Genetic diversity of Collaborative Cross mice enables the establishment of a novel Chlamydia muridarum female genital tract infection model

Chlamydia, mouse genital infection model, Collaborative Cross, gene expression, 22 immunopathogenesis 23 24 25 26 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint

(211 words) 27 Chlamydial infection in women displays wide variation in bacterial burden, persistence, and risk 28 of upper genital tract pathology, yet the host genetic factors underlying this heterogeneity remain 29 poorly defined. We evaluated genital tract infection with Chlamydia muridarum across 20 30 Collaborative Cross (CC) strains, a recombinant inbred mouse panel that captures broad genetic 31 diversity with high within-strain reproducibility. CC strains exhibited striking differences in early 32 bacterial burden, time to clearance, and oviduct pathology, including prolonged low-inflammatory 33 infections and burden –pathology discordance that mirror key features of human disease. 34 Heritability analyses demonstrated that host genetics accounted for most of the variation 35 observed in Day 7 bacterial burden and pathological outcomes. Genome-wide scans identified 36 suggestive quantitative trait loci associated with both traits. Genes within the burden-associated 37 locus converged on host pathways implicated in chlamydial intracellular growth, including 38 membrane dynamics and lipid metabolism, ubiquitin signaling, host cell survival mechanisms, and 39 immune regulatory signaling . In contrast, genes within the pathology -associated locus were 40 enriched for pathways regulating inflammatory cell recruitment, inflammasome activation, and 41 tissue remodeling, processes central to genital tract damage following infection. Cervical 42 transcriptional profiling further revealed strain-dependent innate and adaptive immune programs 43 associated with bacterial burden and disease phenotype. Together, these findings establish the 44 CC as a powerful platform for dissecting the genetic architecture of chlamydial 45 immunopathogenesis, and for improving preclinical evaluation of vaccines and therapeutics. 46 47 48 49 50 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint

51 Chlamydia trachomatis (CT) is the most common sexually transmitted bacterial pathogen 52 worldwide. More than 70% of infections in women and 50% in men are asymptomatic (1), 53 contributing to underdiagnosis , prolonged infection, and sustained transmission. In women, 54 untreated infection can ascend to the upper genital tract and, in ~10% of cases, cause acute 55 pelvic inflammatory disease (PID) characterized by vaginal discharge, lower abdominal or pelvic 56 pain, dyspareunia, and cervical or adnexal tenderness on examination (2). However, most women 57 have no overt symptoms despite ongoing infection and upper tract inflammation, a condition 58 termed subclinical or “silent” PID (3). Both acute and subclinical PID can lead to serious 59 reproductive consequences, including chronic pelvic pain, tubal factor infertility, and ectopic 60 pregnancy (4). Many women with tubal factor infertility do not recall a prior diagnosis of PID or CT 61 infection but show serologic evidence of prior CT exposure (5, 6) . These varied clinical 62 presentations and severe morbidities highlight the urgent need to better define the pathogenesis 63 of chlamydial disease and to develop effective interventions —particularly vaccines capable of 64 reducing bacterial burden, accelerating clearance, and preventing transmission and upper genital 65 tract damage. 66 Excessive inflammation during infection can lead to tubal scarring and post-obstructive 67 dilation of the oviduct , or hydrosalpinx , in women and female mice. However, human C. 68 trachomatis strains are rapidly cleared by laboratory mice, limiting their utility for studying the 69 mechanisms that drive oviduct scarring. To overcome this, investigators commonly use 70 Chlamydia muridarum, a natural respiratory pathogen of mice (7), as a genital tract infection 71 model via intravaginal inoculation. C. muridarum has evolved mechanisms to evade interferon 72 gamma ( IFN-γ)-induced GTPase -mediated restriction of its growth (8), enabling a productive 73 infection that typically lasts for ~28 days (9). Although this model elicits marked inflammation and 74 severe upper genital tract pathology (9), it does not reflect the chronic, low-grade infection that is 75 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint more typical for humans (10). These limitations emphasize the need for animal models that more 76 accurately capture the heterogeneity and chronicity of human chlamydial infection. 77 Host genetic variation plays a substantial role in shaping chlamydial disease outcomes, 78 with evidence from both human (11-16) and animal studies (17-19). We previously reported 79 marked strain-dependent differences in the course and severity of C. muridarum genital tract 80 infection among three inbred mouse strains, that were associated with distinct immune response 81 kinetics (9). These findings were extended by Chen et al., who examined 11 inbred strains and 82 reported gross hydrosalpinx incidences ranging from 10% to 87% (20). Together, these studies 83 underscore the strong influence of host genetics on infection trajectory and pathology. 84 Despite this, a persistent challenge in modeling chlamydial genital tract infection is the 85 striking variability in pathologic outcomes, even when infectious burden and clearance kinetics 86 are comparable (9, 20, 21). Substantial variation occurs both across classical inbred strains and 87 within individual strains. For example, Chen et al. reported broad intra-strain dispersion in oviduct 88 dilatation scores, with mean severity ranging from 5.3 ± 3.2 in a highly susceptible strain to 0.6 ± 89 1.4 in a resistant strain (20). Similar inconsistency is observed across experiments : in our 90 laboratory, ten independent preclinical vaccine studies using genetically identical C57BL/6 mice 91 under standardized conditions yielded gross hydrosalpinx rates ranging from 30% to 100%, with 92 correspondingly wide variation in histologic pathology (22, 23)(unpublished data). This degree of 93 intrastrain and inter-experimental variability limits the precision of genetic mapping and constrains 94 the accuracy of vaccine efficacy assessments. 95 The Collaborative Cross (CC) genetic reference population offers a powerful strategy to 96 address variability in chlamydial disease modeling. CC strains were generated by interbreeding 97 eight genetically diverse founder strains —five classical inbred strains (A/J, C57BL/6J, 98 129S1/SvImJ, NOD/ShiLtJ, and NZO/HlLtJ) and three wild-derived strains (CAST/EiJ, PWK/PhJ, 99 and WSB/EiJ)—followed by inbreeding to produce stable recombinant lines (24). This multi-allelic 100 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint architecture preserves the experimental reproducibility of inbred mice while buffering against the 101 effects of single detrimental variants and capturing a level of genetic diversity that more closely 102 reflects human populations (25). As a result, CC panels are well suited for dissecting polygenic 103 traits and enabling high-resolution mapping of loci that influence infectious disease susceptibility 104 and vaccine responsiveness (26). 105 To test the utility of this approach, we infected CC mice intravaginally with C. muridarum 106 and evaluated 20 strains selected for documented variation in infection susceptibility or immune-107 mediated disease (27-30). Across the panel, we observed marked strain-dependent differences 108 in bacterial burden, infection duration, and gross and histologic pathology , with low within-strain 109 variability. Notably, several disease -resistant strains exhibited prolonged or chronic infection 110 kinetics, paralleling features of C. trachomatis infection in women. Quantitative trait locus (QTL) 111 mapping identified loci associated with bacterial burden and pathological severity, providing 112 insight into host determinants of chlamydial disease, and demonstrating the value of CC strains 113 for capturing the heterogeneity of infection outcomes, and complementing traditional inbred 114 mouse systems used for preclinical vaccine evaluation. 115

