andrologia 34, 373–378 (2002)
Accepted: August 29, 2002
First-trimester pregnancy loss and active Chlamydia trachomatis infection: correlation and ultrastructural evidence P. Vigil1, A. Tapia3, S. Zacharias2, R. Riquelme1, A. M. Salgado2 and J. Varleta3 1 Unit of Reproduction and Development, Faculty of Biological Sciences, Pontifical Catholic University of Chile, Santiago, Chile; 2Faculty of Medicine, Pontifical Catholic University of Chile, Santiago, Chile; 3Biomedical Sciences, Faculty of Medicine, University of Chile, Santiago, Chile
Key words. Chlamydia trachomatis—hamster oocyte—human spermatozoa—spontaneous abortion
Summary. The incidence of Chlamydia trachomatis (Ct) infection and the possible correlation between couples presenting with first-trimester spontaneous abortions and active Ct infection was assessed. Additionally, the ability of Ct to infect zona-free hamster oocytes was explored by incubating the oocytes with spermatozoa from infected patients. A total of 961 women and 750 men consulting our reproductive medicine centre were screened for Ct using direct immunofluorescence. The general incidence of Ct infection was 9.4% in females (90 of 961) and 13.9% in males (104 of 750). In women with spontaneous abortions the incidence of Ct was 21.0% (14 of 66) compared with 8.9% (23 of 59) for women without spontaneous abortions and term pregnancies (chi-square, P < 0.05). When both partners of the couples were considered (one or both partners infected), the incidence rose to 68.8% (22 of 32) (chi-square, P < 0.001). In vitro studies using electron microscopy demonstrated the presence of Ct on the surface of and inside the oocyte. These results indicate a correlation between an active Ct infection and spontaneous abortion. Electron microscopy studies suggested the possibility of direct oocyte infection by Ct. Two models are proposed for the pathogenesis of Ct-related early abortions: (i) direct zygote infection, and (ii) immune response to heat shock proteins expressed by the zygote and triggered by previous Ct infections.
Correspondence: Dr. P. Vigil, Unit of Reproduction and Development, Faculty of Biological Sciences, Pontifical Catholic University of Chile, Casilla 114-D, Santiago, Chile. E-mail:
[email protected]
Introduction Chlamydia trachomatis (Ct) infection is one of the most common sexually transmitted diseases. Ct causes urogenital tract infections that are frequently asymptomatic, thus contributing to dissemination through sexual contact. Furthermore, Ct can provoke nongonococcal urethritis and cervicitis in women, which in turn can lead to severe complications such as pelvic inflammatory disease, salpingitis, ectopic pregnancies and infertility. Ct is also the main cause of nongonococcal urethritis and epididymitis in men. Incidence studies are required to quantify the extent of Ct infection, particularly in asymptomatic patients who are a source of transmission of the organism. Spontaneous abortions have also been attributed to reproductive-tract infections; immune, anatomic and genetic factors; and endocrine disorders; but in 30 to 40% of cases the cause remains unexplained (Stirrat, 1990; Katz & Kuller, 1994). Different studies have shown an increased prevalence of plasma immunoglobulin-G (IgG) against Ct in women suffering spontaneous abortions, but no Ct infection was detected in cultures from the male and female partners of the couples reporting spontaneous abortions. It was concluded that Ct infection was not the direct cause of such abortions. These studies suggest an association between spontaneous abortions and post-Ct infection (Quinn et al., 1987; Witkin & Ledger, 1992). In a previous study we found that the incidence of Ct infection among male partners of infertile couples was 38.6%, while assays reflecting the real fertilizing potential of the spermatozoa (spermatozoa-human zona pellucida binding, follicular fluid-induced acrosomal reaction and zona-free
U.S. Copyright Clearance Center Code Statement: 0303-4569/2002/3406-0373 $ 15.00/0
www.blackwell.de/synergy
374
P. Vigil et al.
hamster oocyte penetration bioassays) showed no significant differences between infected partners of infertile couples and normal patients (Vigil et al., 2002). A comparison of classic sperm analyses of men infected with Ct to those of men not infected with Ct yielded no significant differences (Close et al., 1987). In addition, men infected with Ct showed no negative effects in a study designed to assess the ability of spermatozoa to penetrate cervical mucus. The only observation was a reduced and altered sperm motility pattern (Diquelou et al., 1989; Wolff et al., 1991). These studies suggest that Ct could be acting not at a pre-fertilization or fertilization level but at the post-fertilization stages. According to reported observations, the elementary and reticular body forms of Ct appear to be associated with the spermatozoa (Wolner-Hansen & Mardh, 1984; Villegas et al., 1991; Erbengi, 1993; Vigil et al., 2002). The objective of this study was to determine the incidence of Ct infection amongst patients consulting a reproductive medicine centre; to compare the incidence of Ct infection amongst women with and without first-trimester spontaneous abortions; and to investigate the possible role of Ct-infected spermatozoa as vectors and the ability of Ct to infect oocytes.
