Chlamydia trachomatis serovar L2-wild type and L2-ΔCT228 (LGV 434/Bu, originating from the Hackstadt Lab at Rocky Mountain Laboratories, Hamilton, MT) were propagated in HeLa 229 cells and purified by Renografin density gradient centrifugation as previously described (Caldwell et al., 1981). HeLa were grown in RPMI 1640 + 5% fetal bovine serum (FBS) at 37°C with 5% CO₂ and 1 μg/mL of cycloheximide added as needed.

The chlamydial TargeTron vector pDFTT3(bla) was modified to allow for spectinomycin selection by replacing the bla cassette with the aadA cassette (Lowden et al., 2015). The TargeTron algorithm from Sigma-Aldrich was used to predict CT228 insertion sites and primers were chosen that would target intron insertion at bp 295 (A in ATG used as position 1) in an antisense orientation (E-value of 0.200). The vector was then retargeted for insertion into CT228 using PCR with Phusion High-Fidelity PCR Master Mix (Thermo-Fisher) and primers 228A, 228B, 228C, and EBS universal (Integrated DNA Technologies, Coralville, IA; Table S1) followed by DNA restriction digestion as previously described (Johnson and Fisher, 2013). The pDFTT295 vector was propagated in Escherichia coli DH5α grown at 30°C in LB supplemented with 20 μg/ml chloramphenicol. Chlamydia trachomatis transformation and mutant selection was carried out as described in Lowden et al. using 500 μg/ml spectinomycin (Lowden et al., 2015). The passage 4 mutants were plaque purified to obtain a clonal L2-ΔCT228 strain. The plaque isolate was serially expanded in cell culture using HeLa cells and stored in sucrose-phosphate-glutamic acid (SPG) buffer at −80°C. Stocks were titered using the infectious progeny quantitation assay.

Genomic DNA was harvested from L2-wild type and L2-ΔCT228 EB stocks using the DNeasy Blood and Tissue Kit from Qiagen. PCR reactions were performed with 50 ng of genomic DNA or 5 ng of plasmid DNA and Phusion High-Fidelity PCR Master Mix. The primers used are provided in Table S1. PCR products were run on 1% agarose gels, stained with ethidium bromide, and viewed using UV transillumination. To confirm the location of the GII insertion in CT228, the PCR product generated from primers CT228-2F and CT228-2R was extracted from the agarose gel using the GeneJET gel extraction kit (Thermo-Fisher) and ligated into pJET (Thermo-Fisher) to generate pJET::CT228GIIaadA. Ligation products were transformed into E. coli DH5α and transformants were selected at 37°C on Luria-Bertani (LB) agar plates supplemented with 100 μg/ml of ampicillin. A colony was then selected for overnight growth in LB with ampicillin and the plasmid was extracted using the GeneJET Plasmid Miniprep kit (Thermo-Fisher). The gene insert in pJET::CT228GIIaadA was sequenced using the vector specific primers JETF and JETR via Sanger sequencing performed by Macrogen USA. Sequencing results were analyzed using BioEdit (TA Hall 1999) and Clone Manager (Sci-Ed Software).

Total RNA was harvested from C. trachomatis serovar L2-wild type and L2-ΔCT228 (LGV 434) infected Hela cells at 24 h post-infection using Tri Reagent (Ambion, Austin, TX). To remove contaminating DNA, all RNA samples were treated with RQ1 DNase (Promega, Madison, WI). Removal of contaminating DNA was confirmed using PCR. Reverse Transcriptase (RT)-PCR analysis was carried out using the Access Quick RT-PCR Kit (Promega, Madison, WI) following the manufacturer's instructions. All oligonucleotide primers used in this study are shown in Table S1. RT-PCR analysis of groEL (control), CT229, CT228, CT227, CT226, CT225, and CT224 open reading frames (ORFs) was performed using oligonucleotide primer pairs within each ORF, respectively. For CT228, primers pairs designed upstream (5′), downstream (3′), or spanning the TargeTron insert were employed. Amplicons were separated by 1.5% agarose gel electrophoresis. Ethidium bromide stained gels were visualized on a BioRad Imager (BioRad, Hercules, CA). Genomic DNA samples were analyzed in parallel during each assay as a PCR positive control.