116 Ethics statement 117 All animal procedures were approved by the Institutional Animal Care and Use Committee 118 (IACUC) at the University of North Carolina at Chapel Hill (UNC-CH) and all experiments involving 119 C. muridarum were conducted in accordance with institutional biosafety guidelines. 120 Mouse model of Chlamydia infection 121 Collaborative Cross mice were obtained from the Systems Genetics Core Facility at UNC -CH 122 (https://csbio.unc.edu/CCstatus/index.py). A total of 154 female mice from 20 CC strains (6 –9 123 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint mice per strain) were evaluated over 25 time points (see Fig. 1 for experimental workflow). Strains 124 studied were CC001, CC002, CC004, CC005, CC006, CC012, CC013, CC019, CC023, CC024, 125 CC027, CC030, CC031, CC036, CC037, CC041, CC051, CC065, CC068, and CC078. Mice were 126 age-matched and used between 8 and 12 weeks of age. Mice were housed in the pathogen-free 127 animal facility at UNC-CH and provided food and water ad libitum in an environmentally controlled 128 room with a cycle of 12 hours of light and 12 hours of darkness. All mice were administered 2.5 129 mg of depot medroxyprogesterone acetate (Depo -Provera; Pfizer) subcutaneously to induce 130 anestrus. 131 Seven days post -progesterone treatment, 79 mice (3 –6 per strain) were anesthetized with 132 Nembutal (sodium pentobarbital, 50 mg/ml), diluted 1:10 in sterile PBS, delivered 133 intraperitoneally, in a volume of 10 μl/g mouse weight, and intravaginally inoculated with 5 × 10⁵ 134 inclusion-forming units (IFU) of C. muridarum CM006. CM006, a plaque -purified clonal isolate 135 derived from the parental Nigg stock (31), was delivered in 20 μl of sucrose–sodium phosphate–136 glutamic acid (SPG) buffer containing 250 mM sucrose, 10 mM sodium phosphate, 5 mM l-137 glutamic acid (pH 7.2). The remaining 75 mice received phosphate-buffered saline (PBS) as mock 138 controls. Cervical swabs were collected pre-infection (day −2) and on post-infection days 2 to 10, 139 14, 17, 21, 24, 28, 31, 35, 38, 42, 45, 49, 52, 56, 59, and 63. Swabs were stored in 1 mL of 140 DNA/RNA Shield (Zymo Research, Irvine, California, USA) at −80°C for transcriptional response 141 profiling and determination of chlamydial burden. On day 63, mice were euthanized, and 142 reproductive tracts were harvested en bloc for gross and histopathologic analysis. 143 Gross and histopathological assessment 144 In situ examination of the reproductive tract was performed by a trained technician blinded to 145 experimental groups. Gross pathology was scored as follows: 0, normal; 1, enlarged cervix; 2, 146 swollen or discolored uterine horn (left and/or right horn); 3, unilateral hydrosalpinx; and 4, 147 bilateral hydrosalpinx. The genital tract was then collected in its entirety with careful removal of 148 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint adipose as needed, and mounted on thin cardboard, then fixed in 10% formalin in PBS for 48 h 149 and stored in PBS prior to paraffin embedding. Sections were cut at 4 μm distally, from oviduct to 150 cervix, and stained with hematoxylin and eosin. Oviduct regions of histological samples were 151 evaluated in a masked fashion by a board-certified pathologist, with an assessment of leukocyte 152 infiltration (neutrophils, mononuclear cells, and plasma cells) and dilation reported for each 153 oviduct using a four -tiered semi-quantitative scoring system; 0 = n ormal or none, 1 = mild, 2 = 154 moderate, 3 = marked, 4 = severe (9). 155 Nucleic acid isolation and quantification of bacterial load. 156 DNA and RNA were co-extracted from cervical swabs as previously described (32) using a Quick-157 DNA/RNA Miniprep Plus Kit (Zymo Research). Chlamydial loads from samples obtained on days 158 −2, 7, 10, 35 and 59 were measured by quantitative PCR using primers (23S_F1 5’ 159 GCTCACGTTCGGAAAGGATAA 3’ and 23S_R1 5’ GTGCTTACACCTCCAACCTATC 3”) that 160 targeted the C. trachomatis 23S rRNA loci using the following amplification conditions: 95°C for 161 15’, 60°C for 45’ for 40 cycles using SsoAdvanced Universal SYBR Green Supermix (BioRad Life 162 Science, Hercules, CA) followed by melt curve analysis . Each specimen was analyzed in a 163 triplicate. 164 Gene expression profiling: Six CC strains (CC005, CC012, CC023, CC030, CC031, CC041) 165 representing a range of phenotypes were selected for targeted transcriptional analysis using 166 probe-based quantitation (33, 34) . RNA extracted from swab eluates (3 –5 mice per strain) 167 obtained on days −2, 3, 5, 7, and 35 was analyzed using nCounter® Gene Expression Assay (35) 168 (NanoString Technologies, Seattle, WA) at the UNC Lineberger Comprehensive Cancer Center 169 Translational Genomics Lab oratory. A custom panel comprised of 51 immune -related mouse 170 genes and 6 internal references (Gapdh, Hprt, Cltc, Gusb, Pgk1, and Tubb5) was chosen. Probes 171 targeting chlamydial RNAs (23S rRNA, omcA, pGP8 anti -sense RNA) were also designed and 172 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint included in the assay to enable monitoring of chlamydial RNA abundance (Supplementary Table 173 1). 174 Statistical analyses: 175 Heritability estimates : Broad-sense heritability (H²) and narrow -sense heritability (h²) were 176 calculated to assess the contribution of genetic variation to phenotypic traits. For each phenotype, 177 genetic variance was determined by subtracting within -strain variance, primarily influenced by 178 environmental factors, from between -strain variance, which includes both genetic and 179 environmental factors. H² accounts for additive, dominance, and epistatic effects (36, 37) whereas 180 h² reflects only additive effects (38). Estimates were derived using the est_herit function in the 181 qtl2 R package (v3.1 -0), incorporating a kinship matrix and variance estimated from a mixed-182 effects model (39). 183 Quantitative trait locus (QTL) mapping : QTL mapping for bacterial burden and pathology 184 scores was performed using CC genotypes obtained from the QTL Archive (https://qtlarchive.org) 185 as implemented in the qtl2 R package (39). Founder haplotype probabilities were calculated via 186 a hidden Markov model (40). A genome-wide (GRCm38, bioproject PRJNA20689) mixed-effects 187 model scan was conducted, and significance thresholds for Logarithm of Odds (LOD) scores were 188 established using 1,000 permutations. Confidence intervals were defined using a Bayesian 189 credible interval estimation implemented in R/qtl2 package. Genes within each QTL interval were 190 annotated using Mouse Genome Informatics (MGI) (41) and the UCSC Genome Browser (42). 191 For both burden - and pathology -associated loci, genes located within 10 Mb upstream and 192 downstream of the peak were examined, and targeted literature searches were performed to 193 identify published evidence that linked these genes with chlamydial growth, replication, or disease 194 pathogenesis. 195 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint Intraclass correlation and comparative analysis: Within-strain reproducibility of hydrosalpinx 196 (a binary outcome: presence vs. absence) was assessed using intraclass correlation coefficients 197 (ICCs). The ICC is the proportion of total variance in the outcome that is due to differences 198 between strains, with total variance defined as the sum of variance between and within strains. 199 ICCs for the 20 CC strains analyzed in this study and the 11 inbred strains (data from Table 3 of 200 Chen et al. (20)) were estimated separately using logistic mixed -effects models in the lme4 R 201 package (43), with a random intercept for strain. To test whether ICCs differed between studies, 202 we applied a hierarchical bootstrap (44) that resampled strains with replacement within each 203 study, followed by resampling mice within strains , and computed the ICC difference . This 204 approach preserves the nested data structure and accounts for unbalanced designs, including 205 differences in the number of strains and the number of replicate mice per strain. The bootstrap 206 distribution of ICC differences was used to construct percentile 95% confidence intervals and 207 assess statistical significance. We also performed a secondary analysis , a meta -analytic 208 comparison of ICCs , as a sensitivity analysis to assess the robustness of our findings to 209 unbalanced sample sizes. A random -effects meta-analysis with inverse-variance weighting was 210 used to account for both between -study heterogeneity and differences in sample sizes across 211 comparisons. Analyses were conducted using the metafor R package (45). 