Chlamydia trachomatis infection in neonates, not adults, has been associated with the development of chronic respiratory sequelae. Adult chlamydial infections induce Th1-type responses that subsequently clear the infection, whereas the neonatal immune milieu in general has been reported to be biased toward Th2-type responses. We examined the protective immune responses against intranasal Chlamydia muridarum challenge in 1-day-old C57BL/6 and BALB/c mice. Infected C57BL/6 pups displayed earlier chlamydial clearance (day 14) compared with BALB/c pups (day 21). However, challenged C57BL/6 pups exhibited prolonged deficits in body weight gain (days 12–30) compared with BALB/c pups (days 9–12), which correlated with continual pulmonary cellular infiltration. Both strains exhibited a robust Th1-type response, including elevated titers of serum antichlamydial IgG2a and IgG2b, not IgG1, and elevated levels of splenic C. muridarum-specific IFN-γ, not IL-4, production. Additionally, elevated IFN-γ, not IL-4 expression, was observed locally in the infected lungs of both mouse strains. The immune responses in C57BL/6 pups were significantly greater compared with BALB/c pups after chlamydial challenge. Importantly, infected mice deficient in IFN-γ or IFN-γ receptor demonstrated enhanced chlamydial dissemination, and 100% of animals died by 2 wk postchallenge. Collectively, these results indicate that neonatal pulmonary chlamydial infection induces a robust Th1-type response, with elevated pulmonary IFN-γ production, and that endogenous IFN-γ is important in protection against this infection. The enhanced IFN-γ induction in the immature neonatal lung also may be relevant to the development of respiratory sequelae in adult life.
The high prevalence of genital chlamydial infection in women of child-bearing age results in exposure of an estimated 100,000 neonates to this pathogen in the U.S. annually (1). Chlamydia trachomatis is the causative agent of 16–22% of all reported cases of pneumonia in the U.S. occurring within 4–6 wk after birth (2). Neonatal C. trachomatis pneumonia is characterized by an insidious onset, followed by an afebrile course associated with paroxysms of coughing, tachypnoea, and diffuse bilateral pulmonary cellular infiltrates (3). Antimicrobial drugs have been effective in management of acute neonatal C. trachomatis pneumonia (4). However, an association of neonatal C. trachomatis pulmonary infection and subsequent development later in life of abnormal pulmonary function has been reported (5). The significant morbidity caused by this infection and possible sequelae underscore the need for a better understanding of the pathogenesis and immunity to C. trachomatis infection in newborns.
Adult mouse models have been extensively used to study the pathogenesis and immunity against chlamydial infection (6, 7). Nude athymic mice have been shown to be highly susceptible to pulmonary chlamydial challenge, suggesting the important role of T cells in resolution of chlamydial infections (8). MHC class II-deficient mice or wild-type mice depleted of CD4+ T cells have been shown to be incapable of chlamydial clearance (9), demonstrating an important role of CD4+ T cells in this process (10). Th1 and IFN-γ secretion (11), but not Th2, Chlamydia-specific CD4+ T cell clones have been shown to be important in optimal resolution and prevention of dissemination of the bacterium in adult mice (12, 13, 14). However, the role of IFN-γ in chlamydial clearance in newborn mice has not been evaluated. To this end, multiple studies have suggested that newborns develop biased Th2-type responses (15, 16) and have a transient defect in development of Th1-type immunity involving IFN-γ (17). However, there also is evidence to demonstrate that newborns are highly responsive to immune modulators, such as CpG-containing oligonucleotides (18) or IL-12 (19), to induce Th1-type immune responses. Given the conflicting evidence regarding induction of IFN-γ in the neonatal period and the importance of this cytokine for antichlamydial immunity in adult animals, we sought to resolve the role of endogenous IFN-γ production in protective immunity against neonatal chlamydial infection.
We have established a neonatal pulmonary infection model with C. trachomatis mouse pneumonitis (MoPn; recently designated Chlamydia muridarum) and characterized protective immunity against this organism using two genetically distinct strains of mice: C57BL/6 (H-2b) and BALB/c (H-2d) mice. Using this model, we have determined that intranasal (i.n.)3 C. muridarum challenge with 100 inclusion forming units (IFU) in 1-day-old C57BL/6 or BALB/c pups induces a self-limiting infection and acute bronchopneumonia that are resolved by ∼3 wk. Chlamydia-infected C57BL/6 mice displayed earlier bacterial clearance (day 14) compared with BALB/c mice (day 21). Moreover, C57BL/6 mice exhibited prolonged pulmonary inflammatory response (day 21) compared with BALB/c mice (day 14), which correlated with deficits in body weight gain. Pulmonary IFN-γ production and Th1-type C. muridarum-specific systemic cellular and humoral responses were detected in both strains of challenged pups, with greater elevations occurring in C57BL/6 compared with BALB/c mice. Mice deficient in IFN-γ (BALB/c IFN-γ−/− mice) and IFN-γ receptor (C57BL/6 IFN-γR−/− mice), but not the corresponding wild-type mice, succumbed to the chlamydial challenge. Taken together, these results suggest that newborn mice are capable of mounting a robust IFN-γ response, and that endogenous IFN-γ is important in protective immunity against neonatal pulmonary chlamydial infection.
