Effect of Time of Day of Infection on Chlamydia Infectivity and Pathogenesis

Genital chlamydia infection in women causes complications such as pelvic inflammatory disease and tubal factor infertility, but it is unclear why some women are more susceptible than others. Possible factors, such as time of day of chlamydia infection on chlamydial pathogenesis has not been determined. We hypothesised that infections during the day, will cause increased complications compared to infections at night. Mice placed under normal 12:12 light: dark (LD) cycle were infected intravaginally with Chlamydia muridarum either at zeitgeber time 3, ZT3 and ZT15. Infectivity was monitored by periodic vaginal swabs and chlamydiae isolation. Blood and vaginal washes were collected for host immunologic response assessments. The reproductive tracts of the mice were examined histopathologically, and fertility was determined by embryo enumeration after mating. Mice infected at ZT3 shed significantly more C. muridarum than mice infected at ZT15. This correlated with the increased genital tract pathology observed in mice infected at ZT3. Mice infected at ZT3 were less fertile than mice infected at ZT15. The results suggest that the time of day of infection influences chlamydial pathogenesis, it indicates a possible association between complications from chlamydia infection and host circadian clock, which may lead to a better understanding of chlamydial pathogenesis.


Results
Effect of Time of Day on C. muridarum infectivity. Female mice housed under 12:12 light: dark (12:12LD) were infected with C. muridarum at either 3-hours after lights-on, Zeitgeber Time 3, (ZT3), or 3-hours after lights-off (ZT15). These time-points were selected because they coincide with the early rest period (ZT3) and early active period (ZT15) of the nocturnal mice. When the infectivity was measured by measuring shedding of chlamydiae into the cervico-vaginal vault, mice infected at ZT3 had significantly higher infectivity compared to the mice infected at ZT15 from days 12-24 post-infection (p < 0.0001) (Fig. 1A). These results suggest that mice infected at ZT15 had lower infectivity from days 12-24 and had cleared the infection by day 24 while mice infected during the day ZT3 had higher infectivity from days 12-24 and did not clear the infection until day 27. The results imply that the time of day in which mice are exposed to pathogen plays an important role in C. muridarum infectivity.

Effect of Time Day of C. muridarum infection on the female reproductive tract pathology.
Female mice housed under LD conditions were infected with C. muridarum at either ZT3 or ZT15. The incidence and severity of gross lesions on reproductive tract were more common in mice infected at ZT3 than those infected at ZT15 (Fig. 1B). The lesions consisted of increased numbers of uterine dilations and paraovarian cysts. Histopathologically, mice infected at ZT3 had evidence of increased incidence and severity of uterine inflammation, hyperplasia, and apoptotic necrosis when compared to mice that were infected at ZT15 ( Fig. 2A-K). Mice infected at ZT15, however, experienced some uterine dilations but no histopathological evidence of significant genital tract pathology. These results indicate that the increase in infectivity observed in the mice infected during the early rest period has a positive correlation to the lesions observed in them.

Effect of Time of Day of C. muridarum infection on cytokine production.
We determined the influence of time of day of chlamydia infection on the immune response to C. muridarum by analysing the cytokines, chemokines and antibodies in vaginal lavages and serum samples. Under LD conditions, mice infected during the day had significantly higher levels of CXCL-1 during the first week of infection compared to mice infected at ZT15 (p < 0.01) (Fig. 3A). During the first and second weeks of infection TNF-α expression was higher in mice infected at ZT3 compared to mice infected at ZT15; however, this difference was not significant (Fig. 3B). Interleukin-1 β (IL-1β), Interleukin-10 (IL-10) and Interferon-γ (IFN-γ) were significantly higher in the first week of infection in mice infected at ZT3 compared to mice infected at ZT15 (p < 0.01, p < 0.01, p < 0.01 respectively) (Fig. 3C-E). The results reveal that there was a greater host inflammatory response to C. muridarum when infection took place during the early rest period than during the early active period. The observation corroborates the hypothesis that the host's physiologic state at the time of infection, which might be controlled by circadian clocks, could influence the intensity and pathologic outcomes of genital chlamydia infection.