Chlamydia trachomatis L2-wild type and L2-ΔCT228 were used to infect HeLa cell monolayers at 90% confluency in 24 well plates at a MOI of ~0.5; inoculated plates were centrifuged for 1 h at 700 × g then any free EBs were washed away prior to feeding cells. At 0, 6, 12, 24, and 48 h post-infection cells were lysed with water, lysates were serially diluted in Hank's Balanced Salt Solution and used to infect HeLa cells in 96 well plates. At 24 h post-infection, monolayers were fixed in methanol and stained with rabbit anti-EB antibody followed by anti-rabbit DyLight 488 (Jackson ImmunoResearch Laboratories, Westgrove PA). The number of inclusions per field of view were counted in 20 fields for each sample using a Leica DMI600B fluorescent microscope. Total Infectious Forming Units (IFUs)/ml were calculated for each sample. Titers for both the C. trachomatis L2-wild type and L2-ΔCT228 were repeated 3 times to ensure 5 μL contained 1 × 10⁶ EBs.

Extrusions were enumerated by a method similar to that described by Lutter et al. (2013). HeLa cell monolayers (50% confluence) were plated into 24-well plates and incubated overnight with MYPT1 or Scramble Targetplus Smartpool siRNA (Dharmacon, Lafayette, CO.), complexed with DharmaFECT1 in RPMI with 5% FBS. At 48 h post-transfection, cells were infected in triplicate with C. trachomatis L2-wild type and L2-ΔCT228 at a MOI of ~0.5. At 48 h post-infection, the supernatants were removed and gently pelleted by centrifugation at 300 x g with the resulting pellet gently resuspended in 50 μL of media. Trypan blue (ThermoFisher Scientific) and NucBlue Hoechst live-cell stain (ThermoFisher Scientific) were added and 10 μL samples were loaded onto a Hausser Bright-line hemacytometer and a total of 16 grids were counted for each condition. Extrusions free of host cell nuclei (to distinguish from an infected cell) were enumerated on a Leica DMI6000B fluorescent microscope. The entire experiment was repeated on 3 separate days to ensure consistent data across multiple trials.

C. trachomatis L2-wild type and L2-ΔCT228 were used to infect HeLa cell monolayers on glass coverslips in 24 well plates (CellTreat Scientific, Pepperell MA) at a MOI of ~0.5 (performed in technical triplicates). At 18 h post-infection supernatants were removed and cells were fixed in cold methanol. Recruitment of host proteins was tested with primary antibody staining to MYPT1 (United States Biological Life Sciences, Salem, MA), phospho (pSer-19)-MLC2 (Abcam, Cambridge, MA), phospho (pTyr419) Src (Millipore Sigma, Burlington, MA), Myosin IIa (ThermoFisher Scientific), Myosin IIb (ThermoFisher Scientific), and phospho (pTyr 471) Myosin Light Chain Kinase (Santa Cruz). Chlamydiae were detected with anti-Chlamydia LPS (ThermoFisher Scientific). Fluorescent secondary antibodies, anti-mouse or anti-rabbit DyLight 594 and DyLight 488 (Jackson ImmunoResearch Laboratories, West Grove, PA) were used for indirect immunofluorescence. The entire experiment was repeated on three separate occasions (n = 3 biological replicates). Twelve images were captured per condition using a Leica DMI600B fluorescent microscope and a representative image was selected for presentation.

HeLa cells were infected with C. trachomatis L2-wild type and L2-ΔCT228; at different time points during infection, cells were lysed with 2X Laemmli Sample Buffer (Biorad, Hercules, CA). Protein samples were separated by SDS-PAGE electrophoresis (12 or 15% gel) and transferred to 0.2 um nitrocellulose membrane (Bio-Rad, Hercules, CA). Membranes were blocked with 5% non-fat dry milk/5% BSA in 1X Tris-buffered saline containing 0.1% Tween-20 (TBST) for 1 h at room temperature. After blocking, membranes were incubated with primary antibodies diluted in 5% non-fat milk/5% BSA in 1X TBST at 4°C overnight followed by incubation with horseradish peroxidase conjugated anti-rabbit/mouse antibodies (Cell Signaling, Danvers, MA) for 1 h at room temperature. Detection was performed by enhanced chemiluminescence using SignalFire ECL reagents (Cell Signaling Technology, Danvers, MA). Immunoblot images were acquired using the Fluorchem E FE0622 system (ProteinSimple, San Jose, CA). Primary antibodies to MLC2 (Millipore Sigma, Burlington, MA), p-MLC2 (S19) (ThermoFisher Scientific), HsP60 (ThermoFisher Scientific), GAPDH (Santa Cruz), and MYTP1 (Cell Signaling, Danvers, MA) were used herein.

