CHLAMYDIA TRACHOMATIS is the most common cause of sexually transmitted disease,1 a common cause of ocular infection,2 and has been associated with the acquisition of human immunodeficiency virus.3–5 Its sequelae include infertility and ectopic pregnancy in women6 and blindness.2
Both protective and deleterious immune responses follow exposure to C. trachomatis. Previous studies in both animals and humans show that a transitory protective immunity to C. trachomatis develops after an initial infection.7–10 Although vaccination with killed C. trachomatis tends to induce a short-lived protective immune response, vaccination also induces a hypersensitivity to infection that is persistent.11 Pelvic inflammatory disease leading to infertility in women is believed to be related to the inflammation after repeated chlamydial exposure.6 In animal models, increased deleterious genital inflammation occurs after repeated chlamydial infections.12–14
Repeated exposure to various chlamydial antigens is associated with an increased inflammatory response.15–17 In a recent study in our laboratory, we showed that mice sensitized by inoculation with a Sarkosyl-Dithiothreitol chlamydial extract and then genitally infected by Chlamydia had increased genital inflammation, and, in particular, increased eosinophilia.15 Increased protective immunity, however, did not accompany the enhanced inflammation. Other studies also have identified an eosinophilic response to chlamydial infection that is associated with repeated exposure to various chlamydial antigens.12,13,17,18 Because interleukin-5 is the principle cytokine associated with the recruitment of eosinophils,19 we hypothesized that enhanced interleukin-5 production would be associated with previous exposure to chlamydial antigen.
Previous studies suggest that interferon-γ plays a role in mounting an immune response against primary chlamydial infection.20–22 We found no difference in the protective immune response of extract-sensitized mice compared to control mice.15 Thus, we hypothesized that during chlamydial genital infection, there would be no significant difference between genital interferon-γ levels in extract-sensitized and sham-sensitized mice. Because mice infected previously show strong protective immunity against chlamydial genital reinfection,11 we expected a significant increase of genital interferon-γ levels in these mice.
We chose three groups of mice to study interferon-γ and interleukin-5 levels during chlamydial genital infection: naive mice (sham-sensitized control mice), mice sensitized to chlamydial detergent extract antigens (associated with enhanced inflammation without protective immunity), and mice infected previously (associated with strong protective immunity).
The Mouse Pneumonitis (Mopn) agent of Chlamydia trachomatis (ATCC VR-123, Rockville, MD) was grown on McCoy cell (ATCC CRL 1696) monolayers and purified using procedures described previously.15 Aliquots for later infection were frozen in 0.22-M sucrose, 10-mM NaH2PO4, 3.8-mM KH2PO4, 5-mM glutamic acid, pH 7.4 (sucrose phosphate glutamic acid buffer) and stored at −70 °C. Aliquots were thawed at 37 °C immediately before use. The inclusion forming units of the aliquots were enumerated as described.15
Specific pathogen-free female 6- to 8-week-old Balb/cByJ mice were obtained from Jackson Laboratories (Bar Harbor, ME). Mice were housed in filter top cages and supplied food and water ad libitum in an environmentally controlled room at 22° C with a 12-hour light and 12-hour dark cycle. The animals were observed in the vivarium for signs of illness before use.
Preparation of the Extract ofC. Trachomatis
The chlamydial detergent extract was prepared as described.15 In brief, renograffin-purified elementary bodies were incubated in 0.5% Sarkosyl, 10-mM Dithiothreitol extract. The extracted elementary bodies were pelleted by ultracentrifugation. The supernate was diluted below its critical micellar concentration, extensively dialyzed against 0.008 Na2HPO4, 0.002 NaH2PO4, 0.015-M sodium chloride, pH 7.4 (phosphate-buffered solution [PBS]) at 4 °C, and concentrated with a 30-kDa MWt cutoff Centriprep-30 concentrator (Amicon, Beverly, MA). A bicinchoninic acid protein assay (Pierce Chemical Co., Rockford, IL) was used to assess concentration. The chlamydial proteins contained in this extract have been described previously.15
Sensitization of Mice with the Extract ofC. Trachomatis
We sensitized mice by subcutaneous inoculation of 50 μg of chlamydial extract mixed in incomplete Freund adjuvant and resensitized 3 weeks later with 50 μg mixed in incomplete Freund adjuvant. Control mice were shamsensitized with PBS diluting buffer mixed in incomplete Freund adjuvant. Mice were infected 3 weeks after the booster dose. Genital levels of cytokines were examined immediately before and during the week after infection.