Effect of Time of Day of C. muridarum infection on anti-chlamydial antibody production. Under
LD conditions, in the first week of infection, chlamydia specific Immunoglobulin (Ig) IgG expression was higher in mice infected at ZT15 compared mice infected at ZT3 (p < 0.01). While, in second and third weeks of infection, chlamydia specific IgG secretion was significantly higher in mice infected at ZT3 compared to mice infected at ZT15 (p < 0.001) (Fig. 4A). In the first, second and third weeks of infection, there was a significantly higher secretion of chlamydia specific IgG2C in mice infected at ZT3 compared to mice infected at ZT15 (p < 0.001 and p < 0.01) (Fig. 4B). There was no significant difference in the secretion of serum chlamydia specific IgA at ZT3 and ZT15 infected mice (p > 0.05) (Fig. 4C). However, there was a noticeable difference in IgA secretion in the vaginal wash samples, with a significant increase by the fourth week of infection in mice infected at ZT 15 (p < 0.01) (Fig. 4D). The results show that total IgG in serum is important in mice infected during the early active period in the first week of infection. However, IgG2C seems more relevant in mice infected during the early rest period. While mucosal IgA in mice infected during the early active can be associated with chlamydia clearance.
Effect of Time of Day on C. muridarum reinfection. In humans, complications due to genital chlamydia infection often occurs after repeat or untreated infections. Due to the differences observed in infectivity, pathology, and immune response associated with time of day of chlamydia infection, we wanted to determine if these differences would also be observed in mice with repeat infections. Female mice housed under LD conditions were reinfected with C. muridarum at either ZT3 or ZT15, to determine if repeat infection at the same time as the first infection had worse outcomes. The difference in bacterial burden following reinfection, between ZT3 and ZT15 infection was comparable to what we observed in a single infection. However, mice reinfected at ZT3 had more gross lesions compared to mice reinfected at ZT15, and only mice reinfected at ZT3 had paraovarian cysts ( Fig. 5A) with half of the mice reinfected at ZT3 having uterine tubal dilations or hydrometra. There were no observable differences in uterine histopathology when comparing mice infected at either ZT3 or ZT15 with both groups presenting mild endometrial hyperplasia (increased numbers of cell in the endometrial lining), endometrial apoptotic necrosis (cell death) and hydrometra (accumulation of fluid in the cavity of the uterus). One mouse Figure 1. Chlamydia Infectivity and gross pathology in female mice housed under LD conditions. Mice (n = 12 per group) were infected with C. muridarum at either ZT3 or ZT15. Experiment was repeated twice. (A) Mice infected at ZT3 had a significantly higher bacterial burden compared to mice infected at ZT15 from days 12 and 24 (***p < 0.001). Data was analysed using two-way analysis of variance (ANOVA). (B) Gross pathology in female mice housed under LD conditions. Note, mice infected ZT3 had paraovarian (next to the ovary) cysts as indicated by the white arrow, mice infected at ZT15 did not have paraovarian cysts.