Materials and Methods
C. muridarum was grown in HeLa cell monolayers and purified as described previously (20, 21). Briefly, the chlamydial elementary bodies were harvested by lysing the infected HeLa cells using a sonicator (Fisher Scientific), and the elementary bodies were purified on renograffin gradients. Aliquots of bacteria were stored at −70°C in sucrose-phosphate-glutamine buffer until further use.
C57BL/6, BALB/c, and C57BL/6 IFN-γR-deficient (IFN-γR−/−) and BALB/c IFN-γ-deficient (IFN-γ−/−) mice (4–6 wk old) were purchased from The Jackson Laboratory. In some experiments, 4–5-wk-old female C57BL/6 and BALB/c mice were used. Mice were housed and bred at the University of Texas Animal Care Facility and provided food and water ad libitum, and day-old pups were used for all experiments. Animal care and experimental procedures were performed in compliance with the Institutional Animal Care and Use Committee guidelines.
Intranasal chlamydial challenge
Groups of 1-day-old mice were challenged i.n. with infectious doses varying logarithmically from 10 to 105 IFU of C. muridarum in 5 μl of sterile PBS. Mock-challenged mice received sterile PBS. The mice were monitored daily for survival, and body weights were measured every third day for a period of 1 mo. Mice challenged with 103, 104, and 105 chlamydial IFU exhibited 100% mortality within 2 wk after challenge (data not shown). The mice challenged with 10 and 100 IFU survived the monitoring period of 1 mo. Since mice challenged with 100 IFU demonstrated significant morbidity, but not mortality, this infectious dose was used for subsequent analyses. The lung weights (0.3 ± 0.042 g) of 4–5-wk-old mice were ∼10-fold greater than those of 1-day-old neonates (0.03 ± 0.005 g); therefore, the inoculum for challenge of adult mice was normalized to the lung weights, and a dose of 1000 IFU (compared with 100 IFU in neonatal mice) was used.
Chlamydial burden in tissues
One-day-old mice were challenged i.n. with 100 IFU of C. muridarum or PBS (mock), and lungs, spleen, and liver were collected at various times as indicated. For experiments with 4–5-wk-old mice, C57BL/6 and BALB/c mice were challenged i.n with 1000 IFU of C. muridarum, and chlamydial recovery from the lungs was quantitated at the indicated periods. The tissues were homogenized, centrifuged (250 × g for 5 min), and supernatants were incubated for 24 h with HeLa cells grown on culture coverslips in 24-well plates and stained for chlamydial inclusions as described previously (22). Briefly, the infected HeLa cells were fixed with 2% paraformaldehyde, permeabilized with 1% saponin, and blocked with 10% FBS for 1 h to prevent nonspecific binding. The cells were then incubated with polyclonal rabbit anti-Chlamydia genus-specific Ab (22) for 1 h, and then incubated for an additional 1 h with goat anti-rabbit Ig conjugated to FITC (Sigma-Aldrich) plus Hoechst nuclear stain. The HeLa cell cultures on coverslips were mounted onto microscope slides using FluorSave reagent (Calbiochem). Chlamydial inclusions were enumerated using a Zeiss Axioskop 2 Plus research microscope at ×40 magnification and are expressed as means ± SD per group.
Tissue sections were stained using H&E as described previously (20). One-day-old mice were challenged i.n with C. muridarum or with PBS (mock) and euthanized at various times. Neutral formalin (10%) (fixative) was instilled intratracheally into the lungs in situ. The lung tissues were further fixed in neutral formalin overnight, dehydrated, and embedded in paraffin, and 5-μm-thick serial horizontal sections were prepared and mounted on silane-coated slides (VWR International). The tissue sections were stained with H&E and visualized using a Zeiss Axioskop 2 Plus research microscope, and images were acquired using an AxioCam digital camera (Zeiss). Pulmonary pathology was scored in a blinded fashion with images acquired at ×25 total magnification using a scoring scheme as follows: 0, no inflammation; 1, dispersed cellular infiltration; 2, rim of peribronchiolar infiltration; 3, one to two foci (≤200 μm) of peribronchiolar infiltration; 4, more than two foci (≤200 μm) of peribronchiolar infiltration; 5, one to two foci (>200 μm) of peribronchiolar infiltration; 6, more than two foci (>200 μm) of peribronchiolar infiltration; 7, one partially consolidated lobe of the lung; 8, one fully consolidated lobe of the lung; 9, more than one partially consolidated lobe of the lung; 10, more than one fully consolidated lobe of the lung. A score of 2 was added if luminal cellular plugs were detected in more than two bronchioles.
