Induction of protective immunity against Chlamydia muridarum intravaginal infection with a chlamydial glycogen phosphorylase

Abstract

We evaluated 7 C. muridarum ORFs for their ability to induce protection against chlamydial infection in a mouse intravaginal infection model. These antigens, although encoded in C. muridarum genome, are transcriptionally regulated by a cryptic plasmid that is known to contribute to C. muridarum pathogenesis. Of the 7 plasmid-regulated ORFs, the chlamydial glycogen phosphorylase or GlgP, when delivered into mice intramuscularly, induced the most pronounced protective immunity against C. muridarum intravaginal infection. The GlgP-immunized mice displayed a significant reduction in vaginal shedding of live organisms on day 14 after infection. The protection correlated well with a robust C. muridarum-specific antibody and a Th1-dominant T cell responses, which significantly reduced the severity but not overall incidence of hydrosalpinx. The GlgP-induced partial protection against upper genital tract pathology suggests that GlgP may be considered a component for a multi-subunit vaccine. These results have demonstrated that intramuscular immunization of mice with purified proteins can be used to identify vaccine antigens for preventing intravaginal infection with C. trachomatis in humans.

Figure 1. Evaluation of 7 plasmid-regulated antigens for inducing protective immunity against C. muridaum intravaginal infection.

(A) Groups of female Balb/c mice with 5/each were immunized intramuscularly with three doses of corresponding antigens plus adjuvant (CpG+IFA) as indicated in the figure. One month following the final immunization, mice were challenged intravaginally with 2×10⁴ IFUs of C. muridarum organisms. Vaginal swabs were taken weekly as indicated along the x-axis for measuring the number of live organisms (IFUs). The IFUs from each swab was converted into Log10, and the log10 IFUs were used to calculate mean and standard deviation for each mouse group as displayed along the y-axis. The log10 IFUs were compared between 9 groups of mice at each time point using ANOVA test, followed by a two-tailed Student's t-test for comparing between the GST control group and a test antigen group. ** indicates p<0.01 while *, p<0.05. Although three antigen groups (TC0075, TC0419 & TC0519 or GlgP) displayed significant reduction in live organism shedding, only the GlgP-immunized group maintains a statistic difference after correcting multiple group comparison. (B) Number of mice with detectable infectious units (IFUs) from each group and at each time point was listed. The rates of IFU positivity were compared between the GST control and a test group using the Fisher's exact test. * indicates p<0.05. Three of the 5 mice immunized with GlgP cleared infection on day 14 post infection.

Figure 2. Effect of immunization on genital tract pathologies induced by intravaginal chlamydial infection.

The 9 groups of mice described in Fig. 1 legend were sacrificed for evaluating pathologies of the mouse genital tract tissues. (A) The genital tract gross appearance images from all 9 groups of mice were presented with each group marked with the name of the corresponding antigens used to immunize the groups. The entire genital tract from vagina to ovary was displayed from left to right (left panels) and the oviduct and ovary regions were amplified (right panels). Each oviduct was scored for hydrosalpinx severity under naked eye and a numerical score was assigned for each oviduct as shown in the corresponding images. The scoring criteria were described in the method section. (B) The number of mice with unilateral or bilateral hydrosalpinx from each group was summarized. Note that 4 of 5 mice in GST while only 1 of 5 in EB-immunized groups developed hydrosalpinx. (C) The hydrosalpinx severity scores from each group of mice (indicated along the x-axis) were displayed along the Y-axis. Note that both the EB- and GlgP-immunized groups developed significantly reduced oviduct pathologies. The hydrosalpinx severity scores were compared between different groups using ANOVA followed by the two-tailed Student t test. * indicates p<0.05.

Figure 3. Enhanced resolution of C. muridarum genital tract infection by GlgP immunization.

GlgP along with two control groups (GST as negative and EB as positive controls) of mice (15∼17mice/group) were immunized and infected as described in Fig. 1 legend. (A) Both the EB- and GlgP-immunized groups displayed significantly reduced levels of live organism shedding on day 14 postinfection and 9 of 15 mice from the EB-immunized group cleared infection on this day. (B) The incidence of hydrosalpinx from the 3 groups of mice was summarized. (C) The hydrosalpinx scores from each group were displayed along the Y-axis. ** indicates p<0.01 while * for p<0.05. The quantitative IFU and semi-quantitative hydrosalpinx score data were analyzed with Student t test while number of mice with positive IFUs or with hydrosalpinx were analyzed with Fisher's Exact Test.

Figure 4. Induction of C. muridarum-specific immune responses with GlgP immunization.

Three groups of mice with 5 in each group immunized with GST, C. muridarum (or MoPn) EB or GlgP as described in Fig. 1 legend were sacrificed prior to challenge infection for collecting blood and splenocytes. (A) The antisera after pooling from each group were used to detect the endogenous C. muridarum antigens in the infected cells using an immunofluorescence assay. The mouse antibody binding (red) was visualized with a goat anti-mouse Cy3 conjugate (red) and co-labeled with a rabbit anti-chlamydial organism antibody visualized with a goat anti-rabbit Cy2 conjugate (green) and a DNA dye (blue). To confirm the staining specificity, the antisera from GlgP- (panels b, d & f) and EB (panels c, e & g)-immunized groups were pre-absorbed with either GST-GlgP (panels d & e) or C. muridarum-infected cell lysates (f & g). The pooled anti-GlgP antisera were no longer able to label any endogenous antigens after pre-absorption with either GST-GlgP or C. muridarum-HeLa lysates. However, the binding to endogenous antigens by the pooled anti-EB antiserum was only blocked by C. muridarum-HeLa lysates. (B) Individual antisera from the 3 groups of mice, after serial dilution, were reacted with purified EBs coated onto ELISA plates. Antisera from the GlgP-immunized mice significantly recognized the whole organisms although with much lower titers comparing to the EB-immunized mice. (C) The reactivity of individual antisera from EB (1∶800 dilution) or GlgP (1∶400) -immunized mice with the whole organisms coated onto the ELISA plates was also probed with goat anti-mouse Ig isotype-specific secondary antibodies. Both EB- and GlgP-immunized mice preferentially produced higher levels of IgG2a, indicating the dominance of Th1 cytokines in these mice. ** indicates p<0.01.

Figure 5. Detection of a Th1-dominant cellular response in GlgP-immunized mice.

Splenocytes harvested from EB (open bar) or GlgP (hatched bar) -immunized mice as described in Fig. 4 legend were in vitro re-stimulated with UV-inactivated Chlamydia muridarum EB organisms at 1×10⁶ IFUs per well, 10 µg/ml GlgP or medium alone (none) as indicated at the bottom of the figure. Three days after the stimulation, the culture supernatants were collected for IFNg and IL-5 detection and the results were expressed as pg or ng/ml as displayed along the Y-axis (mean ± SD). Note that both EB and GlgP-immunized mice produced much higher concentrations of IFNg than IL-5, indicating a Th1 dominant cellular response in theses mice.

Figure 6. Minimal cross-reactivity of chlamydial GlgP-induced mouse antibodies with mouse tissue GlgP molecules.

C. muridarum-infected HeLa lysate (lane 1) or extracts from mouse liver (lanes 2–4), muscle (5–7) and brain (8–10) were resolved in a SDS polyacrylamide gel and the protein bands were blotted onto nitrocellulose membrane for Western blot detection with an rabbit anti-mouse GlgP (panel a) or the pooled mouse anti-chlamydial GlgP (b). Note that the mouse anti-chlamydial GlgP antibodies preferentially recognized chlamydial GlgP without any significant cross-reactivity with mouse tissue GlgP.