212 Association of cervical cytokine mRNA expression with burden and pathology: NanoString 213 expression data were processed using nSolver v3.0 (46). Quality control filtering excluded flagged 214 or low -expression genes (defined as counts < mean + 3 SD of negative controls). Technical 215 normalization was performed using spike -in control probes, with lane -specific scaling factors 216 derived from their geometric means. Sample-to-sample normalization was then performed using 217 housekeeping genes, applying the same geometric mean approach. Associations between gene 218 expression and 23S RNA load were analyzed using linear mixed -effects models with CC strain 219 as a random effect. Aligned Rank Transform (ART) ANOVA (47, 48) and ARTool R package (49) 220 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint were used to assess associations with gross pathology scores. To evaluate whether cervical 221 cytokines were associated with pathology independent of bacterial burden, we used a 222 nonparametric factorial ANOVA with align -and-rank transformation. This approach 223 accommodates non-normal distributions and unequal variances while allowing adjustment for 224 bacterial load as a covariate. P -values were obtained from rank -transformed models, and 225 multiple-testing corrections were performed using the Benjamini–Hochberg procedure. 226 Time-series co-expression analysis: Temporal co-expression patterns among host genes were 227 analyzed in the 6 selected CC strains using Lag Penalized Weighted Correlation (LPWC) (50). 228 The optimal number of gene clusters was determined using the Gap Statistic (51). To characterize 229 temporal patterns of cervical host responses, we used the Lag Penalized Weighted Correlation 230 (LPWC) algorithm, which clusters genes based on similarity in expression trajectories while 231 allowing for temporal lags between samples. LPWC was applied to normalized gene-expression 232 data from Days 0, 3, 7, 21, 35, and 42. Cluster stability was assessed using the default penalty 233 and weighting parameters. Gene clusters were interpreted using functional enrichment and 234 manual annotation of immune-related pathways. 235

236 Chlamydial burden and pathology vary independently across CC strains 237 We intravaginally inoculated 20 CC strains with C. muridarum or sham-infected controls 238 with PBS and monitored them through day 63 for infection outcomes, including chlamydial burden, 239 time to clearance, and genital tract pathology. For analysis, log10-transformed cervical chlamydial 240 loads were categorized as: very low (6). Infection 241 duration was classified as: normal clearance: (no detectable Cm genomic DNA by qPCR on D35, 242 extended (DNA detectable on D35), and prolonged (DNA detected through D59). 243 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint All CC strains became infected following the challenge. Bacterial loads detected in the 244 lower genital tract and infection duration varied widely across strains (Fig. 2, and Supplementary 245 Fig. 1) but were highly consistent within each strain (Fig. 3C, D). One strain, CC004, clear ed 246 infection rapidly, with all mice reaching the limit of detection by day 21 (Fig. 2B). In contrast, mice 247 from CC013, CC030 and CC019 showed prolonged infection , with detectable chlamydial DNA 248 still present on day 59. Among the remaining strains, five cleared infection by day 35, and ten 249 strains appeared to clear between days 35 and 59 (Fig. 2C and Supplementary Fig. 1). 250 Across the 20 CC strains the overall hydrosalpinx incidence was 21.5% at tissue 251 harvest. Twelve strains showed no hydrosalpinx; one showed low incidence despite prolonged 252 infection (CC019: 16.7%); another three showed moderate incidence (CC001, CC004, and 253 CC068: 25 –33.3%), and four strains displayed high incidence (CC005, CC012, CC023, and 254 CC041: 60-100%). Six CC strains (CC002, CC006, CC024, CC051, CC065, and CC078) showed 255 no gross pathology of the upper or lower genital tract (mean pathology score = 0) despite medium 256 to high Day 7 bacterial loads (Fig. 2C). Histopathology supported these findings with none of the 257 oviducts showing a dilatation score > 3. Most mice of CC002, CC006, CC051, and CC078 strains 258 showed no oviduct inflammation, with only occasional mild or moderate mononuclear and plasma 259 cell infiltrates in the oviduct or mesosalpingeal tissues. Strains CC013, CC027, and CC030, had 260 low mean gross pathology scores (1.3- 1.5), no gross hydrosalpinx, and no oviduct dilatation by 261 histology, despite high early bacterial burden and prolonged infection. Despite the absence of 262 gross oviduct dilatation, CC013 mice that were confirmed C. muridarum positive by cervical PCR 263 on day 59, exhibited moderate to severe histiocytic inflammation in the bursal and mesosalpingeal 264 tissues, a pathologic change not seen in any PBS-inoculated CC013 controls. 265 The CC004 strain, despite clearing infection rapidly , developed mild to marked oviduct 266 dilatation; one mouse of three exhibited bilateral hydrosalpinx along with mild inflammatory 267 infiltrates in the mesosalpinx and bursa. CC023 mice had a relatively high mean gross pathology 268 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint score (2.6), with hydrosalpinx in 5 of 10 oviducts and corresponding histologic dilatation scores 269 of 3-4, despite relatively low bacterial burdens (Fig. 2C and Supplementary Fig. 1). None of the 270 PBS-inoculated controls developed hydrosalpinx, although occasional mild uterine hyperemia or 271 hydrometra was observed , likely due to Depo -Provera treatment. Together, these findings 272 demonstrate that in genetically diverse CC mice, neither chlamydial burden nor infection duration 273 reliably predicted pathological outcomes. 274 Quantifying genetic influence on chlamydia burden and disease 275 Heritability describes the proportion of variation in a trait that is explained by genetic differences 276 rather than environmental factors. Narrow -sense heritability reflects the proportion of variance 277 attributable to additive genetic effects , those that sum predictably across alleles. Broad -sense 278 heritability captures all genetic contributions, including additive effects as well as dominance and 279 epistatic (gene–gene interaction) effects. Among the 20 CC strains, heritability of bacterial burden 280 was highest on Day 7, with narrow -sense and broad -sense estimates of 69.3% and 76.4%, 281 respectively ( Table 1 ). For gross pathology scores, narrow -sense heritability was 50.1% and 282 broad-sense heritability was 57.5% (Table 1). These findings indicate that host genetic factors 283 were major contributors to variation in both bacterial load and pathology following C. muridarum 284 infection. 285 QTL mapping reveals loci associated with chlamydial burden and pathology 286 We proceeded to examine the influence of genotypes on bacterial load and pathology using a 287 genome-wide scan with CC genotype probabilities obtained from the QTL Archive 288 (https://qtlarchive.org). Although no genome-wide significant QTLs were detected for either trait 289 after permutation correction (p 4.5 for Day 7 bacterial load (Fig. 3A) and 291 4.15 for pathology ( Fig. 3B). Such LOD score s correspond to odds >10,000:1 in favor of 292 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint association compared to chance alone, providing strong suggestive evidence that these regions 293 may contribute to variation in chlamydial infection outcomes. 294 Genes within the locus associated with chlamydial burden mapped to host pathways previously 295 implicated in chlamydial intracellular development an d host control (Supplementary Table 2). 296 Several candidates localized to membrane dynamics and lipid metabolism, processes central to 297 inclusion formation and maintenance. Notably, Pi4ka, encoding phosphatidylinositol 4-kinase α, 298 lies within a pathway shown to support inclusion membrane biogenesis through host 299 phosphatidylinositol-4-phosphate supply (52, 53) while Smpd4, a sphingomyelinase, maps to 300 sphingolipid metabolic pathways required for inclusion stability and bacterial replication (54). 301 Genes involved in ubiquitin signaling (Ube2l3, Ube2v2) were also detected; ubiquitin-dependent 302 targeting of chlamydial inclusions is part of cell-autonomous host defense (55), and Chlamydia 303 encodes effectors that modulate this process (56, 57), suggesting that host ubiquitination capacity 304 may influence chlamydial burden. 