Our murine model of ascending genital infection has been described previously.15 In brief, the stage of estrus was fixed by pretreating mice with medroxyprogesterone acetate (Upjohn, Kalamazoo, MI). Mice were inoculated subcutaneously with 2.5 mg of medroxyprogesterone acetate a week before and on the day of infection. Mice were anesthetized with 1.5 mg Nembutal (Abbott Laboratories, North Chicago, IL) on the day of infection. The Nembutal induced anesthesia for 45 to 60 minutes. The mice then were intravaginally inoculated with 2.5 × 106 inclusion-forming units of the Mopn agent of C. trachomatis in 40 μl of sucrose phosphate glutamic acid buffer. A Hamilton gastight syringe with a repeating dispenser and point style 5 needle was used to intravaginally inoculate. The mice were kept in a Trendelenburg position until they awakened. Animals that were infected twice were pretreated with progesterone before each infection.
Induction of Protective Immunity by Previous Chlamydial Infection
Naive mice were infected and then reinfected 5 weeks later. Chlamydial genital infection clears 3 to 4 weeks after primary infection. Genital levels of cytokines were examined immediately before and during the week after reinfection.
At various times before and after infection, mice were anesthetized with Nembutal and then killed by cervical dislocation. The genitalia were dissected aseptically en bloc (vagina, cervix, uterus, and ovaries), immediately homogenized using a tissue homogenizer (Tekmar, Cincinnati, OH) with a rotor-stator generator (Cole Parmer, Niles, IL), and frozen at −70 °C until used in the interleukin-5 and interferon-γ enzyme-linked immunosorbent assay.
Cytokine Enzyme-Linked Immunosorbent Assay
Monoclonal anticytokine antibody (Pharmingen, San Diego, CA) at 2 μg/ml in 0.1-M NaHCO3, pH 8.2, was adsorbed overnight in 96-well Immulon plates (Dynatek Laboratories, Chantilly, VA) at 4 °C. The wells were washed with PBS-0.05% Tween 20 and blocked with PBS-10% calf serum (wash buffer). Homogenates of the genitalia or a known quantity of the appropriate cytokine (Pharmingen) were diluted serially in wash buffer. The dilutions were incubated in the wells overnight at 4 °C, and then the wells were washed. The appropriate biotin conjugated monoclonal anticytokine antibody (Pharmingen), which recognizes a different site on the same cytokine as the first monoclonal, was reacted 45 minutes at room temperature and washed. Avidin-peroxidase (Sigma Chemical Co., St. Louis, MO) was diluted 1:2000 and reacted 30 minutes. The wells were reacted with 2,2′-azino-bis(3-ethylbenzthiazoline) in 0.1-M citric acid, pH 4.35, and hydrogen peroxide as described by the manufacturer (Biorad, Melville, NY). The plates were read at 410 nm in an MR 600 microplate reader (Dynatech Instruments, Torrance, CA). The quantity of cytokine in the experimental wells was calculated from a standard curve obtained by titration of authentic cytokine. As little as 50 pg of interferon-γ and 8 pg of interleukin-5 secretion was detectable.
Because progesterone affects immune response,23,24 we examined whether progesterone pretreatment alone affects interferon-γ or interleukin-5 production. Mice were inoculated with 2.5 mg of medroxyprogesterone acetate twice, 1 week apart. The mice were not infected, and cytokines in the genitalia were evaluated. Three mice were examined at each timepoint: before or 1, 2, 3, and 7 days after the second dose of medroxyprogesterone acetate. Overall, progesterone pretreatment did not stimulate significant interferon-γ or interleukin-5 production (data not shown).
We examined homogenized genitalia's capacity to alter the levels of interleukin-5 and interferon-γ detected by our enzyme-linked immunosorbent assay by mixing known quantities of cytokine with genital homogenates. Overall, there was no significant effect on the level of these cytokines by homogenized genitalia (data not shown).