Serum Ab isotype levels by ELISA
One-day-old mice were infected i.n with C. muridarum or with PBS (mock) and bled on day 14 or 30 after challenge. Antichlamydial Ab titers in the sera were measured by ELISA as described previously (22, 23). Briefly, UV-inactivated C. muridarum at 105 IFU per well were coated onto 96-well microtiter plates and incubated overnight at 4°C in sodium bicarbonate buffer (pH 9.5). The plates were washed and blocked for 1 h at room temperature with 10% FBS solution. Serial dilutions of sera were added to the wells and incubated at room temperature for 2 h. The plates were then washed and incubated for an additional 1 h with anti-mouse total Ig, IgG1, IgG2a, IgG2b, IgM, or IgA conjugated to alkaline phosphatase (Southern Biotechnology Associates). Following incubation for 1 h, the plates were washed and p-nitrophenyl phosphate substrate was added for color development. Absorbance was measured at 405 nm using an ELISA microplate reader (BioTek Instruments), and reciprocal serum dilutions corresponding to 50% maximal binding were used to obtain titers.
Cellular cytokine responses
One-day-old mice were challenged i.n with C. muridarum or with PBS (mock) and euthanized on day 14 after challenge. Spleens or cervical lymph nodes were removed, single cell suspensions made, and cellular cytokine responses were measured as described previously (22, 23). Briefly, splenocytes (106/well) or lymph node cells (5 × 106/ml) were incubated with 104 or 105 IFU of UV-inactivated C. muridarum for 72 h at 37°C. Supernatants were then collected and analyzed by ELISA for IL-2, IFN-γ, or IL-4 production using BD OptEIA kits (BD Pharmingen) according to the manufacturer’s instructions. The limits of detection of the ELISAs performed were 3.1, 31.25, and 7.81 pg/ml for IL-2, IFN-γ and IL-4, respectively.
Quantization of pulmonary cytokines by real-time PCR and ELISA
Animals were infected i.n with C. muridarum or PBS (mock) and euthanized at the indicated times. Lungs were collected, snap frozen using liquid nitrogen, and total RNA was obtained using TRIzol reagent (Invitrogen) according to the manufacturer’s instructions. RNA was quantitated and first-strand cDNA synthesis was done using SuperScript II (Invitrogen). Real-time PCR of the cDNA products was performed as described previously (20) using DyNAmo SYBR green qPCR kit (Finnzymes) and DNA Engine Opticon II continuous fluorescence detection system (MJ Research). Sense and anitsense primers used were as follows: murine IFN-γ, 5′-TCAAGTGGCATAGATGTGGAAGAA-3′ and 5′-TGGCTCTGCAGGATTTTCATG-3′; and GADPH, 5′-TTCACCACCATGGAGAAGGC-3′ and 5′-GGCATGGACTGTGGTCATGA-3′. The cDNA was amplified for 40 cycles using the following conditions: 1 s of denaturation at 95°C, 10 s of primer annealing at 62.5°C, and 20 s of elongation at 72°C. Quantification was conducted at 78°C for IFN-γ and 72°C for GADPH at the end of each elongation cycle. The fluorescent dye SYBR green was used to monitor the RNA induction pattern. Relative quantification was conducted by normalizing the levels of IFN-γ to the housekeeping gene GADPH, and to baseline levels of cytokine gene expression in uninfected mice at the corresponding time periods. The values obtained were expressed as fold changes of gene expression. Additionally, C. muridarum-infected lung tissues from neonatal C57BL/6 and BALB/c mice were collected at the indicated time periods, homogenized in 500 μl of sterile PBS, centrifuged, and the supernatants were stored at −80°C until further analyses. The collected supernatants were then analyzed by ELISA for IFN-γ production using BD OptEIA kits according to manufacturer’s instructions.
SigmaStat (Systat Software) was used to perform all tests of significance. Student’s t test was used for comparisons between two groups, and ANOVA was used when more than two groups were compared. Differences were considered statistically significant for p values <0.05. All data are representative of at least two to three independent experiments, and each experiment was analyzed independently.