with hydrosalpinx associated with dilated oviductal ampulla and two mice with Bursal cysts were observed in the mice reinfected during the early rest period, and in this group had similar incidence of inflammation in their ovary compared to mice reinfected during the early rest period. An interesting observation was that between the ZT3 and ZT15 uninfected control groups, the incidence of endometrial hyperplasia was higher in the ZT3 group than in the ZT15 group. In contrast, the incidence of endometrial apoptotic necrosis was higher in the ZT15 control group than in the ZT3 control group. Following reinfection, control, ZT3 and ZT15 reinfected mice were mated to determine the effect of time of day of infection on pregnancy and fertility rate. There was no significant difference in the pregnancy rate between the ZT3 and ZT15 reinfected mice (Fig. 5B), however, mice infected during the early active period had a significantly greater number of pups compared to mice infected during the early rest period. Mice reinfected during the early active period had an average of eight pups while mice infected during the early rest period had an average of 5 (p < 0.05) (Fig. 5C). The results suggested that the effect of time of day on chlamydial pathogenesis is the same during a repeat infection, furthermore time of day of infection influences the overall fertility rate of the infected mice. LGV) at either ZT3 or ZT15. Like the mice infected with C. muridarum, mice infected with LGV at ZT3 had significantly higher infectivity compared to mice infected at ZT15 from days 19 and 26 (p < 0.0001) (S. Fig. 1). Mice infected at ZT15 cleared the infection by day 26 while mice infected at ZT3 did not fully clear their infection even at 26 days post infection. On histopathology, however, there were no significant differences in the incidence and severity of lesions observed between control mice and mice infected during early active or early rest periods. Overall the uterus of mice infected during the early rest period and mice infected during the early active period were not significantly different histopathologically, however the ovaries of mice infected during the early rest period had more inflammation (S. Fig. 2). Following reinfection, we observed a trend in fertility in LGV infected mice like C. muridarum infected mice. Mice infected during the early rest period had a lower pregnancy rate than mice infected during the early active period and they had significantly less pups (p < 0.05) (S . Figs 3 and 4). The results suggest that the effect of time of day on infection was not strain specific, implying that human chlamydia infection is under the effect of time of day of infection.

Effect of time of day of infection on chlamydia infectivity and pathogenesis in older mice.
It has been reported that as mice age, the duration of their chlamydia infection is shorter 50 which mimics the lower incidence of genital chlamydia infection in older human beings 51 . To test the effect of time of day of chlamydia infection on the intensity of infection and development of lesions in older mice, 15-week-old female mice housed under LD conditions were infected and reinfected with C. muridarum at either ZT3 or ZT15. The results showed that there was no significant difference in infectivity between ZT3 and ZT15 infected older mice with the duration of the infection being shorter compared to the younger mice (Fig. 6A). Note that the infection was cleared by day 24 in most of the mice, compared to the younger mice that cleared their infection from day 27. In addition, the incidence and severity of significant gross pathology and histopathological changes were greatly diminished when comparing mice infected in the early rest and early active periods (Fig. 6B). The results indicated that the effect of time of day of infection is not apparent in older mice, thus implying that age is an important factor in determining the effect of time of day of infection.

Discussion
Genital chlamydia infection, a disease caused by infection with C. trachomatis, is the most frequently reported bacterial STD in the United States 1,2 . We still do not fully understand why some women are more likely to develop an asymptomatic infection, have severe infection leading to PID and tubal factor infertility, or after exposure to C. trachomatis stay uninfected 3,6,7 . We hypothesize that some of these differences amongst individuals infected with C. trachomatis might be due to the time of day they were exposed to it.
Our results revealed that under normal LD conditions, mice infected at the early active period shed less bacteria and cleared the infection faster compared to mice infected during the early rest period. The results from our study is corroborated by other studies conducted with pathogens such as herpes simplex virus I and IV, influenza virus and Salmonella, which revealed that mice infected during the active period (at night) had significantly less pathogen burdens compared to the mice that were infected during the early rest period (day) 18,19 . Also, under LD conditions, mice infected with C. muridarum during the early active period showed reduced gross lesions with less severe uterine inflammation and dilation compared to mice infected during the early rest period. In addition, development of hydrosalpinx and Bursal cysts was only observed in mice infected during the early rest period. These results corroborate earlier studies, which showed more deaths from endotoxin and pneumococcal infections in nocturnal rodents infected during the late rest period 16,17 . The results reveal that mice can resist genital C. muridarum infection when they are infected in the early active period than during the early rest period, indicating a direct linkage between the lesions observed after infection with the time of day of infection. The reason for this difference in pathology could be due to the immune response generated by the chlamydia at the time of infection.