305 Additional candidates were linked to host cell stress responses and survival, including Dnm1l 306 (DRP1), which regulates mitochondrial fission and apoptosis susceptibility , and is inhibited by 307 chlamydial infection (58). Genes involved in DNA damage sensing and repair ( Prkdc, Ercc4, 308 Mcm4, Spidr, Top3b) were also detected. Chlamydia infection induces host DNA damage while 309 altering repair and checkpoint responses (59), making host DNA repair capacity a reasonable 310 modifier of cellular survival and consequently, chlamydial replication. The locus also contained 311 centrosome- and spindle-regulating genes (Cep20, Mzt2), processes disrupted during chlamydial 312 infection and associated with enhanced host cell survival and inclusion stabilit y (60). Finally, 313 immune-regulatory genes (Ciita, Mapk1, Socs1) were present, implicating variation in interferon-314 responsive and inflammatory signaling pathway s as additional contributors to differences in 315 bacterial burden across Collaborative Cross strains. The identification of Ciita, the master 316 regulator of MHC class II expression, is particularly relevant given the established requirement 317 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint for CD4⁺ T cell–mediated immunity in controlling chlamydial infection . While these associations 318 do not establish causal roles for individual genes, the convergence of multiple candidates within 319 host pathways known to support intracellular growth provides a biologically coherent framework 320 for interpreting genetic effects on chlamydial burden. A full list of genes in this locus is provided 321 in Supplementary Table 3. 322 We next assessed the effects of the eight founder alleles at this locus on Day 7 chlamydial load 323 (Fig. 3C). The allelic effects clustered into three groups. Mice carrying the NOD/ShiLtJ (D) , 324 WSB/EiJ (H), A/J (A), or C57BL/6J (B) alleles showed the highest bacterial loads, with log10 325 values >6. In contrast, mice with the PWK/PhJ (G) or CAST/EiJ (F) alleles had lower burdens, 326 with log10 values of 4 –6 and 0 –5, respectively. The 129S1/SvImJ (C) allele produced an 327 intermediate phenotype, with bacterial loads ranging from log10 ~4.5 to 7. 328 The pathology -associated locus on Chromosome 5 encompasses multiple genes 329 (Supplementary Table 4) with established roles in host immune responses and genital tract 330 pathology following chlamydial infection. Included in this locus were genes encoding ELR+ CXC 331 chemokines (Cxcl1, Cxcl2, Cxcl3, Cxcl5 and Cxcl15), key mediators of neutrophil recruitment to 332 sites of injury or infection (61). Neutrophil influx has been consistently linked to tissue damage in 333 murine models of chlamydial genital tract infection (62-64). Also present were genes encoding 334 IFN-inducible chemokines, (Cxcl9, Cxcl10 and Cxcl11), that attract CXCR3+ Th1 and cytotoxic T 335 cells to aid pathogen clearance but also implicated in immunopathology (62, 65, 66) . Cxcl13, 336 which supports ectopic lymphoid follicle formation , contributes to persistent inflammation and 337 fibrosis in chronic infection settings (66, 67) . Bmp3, part of the BMP/TGF -β superfamily, 338 modulates tissue remodeling and dysregulated BMP signaling has been linked to fallopian tube 339 scarring and infertility in human studies (15, 68). Additional genes in this locus include heparinase 340 (Hpse) which influences extracellular matrix remodeling (69); multiple guanylate-binding protein 341 (Gbp) family members involved in inflammasome activation during chlamydial infection (70, 71); 342 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint and osteopontin ( Spp1), a multifunctional cytokine linked to inflammatory cell recruitment and 343 tissue remodeling (72, 73) . A complete list of genes within this interval is provided in 344 Supplementary Table 5. 345 We next examined the effects of the eight founder alleles at this locus on pathology scores (Fig. 346 3D). The allelic effects grouped into three tiers. Mice carrying the NOD/ShiLtJ (D) or CAST/EiJ 347 (F) alleles showed the highest pathology scores. In contrast, the WSB/EiJ (H), 129S1/SvImJ (C), 348 or NZO/HlLtJ (E) alleles were associated with the lowest pathology. The C57BL/6J (B) and 349 PWK/PhJ (G) alleles produced intermediate pathology scores , with the C57BL/6J (B) allele 350 showing the greatest variability both within and across CC strains. 351 Within-strain reproducibility of hydrosalpinx rates in CC mice 352 To quantify within-strain consistency, we calculated the intraclass correlation coefficient (ICC) for 353 hydrosalpinx. The ICC was 0.598 across the 20 CC strains examined here, compared with 0.275 354 across 11 previously profiled inbred strains (20). To account for unequal sample sizes, we applied 355 both bootstrap and meta-analytic approaches, which yielded a bootstrap P = 0.0003 and a meta-356 analysis P = 0.007. Together, these findings show that hydrosalpinx outcomes are markedly more 357 consistent within CC strains than within commonly used inbred laboratory strains. 358 Relationship between cervical cytokine mRNA expression with bacterial burden and 359 pathology 360 Human (74-76) and murine (62, 77 -79) studies have shown that pro-inflammatory responses 361 contribute substantially to immune-mediated pathology following chlamydial infection. To explore 362 how host genetics shape these responses, we assessed the relationship between expression of 363 51 cervical cytokine genes and bacterial burden in six CC strains (CC005, CC012, CC023, 364 CC030, CC031, and CC041) spanning a broad range of bacterial loads, infection durations, and 365 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint gross pathology. Cytokine genes were significantly associated with bacterial burden on days 3, 366 5, and 7, (listed in Supplementary Tables 6-8). Notably, no cytokine gene displayed a significant 367 negative association with burden at any early time point, and no significant cytokine –burden 368 associations were detected on day 35. 369 Of the 51 genes measured, 25, 8, and 11 were significantly upregulated in association with 370 increased chlamydial burden on days 3, 5, and 7, respectively. Six genes, Il17a, Il22, Il10, Il12a, 371 Ifng, and Ifnb1, showed significant positive associations with bacterial burden at all three early 372 time points. These include d both pro-inflammatory (Il17a, Il22, Ifng, Ifnb1) and regulatory/anti -373 inflammatory (Il10, Il12a) mediators, with Il12a also contributing to Th1 differentiation and the 374 production of IFN-γ and TNF. Cxcl10, a chemokine that promotes T cell recruitment, and Eomes, 375 a transcription factor critical for T cell differentiation , were significantly associated with bacterial 376 burden on days 3 and 5. Several genes , Il23, Cxcl1, Stat3, Ccl4, Ccl3, Tlr2, Il6, Il1rn, Tnf, Il15, 377 Il18, and Gata3, were uniquely associated with increased bacterial burden on day 3, whereas Ltb 378 showed a unique association on Day 7. 379 We also sought to determine whether cytokine mRNA expression was associated with pathology 380 scores in these 6 CC strains. After adjusting for bacterial burden, Ccl4, Cxcl9, and IL1ra transcript 381 levels were modestly higher in mice with hydrosalpinx compared to those without (P = 0.039, 382 0.049, and 0.068, respectively). However, none of these associations were statistically significant 383 after correction for multiple testing. 384 Temporal co-expression patterns of cervical host genes in 6 selected CC strains 385 Because the timing and coordination of immune gene expression are critical determinants of 386 infection outcome, we examined temporal patterns of cervical gene expression using a clustering 387 approach that accommodates differences in activation timing. This analysis resolved two major 388 trajectories: an early-response cluster enriched for genes involved in initiating and recruiting a T-389 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint cell response (Fig. 4A), and a later -response cluster dominated by markers of T -cell infiltration 390 (e.g., CD4, CD3), differentiation ( Tbx21, Foxp3), and effector function ( IL-2, IL-21, IL-16) (Fig. 391 4B). 392