Clinical course of illness and bacterial clearance in neonatal C57BL/6 and BALB/c mice following i.n. C. muridarum challenge
Groups of 1-day-old C57BL/6 and BALB/c mice (n = 3) were challenged i.n. with 100 IFU of C. muridarum (a challenge dose that was determined to cause morbidity, but not mortality) or PBS (mock), and body weights were measured every third day for a period of 1 mo to monitor the clinical course of illness. As shown in Fig. 1, C57BL/6 mice exhibited significant reduction in body weight gain between days 12 and 30 (p < 0.001) after challenge when compared with mice treated with PBS (mock). In contrast, the body weights of C. muridarum-challenged BALB/c mice were comparable to PBS (mock)-treated mice at each time point, except on days 9 and 12 (p < 0.001) after challenge. These analyses revealed that C. muridarum-challenged C57BL/6 mice display a prolonged clinical course of greater severity (p < 0.01) than do similarly treated BALB/c mice.
To evaluate the kinetics of bacterial clearance, the C. muridarum-challenged mice were euthanized at the indicated intervals, and chlamydial recovery from the lungs, liver, and spleen was measured. As shown in Fig. 2, C57BL/6 mice exhibited high numbers of Chlamydia in the lungs on day 4 (1020 ± 99 IFU), with progressively reducing numbers on days 7 (489 ± 92 IFU) and 10 (310 ± 56 IFU) after challenge. These mice demonstrated chlamydial dissemination to the liver on days 7 (176 ± 19 IFU) and 10 (110 ± 15 IFU) and to the spleen on day 10 (85 ± 12 IFU) postchallenge. The infection was completely resolved in all of the C57BL/6 mice between days 14 and 18. Similarly challenged BALB/c mice exhibited high numbers of Chlamydia in the lungs on day 4 (1500 ± 90 IFU), with progressively reducing numbers on days 7 (992 ± 78 IFU), 10 (598 ± 67 IFU), 14 (131 ± 34 IFU), and 18 (34 ± 13 IFU) after challenge. BALB/c mice also exhibited chlamydial dissemination to the liver by days 7 (342 ± 12 IFU), 10 (189 ± 10 IFU), and 14 (85 ± 9 IFU), and to spleen by day 10 (24 ± 8 IFU) postchallenge. The infection was completely resolved in all of the BALB/c mice by day 21 after i.n. C. muridarum challenge. These results demonstrate that both groups of mice exhibit comparable kinetics of bacterial dissemination and resolution within 3 wk after challenge, and that BALB/c mice display greater lung chlamydial burden and time to bacterial clearance compared with C57BL/6 animals.
Pulmonary pathology in C. muridarum-challenged C57BL/6 and BALB/c mice
We evaluated the pulmonary pathology in newborn mice following i.n. chlamydial challenge. Lung sections from C. muridarum-infected C57BL/6 and BALB/c mice (n = 3) were examined at different intervals and scored in a blinded fashion, and the individual scores for the mice in each group are shown in Fig. 3. The H&E-stained lung sections from both groups of C. muridarum-infected mice displayed diffuse interstitial edema and a thin rim of peribronchiolar inflammatory cellular infiltration on day 4 following challenge (C57BL/6: 1.67 ± 0.6; BALB/c: 2.33 ± 0.6). On day 7, both groups of mice displayed inflammatory cellular infiltrates that were bilateral, patchy, and found in the peribronchiolar and perivascular regions within both groups of mice (C57BL/6: 4.67 ± 2; BALB/c: 8.67 ± 2.3). Additionally, BALB/c mice displayed partial or complete consolidation of one or more lobes of the lung. The infiltrates up to this time period were composed predominantly of mononuclear cells and polymorphs, with abundant cellular debris at the foci of inflammation. On day 10 after challenge, multiple foci of peribronchiolar cellular infiltration were found in both groups of animals (C57BL/6: 5 ± 1.7; BALB/c: 6.33 ± 2.5). Moreover, one of the three BALB/c mice displayed partial consolidation of two lobes of the lung. On days 14 and 21, there were minimal scattered foci of peribronchiolar infiltration in the infected BALB/c mice (1.67 ± 1.15 and 1.67 ± 1.16, respectively). In contrast, C57BL/6 mice continued to display multiple, albeit progressively receding, peribronchiolar infiltrates through days 14 and 21 after challenge (5.33 ± 0.6 and 5 ± 1.73, respectively). Interestingly, all C57BL/6 mice displayed intraluminal cellular impactions in most of the inflamed bronchioles, which were composed predominantly of mononuclear lymphocytes and cellular debris on day 14 after challenge, and pathology scores were significantly greater (p = 0.008) than BALB/c mice at this time point. On day 28 after challenge, minimal inflammation was found in the lungs of both groups of mice (C57BL/6: 2.33 ± 1.15; BALB/c: 1.33 ± 0.6), with some scattered foci of peribronchiolar inflammation apparent in C57BL/6 mice. The inflammatory cellular infiltration in both groups of mice between days 7 and 28 was composed predominantly of mononuclear lymphocytes. Collectively, these histopathological analyses demonstrate that BALB/c mice exhibit greater lung pathology on days 7 and 10 compared with C57BL/6 animals. However, the cellular infiltrates in C57BL/6 mice persisted for longer periods of time after infection compared with BALB/c mice.