Six-week old female inbred C3H/HeJ mice were purchased from Jackson Laboratories (Bar Harbor, Maine). All animal work was performed according to The Animal Care and Use Guidelines and approved by the Institution of Animal Care and Use Committee at Oklahoma State University. At 7 and 3 days prior to infection, all mice were subcutaneously injected with a depot formulation of 2.5 mg of medroxyprogesterone acetate (Upjohn, Kalamazoo, MI) to synchronize estrous prior to infection. Mice were infected intravaginally with 1 × 10⁶ C. trachomatis L2-wild type EBs or L2-ΔCT228 EBs in 5 μL sucrose-phosphate-glutamate (SPG) to monitor both the course of infection and immune response (n = 8 infected/colonized with L2-wild type; n = 9 infected/colonized with L2-ΔCT228) and histopathology (n = 5 for L2-wild type and L2-ΔCT228). Due to the inherent technical imperfection of the intra vaginal route of infection using small volumes (e.g., 5 μL) and that the human L2 serovar produces less robust colonization relative to other strains, any mice swabbed early (Day 7) post-infection who shed 0 IFU (100% clear of infection) were not processed any further.

At days 7, 14, 21, 28, and 42 days post-infection, cervicovaginal tracts were swabbed (Puritan Diagnostics, HydraFlock 6″ 15 cm swabs; Guilford, ME) and each swab was added to tubes containing 600 μL SPG and two glass beads on ice, as adopted from previously described studies (Shaw et al., 2001, 2017). Swab samples were vortexed vigorously to liberate EBs from the swab followed by serial dilution and inoculation of confluent HeLa cell monolayers in 96 well plates (CellTreat Scientific, Pepperell MA). Plates were centrifuged at 700 × g for 1 h to promote EB entry, then incubated at 37°C for 30 min. Inoculated HeLa monolayers were washed to remove any extracellular EBs then incubated in RPMI 1640 + 5% FBS + gentamicin (10 μg/mL) for 24 h at 37°C with 5% CO₂. Plates were fixed with methanol and stained with anti-MOMP (courtesy of Dr. Harlan Caldwell) followed by anti-mouse DyLight 488 (Jackson ImmunoResearch). Twenty fields of view were counted using a Leica MI6000B fluorescent microscope. Total recoverable IFUs per mouse were calculated for each sample. While variation ranging across 2 logs of recoverable IFU exists throughout the Chlamydia murine model literature and periodically within the same study, replicated studies using this current L2 stock at 1 × 10⁶ per mouse consistently produces 10⁴-10⁵ recoverable IFU at early time points post-murine infection and is therefore used for internal comparison to mice infected with 1 × 10⁶ L2-ΔCT228 EBs herein.

To measure systemic and mucosal immune responses to infection, sera and vaginal lavages were collected 31 days post-infection for ELISA. Vaginal lavages were obtained by gently washing the vaginal vault with 60 μL 0.5% BSA-PBS twice and stored at −20°C until the ELISA was performed. Briefly, 96-well polystyrene plates (Immulon 2HB; Thermo, Milford, MA) were coated with 1 μg of formalin fixed EBs (serovar L2) per well in 100 μL TBS (50 mM Tris buffer pH 7.5, 0.15 M NaCl) overnight at 4°C. Following adsorption of fixed EBs, wells were washed to remove unbound EBs then blocked with 200 μL 2% BSA in 0.012 M Tris pH 7.4, 0.14 M NaCl, 3.0 mM KCl, 0.05% Tween 20 for 90 min at 37°C. Blocking solution was removed and wells were washed, then serial dilutions of sera and vaginal washes were added and incubated for 90 min at 37°C. Chlamydia-specific antibody titers were measured using alkaline phosphatase-conjugated anti-mouse IgG2a (sera samples) and anti-mouse IgG and IgA antibodies (vaginal lavage samples) (Southern Biotech Associates, Birmingham, AL). P-nitrophenyl phosphate (pNPP) was used as a substrate and resulting absorbance was read at 405 nm (BioTek Synergy, Winooski, VT). Antibody titers were considered positive at the highest sample dilution that displayed an absorbance that was ≥3X the absorbance of concentrated sham (uninfected) samples.