During chlamydia infection TLRs found on immune cells particularly TLR2, TLR4 and TLR9 which are important in recognizing and sensing chlamydia, are highly expressed in the female genital tract where they mediate immune cell response to chlamydia [52][53][54] . TLR4, TLR5 and TLR9 have been reported to show daily variation in response to their antigens 29,55,56 . We analysed the host immune responses to the infection by measuring the cytokine/chemokine response in the genital washes and chlamydia-specific antibody production. We observed differences in immune response based on the time of day that the mice were infected. For instance, mice infected during the early rest period had significantly higher levels of CXCL-1 in the first week of infection compared to mice infected during the early active period. That implies that more neutrophils were called early to the site of infection, and it has been reported that increased neutrophils are associated with increased pathology 57 , with the depletion of neutrophils associated with reduction in genital tract immunopathology 58 . In addition, neutrophils have been shown to be recruited in the early rest period during an inflammatory response and this rhythmic . Anti-chlamydia antibody determination in mice housed under LD conditions. Mice were infected with C. muridarum at either ZT3 or ZT15 and serum samples were collected at different time points after infection. (A) IgG was significantly higher in mice infected at ZT15 in the first week of infection compared to mice infected at ZT3 (*p < 0.05). However, during weeks two and three of infection, IgG was significantly higher in mice infected at ZT3 compared to mice infected at ZT15 (**p < 0.01). (B) IgG2C was significantly higher in mice infected at ZT3 compared to mice infected at ZT15 (**p < 0.01, ****p < 0.0001). (C) No significant difference in IgA levels between mice infected at ZT3 or ZT15. (D) Significant increase in IgA levels in mice infected at ZT15 in the fourth week of infection compared to mice infected at ZT3 (**p < 0.01). The data was analysed using a one-way ANOVA and post hoc test. (2019) 9:11405 | https://doi.org/10.1038/s41598-019-47878-y www.nature.com/scientificreports www.nature.com/scientificreports/ recruitment was associated with epithelial cells producing chemokines 29 . It should be noted that the epithelial cells in the female genital tract are referred to as sentinels of immune protection recognizing pathogens and signalling underlying immune cells 59,60 . Here we show that clusters of neutrophils in infected mice persisted in the genital tract tissue all through the infection. Mice infected during the early rest period had higher proinflammatory cytokine secretion compared to mice infected during the early active period and significantly higher concentrations of IL-10, IFN-γ, and IL-1β. We do not know the exact cells responsible for producing the cytokines, but this could be attributed to the innate cells recruited early to the site of infection; mitigated by the expression of TLRs in the genital tract; such as the natural killer cells (NK) cells, monocytes, macrophages, dendritic cells and the epithelial cells that are already part of the physical barrier providing protection in the genital tract 52,56,58,59,61-66 . IFN-γ and IL-1β are important in clearing and reducing the bacterial burden, by helping to elicit a proper Th1 response [23][24][25] . IFN-γ is produced early by NK and other immune cells, and may be modulating DC function towards Th1 activation 67 . However, very high amounts of IL-1β and other pro-inflammatory cytokines have been associated with detrimental pathologies observed during chlamydia infection 57,68 . There is often a high concentration of anti-inflammatory cytokine IL-10 to counter the effects of high concentrations of proinflammatory cytokines, which we observed in the mice infected during the early rest period 69 . In general, a high inflammatory response is beneficial in clearing an infection, but prolonged inflammation can lead to tissue damage 57 . We believe that the differences observed in the lesions of mice infected during the early rest or active periods are associated with cytokine response.