393 Using 20 CC mouse strains infected with C. muridarum , we observed broad , genetically 394 reproducible variation in infection outcomes that parallel the heterogeneity of human chlamydial 395 disease. Several strains exhibited prolonged or near -chronic infection with minimal pathology, 396 resembling persistent, asymptomatic infections in women (3, 10) , whereas o thers developed 397 hydrosalpinx despite low bacterial burden. This burden–pathology discordance, previously noted 398 in select inbred strains (20, 53) , was consistently reproduced across the CC panel, enabling 399 systematic genetic dissection of mechanisms that uncouple pathogen load from tissue damage. 400 Together, these findings indicate that host genetic background, rather than bacterial burden 401 alone, is a major determinant of inflammatory regulation and disease severity. 402 Across all CC strains, hydrosalpinx occurred in ~20% of mice, comparable to infertility risk 403 estimates following chlamydial PID in women (80, 81), underscoring the translational relevance 404 of the model. Notably, pathology outcomes were far more consistent within CC strains than within 405 classical inbred strains (ICC 0.598 vs. 0.275; p < 0.001). Although inbred strains are often 406 presumed to exhibit phenotypic stability, prior studies demonstrate substantial variability in 407 pathological responses (24, 25, 82), consistent with our observations in C57BL/6 mice and with 408 data reported by Chen et al. (20). Fixed deleterious or threshold -sensitive alleles and limited 409 redundancy in immune regulatory pathways likely render inbred strains vulnerable to stochastic 410 influences. In contrast, CC strains inherit diverse combinations of alleles from eight founders, 411 buffering single-variant effects and producing tighter phenotypic clustering, thereby enhancing 412 reproducibility for genetic mapping and vaccine studies. 413 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint Heritability estimates were high for both Day 7 bacterial burden (narrow -sense 69.3%; broad -414 sense 76.4%) and pathology (narrow -sense 50.1%; broad -sense 57.5 %), indicating that host 415 genetics accounted for most observed variation. The modest difference between broad - and 416 narrow-sense estimates suggests that additive effects predominate, with a smaller contribution 417 from non-additive interactions. These findings validate the use of CC strains for detecting genetic 418 loci underlying complex traits such as clearance and immunopathology and indicate that 419 environmental noise is unlikely to obscure genetic signals within strains. Importantly, these results 420 align with human genetic–epidemiologic data, including twin studies showing substantial heritable 421 contributions to immune responses against C. trachomatis (83), reinforcing the translational 422 relevance of genetically diverse model systems. 423 Genome-wide scans identified distinct suggestive QTLs for bacterial burden (Chromosome 16) 424 and pathology (Chromosome 5). Genes within the burden -associated locus converged on host 425 pathways that support chlamydial intracellular growth, including membrane dynamics and lipid 426 metabolism, ubiquitin signaling, host cell survival , DNA damage responses, centrosome 427 regulation, and immune -regulatory signaling , processes that are actively modulated during 428 infection and influence inclusion stability and bacterial replication. Although individual causal 429 genes were not resolved, pathway-level convergence provides a coherent biological framework 430 linking host genetic variation to differences in chlamydial burden. 431 Genes within the pathology -associated locus converged on host pathways that regulate 432 inflammatory cell recruitment, tissue remodeling, and chronic immune activation , processes 433 central to genital tract damage following chlamydial infection. Prominent representation of 434 chemokine networks governing neutrophil and Th1 cell trafficking, together with genes linked to 435 ectopic lymphoid structure formation, supports a model in which genetically determined variation 436 in immune cell localization and persistence contributes to pathological outcomes. The presence 437 of additional candidates involved in extracellular matrix remodeling, inflammasome activation, and 438 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint BMP/TGF-β signaling further implicates dysregulated host repair and fibrotic responses as key 439 modifiers of disease severity. 440 Founder allele analysis further demonstrated dissociation between genetic effects on burden and 441 pathology, reinforcing the concept that pathogen control and tissue damage are partially 442 independent of host processes. Variability associated with certain founder alleles, including 443 C57BL/6J, underscores the limitations of relying on conventional inbred strains and highlights the 444 strength of the CC for uncovering genetic effects otherwise obscured. 445 Cervical cytokine profiling provided additional insight into immune dynamics. Early expression of 446 inflammatory and regulatory cytokines correlated positively with bacterial burden, suggesting that 447 early inflammation primarily reflects ongoing replication rather than effective control, consistent 448 with observations in women (84). Temporal trajectory analyses revealed a shift from innate to 449 adaptive immune programs, aligning with the known requirement for CD4 Th1 –mediated IFN-γ 450 responses for clearance while suggesting that genetic background influences how effectively 451 early innate cues shape protective T-cell responses (16). 452 Although no cytokine remained significant after multiple -testing correction in this small dataset, 453 CXCL9, CCL4, and IL1RA emerged as the strongest burden -adjusted predictors of pathology . 454 CXCL9, produced by IFN -γ–stimulated immune and epithelial cells, recruits CXCR3 ⁺ effector T 455 cells to inflamed tissues (62, 85). CCL4, generated by innate immune cells and activated T cells, 456 recruits CCR5 ⁺ effector T cells, amplifying local inflammatory responses (86-88). In contrast, 457 IL1RA restrains inflammation by blocking IL -1 signaling (89), pointing to a balance between 458 chemokine-driven effector recruitment and inflammatory restraint as a key modifier of tissue 459 damage. 460 Collectively, these findings demonstrate that CC mice offer a robust and clinically relevant 461 platform for modeling chlamydial disease. Their genetic diversity yields phenotypes that mirror 462 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint human variability, while fixed genomes ensure high reproducibility. This combination has 463 important implications for vaccine development, particularly for evaluating interventions across 464 genetically distinct hosts that dissociate clearance from pathology. Prioritizing CC strains with 465 extreme but reproducible phenotypes may improve power to detect true vaccine effects and 466 enable mechanistic dissection of protective versus pathogenic immune programs. 467 Several limitations warrant consideration. C. muridarum differs biologically from C. trachomatis, 468 and larger cohorts or expansion to CC-F2 or Diversity Outbred populations will increase power 469 for genome-wide significance. Cytokine profiling was limited to six strains, and broader sampling 470 may reveal additional immune architectures. Integration of single-cell transcriptomics, proteomics, 471 and causal network modeling (90) will be essential for pinpointing molecular drivers of protection 472 and pathology. 473 In summary, this study demonstrates that host genetics profoundly influences chlamydial burden, 474 persistence, and immunopathology and that CC mice recapitulate key features of human disease 475 more comprehensively than conventional inbred strains. By delineating genetically distinct axes 476 of pathogen control and tissue damage, the CC provides a powerful framework for mechanistic 477 discovery and for advancing vaccines and therapeutics in a genetically diverse, clinically relevant 478 context. 479

480 We would like to acknowledge the histopathology scoring assessments of Rani Sellers, PHD, 481 DVM, DACVP, UNC Pathology Services Core and the UNC-CH Systems Genetics Core Facility 482 for the maintenance of the Collaborative Cross lines used in these studies. The UNC-CH Systems 483 Genetics Core Facility CC Pilot Program (CC004) provided funding support. 484 485 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint

486 487 1. Centers for Disease Control and Prevention. Sexually Transmitted Infections 488 Surveillance, 2024 (Provisional). https://www.cdc.gov/sti-statistics/annual/index.html 489 accessed December 17, 2025. 490 2. Oakeshott P, Kerry S, Aghaizu A, Atherton H, Hay S, Taylor-Robinson D, Simms I, Hay 491 P. 2010. Randomised controlled trial of screening for Chlamydia trachomatis to prevent 492 pelvic inflammatory disease: the POPI (prevention of pelvic infection) trial. BMJ 493 340:c1642. 494 3. Wiesenfeld HC, Hillier SL, Meyn LA, Amortegui AJ, Sweet RL. 2012. Subclinical pelvic 495 inflammatory disease and infertility. Obstet Gynecol 120:37-43. 496 4. Brunham RC, Gottlieb SL, Paavonen J. 2015. Pelvic inflammatory disease. N Engl J 497 Med 372:2039-48. 498 5. Westrom LV. 1996. Chlamydia and its effect on reproduction. J Br Fer Soc 1:23-30. 499 6. Tubal infertility: serologic relationship to past chlamydial and gonococcal infection. World 500 Health Organization Task Force on the Prevention and Management of Infertility. Sex 501 Transm Dis 22:71-7. 502 7. Nigg C. 1942. An unidentified virus which produces pneumonia and systemic infection in 503 mice. Science 95:49-50. 504 8. Nelson DE, Virok DP, Wood H, Roshick C, Johnson RM, Whitmire WM, Crane DD, 505 Steele-Mortimer O, Kari L, McClarty G, Caldwell HD. 2005. Chlamydial IFN-gamma 506 immune evasion is linked to host infection tropism. Proc Natl Acad Sci USA 102:10658-507 10663. 508 9. Darville T, Andrews CW, Jr., Laffoon KK, Shymasani W, Kishen LR, Rank RG. 1997. 509 Mouse strain-dependent variation in the course and outcome of chlamydial genital tract 510 infection is associated with differences in host response. Infect Immun 65:3065-3073. 511 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint 10. Molano M, Meijer CJ, Weiderpass E, Arslan A, Posso H, Franceschi S, Ronderos M, 512 Munoz N, van den Brule AJ. 2005. The natural course of Chlamydia trachomatis 513 infection in asymptomatic Colombian women: a 5-year follow-up study. J Infect Dis 514 191:907-916. 515 11. den Hartog JE, Ouburg S, Land JA, Lyons JM, Ito JI, Pena AS, Morre SA. 2006. Do host 516 genetic traits in the bacterial sensing system play a role in the development of 517 Chlamydia trachomatis-associated tubal pathology in subfertile women? BMC Infect Dis 518 6:122. 519 12. Ouburg S, Spaargaren J, den Hartog JE, Land JA, Fennema JS, Pleijster J, Pena AS, 520 Morre SA. 2005. The CD14 functional gene polymorphism -260 C>T is not involved in 521 either the susceptibility to Chlamydia trachomatis infection or the development of tubal 522 pathology. BMC Infect Dis 5:114. 523 13. Morre SA, Murillo LS, Bruggeman CA, Pena AS. 2003. The role that the functional 524 Asp299Gly polymorphism in the toll-like receptor-4 gene plays in susceptibility to 525 Chlamydia trachomatis-associated tubal infertility. J Infect Dis 187:341-2; author reply 526 342-3. 527 14. Barr EL, Ouburg S, Igietseme JU, Morre SA, Okwandu E, Eko FO, Ifere G, Belay T, He 528 Q, Lyn D, Nwankwo G, Lillard J, Black CM, Ananaba GA. 2005. Host inflammatory 529 response and development of complications of Chlamydia trachomatis genital infection 530 in CCR5-deficient mice and subfertile women with the CCR5delta32 gene deletion. J 531 Microbiol Immunol Infect 38:244-254. 532 15. Zheng X, Zhong W, O'Connell CM, Liu Y, Haggerty CL, Geisler WM, Anyalechi GE, 533 Kirkcaldy RD, Wiesenfeld HC, Hillier SL, Steinkampf MP, Hammond KR, Fine J, Li Y, 534 Darville T. 2021. Host genetic risk factors for Chlamydia trachomatis-related infertility in 535 women. J Infect Dis 224:S64-s71. 536 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint 16. Zhong W, Kollipara A, Liu Y, Wang Y, O'Connell CM, Poston TB, Yount K, Wiesenfeld 537 HC, Hillier SL, Li Y, Darville T, Zheng X. 2022. Genetic susceptibility loci for Chlamydia 538 trachomatis endometrial infection influence expression of genes involved in T cell 539 function, tryptophan metabolism and epithelial integrity. Front Immunol 13:1001255. 540 17. Ouburg S, Lyons JM, Land JA, den Hartog JE, Fennema JS, de Vries HJ, Bruggeman 541 CA, Ito JI, Pena AS, Lundberg PS, Morre SA. 2009. TLR9 KO mice, haplotypes and 542 CPG indices in Chlamydia trachomatis infection. Drugs Today 45 Suppl B:83-93. 543 18. Lichtenwalner AB, Patton DL, Cosgrove Sweeney YT, Gaur LK, Stamm WE. 1997. 544 Evidence of genetic susceptibility to Chlamydia trachomatis-induced pelvic inflammatory 545 disease in the pig-tailed macaque. Infect Immun 65:2250-2253. 546 19. Tuffrey M, Alexander F, Woods C, Taylor-Robinson D. 1992. Genetic susceptibility to 547 chlamydial salpingitis and subsequent infertility in mice. J Reprod Fertil 95:31-38. 548 20. Chen J, Zhang H, Zhou Z, Yang Z, Ding Y, Zhou Z, Zhong E, Arulanandam B, Baseman 549 J, Zhong G. 2014. Chlamydial induction of hydrosalpinx in 11 strains of mice reveals 550 multiple host mechanisms for preventing upper genital tract pathology. PLoS One 551 9:e95076. 552 21. Ito JI, Jr., Lyons JM, Airo-Brown LP. 1990. Variation in virulence among oculogenital 553 serovars of Chlamydia trachomatis in experimental genital tract infection. Infect Immun 554 58:2021-2023. 555 22. Poston TB, Girardi J, Kim M, Zwarycz P, Polson AG, Yount KS, Hanlan C, Jaras Salas I, 556 Lammert SM, Arroyo D, Bruno T, Wu M, Rozzelle J, Fairman J, Esser-Kahn AP, Darville 557 T. 2025. Intranasal immunization with CPAF combined with ADU-S100 induces an 558 effector CD4 T cell response and reduces bacterial burden following intravaginal 559 infection with Chlamydia muridarum. Vaccine 43:126526. 560 23. Poston TB, Girardi J, Polson AG, Bhardwaj A, Yount KS, Jaras Salas I, Trim LK, Li Y, 561 O'Connell CM, Leahy D, Harris JM, Beagley KW, Goonetilleke N, Darville T. 2024. Viral-562 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint vectored boosting of OmcB- or CPAF-specific T-cell responses fail to enhance protection 563 from Chlamydia muridarum in infection-immune mice and elicits a non-protective CD8-564 dominant response in naive mice. Mucosal Immunol doi:10.1016/j.mucimm.2024.06.012. 565 24. Noll KE, Ferris MT, Heise MT. 2019. The Collaborative Cross: A Systems Genetics 566 Resource for Studying Host-Pathogen Interactions. Cell Host Microbe 25:484-498. 567 25. Graham JB, Swarts JL, Mooney M, Choonoo G, Jeng S, Miller DR, Ferris MT, 568 McWeeney S, Lund JM. 2017. Extensive homeostatic T cell phenotypic variation within 569 the Collaborative Cross. Cell Rep 21:2313-2325. 570 26. Srivastava A, Morgan AP, Najarian ML, Sarsani VK, Sigmon JS, Shorter JR, Kashfeen 571 A, McMullan RC, Williams LH, Giusti-Rodríguez P, Ferris MT, Sullivan P, Hock P, Miller 572 DR, Bell TA, McMillan L, Churchill GA, de Villena FP. 2017. Genomes of the mouse 573 Collaborative Cross. Genetics 206:537-556. 574 27. Keele GR, Zhang T, Pham DT, Vincent M, Bell TA, Hock P, Shaw GD, Paulo JA, 575 Munger SC, de Villena FP, Ferris MT, Gygi SP, Churchill GA. 2021. Regulation of 576 protein abundance in genetically diverse mouse populations. Cell Genom 1:100003. 577 28. Nagarajan A, Scoggin K, Gupta J, Aminian M, Adams LG, Kirby M, Threadgill D, 578 Andrews-Polymenis H. 2024. Collaborative Cross mice have diverse phenotypic 579 responses to infection with Methicillin-resistant Staphylococcus aureus USA300. PLoS 580 Genet 20:e1011229. 581 29. Kollmus H, Pilzner C, Leist SR, Heise M, Geffers R, Schughart K. 2018. Of mice and 582 men: the host response to influenza virus infection. Mamm Genome 29:446-470. 583 30. Pérez Gómez AA, Karmakar M, Carroll RJ, Lawley KS, Amstalden K, Young CR, 584 Threadgill DW, Welsh CJ, Brinkmeyer-Langford C. 2022. Serum cytokines predict 585 neurological damage in genetically diverse mouse models. Cells 11:2044. 586 31. Poston TB, O'Connell CM, Girardi J, Sullivan JE, Nagarajan UM, Marinov A, Scurlock 587 AM, Darville T. 2018. T Cell-independent gamma interferon and B Cells cooperate to 588 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint prevent mortality associated with disseminated Chlamydia muridarum genital tract 589 infection. Infect Immun 86 e00143-18. 590 32. O'Connell CM, Brochu H, Girardi J, Harrell E, Jones A, Darville T, Sena AC, Peng X. 591 2019. Simultaneous profiling of sexually transmitted bacterial pathogens, microbiome, 592 and concordant host response in cervical samples using whole transcriptome 593 sequencing analysis. Microb Cell 6:177-183. 594 33. Veldman-Jones MH, Brant R, Rooney C, Geh C, Emery H, Harbron CG, Wappett M, 595 Sharpe A, Dymond M, Barrett JC, Harrington EA, Marshall G. 2015. Evaluating 596 robustness and sensitivity of the NanoString Technologies nCounter platform to enable 597 multiplexed gene expression analysis of clinical samples. Cancer Res 75:2587-93. 598 34. Nielsen T, Wallden B, Schaper C, Ferree S, Liu S, Gao D, Barry G, Dowidar N, Maysuria 599 M, Storhoff J. 2014. Analytical validation of the PAM50-based Prosigna Breast Cancer 600 Prognostic Gene Signature Assay and nCounter Analysis System using formalin-fixed 601 paraffin-embedded breast tumor specimens. BMC Cancer 14:177. 