Mouse uteri were excised and immersion fixed in 10% buffered neutral formalin on days 23 and 64 post-infection for scoring inflammatory damage. Following fixation, they were processed in whole and embedded en bloc into paraffin followed by sections cut at 4 μm and stained with hematoxylin and eosin (H&E). Sections were examined and scored via light microscopy by an American College of Veterinary Pathologists (ACVP) certified veterinary pathologist with the following numerical designations: 0, normal; 1, minimal change; 2, mild change; 3, moderate change; 4, severe change. Scored parameters (mean ± SE) included an overall impression, periglandular mucinous change, hydrosalpinx, uterine tubal dilation, and luminal PMN. Morphometric analyses of endometrial luminal epithelial height were determined using calibrated measurements via Olympus CellSens software coupled to an Olympus DP70 microscope camera. Each reproductive tract was evaluated regionally and morphometrically at six locations (proximal, mid-, and distal locations of the two uterine horns) determined to be uniform and free of tissue bends and curves.

Student's unpaired two-tailed t-test with equal variance was used to compare mean ± SE growth (IFUs) and extrusion numbers between L2-wild type and L2-ΔCT228. One-way ANOVA repeated measures was used to analyze the recoverable log IFU data across four time points following murine infection with L2-wild type or L2-ΔCT228. Student's unpaired two-tailed t-test was used to analyze systemic (IgG2a) and mucosal (IgA, IgG) mean ± SE antibody titers in mice infected with L2-wild type vs. L2-ΔCT228 and performed on clinical scores for histopathology between L2-wild type and L2-ΔCT228 infected mice. Statistical analyses were performed using Prism 5.0 and data were considered significant at p < 0.05.

The chlamydial TargeTron vector pDFTT3(bla) was modified to allow for spectinomycin selection by replacing the bla cassette with the aadA cassette. The vector was then retargeted to insert into CT228 as previously described (Johnson and Fisher, 2013). The predicted insertion site was at bp 295 within CT228 in an antisense orientation. Mutants were selected using 500 μg/ml spectinomycin and plaque purified to obtain a clonal L2-ΔCT228 strain. The CT228 locus map, TargeTron insertion site, and PCR validation of the intron insertion are shown in Figures 1A–C. Sequencing of the L2-ΔCT228 mutant CT228::GII(aadA) locus confirmed intron insertion at the predicted location. HeLa cell monolayers were infected with both L2-wild type and L2-ΔCT228 for 18 h and processed for immunofluorescence using anti-CT228 and anti-LPS antibodies. CT228 is detected surrounding the chlamydial inclusion in L2-wildtype, however is absent in the L2-ΔCT228 inclusion membrane confirming the inactivation of CT228 (Figure 1D). Southern blotting confirmed a single TargeTron insertion in the L2-ΔCT228 mutant (Figure S1).

The CT229 through CT224 gene linkage map suggests that these genes are transcriptionally linked as an operon (Stephens et al., 1998). In effort to determine whether the TargeTron insertion in CT228 may exert polar effects on the transcription of other genes within the putative operon, RT-PCR analysis was performed using oligonucleotide primers (Table S1) designed within each ORF as well as spanning the TargeTron insertion site (Figure 2A). The green and red arrows in Figure 2A indicate the position of forward and reverse primers, respectively. Using total RNA harvested from Hela cells infected with C. trachomatis L2-wild type as template, amplification products were observed (Figure 2B) for each ORF (CT229, CT228, CT227, CT226, CT225, and CT224). In total RNA from C. trachomatis L2-ΔCT228 infected cells, robust amplification products were observed (Figure 2B) for CT229, CT227, CT226, CT225, and CT224. Of particular note is the observation that the region of CT228 that lies upstream of the TargeTron insertion also produced a robust amplification product, while primers spanning the insertion site do not. It was also observed that RT-PCR using primers situated downstream of the TargeTron insertion site, yet within the CT228 ORF, produce a far less robust amplification product (Figure 2B). Taken together, the data indicate that the TargeTron insertion within CT228 does not cause a polar elimination of CT227, CT226, CT225, and CT224 transcripts while it does interrupt the production of a complete CT228 transcript.