We determined whether interleukin-5 was responsible for the enhanced eosinophilia detected in extract-sensitized mice. Two groups of 11 mice were extract sensitized, and a third group was sham sensitized (control mice). Three weeks after the booster dose, each group was infected intravaginally with the Mopn agent. One group of extract-sensitized mice was depleted of interleukin-5 by intraperitoneal inoculation of 50 μg of purified antiinterleukin-5 monoclonal antibody (Pharmingen) before infection and 2 and 4 days after infection. The other group of extract-sensitized mice and control mice was sham depleted by inoculation with diluting buffer (PBS). One week after infection, the mice were killed, and the genitalia were examined histopathologically.
For histopathologic analysis, the genitalia were removed en bloc, fixed in PBS-buffered formaldehyde, embedded in paraffin, sectioned longitudinally, stained with hematoxylin and eosin, and examined microscopically as described previously.15 The evaluation was single blinded. The genital sections examined include the cervix, endometrium, myometrium, Fallopian tubes, and ovaries. We assessed qualitatively eosinophils, lymphocytes, and polymorphonuclear neutrophils on a scale of 0 to +4. The approximate range of cells that this scale represents per high-powered field for each section examined was as follows: 0 = no cells; +1 = 1 to 5 cells; +2 = 6 to 10 cells; +3 = 11 to 25 cells; and +4 = more than 25 cells. We calculated the inflammation index by dividing the sum of inflammation scores (eosinophils, lymphocytes, polymorphonuclear neutrophils) for each section of the genitalia (cervix, endometrium, myometrium, Fallopian tubes, and ovaries) by the number of genital sections examined.
The results were analyzed using Sigmastat statistical software (Jandel Scientific, San Rafael, CA). The Mann-Whitney rank-sum test was used to compare median cytokine levels among groups because the data did not fit the criteria for normality. Significance was defined as P < 0.05.
Interferon-γ Levels in the Genitalia After Chlamydial Genital Infection
We evaluated interferon-γ levels in the genitalia during primary infection in extract-sensitized and sham-sensitized control mice and during reinfection among mice infected previously. The level of interferon-γ in the genitalia of extract-sensitized and control mice did not differ before infection (Figure 1). There was, however, significantly more interferon-γ in the genitalia of mice infected previously than in control mice before infection. The level of interferon-γ peaked 72 hours after infection in sham-sensitized control mice, 48 hours after infection in extract-sensitized mice, and 24 hours after infection in reinfected mice. The level of interferon-γ, however, peaked and then declined among all three groups of mice. Although the level of interferon-γ in the genitalia of extract-sensitized mice was greater than in control mice during the 7-day observation period, the difference was not significant. Although mice infected previously showed a peak of interferon-γ in the genitalia 24 hours after infection that exceeded levels in extract-sensitized and control mice, the difference was not significant. Overall, among all three groups, there was a substantial increase of interferon-γ that peaked between 1 and 3 days after infection and diminished toward baseline during the observation period, but remained significantly elevated among mice infected previously, even 5 weeks after infection.
Interleukin-5 Levels in the Genitalia of Extract-Sensitized Mice
We evaluated interleukin-5 levels in the genitalia during primary infection in extract-sensitized and sham-sensitized control mice and during reinfection in mice infected previously. The level of interleukin-5 in the genitalia of extract-sensitized and control mice did not differ before infection (Figure 2). Extract-sensitized mice had greater levels of interleukin-5 after infection compared to those of control mice. The difference achieved significance at Day 1 and was not quite significant at Day 2. By Day 3, there was no significant difference among the groups, although the level in the extract-sensitized mice still was greater. Overall, the increase in genital interleukin-5 levels from before infection to 1 week after infection was greater than 5 times in sham-sensitized control mice and almost 8 times in extract-sensitized mice. In both groups of mice, the interleukin-5 levels continued to increase during the 7-day observation period, with the highest level observed on Day 7.
Interleukin-5 levels in mice infected previously were significantly greater before reinfection (5 weeks after primary infection) than the maximum levels observed on Day 7 of infection for sham-sensitized control mice. Thus, interleukin-5 levels continued to increase over the 5-week period after primary chlamydial infection. During the week of observation of chlamydial genital infection, interleukin-5 levels among reinfected mice were significantly greater than were levels in extract-sensitized and sham-sensitized mice. Interleukin-5 levels among reinfected mice, however, decreased during the first day after reinfection before rising again. Thus, among all groups of mice, interleukin-5 levels in the genitalia rose more slowly and persistently compared to those of interferon-γ.