602 35. Geiss GK, Bumgarner RE, Birditt B, Dahl T, Dowidar N, Dunaway DL, Fell HP, Ferree S, 603 George RD, Grogan T, James JJ, Maysuria M, Mitton JD, Oliveri P, Osborn JL, Peng T, 604 Ratcliffe AL, Webster PJ, Davidson EH, Hood L, Dimitrov K. 2008. Direct multiplexed 605 measurement of gene expression with color-coded probe pairs. Nat Biotechnol 26:317-606 25. 607 36. Scoggin K, Gupta J, Lynch R, Nagarajan A, Aminian M, Peterson A, Adams LG, Kirby M, 608 Threadgill DW, Andrews-Polymenis HL. 2022. Elucidating mechanisms of tolerance to 609 Salmonella Typhimurium across long-term infections using the Collaborative Cross. 610 mBio 13:e0112022. 611 37. Scoggin K, Lynch R, Gupta J, Nagarajan A, Sheffield M, Elsaadi A, Bowden C, Aminian 612 M, Peterson A, Adams LG, Kirby M, Threadgill DW, Andrews-Polymenis HL. 2022. 613 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint Genetic background influences survival of infections with Salmonella enterica serovar 614 Typhimurium in the Collaborative Cross. PLoS Genet 18:e1010075. 615 38. Iraqi FA, Athamni H, Dorman A, Salymah Y, Tomlinson I, Nashif A, Shusterman A, 616 Weiss E, Houri-Haddad Y, Mott R, Soller M. 2014. Heritability and coefficient of genetic 617 variation analyses of phenotypic traits provide strong basis for high-resolution QTL 618 mapping in the Collaborative Cross mouse genetic reference population. Mamm 619 Genome 25:109-19. 620 39. Broman KW, Gatti DM, Simecek P, Furlotte NA, Prins P, Sen Ś, Yandell BS, Churchill 621 GA. 2019. R/qtl2: Software for mapping quantitative trait loci with high-dimensional data 622 and multiparent populations. Genetics 211:495-502. 623 40. Hampton BK, Plante KS, Whitmore AC, Linnertz CL, Madden EA, Noll KE, Boyson SP, 624 Parotti B, Xenakis JG, Bell TA, Hock P, Shaw GD, de Villena FP, Ferris MT, Heise MT. 625 2022. Forward genetic screen of homeostatic antibody levels in the Collaborative Cross 626 identifies MBD1 as a novel regulator of B cell homeostasis. PLoS Genet 18:e1010548. 627 41. Blake JA, Baldarelli R, Kadin JA, Richardson JE, Smith CL, Bult CJ. 2021. Mouse 628 Genome Database (MGD): Knowledgebase for mouse-human comparative biology. 629 Nucleic Acids Res 49:D981-d987. 630 42. Perez G, Barber GP, Benet-Pages A, Casper J, Clawson H, Diekhans M, Fischer C, 631 Gonzalez JN, Hinrichs AS, Lee CM, Nassar LR, Raney BJ, Speir ML, van Baren MJ, 632 Vaske CJ, Haussler D, Kent WJ, Haeussler M. 2025. The UCSC Genome Browser 633 database: 2025 update. Nucleic Acids Res 53:D1243-d1249. 634 43. Bates D, Mächler M, Bolker B, Walker S. 2015. Fitting Linear Mixed-Effects Models 635 Using lme4. Journal of Statistical Software 67:1 - 48. 636 44. Efron B, Tibshirani R. 1994. An Introduction to the Bootstrap, Chapman and Hall/CRC, 637 New York. 638 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint 45. Viechtbauer W. 2010. Conducting Meta-Analyses in R with the metafor Package. J Stat 639 Softw 36:1 - 48. 640 46. https://nanostring.com/products/ncounter-analysis-system/ncounter-analysis-641 solutions/nsolver-data-analysis-support/. Accessed December 2, 2025. 642 47. Wobbrock JO, Findlater L, Gergle D, Higgins JJ. 2011. The aligned rank transform for 643 nonparametric factorial analyses using only anova procedures, abstr Proceedings of the 644 SIGCHI Conference on Human Factors in Computing Systems, May 7-12, Vancouver, 645 BC, Canada, Association for Computing Machinery. 646 48. Elkin LA, Kay M, Higgins JJ, Wobbrock JO. 2021. An aligned rank transform procedure 647 for multifactor contrast Tests, abstr The 34th annual ACM symposium on user interface 648 software and technology, October 10-14, Virtual Event, USA, Association for Computing 649 Machinery. 650 49. Kay M, Elkin L, Higgins JJ, Wobbrock JO. 2025. mjskay/ARTool: ARTool 0.11.2, 651 vv0.11.2. Zenodo, http://dx.doi.org/10.5281/zenodo.15192343. 652 50. Chandereng T, Gitter A. 2020. Lag penalized weighted correlation for time series 653 clustering. BMC Bioinform 21:21. 654 51. Tibshirani R, Walther G, Hastie T. 2001. Estimating the number of clusters in a data set 655 via the Gap Statistic. J. R. Stat. Soc., B: Stat. Methodol. 63:411-423. 656 52. Moorhead AM, Jung JY, Smirnov A, Kaufer S, Scidmore MA. 2010. Multiple host 657 proteins that function in phosphatidylinositol-4-phosphate metabolism are recruited to 658 the chlamydial inclusion. Infect Immun 78:1990-2007. 659 53. Clayton EL, Minogue S, Waugh MG. 2013. Mammalian phosphatidylinositol 4-kinases as 660 modulators of membrane trafficking and lipid signaling networks. Prog Lipid Res 52:294-661 304. 662 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint 54. Robertson DK, Gu L, Rowe RK, Beatty WL. 2009. Inclusion biogenesis and reactivation 663 of persistent Chlamydia trachomatis requires host cell sphingolipid biosynthesis. PLOS 664 Pathogens 5:e1000664. 665 55. Finethy R, Coers J. 2016. Sensing the enemy, containing the threat: cell-autonomous 666 immunity to Chlamydia trachomatis. FEMS Microbiol Rev 40:875-893. 667 56. Haldar AK, Piro AS, Finethy R, Espenschied ST, Brown HE, Giebel AM, Frickel EM, 668 Nelson DE, Coers J. 2016. Chlamydia trachomatis is resistant to inclusion ubiquitination 669 and associated host defense in gamma interferon-primed human epithelial cells. mBio 7: 670 e01417-16. 671 57. Hausman JM, Kenny S, Iyer S, Babar A, Qiu J, Fu J, Luo ZQ, Das C. 2020. The two 672 deubiquitinating enzymes from Chlamydia trachomatis have distinct ubiquitin recognition 673 properties. Biochemistry 59:1604-1617. 674 58. Yang Y, Lei W, Zhao L, Wen Y, Li Z. 2022. Insights into mitochondrial dynamics in 675 chlamydial infection. Front Cell Infect Microbiol 12:835181. 676 59. Mi Y, Gurumurthy RK, Zadora PK, Meyer TF, Chumduri C. 2018. Chlamydia trachomatis 677 inhibits homologous recombination repair of DNA Breaks by interfering with PP2A 678 Signaling. mBio 9: e01465-18. 679 60. Johnson KA, Tan M, Sutterlin C. 2009. Centrosome abnormalities during a Chlamydia 680 trachomatis infection are caused by dysregulation of the normal duplication pathway. 681 Cell Microbiol 11:1064-73. 682 61. Rajarathnam K, Schnoor M, Richardson RM, Rajagopal S. 2019. How do chemokines 683 navigate neutrophils to the target site: Dissecting the structural mechanisms and 684 signaling pathways. Cell Signal 54:69-80. 685 62. Lijek RS, Helble JD, Olive AJ, Seiger KW, Starnbach MN. 2018. Pathology after 686 Chlamydia trachomatis infection is driven by nonprotective immune cells that are distinct 687 from protective populations. Proc Natl Acad Sci U S A 115:2216-2221. 688 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint 63. Scurlock AM, Frazer LC, Andrews CW, Jr., O'Connell CM, Foote IP, Bailey SL, Chandra-689 Kuntal K, Kolls JK, Darville T. 2011. Interleukin-17 contributes to generation of Th1 690 immunity and neutrophil recruitment during Chlamydia muridarum genital tract infection 691 but is not required for macrophage influx or normal resolution of infection. Infect Immun 692 79:1349-62. 693 64. Frazer LC, O'Connell CM, Andrews CW, Jr., Zurenski MA, Darville T. 2011. Enhanced 694 neutrophil longevity and recruitment contribute to the severity of oviduct pathology during 695 Chlamydia muridarum infection. Infect Immun 79:4029-41. 696 65. Olive AJ, Gondek DC, Starnbach MN. 2011. CXCR3 and CCR5 are both required for T 697 cell-mediated protection against C. trachomatis infection in the murine genital mucosa. 698 Mucosal Immunol 4:208-16. 699 66. King M, Poya H, Rao J, Natarajan S, Butch AW, Aziz N, Kok S, Chang MH, Lyons JM, 700 Ault K, Kelly KA. 2009. CXCL13 expression in Chlamydia trachomatis infection of the 701 female reproductive tract. Drugs Today (Barc) 45 Suppl B:125-34. 702 67. Marchesi F, Martin AP, Thirunarayanan N, Devany E, Mayer L, Grisotto MG, Furtado 703 GC, Lira SA. 2009. CXCL13 expression in the gut promotes accumulation of IL-22-704 producing lymphoid tissue-inducer cells, and formation of isolated lymphoid follicles. 705 Mucosal Immunol 2:486-494. 706 68. Taylor BD, Zheng X, Darville T, Zhong W, Konganti K, Abiodun-Ojo O, Ness RB, 707 O'Connell CM, Haggerty CL. 2017. Whole-exome sequencing to identify novel biological 708 pathways associated with infertility after pelvic inflammatory disease. Sex Transm Dis 709 44:35-41. 710 69. Masola V, Bellin G, Gambaro G, Onisto M. 2018. Heparanase: a multitasking protein 711 involved in extracellular matrix (ECM) remodeling and intracellular events. Cells 7:236. 712 70. Finethy R, Jorgensen I, Haldar AK, de Zoete MR, Strowig T, Flavell RA, Yamamoto M, 713 Nagarajan UM, Miao EA, Coers J. 2015. Guanylate binding proteins enable rapid 714 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint activation of canonical and noncanonical inflammasomes in Chlamydia-infected 715 macrophages. Infect Immun 83:4740-9. 