Antichlamydial immune responses in neonatal C57BL/6 and BALB/c mice
Humoral responses were measured in the sera from C57BL/6 and BALB/c mice (n = 3) on days 14 and 30 after challenge. Elevated titers of anti-Chlamydia total Ab (C57BL/6: 4101 ± 1100, BALB/c: 3384 ± 1044) were detected as early as day 14 (Fig. 4,A). The isotype of the Ab at this time point was predominantly IgM (C57BL/6: 3276 ± 1110, BALB/c: 876 ± 110). There were minimal levels of antichlamydial IgG1, IgG2a, IgG2b, and IgA in both groups of mice at this time point. On day 30 after challenge, both C57BL/6 and BALB/c mice displayed enhanced titers of anti-C. muridarum total Ab (C57BL/6: 8406 ± 1287, BALB/c: 2825 ± 81). Moreover, there was evidence of Ab isotype class switching with increases in IgG2a (C57BL/6: 4200 ± 120, BALB/c: 3711 ± 110) and IgG2b (C57BL/6: 2339 ± 250, BALB/c 1939 ± 259), but minimal levels of IgG1, IgM, and IgA (Fig. 4 B). PBS (mock)-challenged animals displayed minimal Chlamydia-specific Ab responses. No binding of sera was detected in plates coated with the unrelated Ag, hen egg lysozyme (HEL). These results indicate the induction of anti-C. muridarum total Ab and IgM from both groups of mice on day 14 postchallenge and elevated levels of IgG2a and IGg2b Abs, with minimal levels of IgG1, by day 30 postchallenge.
The cellular immune responses also were analyzed in C57BL/6 (n = 5) and BALB/c (n = 6) mice 14 days after i.n. C. muridarum challenge. Collected splenocytes were stimulated in vitro with 104 or 105 IFU of UV-inactivated C. muridarum, and the production of IL-2, IFN-γ, and IL-4 were measured by cytokine-specific ELISA. These analyses revealed that splenocytes from C. muridarum-infected C57BL/6 and BALB/c mice displayed dose-dependent induction of Ag-specific IFN-γ production on day 14 after bacterial challenge (Fig. 4,C). Splenocytes from C57BL/6 mice produced significantly (p ≤ 0.01) higher levels of IFN-γ (5.5 ± 1.2 ng/ml at 104 IFU and 12.72 ± 2.34 ng/ml at 105 IFU) when compared with BALB/c mice (2.2 ± 1.4 ng/ml at 104 IFU and 7.7 ± 1.7 ng/ml at 105 IFU). However, comparable levels of IL-2 production upon C. muridarum stimulation (Fig. 4,D) were observed in C57BL/6 (28 ± 2.34 pg/ml at 104 IFU and 37.45 ± 4.2 pg/ml at 105 IFU) and BALB/c (29 ± 7.8 pg/ml at 104 IFU and 29.2 ± 6.8 pg/ml at 105 IFU) mice. A similar pattern of Ag-specific IFN-γ was detected in cervical lymph node cells on day 14 after challenge, with C57BL/6 mice exhibiting significantly (p < 0.05) higher levels of IFN-γ (200 ± 10 pg/ml at 104 IFU and 800 ± 8 pg/ml at 105 IFU) when compared with BALB/c mice (119 ± 10 pg/ml at 104 IFU and 659 ± 8 pg/ml at 105 IFU) (Fig. 4 E). We and others (24) have found comparable frequencies of CD4+ and CD8+ T cells in the spleens of neonatal C57BL/6 and BALB/c mice by flow cytometry (25) (data not shown), suggesting that the induction of lower levels of IFN-γ by individual cells, rather than differences in cell frequencies, account for reduced cytokine levels in BALB/c compared with C57BL/6 cells. Splenocytes or lymph node cells from PBS (mock)-treated animals, or from any animal group stimulated with medium alone or an unrelated Ag HEL, did not induce detectable cytokine production. Additionally, there were minimal levels of IL-4 production from each animal group (data not shown). These results indicate the induction of a C. muridarum-specific Th1-type cellular response from both groups of mice after challenge, with a greater degree of response in C57BL/6 compared with BALB/c mice.