Bacterial growth was measured at 0, 6, 12, 24, and 48 h post-infection of HeLa cell monolayers with C. trachomatis L2-wild type and L2-ΔCT228. There were no statistically significant differences between growth curves of the different strains and the curves followed the expected pattern for growth (Figure 2C). L2-wild type and L2-ΔCT228 were assessed for recruitment of MYPT1 and myosin phosphatase pathway host proteins at 18 h post-infection in HeLa cell monolayers. While the L2-wild type strain recruited MYPT1 to the periphery of the inclusion (Figure 3A), the L2-ΔCT228 failed to do so (Figure 3B). However, both strains demonstrate recruitment of MLCK (pY474), MLC2 (pS19), and Myosin IIa and Myosin IIb to the inclusion co-localizing with active Src Y474 in the microdomains at 18 h post-infection (Figures 3A,B). Recruitment of MLCK (pY474), MLC2 (pS19), and Myosin IIa and Myosin IIb at 48 h post-infection was also visualized confirming recruitment for the duration of the infection (Figure S2). Western blot analysis of MLC2 (pS19) and endogenous MLC2 levels did not show any differences during the course of infection at 24 and 48 h post-infection (Figure 3D). The number of extrusions produced by both L2-wild type and L2-ΔCT228 was quantified at 48 h post-infection in Scramble or MYPT1 siRNA depleted HeLa cells [successful MYPT1 knockdown was verified by western blot (Figure 3E)]. In the Scramble treated cells, L2-ΔCT228 produced significantly more extrusions than the L2-wild type (Figure 3E, p < 0.0001) however, there was no difference in extrusion production between the L2-wild type and L2-ΔCT228 in the MYTP1 siRNA treated cells (Figure 3E). Analysis of total Chlamydia IFUs was comparable between strains and conditions (Figure 3F).

Bacterial burden in mice infected intravaginally with either 1 × 10⁶ L2-wild type EBs (n = 8) or L2-ΔCT228 EBs (n = 9) was quantitated at days 7, 14, 21, and 28 via swabbing the cervicovaginal tract and culturing on HeLa cell monolayers. The calculated numbers of recoverable log IFUs are shown in Figure 4. Early L2-wild type (Figure 4A) and L2-ΔCT228 (Figure 4B) infection in mice at day 7 produced an infectious burden that was not statistically different from one another. However, there was a statistically significant effect of time on the clearance of L2-wild type (p = 0.006) that was not present in response to L2-ΔCT228 (p = 0.371).

Mean systemic anti-chlamydia IgG2a antibody titers were significantly higher (p = 0.041) in mouse sera at day 31 post-infection with L2-wild type (n = 8) relative to L2-ΔCT228 (n = 9) as shown in Figure 5A. As expected for serovar L2, mucosal anti-Chlamydia antibody responses were low overall. There were negligible mucosal anti-Chlamydia IgG and IgA titers in mice 31 days post-infection with L2-ΔCT228 relative to L2-wild type (Figures 5B,C).

Whole reproductive tracts were collected 23 and 64 days post-infection (L2-wild type vs. L2-ΔCT228) and formalin fixed for histological analyses. Sections of uterine tissue were examined and the scores assigned by an AVCP certified veterinary pathologist via light microscopy. Individual clinical scores ranged from normal to moderate pathology for the infected mice (Table 1, raw scores provided in Table S2). As previously shown, intravaginal infection with serovar L2 overall produces less histopathology relative to other human serovars (e.g., serovar D-LC) and the mouse-specific pathogen C. muridarum (MoPn) (Shaw et al., 2017). A single infection with serovar L2 produces only minimal to mild histopathology, however it was important to assess any inflammatory damage following infection with L2-ΔCT228 since this strain had never previously been tested in vivo and studies support the concept that chlamydial genotypes may alter the capacity for ascension and survival within the upper genital tract (Menon et al., 2015). Histological parameters compared between L2-wild type and L2-ΔCT228 revealed no statistical differences with the exception of fewer mucinous changes observed in the L2-ΔCT228 infected mice (Figures 6A–D and Table 1, p = 0.02).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.