Materials and methods Patients A total of 1711 patients (961 female and 750 male) consulting our Reproductive Medicine Center (CEBRE, Santiago, Chile) was evaluated between 1998 and 1999. All patients signed an informed consent form authorizing the use of their test results in the study. Cervical specimens were taken from the female patients and seminal or urethral specimens from the male patients. The sociodemographic characteristics of female subjects in the spontaneous abortion group and the control group were equivalent. In order to detect Ct infection, the specimens were analysed by direct immunofluorescence using the MicroTrak C. trachomatis Direct Specimen Test Kit (Syva Company, Behring Diagnostics Inc., San Jose, CA). Women with spontaneous abortions group Sixty-six women with a history of one or more firsttrimester spontaneous abortions, 32 of whom were controlled with their male partners. All patients underwent hormone, genetic, immune and anatomical studies in order to rule out these factors as possible abortion causes.
Control women group Two hundred and fifty-nine women with children and no history of spontaneous abortions, who were seen in CEBRE for their annual gynaecological examination. Control couples group Seventeen couples with children and no history of spontaneous abortions, who were seen at CEBRE for routine examinations. Statistical analysis The chi-squared test with Yates’s correction was employed to estimate the significance of differences in the prevalence of infection between female patients with spontaneous abortions and control patients. Ct detection The presence of Ct infection was detected by direct immunofluorescence, using the MicroTrak C. trachomatis Direct Specimen Test Kit. All samples were analysed by a single laboratory assistant, who was totally blinded with respect to the origin of the samples (they were numbered). Cervical and seminal, or urethral, smears were dried at room temperature and fixed with 0.5 ml of 100% methanol. The smears were then covered with 30 ll of a preparation of monoclonal antibodies to the major outer membrane protein (MOMP) coded by the omp 1 gene, which is expressed in the 15 known human Ct serotypes (or serovars) in the two forms of the organism: the infectious elementary body and the metabolically active replicative reticular body. The antibodies, which were conjugated with fluorescein isothiocyanate, were detected under a fluorescence microscope after the smears were rinsed with distilled water to eliminate unbound antibodies and mounted with the kit’s mounting fluid. The Ct-positive specimens contained apple-green elementary or reticular bodies contrasting with the reddish-brown background of counterstained cells. Specimens containing ten, or more, Ct elementary bodies per smear were considered positive; they were observed under ·40 and ·100 zoom lenses as spherical corpuscles with a smooth surface measuring 0.4–1 lm in diameter. Hamster test Semen samples were obtained from Ct infected patients after 2-5 days of sexual abstinence. The patients volunteered to participate in the study for ANDROLOGIA 34, 373–378 (2002)
Chlamydia trachomatis and spontaneous abortion
which they signed an informed consent form. The ejaculates were examined within 2 h for sperm concentration, motility and morphology. Only apparently normal semen specimens were used in our study. Zona-free hamster oocytes were obtained as described previously (Yanagimachi et al., 1976; Vigil et al., 1990). One hundred ll (10 · 106 cells ml)1) sperm aliquots were incubated at 37 C and 5% CO2 with 30–40 zona-free hamster oocytes in 0.35% Biggers-Whitten–Whittingham’s-human serum albumin (BWW-HSA) medium under mineral oil. After 6 h, the percentage of fertilized oocytes was determined either by counting the number of sperm chromatin decondensations in the ovum cytoplasm with phase contrast microscopy, or by detecting Ct with direct immunofluorescence and scanning electron microscopy (SEM). For transmission electron microscopy (TEM) studies, the oocytes were incubated with the Ct-infected spermatozoa for 24 h, to allow a replicative Ct cycle and increase the chances of detecting Ct inside the oocytes. Scanning electron microscopy Ct was detected with SEM in patient spermatozoa and hamster oocytes found to be positive using the direct immunofluorescence technique. To prepare the spermatozoa and oocytes for SEM, oocytes or seminal smears were placed on slides covered with 10% polylysin, allowed to dry, and then fixed with 2% glutaraldehyde in sodium cacodylate buffer (0.24 m, pH 7.3) for 30 min. The specimens were subsequently washed in the buffer and dehydrated in increasing concentrations of acetone using a Sorvall critical point drying apparatus. Then, the preparations were shadowed with palladium gold, as described previously (Ebensperger & Barros, 1984), and observed under a Jeol JSM II SEM ( Jeol Ltd, Tokyo, Japan).