Histopathologic Effect of Interleukin-5 Depletion
We determined whether interleukin-5 was responsible for the enhanced eosinophilia associated with extract sensitization. Extract-sensitized mice that were sham depleted of interleukin-5 showed significantly greater eosinophils and lymphocytes in the genitalia compared to sham-sensitized, sham-depleted mice (control mice) (Figure 3). In contrast, extract-sensitized mice that were depleted of interleukin-5 showed a significant decrease of genital eosinophils. Extract-sensitized, interleukin-5 depleted mice, however, had significantly increased lymphocytes compared to extract-sensitized, sham-depleted mice. Thus, interleukin-5 depletion abrogated the enhanced eosinophilia found in the genitalia of extract-sensitized mice, but also resulted in enhanced lymphocyte counts in the genitalia.
Both transitory protective and persistent deleterious immune responses follow exposure to C. trachomatis. Inflammation from repeated chlamydial exposure is thought to lead to such deleterious sequelae as infertility in women.11 Mice that have been sensitized with a chlamydial detergent extract and then genitally infected with the Mopn agent show increased genital inflammation and, in particular, increased genital eosinophils 1 week after chlamydial genital infection.15 These mice, however, show no increase in protective immunity.
In the present study, we sought to evaluate differences between the protective immune response and the nonprotective immune response. We compared the immune response among sham-sensitized control mice, extract-sensitized mice, and mice infected previously. We studied sham-sensitized mice because these mice have no previous chlamydial exposure and are susceptible to chlamydial infection. We studied extract-sensitized mice because these mice show an enhanced immune response, particularly an enhanced eosinophilia, to reexposure to chlamydial antigens, but no enhanced protective immunity. We studied mice infected previously because these mice have strong protective immunity against chlamydial reinfection. In particular, we studied the interleukin-5 response because it is associated with eosinophilia19 and the interferon-γ response because it has been associated with protective immunity.21,22 We chose to use our standard infecting dose in these experiments. Whether the immune response is different using different infecting doses or different levels of chlamydial antigen sensitization remains to be explored.
Mice sensitized with a chlamydial detergent extract and then genitally infected with the Mopn agent of C. trachomatis showed enhanced production of interleukin-5 during the first few days of infection compared to sham-sensitized control mice. This early increase in interleukin-5 levels in the genitalia is associated with enhanced genital eosinophilia reported by us in a previous study.15 The absence of enhanced protective immunity, however, suggests that the enhanced interleukin-5 response and the enhanced eosinophilic response may not have contributed to protective immunity. Thus, interleukin-5 production is associated with enhanced inflammation that is nonprotective.
Unlike interferon-γ, interleukin-5 levels in the genitalia continued to increase at Day 7 postinfection in all three groups. Mice genitally infected with the Mopn agent clear infection after 2 to 3 weeks. Yet, 5 weeks after primary infection, before reinfection, mice showed elevated interleukin-5 levels. Thus, interleukin-5 levels in the genitalia persist despite the absence of detectable infection. Although controversial, data from animal and human chlamydial infections suggest that Chlamydia may persist in vivo despite the inability to detect infection by culture techniques.25–27 Whether persistent C. trachomatis or chlamydial antigens contribute to the continued presence of interleukin-5 in the genitalia has not yet been determined. Nevertheless, this is the first time we are aware of a report of a mechanism that may underlie a persistent inflammatory response after chlamydial infection. Whether and how these observations contribute to the sequelae of chlamydial infection is as yet unexplored.
Interleukin-5 is produced by T cells and mast cells.28 Interleukin-5 recruits eosinophils, prolongs eosinophil survival, enhances eosinophil degranulation, and promotes superoxide anion production. In the mouse, it acts on B cells and enhances immunoglobulin A and immunoglobulin M response, although it is not active on human B cells. Human interleukin-5 acts on basophils to release histamine and leukotriene. Eosinophils are cytotoxic cells most abundant in tissues with an epithelial surface such as the lower genitourinary tract.19 Eosinophils survive longer than do neutrophils and may persist for weeks in tissue. They are abundant in toxic substances, including cationic proteins that are cytotoxic to host cells. Eosinophils secrete cytokines and also may serve as antigen-presenting cells. Whether the effector activities of the eosinophil contribute to the genital inflammatory response to chlamydial infection will be the subject of future studies.