The presence of anti-chlamydia IgG, IgG2C, and IgA are indicative of a current or chronic chlamydia infection and to a certain extent protection from chlamydia infection 70 . Under normal light conditions serum levels of IgG were significantly higher in mice infected during the early active period in the first week of infection. However, in the second and third weeks of infection, mice infected during the early rest period had significantly higher IgG. It is possible that having an infection during the early active period elicits IgG quickly, which leads to the faster clearance of infection compared to mice infected during the early rest period. In the first, second and third weeks of infection, mice infected during the early rest period had significantly higher IgG2C than mice infected during the early active period. It should be noted that the IFN-γ produced during the early stages of chlamydia infection has been associated with the production of the Th1 cytophilic antibody IgG2a (in BL6 mice; IgG2c) 67 . Here we show a strong correlation between IFN-γ and IgG2c secretion in mice infected at ZT3. This Th1 associated antibody might be important in process of reducing the high bacteria burden in mice infected during the early www.nature.com/scientificreports www.nature.com/scientificreports/ rest period. There was no significant difference in the expression of chlamydia specific serum IgA between mice infected during the early rest or active periods. However, we noticed a dramatic increase in mucosal IgA in mice infected during the early active period, four weeks after the infection, this might be associated with chlamydia clearance and longer protective immunity.
To buttress the effect of time of day of infection on infectivity and pathology, we decided to infect mice twice with C. muridarum. Following reinfection, a pattern like we had observed after a single infection, with mice infected during the early rest period having more infectivity and pathology than mice infected during the early active period. Mice are nocturnal and are more active at night 71 , and we infected about 3 hours into the mice active period. Based on the infectivity and pathology data, mice infected with C. muridarum during the active period results in less severe lesions and disease outcomes compared to having an infection during their early rest period. However, in humans the reverse might be the case since we are diurnal organisms. This suggests that in humans, infection with C. trachomatis during the day might result in less severe infection and reduced complications when compared to night infection. This would have ramifications in the public health status of individuals especially those who are more liable to be active at night such as prostitutes and maybe even shift workers.
Women who are infected with C. trachomatis have a greater chance of becoming infertile 6,7 . Reinfection leads to increased pathology and infertility, here we used mice that were reinfected with C. muridarum to determine the effect of time of day of chlamydia infection on fertility outcomes. There was no significant difference in the fertility rate between the groups; however, there was a significant difference in the number of pups, with mice infected during the early active period having a significantly greater number of pups than mice infected during the early rest period. This is intriguing as there appears to be a direct relationship between time of infection and number of pups; fertility rate. We do not know if this relationship exists in women of childbearing age infected with C. trachomatis.
Our results suggest that the effect of time of day of infection was not strain specific or age dependent. Like C. muridarum infected mice, mice infected with human chlamydia serovar, LGV, had higher infectivity if the infection was during the early rest period compared to infection during the early active period. They were also less fertile if they were infected during the early rest period. In addition, we investigated the effect of age on this phenomenon, since it is known that older mice have a shorter duration of genital chlamydia infection 50 . Our results showed that older mice in general cleared their infections faster, however, the trend was similar with mice infected during the early active period having lower bacterial burden than mice infected during the early rest period.
Time of day of chlamydia infection appears to have an influence on its pathogenesis and disease progression. The time of day that an individual becomes infected with C. trachomatis could explain why there are differences in susceptibility and severity among individuals. It may also predict why some individuals are asymptomatic while www.nature.com/scientificreports www.nature.com/scientificreports/ others are not. This can be linked to the rhythmic changes in the production of chemokines/cytokines and antibodies. The numbers of immune cells found at the site of an infection have been shown to be regulated by the SCN through humoral or neuronal coordination 33,[47][48][49] . However, we do not know if the production of the immune effectors observed in our study, is controlled by the clock within the immune cells or the oscillators associated with the clocks. Mechanistic studies should be carried out to understand the circadian control of immune cells in chlamydia infection and pathogenesis. Female sex hormones have been postulated to control immune response to chlamydia infection 52 , further work on understanding how this phenomenon happens at different times of the day would be important in elucidating chlamydial pathogenesis. Overall, findings from this study have implications for shift workers and individuals who do not have a normal sleep schedule. The circadian clocks of shift workers are disrupted causing them to potentially become more susceptible to diseases in general. These individuals, especially women, could be more at risk for developing a more severe genital chlamydia infection leading to infertility. This study provides an opportunity from which to start threading together the numerous innovative developments in the circadian field into elucidating the mechanisms underlying chlamydial pathogenesis.