375
This study was carried out in accordance with the ethical principles of our institution. Results Of the 1771 samples from male and female subjects, 194 (11.3%) were positive for Ct infection. Ct incidence was similar in both sexes: 9.4% for women (90 of 961) and 13.9% for men (104 of 750) (Fig. 1). Incidence of Ct infection amongst women with one or more spontaneous abortions was 21.0% (14 of 66), higher than the incidence in the control women group (P < 0.01). In the group of women with spontaneous abortions, 32 provided specimens from their male partners as well. Incidence among women in this group was 21.9% (7 of 32), while incidence when considering both members of the couple as a whole, increased to 59.4% (19 of 32); these two percentages differed from those found in the general population (Table 1). The male partnersÕ incidence of Ct infection in couples presenting with and without spontaneous abortion is outlined in Fig. 2. Electron microscopy studies allowed to observe the two infectious forms of Ct in hamster oocytes following incubation with infected semen samples. The mature infectious forms – the elementary bodies – were seen with an SEM as small round
Transmission electron microscopy TEM was used to assay the semen samples of patients infected with Ct, as well as the zona-free hamster oocytes incubated for 24 h with the spermatozoa of infected patients. For the spermatozoa, a semen aliquot in phosphate buffer saline (pH 7.4) was washed at 300 g for 10 min. Similarly, the oocytes were fixed in 2% glutaraldehyde in sodium cacodylate buffer (0.24 m, pH 7.3) for 30 min and post-fixed with 1% osmium tetroxide for 30 min. Next, the specimens were stained with 1% aqueous uranyl acetate, dehydrated in increasing concentrations of acetone, embedded in low-viscosity Spurr resin and examined under a Philips (Eindhoven, The Netherlands) Tecnaik 12 (an 80 kV TEM). ANDROLOGIA 34, 373–378 (2002)
Figure 1. General incidence of Chlamydia trachomatis infection in females (a) and males (b) consulting a reproductive medicine centre. Genital samples were tested using direct immunofluorescence.
Table 1. Incidence of Chlamydia trachomatis among women and couples, as detected with direct immunofluorescence Study group
n
Women with spontaneous 66 abortions Control women 259 Couples with spontaneous 32 abortions Control couples 17
Positive Incidence specimens 14
21.0% (P < 0.01)
23 22
8.9% 68.7% (P < 0.001)
3
17.6%
376
P. Vigil et al.
Figure 2. Incidence of Chlamydia trachomatis in groups of males, females and couples, with or without first-trimester spontaneous abortions. Solid bars: with abortions. Hatched bars: proven fertility without abortions. Significance of differences was calculated using chisquared test with Yate’s correction.
Figure 4. Scanning electron microscopy microphotograph of human spermatozoa from infected male patient, with elementary body of Ct attached to plasmatic membrane. Scale bar represents 1 lm.
Figure 5. Transmission electron microscopy microphotograph of section of zona-free hamster oocyte incubated with human Ct-infected spermatozoa for 24 h. Arrows indicate Ct inside inclusion vesicle in oolemma. Inset: detailed elementary bodies of Ct. Notice characteristic double membrane. Scale bar represents 0.25 lm; in inset, 50 nm. Figure 3. Scanning electron microscopy microphotograph of zonafree hamster oocytes incubated with human Ct-infected spermatozoa for 6 h. Arrows indicate typically rounded elementary body of Ct on oocyte surface. Inset: detail of elementary body of Ct on oocyte surface. Arrowhead shows contaminating bacillus. Scale bar represents 17 lm; in inset, 2 lm.