Unlike control mice and extract-sensitized mice, mice infected previously mount a strong protective immune response that eradicates the Mopn agent from the genitalia such that it is undetectable by culture 7 days after infection. We expect that the cytokines that contribute to such a strong response would show a surge in genital levels immediately after genital infection. Interferon-γ has been associated with protective immunity against C. trachomatis.20–22 Among mice infected previously, interferon-γ levels surged during the first 24 hours after reinfection. After primary infection in extract-sensitized and control mice, interferon-γ levels peak after infection also, but the peak is lower and delayed compared to that shown among mice infected previously. In comparison, although interleukin-5 levels were elevated before reinfection, interleukin-5 levels actually decreased during the first day after reinfection. Although the presence of elevated interleukin-5 levels may have contributed to a milieu that provides protective immunity against reinfection, a decrease in genital levels while a strong protective immune response is mounted suggests that a decline in genital interleukin-5 levels is not unfavorable to the protective immune response. Thus, although interleukin-5 may contribute to a chronic inflammatory response represented by the presence of eosinophils, it may not contribute to the acute protective immune response against reinfection. Interleukin-5 production in the genitalia may, however, contribute to a persistent deleterious inflammatory response and to the sequelae of chlamydial infection.
Genital interferon-γ production in response to secondary chlamydial genital infection differs dramatically from genital interleukin-5 production (Figures 1 and 2). In contrast to interferon-γ, which increases during the first day of reinfection and then decreases, interleukin-5 levels decrease during the first day of reinfection and then increase. The pattern of interleukin-5 and interferon-γ suggests a coordinate regulation of the production of the two cytokines. Interleukin-5 production is characteristic of Th2 cells, and interferon-γ production is characteristic of Th1 cells. Coordinate regulation of Th1 and Th2 characteristic cytokines has been described previously.29 Whether the phenomenon we have observed results from such coordinate activity remains to be proved. By depleting interleukin-5 among extract-sensistized mice, we showed that interleukin-5 elicits the enhanced genital eosinophils. We are not certain why depletion of interleukin-5 led to a significant increase of genital lymphocytes. If coordinate control exists among subsets of T cells, then diminishing interleukin-5 activity may allow greater activity from other cytokines that further enhance lymphocyte proliferation.
We have shown that both interferon-γ and interleukin-5 levels increase in the genitalia after chlamydial genital infection although the patterns differed. Interferon-γ levels surged early after infection and then declined. Interleukin-5 levels, however, increased slowly and were more persistent, although there was an early increase in animals sensitized to chlamydial extract and an early decrease associated with previous chlamydial infection. Five weeks after primary infection, before reinfection, elevated genital interleukin-5 levels persist. Such a response may be associated with persistent inflammation and deleterious inflammatory sequelae.
1. Schachter J, Hanna L, Hill EC, et al. Are chlamydial infections the most prevalent venereal disease? JAMA 1975; 231:1252–1255.
2. Tarizzo ML. Field methods for the control of trachoma. Geneva: World Health Organization, 1973:9.
3. Kirby PK, Munyao T, Kreiss J, Holmes KK. The challenge of limiting spread of human immunodeficiency virus by controlling other sexually transmitted diseases. Arch Dermatol 1991; 127:237–242.
4. Ryder RW, Mibandumba N, Hassig SE, et al. Heterosexual transmission of HIV-1 among employees and their spouses at two large businesses in Zaire. AIDS 1990; 4:725–732.
5. Wasserheit J. Epidemiological synergy: interrelationships between HIV infection and other STDs. In: Chen LC, Sepulvda J, Segal SJ, eds. Aids and Women's Reproductive Health. New York: Plenum Press, 1991:47–52.
6. Schachter J. Overview of Human Diseases. Barron AL, ed. Boca Raton: Microbiology of Chlamydia CRC Press, 1988:153–165.
7. Grayston JT, Woolridge RL, Wang SP. Trachoma vaccine studies on Taiwan. Ann N Y Acad Sci 1962; 98:352–367.
8. Katz BP, Batteiger BE, Jones RB. Effect of prior sexually transmitted disease on the isolation of Chlamydia trachomatis.
Sex Transm Dis 1987; 14:160–164.