Pulmonary IFN-γ induction in neonatal C57BL/6 and BALB/c mice challenged i.n. with C. muridarum
Since we found the robust induction of Th1-type cellular and humoral immune responses, and IFN-γ has been shown earlier to be important in the resolution of chlamydial infection (26, 27), the induction of this cytokine locally in the lungs of infected mice was determined using real-time PCR (Fig. 5,A). IFN-γ gene induction could be measured as early as day 4 after challenge in C57BL/6 (61.5 ± 8-fold increase) and BALB/c (48.5 ± 6-fold increase) mice with a progressively decreasing trend from days 4 through 18 postchallenge. In parallel, the production of IFN-γ in lung homogenates of infected mice also was analyzed by ELISA at various time intervals after challenge. Both groups of infected mice exhibited high levels of IFN-γ on day 4 (1885 ± 55 and 1142 ± 45 pg/ml in C57BL/6 and BALB/c mice, respectively) compared with uninfected mice, with a progressively decreasing trend from days 4 through 18 (72 ± 33 and 82 ± 33 pg/ml in C57BL/6 and BALB/c mice, respectively) postchallenge (Fig. 5 B). Additionally, C57BL/6 mice demonstrated significantly greater (p ≤ 0.05) IFN-γ production on days 4 and 7 when compared with similarly challenged BALB/c mice. Lung homogenates from PBS (mock)-treated animals displayed minimal levels of IFN-γ at all time points examined. Additionally, there were minimal levels of IL-4 mRNA and protein in the lung homogenates from each of the animal groups (data not shown). These results reveal a pattern of IFN-γ induction that parallels bacterial clearance in the lungs.
Susceptibility of neonatal IFN-γ- or IFN-γR-deficient mice to i.n C. muridarum challenge
To evaluate the role of endogenous IFN-γ in neonatal protection against C. muridarum challenge, IFN-γ−/− (BALB/c) or IFN-γR−/− (C57BL/6) mice were infected with 100 IFU of C. muridarum and monitored for survival. As shown in Fig. 6,A, IFN-γ−/− mice (n = 5) exposed to C. muridarum as neonates were highly susceptible to the infection, as evidenced by reduction in body weight gain, with mortality observed as early as day 10 in some animals (40%), with all infected IFN-γ−/− mice succumbing to the bacterial infection by day 17 postchallenge. In contrast, similarly challenged IFN-γ+/+ mice exhibited 100% survival up to day 30, with weight gain comparable to mock-treated animals. IFN-γR−/− animals (Fig. 6 B) also demonstrated greater susceptibility, with all mice progressively losing body weight and reaching mortality as early as day 10 (60%), with all animals succumbing to the infection by day 14, compared with similarly challenged IFN-γR+/+ animals. All of the challenged wild-type IFN-γR+/+ mice survived through day 30 and exhibited much less reduction in body weight. Since both IFN-γ−/− and IFN-γR−/− mice exhibited similar levels of susceptibility to chlamydial challenge as neonates, we used the BALB/c IFN-γ−/− mice for subsequent characterization.
To examine bacterial dissemination in IFN-γ−/− mice, tissues were collected from C. muridarum-challenged mice at the indicated intervals. C. muridarum-challenged IFN-γ−/− mice exhibited a bacterial load of 1690 ± 91 chlamydial IFU in the lungs on day 4, which was comparable to IFN-γ+/+ mice (1920 ± 99 IFU). However, IFN-γ−/− mice exhibited significantly greater (p < 0.05) lung bacterial loads on days 7 (1509 ± 156 IFU), 10 (1013 ± 110 IFU), and 14 (619 ± 7 IFU) when compared with IFN-γ+/+ mice at the corresponding times (day 7: 922 ± 92 IFU; day 10: 498 ± 67 IFU; and day 14: 91 ± 344 IFU) after challenge (Fig. 7). IFN-γ−/− mice also exhibited greater dissemination to the spleen at days 7 (72 ± 2 IFU), 10 (190 ± 10 IFU), and 14 (192 ± 2 IFU) after challenge compared with IFN-γ+/+ mice (day 7: 40 ± 2 IFU; day 10: 52 ± 5 IFU; and day 14: 32 ± 3 IFU). Greater dissemination to the liver also was observed in infected IFN-γ−/− mice on days 4 (150 ± 8 IFU), 7 (201 ± 9 IFU), 10 (410 ± 19 IFU), and day 14 (419 ± 7 IFU) after challenge compared with IFN-γ+/+ mice (day 4: 41 ± 3 IFU; day 7: 188 ± 34 IFU; day 10: 179 ± 9 IFU; and day 14: 50 ± 6 IFU) after challenge. These results demonstrate greater bacterial burden and dissemination of Chlamydia in IFN-γ−/− mice compared with similarly challenged wild-type animals.