particles adhered to the whole surface of the oocyte and the sperm (Figs 3 and 4). The reticular bodies were observed after the infectious elementary body entered the oolemma. Round structures with double membranes characterized the chlamydial elements inside the infectious reticular body (Fig. 5). As stated above, immunofluorescence showed elementary bodies of Ct attached to spermatozoa taken from infected males. Ct was also seen attached to the plasma membranes of zona-free hamster oocytes incubated with human Ct-infected spermatozoa. Discussion In this study we established the incidence of Ct infection by gender and its correlation with sponta-
neous abortion. Earlier studies placed the incidence of infection between 5.4 and 8.2% (Aliaga et al., 1985; Martı´nez et al., 1985; Romero et al., 1997). A significantly increased incidence of Ct infection was observed among women with spontaneous abortions. The increase was even higher among their male partners. Our findings are consistent with previous reports stating that early pregnancy loss could be induced by persistent, asymptomatic Ct infection spreading to the fetal tissue (Gravett et al., 1986). Also, prior exposure to Ct could encourage a chronic inflammatory response by the immune system’s defence mechanisms (Quinn et al., 1987; Witkin & Ledger, 1992) in which persistent antibodies to Ct could attack the embryonic cells. Importantly, not all researchers have found evidence supporting the above hypothesis (Plouffe et al., 1992). Some reports described a total lack of association between recurring spontaneous abortions and Ct infection (Rae et al., 1994; Rivlin et al., 1997; Sozio & Ness, 1998; Paukku et al., 1999). Although a correlation does not necessarily imply a cause-effect relation, it does support the idea that ANDROLOGIA 34, 373–378 (2002)
Chlamydia trachomatis and spontaneous abortion
Ct infection can play a role in spontaneous abortions. Thus, additional studies on the influence of Ct infection on pregnancy are needed. It is well known that Ct attaches to and infects eukaryotic cells. In our research we found Ct adhered to the sperm membrane and inside the oocyte cytoplasm. We also found the replicative reticular body of Ct inside the hamster oocytes incubated in vitro with the spermatozoa of infected patients. In a previous study, we demonstrated the presence of Ct particles inside spermatozoa (Vigil et al., 2002). This suggests that the spermatozoon, in vivo, could be an agent in the transmission of Ct to the oocyte at fertilization, which could result in zygote lysis or chronic blastomere infection in the early embryo. Our study illustrated how the Ct-infected spermatozoon can become an active agent in the dissemination of the disease and a possible vector of Ct to the zygote. An essential consideration at this point is that the hamster oocytes used in the bioassay lacked cumulus oophorus and zona pellucida cells. In oocytes with both cumulus oophorus and zona pellucida cells, it is not certain whether Ct in the spermatozoon will adhere to the plasma membrane and penetrate the cytoplasm. It is also possible that Ct elementary bodies infect the blastomeres of the hatched early embryo while the embryo migrates along the female genital tract in a Ct-infected environment. The evidence collected in this study strongly suggests that Ct is capable of penetrating the ovum cytoplasm, but it has yet to be ascertained whether this effectively occurs in vivo. Finally, to explain a correlation between Ct infection and early abortion we propose two mechanisms. The first one was suggested by Witkin (1999), who thought that chronic Ct infection in the upper genital tract would allow prolonged exposure to the 60 kD heat shock protein (hsp60) of chlamydiae. This, in turn, would lead to immunity against the conserved epitopes of a hsp60 and subsequent immunity to a person’s own hsp60, which would be produced by the early embryo. The second one considers an infected zygote, where either the Ct-infected spermatozoon would directly transmit the organism to the oocyte, or the early embryo would become infected with Ct on its way down the oviduct or inside the uterus. This would result in lysis of the zygote or early embryo. Acknowledgement We wish to thank Ms Paulina del Rı´o for her assistance with the translation of this paper from Spanish into English. ANDROLOGIA 34, 373–378 (2002)
