 2 3 4 1 3

1 0 1 1 1 1 1

1 3 3 1 4 0 0

(23.1) (30.0) (25.0) (42.9) (66.7) (33.3) (33.3)

(7.7) (12.5) (14.3) (16.7) (33.3) (11.1)

(7.7) (30.0) (37.5) (14.3) (66.7)a

5 0.05 compared with all others.

1996). The present observation that systemic immunity to this conserved chlamydial hsp60 epitope was associated with a history of at least two spontaneous abortions is consistent with this finding. The mechanism of this effect remains to be elucidated. As previously postulated (Witkin et al., 1994, 1996), perhaps expression of human HSP60 in the developing embryo or uterine decidua (Bensuade and Morange, 1983; Mincheva-Nilsson et al., 1994) reactivated lymphocytes previously sensitized to the chlamydial hsp60, resulting in proinflammatory cytokine production and immune rejection of the embryo (Askienazy-Elbhar, 1996). The data also suggest that humoral immunity to chsp 260–271 is a better predictor of spontaneous abortion than antibody to C.trachomatis surface components, at least utilizing the assays in this study. Further experimentation measuring humoral immunity to surface components of all chlamydial serovars is required before the validity of this statement can be definitively accepted. Antibodies to chsp 260–271 of the chlamydial hsp60 were also associated with antibodies to the human HSP60. This confirms a recent report in women with PID (Domeika et al., 1998) and is consistent with a previous report demonstrating that antibodies reactive with this epitope were also capable of recognizing the corresponding epitope in human HSP60 (Yi et al., 1993). In contrast, although the E.coli hsp60 epitope corresponding to chsp 260–271 also closely resembles the human epitope, there was no relationship between antibodies to the E.coli hsp60 and antibodies to human HSP60. This reinforces the suggestion that immunity to conserved hsp60 epitopes does not occur after an acute infection but, rather, is 1178

No. positive (%)

a consequence of a prolonged exposure to an hsp60. In the present study there was no relationship between a current lower genital tract infection and antibodies to bacterial or human HSP60, except for a possible association between U.urealyticum and anti-human HSP60 IgG. Development in some women of immune sensitization to their own hsp60 may impair subsequent fertility, as mentioned above. In addition, autoimmunity to hsp60 in man has also been associated with autoimmune diseases (Panchapakesan et al., 1992; Georgopoulos and Macfarland, 1993). Prospective studies are needed to clarify the variables associated with immunity to the C.trachomatis hsp60 and its relationship to development of autoimmunity to human HSP60, pregnancy failure and autoimmune disease. The need for further research on the relationship between immunity to specific epitopes of chlamydial heat shock proteins and immunopathogenesis of this infection has been emphasized in a recent review (Paavonen and Lehtinen, 1996).

Acknowledgements The technical assistance of A.M.Bongiovanni and C.Gautheron and critical review of the manuscript by Andreas Neuer and Jan Jeremias are gratefully acknowledged.

References Amsel, R., Totten, P.A., Spiegel, C.A. et al. (1983) Non-specific vaginitis: diagnostic criteria and microbial and epidemiologic associations. Am. J. Med., 74, 14–22.