The cellular responses in IFN-γ−/− and IFN-γ+/+ mice were analyzed at day 14 following i.n. challenge. Splenocytes from challenged IFN-γ+/+ or IFN-γ−/− mice (n = 3) were stimulated in vitro with 104 or 105 IFU of UV-inactivated C. muridarum, and the production of Ag-specific IL-2, IFN-γ, and IL-4 was quantitated by ELISA. Splenocytes from C. muridarum-infected IFN-γ−/− mice expectedly did not induce IFN-γ, whereas similarly treated IFN-γ+/+ splenocytes produced IFN-γ in a dose-dependent fashion (1.7 ± 0.8 ng/ml at 104 IFU and 5.7 ± 1.7 ng/ml at 105 IFU) (Fig. 8 A). However, splenocytes from both groups of challenged IFN-γ−/− and IFN-γ+/+ mice produced comparable levels (27.18 ± 7.8 and 37.23 ± 8.17 pg/ml at 104 IFU and 37.12 ± 7.23 and 47.23 ± 8.1 pg/ml at 105 IFU) of Ag-specific IL-2. Additionally, challenged IFN-γ+/+ animals did not exhibit detectable IL-4 production, whereas splenocytes from IFN-γ−/− mice produced measurable amounts (20.34 ± 7.1 pg/ml at 104 IFU and 37.23 ± 8.12 pg/ml at 105 IFU) of this cytokine. There was minimal induction of each cytokine induced by splenocytes exposed to media alone or to the unrelated Ag HEL.
The humoral responses in sera from IFN-γ−/− and IFN-γ+/+ mice (n = 3) were measured on day 14 after challenge. Elevated levels of Chlamydia-specific total Ab (3245 ± 210), predominantly of the IgM isotype (5289 ± 421), could be detected on day 14 in IFN-γ−/− mice, which were comparable to the total Ab (4101 ± 1100) and IgM isotype (1224 ± 350) titers in sera from IFN-γ+/+ mice. There were minimal levels of antichlamydial IgG1, IgG2a, IgG2b, and IgA Ab isotypes in both groups of animals, as shown in Fig. 8 B. PBS (mock)-challenged animals displayed minimal Chlamydia-specific Ab responses, and no binding was detected in ELISA plates coated with the unrelated Ag HEL. Collectively, these results demonstrate the induction of comparable levels of C. muridarum-specific IL-2, total Ab, and IgM in IFN-γ−/− and wild-type animals, suggesting that adaptive immune responses are generated in the IFN-γ−/− mice. However, in the absence of IFN-γ, such responses do not confer protection against the infection.
Neonatal pulmonary chlamydial infections have long been reported to be associated with chronic respiratory sequelae in adult life (5). As a first step toward determining a cause-and-effect relationship, and to examine the underlying mechanisms, we established and characterized a model of neonatal pulmonary chlamydial infection in both C57BL/6 and BALB/c mice. Intranasal chlamydial challenge of neonates from both strains of mice induced a self-limiting bronchopneumonia, which was associated with reduced body weight gain and inflammatory cellular infiltrates into the lung, mimicking the infection in humans (3). A predominant Th1-type adaptive immune response, including elevated levels of C. muridarum-specific IFN-γ, but minimal IL-4, production from splenocytes and draining lymph nodes, and elevated levels of serum Chlamydia-specific IgG2a and IgG2b, but minimal IgG1, Ab were induced following challenge. Elevated levels of IFN-γ, but minimal IL-4, production was detected locally in the lungs of infected animals, and the levels of IFN-γ correlated temporally with lung chlamydial burden. C57BL/6 mice exhibited greater Th1 responses, earlier bacterial clearance, and a prolonged period of pulmonary inflammatory cellular infiltration and reduced body weight gain when compared with BALB/c mice after challenge. Importantly, IFN-γ- or IFN-γR-deficient mice displayed significantly increased chlamydial dissemination and mortality compared with similarly challenged wild-type animals. Collectively, these results demonstrate that neonatal pulmonary chlamydial infection induces a robust IFN-γ response that is required for resolution of this infection.