377
References Aliaga P, Bernal J, Martı´nez M, Dabacens A, Ferna´ndez P, Nazer J (1985) Incidencia de Chlamydia trachomatis en el embarazo. Rev Chil Obstet Ginecol 50:140–149. Close CE, Wang SP, Roberts PL, Berger RE (1987) The relationship of infection with Chlamydia trachomatis to the parameters of male infertility and sperm autoimmunity. Fertil Steril 48:880–883. Diquelou JY, Pastorini E, Feneux D, Giecquel JM (1989) The role of Chlamydia trachomatis in producing abnormal movements by spermatozoa. J Gynecol Obstet Biol Reprod 18:615–625. Ebensperger C, Barros C (1984) Changes at the hamster oocyte surface from the germinal vesicle stage to ovulation. Gamete Res 9:387–397. Erbengi T (1993) Ultrastructural observations on the entry of Chlamydia trachomatis into human spermatozoa. Hum Reprod 8:416–421. Gravett MG, Nelson HP, DeRouen Timothy, Critchlow C, Eschenbach DA, Holmes KK (1986) Independent associations of bacterial vaginosis and Chlamydia trachomatis infections with adverse pregnancy outcome. JAMA 256: 1899–1903. Katz VL, Kuller JA (1994) Recurrent pregnancy loss. Am J Perinatol 11:386–397. Martı´nez M, Gubelin W, Pinto M, Ojeda J (1985) Diagnosis of Chlamydia trachomatis by direct specimen test with monoclonal antibodies. Rev Med Chile 113:769–771. Paukku M, Tulppala M, Puolakkainen M, Anttila T, Paavonen J (1999) Lack of association between serum antibodies to Chlamydia trachomatis and a history of recurrent pregnancy loss. Fertil Steril 72:427–430. Plouffe L, White EW, Tho SP, Sweet CS, Layman LC, Whitman GF, McDonough PG (1992) Etiologic factors of recurrent abortion and subsequent reproductive performance of couples. Have we made any progress in the past 10 years? Am J Obstet Gynecol 167:313–321. Quinn PA, Petric M, Barking M, Butany J, Derzko C, Gysler M, Lie KID, Shewchuck AB, Shuber J, Ryan E, Chapman MA (1987) Prevalence of antibody to Chlamydia trachomatis in spontaneous abortion and infertility. Am J Obstet Gynecol 156:261–266. Rae R, Smith I, Listen WA, Kilpatrick DC (1994) Chlamydial serologic studies and recurrent spontaneous abortion. Am J Obstet Gynecol 170:782–785. Rivlin ME, Morrison JC, Grossman JH (1997) Comparison of pregnancy outcome between treated and untreated women with chlamydial cervicitis. J Miss State Med Assoc 38:404– 407. Romero J, Prado J, Gaete V, Martı´nez J, Pizarro E (1997) Chlamydia trachomatis infection in asymptomatic men with urethritis. Usefulness of first catch urine samples. Rev Med Chile 125:1165–1171. Sozio J, Ness RB (1998) Chlamydial lower genital tract infection and spontaneous abortion. Infect Dis Obstet Gynecol 6:8–12. Stirrat GM (1990) Recurrent miscarriage II: clinical associations, causes, and management. Lancet 336:728–733. Vigil P, Herrera E, Leontic E, Bustos-Obregon E, Morales P (1990) Fertilizing ability of human spermatozoa kept for 24 hours in medium with fetal cord serum. Molec Androl 2:169–178. Vigil P, Morales P, Tapia A, Riquelme R, Salgado AM (2002) Chlamydia trachomatis infection in male partners of infertile couples: incidence and sperm function. Andrologia 34: 155–161.
378
P. Vigil et al.
Villegas H, Pinon M, Shor V, Karchmer S (1991) Electron microscopy of Chlamydia trachomatis infection of the male genital tract. Arch Androl 27:117–126. Witkin SS (1999) Immunity to heat shock proteins and pregnancy outcome. Infect Dis Obstet Gynecol 7:35–38. Witkin SS, Ledger WL (1992) Antibodies to Chlamydia trachomatis in sera of women with recurrent spontaneous abortions. Am J Obstet Gynecol 167:135–139. Wolff H, Neubert U, Zebhauser M, Bezold G, Korting H, Meurer M (1991) Chlamydia trachomatis induces an
inflammatory response in the male genital tract and is associated with altered semen quality. Fertil Steril 55: 1017–1019. Wolner-Hansen P, Mardh PA (1984) In vitro tests of adherence of Chlamydia trachomatis to human spermatozoa. Fertil Steril 42:102–107. Yanagimachi R, Yanagimachi H, Rogers BJ (1976) The use of zona-free animal ova as a test system for the assessment of the fertilizing capacity of human spermatozoa. Biol Reprod 15:471–476.
ANDROLOGIA 34, 373–378 (2002)