9. Ramsey KH, Newhall WJ, Rank RG. Humoral immune response to chlamydial genital infection of mice with the agent of mouse pneumonitis. Infect Immun 1989; 57:2441–2446.
10. Schachter J, Cles LD, Ray RM, Hesse FL. Is there immunity to chlamydial infections of the human genital tract? Sex Transm Dis 1983; 10:123–125.
11. Grayston JT, Wang S-P, Yeh L-J, Kuo C-C. Importance of reinfection in the pathogenesis of trachoma. Rev Infect Dis 1985; 7:717–725.
12. Patton DL, Kuo C-C, Wang S-P, Halbert SA. Distal tubal obstruction induced by repeated Chlamydia trachomatis salpingeal infections in pig-tailed macaques. J Infect Dis 1987; 155:1292–1299.
13. Patton DL, Kuo C-C. Histopathology of Chlamydia trachomatis salpingitis after primary and repeated reinfections in the monkey subcutaneous pocket model. J Reprod Fertil 1989; 85:647–656.
14. Tuffrey M, Alexander F, Taylor-Robinson D. Severity of salpingitis in mice after primary and repeated inoculation with a human strain of Chlamydia trachomatis. J Exp Pathol 1990; 71:403–410.
15. Blander SJ, Amortegui AJ. Mice immunized with a chlamydial extract have no increase in early protective immunity despite increased inflammation following genital infection by the Mouse Pneumonitis Agent of Chlamydia trachomatis.
Infect Immun 1994; 62:3617–3624.
16. Taylor HR, Johnson SL, Schachter J, Caldwell HD, Prendergast RA. Pathogenesis of trachoma: the stimulus for inflammation. J Immunol 1987; 138:3023–3027.
17. Watkins NG, Hadlow WJ, Moos AB, Caldwell HD. Ocular delayed hypersensitivity: a pathogenetic mechanism of chlamydial conjunctivitis in guinea pigs. Proc Natl Acad Sci U S A 1986; 83:7380–7384.
18. Abu El-Asrar AM, Van den Oord JJ, Geboes K, Missotten L, Emarah MH, Desmet V. Immunopathology of trachomatous conjunctivitis. Br J Ophthalmol 1989; 73:276–282.
19. Weller PF. The immunobiology of eosinophils. N Engl J Med 1991; 324:1110–1118.
20. Rank RG, Ramsey KH, Pack EA, Williams DM. Effect of gamma interferon on resolution of murine chlamydial genital infection. Infect Immun 1992; 60:4427–4429.
21. Williams DM, Byrne GI, Grubbs B, Marshall TJ, Schachter J. Role in vivo for gamma interferon in control of pneumonia caused by Chlamydia trachomatis
in mice. Infect Immun 1988; 56:3004–3006.
22. Williams DM, Grubbs BG, Schachter J, Magee DM. Gamma interferon levels during Chlamydia trachomatis
pneumonia in mice. Infect Immun 1993; 61:3556–3558.
23. Stern JE, Wira CR. Progesterone regulation of secretory component (SC): Uterine secretory component response in organ culture following in vivo hormone treatment. J Steroid Biochem Mol Biol 1988; 30:233–237.
24. Wira CR, Hyde E, Sandoe CP, Sullivan DA, Spencer S. Cellular aspects of the rat uterine IgA response to estradiol and progesterone. J Steroid Biochem Mol Biol 1980; 12:451–459.
25. Holland SM, Hudso, Bobo L, et al. Demonstration of chlamydial RNA and DNA during a culture-negative state. Infect Immun 1992; 60:2040–2047.
26. Schachter J, Moncada J, Dawson CR, et al. Nonculture methods for diagnosing chlamydial infection in patients with trachoma: a clue to the pathogenesis of disease? J Infect Dis 1988; 158:1347–1352.
27. Ward M, Bailey R, Lesley A, Kajbaf M, Robertson J, Mabey D. Persisting inapparent chlamydial infection in a trachoma endemic community in Gambia. Scand J Infect Dis Suppl 1990; 69:137–148.
28. Takatsu K. Interleukin-5. Curr Opin Immunol 1992; 4:299–306.
29. Fitch FW, McKisic MD, Lancki DW, Gajewskiand TF. Differential regulation of murine T lymphocyte subsets. Annu Rev Immunol 1993; 11:29–48.