Antibodies to chlamydial hsp60 Arno, J.N., Yuan, Y, Cleary, R.E. and Morrison, R.P. (1995) Serologic responses of infertile women to the 60-kD chlamydial heat shock protein (hsp60). Fertil. Steril., 64, 730–735. Askienazy-Elbhar, M. (1996) Immune consequences of Chlamydia infections in pregnancy and in vitro fertilization. Infect. Dis. Obstet. Gynecol., 4, 143–148. Beatty, W.L., Byrne, G.I. and Morrison, R.P. (1993) Morphologic and antigenic characterization of interferon γ-mediated persistent Chlamydia trachomatis infection in vitro. Proc. Natl. Acad. Sci USA, 90, 3998–4002. Bensuade, O. and Morange, M. (1983) Spontaneous high expression of heatshock proteins in mouse embryonal cells and ectoderm from day 8 mouse embryos. EMBO J., 2, 173–177. Brunham, R.C., Maclean, I.W., Binns, B. and Peeling, R.W. (1985) Chlamydia trachomatis: its role in tubal infertility. J. Infect. Dis., 152, 1275–1282. Cates, W. Jr. and Wasserheit, J.N. (1991) Genital chlamydial infections: epidemiology and reproductive sequelae. Am. J. Obstet. Gynecol., 164, 1771–1781. Domeika, M., Domeika, K., Paavonen, J. et al. (1998) Humoral immune response to conserved epitopes of Chlamydia trachomatis and human 60kDa heat-shock protein in women with pelvic inflammatory disease. J. Infect. Dis., 177, 714–719. Georgopoulos, C. and Macfarland, H. (1993) Heat shock proteins in multiple sclerosis and other autoimmune diseases. Immunol. Today, 14, 373–375. Mincheva-Nilsson, L., Baranov, V., Yeung, M.M. et al. (1994) Immunomorphologic studies of human decidua-associated lymphoid cells in normal early pregnancy. J. Immunol., 152, 2020–2032. Morrison, R.P., Manning, D.S. and Caldwell, H.D. (1992) Immunology of Chlamydia trachomatis infections. Immunoprotective and immunopathogenetic responses. In Quinn, T.C. (ed.), Sexually Transmitted Diseases. Raven Press, New York, USA, pp. 57–84. Neuer, A., Lam, K.N., Tiller, F.W. et al. (1997) Humoral immune response to membrane components of Chlamydia trachomatis and expression of human 60 kDa heat shock protein in follicular fluid of in-vitro fertilization patients. Hum. Reprod., 12, 925–929. Nugent, R.P., Krohn, M.A. and Hillier, S.L. (1991) Reliability of diagnosing bacterial vaginosis is improved by a standardized method of Gram stain interpretation. J. Clin. Microbiol., 29, 297–301. Paavonen, J. and Lehtinen, M. (1996) Chlamydial pelvic inflammatory disease. Hum. Reprod. Update, 2, 519–529. Panchapakesan, J., Daglis, M. and Gatenby, P. (1992) Antibodies to the 65 kDa heat shock proteins in rheumatoid arthritis and systemic lupus erythematosus. Immunol. Cell Biol., 70, 295–300. Patton, D.L., Askienazy-Elbhar, M., Henry-Suchet, J. et al. (1994) Detection of Chlamydia trachomatis in fallopian tube tissue in women with postinfectious tubal infertility. Am. J. Obstet. Gynecol., 171, 95–101. Toye, B., Laferriere, C., Claman, P. et al. (1993) Association between antibody to the chlamydial heat-shock protein and tubal infertility. J. Infect. Dis., 168, 1236–1240. Wager, E.A., Schachter, J., Bavoil, P. and Stephens, R.S. (1990) Differential human serological response to two 60,000 molecular weight Chlamydia trachomatis antigens. J. Infect. Dis., 162, 922–927. Witkin, S.S. (1995) Immune pathogenesis of asymptomatic Chlamydia trachomatis infections in the female genital tract. Infect. Dis. Obstet. Gynecol., 3, 169–174. Witkin, S.S., Jeremias, J., Toth, M. and Ledger, W.J. (1993) Cell-mediated immune response to the recombinant 57-kDa heat-shock protein of Chlamydia trachomatis in women with salpingitis. J. Infect. Dis., 167, 1379–1383. Witkin, S.S., Sultan, K.M., Neal, G.S. et al. (1994) Unsuspected Chlamydia trachomatis infection and in vitro fertilization outcome. Am. J. Obstet. Gynecol., 171, 1208–1214. Witkin, S.S., Jeremias, J., Neuer, A. et al. (1996) Immune recognition of the 60kD heat shock protein: implications for subsequent fertility. Infect. Dis. Obstet. Gynecol., 4, 152–158. Yi, Y., Zhong, G. and Brunham, R.C. (1993) Continuous B cell epitopes in Chlamydia trachomatis heat shock protein 60. Infect. Immun., 61, 1117–1120. Received on July 23, 1997; accepted on January 16, 1998

1179