716 71. Haldar AK, Piro AS, Pilla DM, Yamamoto M, Coers J. 2014. The E2-like conjugation 717 enzyme Atg3 promotes binding of IRG and Gbp proteins to Chlamydia- and 718 Toxoplasma-containing vacuoles and host resistance. PLoS One 9:e86684. 719 72. Wang PH, Liu YF, Tsai HT, Tee YT, Lin LY, Hsieh YH, Yang SF. 2010. Elevated plasma 720 osteopontin level is associated with pelvic inflammatory disease. Reprod Sci17:1052-8. 721 73. De Filippis A, Buommino E, Domenico MD, Feola A, Brunetti-Pierri R, Rizzo A. 2017. 722 Chlamydia trachomatis induces an upregulation of molecular biomarkers podoplanin, 723 Wilms' tumour gene 1, osteopontin and inflammatory cytokines in human mesothelial 724 cells. Microbiol (Reading) 163:654-663. 725 74. Kiviat NB, Wolner-Hanssen P, Eschenbach DA, Wasserheit JN, Paavonen JA, Bell TA, 726 Critchlow CW, Stamm WE, Moore DE, Holmes KK. 1990. Endometrial histopathology in 727 patients with culture-proved upper genital tract infection and laparoscopically diagnosed 728 acute salpingitis. Am J Surg Pathol 14:167-75. 729 75. Zheng X, O'Connell CM, Zhong W, Nagarajan UM, Tripathy M, Lee D, Russell AN, 730 Wiesenfeld H, Hillier S, Darville T. 2018. Discovery of blood transcriptional endotypes in 731 women with pelvic inflammatory disease. J Immunol 200:2941-2956. 732 76. Zheng X, O'Connell CM, Zhong W, Poston TB, Wiesenfeld HC, Hillier SL, Trent M, 733 Gaydos C, Tseng G, Taylor BD, Darville T. 2018. Gene expression signatures can aid 734 diagnosis of sexually transmitted infection-induced endometritis in women. Front Cell 735 Infect Microbiol 8:307. 736 77. Dockterman J, Coers J. 2021. Immunopathogenesis of genital Chlamydia infection: 737 insights from mouse models. Pathogens and Disease 79: ftab012. 738 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint 78. Gyorke CE, Kollipara A, Allen Jt, Zhang Y, Ezzell JA, Darville T, Montgomery SA, 739 Nagarajan UM. 2020. IL-1α is essential for oviduct pathology during genital chlamydial 740 infection in mice. J Immunol 205:3037-3049. 741 79. Darville T, O'Neill JM, Andrews CW, Jr., Nagarajan UM, Stahl L, Ojcius DM. 2003. Toll-742 like receptor-2, but not toll-like receptor-4, is essential for development of oviduct 743 pathology in chlamydial genital tract infection. J Immunol 171:6187-6197. 744 80. Anyalechi GE, Hong J, Danavall DC, Martin DL, Gwyn SE, Horner PJ, Raphael BH, 745 Kirkcaldy RD, Kersh EN, Bernstein KT. 2021. High plasmid gene protein 3 (Pgp3) 746 Chlamydia trachomatis seropositivity, pelvic inflammatory disease, and infertility among 747 women, National Health and Nutrition Examination Survey, United States, 2013-2016. 748 Clin Infect Dis 73:1507-1516. 749 81. Wiesenfeld HC. 2017. Screening for Chlamydia trachomatis Infections in Women. N 750 Engl J Med 376:765-773. 751 82. Tuttle AH, Philip VM, Chesler EJ, Mogil JS. 2018. Comparing phenotypic variation 752 between inbred and outbred mice. Nat Methods 15:994-996. 753 83. Bailey RL, Natividad-Sancho A, Fowler A, Peeling RW, Mabey DC, Whittle HC, Jepson 754 AP. 2009. Host genetic contribution to the cellular immune response to Chlamydia 755 trachomatis: Heritability estimate from a Gambian twin study. Drugs Today 45 Suppl 756 B:45-50. 757 84. Poston TB, Lee DE, Darville T, Zhong W, Dong L, O'Connell CM, Wiesenfeld HC, Hillier 758 SL, Sempowski GD, Zheng X. 2019. Cervical cytokines associated with Chlamydia 759 trachomatis susceptibility and protection. J Infect Dis 220:330-339. 760 85. Nagarajan UM, Prantner D, Sikes JD, Andrews CW, Jr., Goodwin AM, Nagarajan S, 761 Darville T. 2008. Type I interferon signaling exacerbates Chlamydia muridarum genital 762 infection in a murine model. Infect Immun 76:4642-8. 763 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint 86. Chen R, Ma L, Jiang C, Zhang S. 2022. Expression and potential role of CCL4 in CD8+T 764 cells in NSCLC. Clin Transl Oncol 24:2420-2431. 765 87. Repeke CE, Ferreira SB, Jr., Claudino M, Silveira EM, de Assis GF, Avila-Campos MJ, 766 Silva JS, Garlet GP. 2010. Evidences of the cooperative role of the chemokines CCL3, 767 CCL4 and CCL5 and its receptors CCR1+ and CCR5+ in RANKL+ cell migration 768 throughout experimental periodontitis in mice. Bone 46:1122-30. 769 88. Blanpain C, Migeotte I, Lee B, Vakili J, Doranz BJ, Govaerts C, Vassart G, Doms RW, 770 Parmentier M. 1999. CCR5 binds multiple CC-chemokines: MCP-3 acts as a natural 771 antagonist. Blood 94:1899-905. 772 89. Garlanda C, Di Ceglie I, Jaillon S. 2025. IL-1 family cytokines in inflammation and 773 immunity. Cell Mol Immunol 22:1345-1362. 774 90. Zhong W, Dong L, Poston TB, Darville T, Spracklen CN, Wu D, Mohlke KL, Li Y, Li Q, 775 Zheng X. 2020. Inferring regulatory networks from mixed observational data using 776 directed acyclic graphs. Front Genet 11:8. 777 778 779 780 781 782 783 784 785 786 787 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint Table 1. Heritability of chlamydial load and gross pathology. Phenotype Heritability (%) Narrow-sense Broad-sense Day 7 chlamydial load 0.69 0.76 Day 10 chlamydial load 0.58 0.67 Day 35 chlamydial load 0.39 0.42 Day 59 chlamydial load 0.22 0.22 Pathology score 0.50 0.58 788 789 790 791 792 793 794 795 796 797 798 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint 799 Figure 1. Mouse experimental scheme. Twenty Collaborative Cross (CC) strains were 800 pretreated with Depo-Provera and intravaginally inoculated with either C. muridarum strain 801 CM006 or PBS. Cervical swabs were collected throughout infection to monitor bacterial 802 load and clearance. Following resolution of infection, mice were sacrificed for evaluation 803 of genital tract pathology, and serum and immune cells were banked for future analyses. 804 Created in BioRender. O’Connell, C. (2026) https://BioRender.com/t0b0v2x. 805 806 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint 807 Figure 2. Collaborative Cross (CC) mice exhibit wide variation in infection dynamics 808 following C. muridarum challenge. Lower genital tract shedding of strain CM006 was 809 quantified by qPCR targeting the 23S rRNA locus. (A) Mean chlamydial burden ± SEM is 810 shown for a representative subset of CC strains (3–6 mice per group). (B) Time to infection 811 clearance was defined as the first day on which chlamydial genomic DNA fell below the 812 assay limit of detection (<8.3 × 10¹ genomes/swab). (C) Heat map summarizing pathology 813 scores, Day 7 bacterial burden, and infection duration measured through day 59 across 814 all twenty CC strains. Strain sample sizes: CC078 (n=3), CC031 (n=4), CC004 (n=3), 815 CC019 (n=6), CC027 (n=3). 816 817 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint 818 Figure 3. QTL mapping reveals loci on Chromosomes 5 and 16 that influence 819 chlamydial infection outcomes in CC mice. Genome-wide QTL scans (LOD score 820 profiles) identified significant associations between host genetic loci and infection 821 phenotypes. (A) Day 7 chlamydial burden showed its strongest association on 822 Chromosome 16; the corresponding QTL spans a 95% Bayesian credible interval from 6.8 823 to 21.4 Mb. (B) Day 63 pathology scores showed their strongest association on 824 Chromosome 5; this QTL spans a 95% Bayesian credible interval from 40.6 to 126.0 Mb. 825 Arrow indicates the position of LOD peak. (C –D) Founder haplotype probabilities at the 826 peak markers for these loci are shown for each mouse within each CC strain, illustrating 827 the distribution of inferred founder alleles. Notably, CC036 exhibits an equal (50/50) 828 probability of A/J and C57BL/6J founder alleles at the Chromosome 16 burden locus, so 829 chlamydial burden values are indicated for each. 830 831 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint 832 Figure 4. Time series co -expression of cervically -expressed host infection 833 responses for 6 selected CC strains. Lag Penalized Weighted Correlation (LPWC) test 834 used for this analysis allows for temporal offsets and LPWC analysis identified co -835 expression groups of immune genes upregulated with infection onset (A) and genes 836 showing increased expression levels 5 days post -infection (B). Selected host transcripts 837 (N=50) were quantified in total RNA extracted from mouse swabs (Days -2, 3, 5, 7, 10, 838 35) using nCounter assay with additional probes used for normalization (host N=6, 839 chlamydiae N=3). 840 841 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted December 24, 2025. ; https://doi.org/10.64898/2025.12.22.695995doi: bioRxiv preprint