Elevated levels of IFN-γ, but not IL-4, expression were detected throughout the course of infection in the lungs of both groups of mice in newborns. The kinetics of pulmonary IFN-γ expression closely paralleled the lung chlamydial burden, with high levels of induction during active infection and minimal levels following complete bacterial clearance. Additionally, neonatal chlamydial challenge in mice deficient in IFN-γ or the IFN-γ receptor resulted in significantly increased bacterial dissemination and mortality. These findings are in agreement with results from murine adult pulmonary and vaginal chlamydial infection models (13, 14, 28), suggesting the induction of a predominantly Th1-type antichlamydial immune response and the importance of IFN-γ in optimal resolution of the infection (27). However, others have shown previously that newborns exhibit greater susceptibility than do adults to infections such as cytomegalovirus (29), herpes simplex virus (30), and Mycobacterium tuberculosis (31) that can be resolved only with a robust Th1-type immune response. The inability of neonates to produce high levels of IFN-γ was attributed to a predominant Th2 phenotype during the neonatal period as a consequence of age-dependent regulation of IFN-γ expression (32, 33). In contrast, newborns also have been shown to mount strong Th1-type immune responses and IFN-γ production when administered with appropriate Ags (e.g., Bacillus Calmette-Guérin vaccine) (34) or Th1-polarizing adjuvants (e.g., IL-12) (19, 35). The induction of a robust Th1 response to lung chlamydial infection in neonatal mice provides compelling support to the ability of mice to mount efficient Th1 responses in the neonatal period. The susceptibility of animals deficient in IFN-γ or IFN-γR to neonatal pulmonary chlamydial infection further supports the pivotal importance of IFN-γ in protective immunity against this pathogen in the neonatal period. Moreover, we have found that the kinetics of chlamydial clearance in each group of 4–5-wk-old mice is comparable to that in the neonates of the corresponding mouse strain. Specifically, 4–5-wk-old C57BL/6 mice challenged with C. muridarum (1000 IFU) displayed significantly reduced lung bacterial burden and earlier resolution of the infection (day 14 after challenge) when compared with similarly treated age-matched BALB/c mice (day 21 after challenge) (Fig. 9). Yang et al. (39) also have shown comparable differences in kinetics of pulmonary chlamydial clearance in 8–12-wk-old adult C57BL/6 vs BALB/c mice as seen with 4–5-wk-old mice in this study. Thus, the murine neonatal Th1 immune response to lung chlamydial infection was comparable to that of the adult mice for chlamydial clearance, and neonatal mice per se may not be more susceptible than are adult mice to Chlamydia during the course of lung infection.
The pulmonary chlamydial infection in both C57BL/6 and BALB/c neonatal mice led to an acute bronchopneumonia. Importantly, the neonatal C57BL/6 mice exhibited significantly lower lung chlamydial burden, early clearance of the bacteria, greater levels of Th1-type response including local IFN-γ production, prolonged period of lung inflammatory cellular infiltration, and reduced body weight gain (until day 30 after challenge) compared with similarly challenged BALB/c mice. The earlier chlamydial clearance in C57BL/6 compared with BALB/c mice is presumably due to the greater Th1 immune response, including lung IFN-γ production. Despite earlier bacterial clearance, C57BL/6 mice exhibited prolonged deficit in body weight gain that may in part be due to an exaggerated pulmonary cellular inflammatory response that persisted even after bacterial clearance had been achieved (days 14–28 after challenge), compared with similarly challenged BALB/c mice. Additionally, elevated levels of serum IL-6, a cytokine associated with muscular wasting and morbidity (36), were detected during the course of the infection (days 4–10 after challenge) in C57BL/6, but not BALB/c, mice (M. Jupelli and P. Arulanandam, unpublished observations). Such differential resolution of chlamydial infection and induction of pathology in different strains of mice have been reported by others (37, 38, 39, 40). The observed strain-specific differences in immune response were not surprising, because the relative Th1 bias in C57BL/6 mice vs a Th2 bias in BALB/c mice also has been reported (41, 42, 43). Collectively, these results suggest that whereas an optimal Th1 response is important for resolution of neonatal pulmonary chlamydial infection, an exaggerated response may in fact be deleterious to the host.
Accumulating evidence suggests that a strong inflammatory cellular and cytokine response, including IFN-γ, to infection in the neonatal period may play a role in the development of respiratory dysfunction later in life. In this context, pulmonary chlamydial infections in human neonates have been associated with the development of long-term sequelae such as decreased lung function (3). Recently, Horvat et al. (44) demonstrated that C. muridarum lung infection in neonatal mice resulted in reduced lung function during the adult mouse life. Such effects of chlamydial infection have been reported to be greatest when neonates are infected early, rather than at later times after birth (44). Furthermore, pulmonary infections in the neonate have been shown to alter the development of normal lung architecture and thus lead to reduced respiratory function later in life (45). Since surfactant proteins have been shown to be required for the normal development and functioning of the lung (46), it is conceivable that alteration of the normal levels of the proteins during lung development may result in long-term effects on respiratory function. Specifically, IFN-γ has been shown to modulate the infiltration of inflammatory cells and/or effects of cytokines during infection, with consequent surfactant suppression and lung dysfunction (47). The role of IFN-γ in the development of respiratory dysfunction in an age-dependent fashion can be further addressed using the murine neonatal chlamydial infection model established in this study.
We thank Michael Pammit and Heather Powell (both at the University of Texas at San Antonio) for technical assistance.
The authors have no financial conflicts of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported by National Institutes of Health Grant SO6GM008194-24.
Abbreviations used in this paper: i.n., intranasal; HEL, hen egg lysozyme; IFU, inclusion forming unit.