Materials and Methods
Animal protocol approval statement. This study was carried out in strict adherence with the recom- C. muridarum infectivity assay. All mice were subcutaneously injected between 10:00 am and 12:00 noon with 2.5 mg/mouse Depo Provera, medroxyprogesterone acetate (Pfizer, New York, NY) in sterile Phosphate Buffer Saline (PBS) to synchronize the estrous cycle. Mice were intravaginally infected seven days later, with 1 × 10 5 C. muridarum at 10:00 am (ZT3, early rest period) or 10:00 pm (ZT15, early active period), which can be interpreted as three hours after the lights were turned on or off in the room (7:00 am lights on, 7:00 pm lights off). The mice were anesthetized using isoflurane during process of infection. For a repeat infection or reinfection, mice were infected 4 weeks after following the same process mentioned above. Infected and reinfected mice were swabbed every three days for 27 days and the bacteria was isolated and cultured to track the progression and clearance of the infection.
Gross pathology. Mice were infected and reinfected intravaginally with C. muridarum (1 × 10 5 IFU per mouse). All mice were sacrificed four weeks after infection, between 10:00 am and 12:00 noon and the entire genital tract from the vagina to the ovary was collected and fixed in neutral-buffered, 10% formalin. Euthanasia was carried out using carbon dioxide and cervical dislocation. Gross pathology was performed by counting and observing the numbers of paraovarian cysts and tubal dilations in the genital tract of chlamydia infected mice. Note that the collection was synchronised to make sure that all samples were collected within the same period.
Histopathology. Histopathology of the genital tract associated with chlamydia infection was investigated.
Mice were infected and reinfected intravaginally with C. muridarum (1 × 10 5 IFU per mouse). All mice were sacrificed four weeks after infection between 10:00 am and 12:00 noon, and the entire genital tract from the vagina to the ovary was collected and fixed in neutral-buffered, 10% formalin solution for less than 1 week. Tissues were routinely processed, embedded in paraffin, cut approximately into 5 μm sections, and stained with hematoxylin and eosin (H and E). Histopathological exam consisted of evaluation of the ovaries, oviducts, and uteri for the incidence (presence or absence), severity, and distribution of inflammation, necrosis, and hyperplasia. Histopathologic severity scores were assigned as grades 0 (no significant histopathological alterations); 1 (minimal); 2 (mild); 3 (moderate); or 4 (severe) based on an increasing extent and/or complexity of change, unless otherwise specified. Lesion distribution was recorded as focal, multifocal, or diffuse, with distribution scores of 1, 2, or 3, respectively. A total severity-and-distribution group score was calculated by adding individual distribution and severity scores. Because the group sizes were uneven, an average severity-and-distribution group score was calculated by dividing the total severity-and-distribution score by the numbers of animals in the group.
Cytokine and chemokine assay. Vaginal lavages were collected every week between 10:00 am and 12:00 noon throughout the duration of the infection and the amount of cytokines (CXCL1, TNF-α, IL-10, IL-1β and IFN-γ) produced was determined using the R&D Systems Magnetic Luminex Assay, Mouse Premixed Multi-Analyte Kit (R&D, Hercules, CA) in accordance with the manufacturer's protocol. The concentration of cytokine in each sample was obtained by extrapolation from a standard calibration curve. The mean and SD of all replicate cultures were calculated.
Enzyme linked immunosorbent assay. Blood samples were collected every week between 10:00 am and 12:00 noon throughout the duration of the infection and serum was collected by centrifuging the blood sample at 2500 rpm for 2 minutes. Determination of concentrations of chlamydia-specific antibody isotypes (IgG, IgG2C, and IgA) in vaginal lavage and serum was measured by a standard ELISA procedure described previously 72 . UV inactivated. C. muridarum elementary bodies (10 µg/ml) in 50 µl of PBS was used to coat 96 -well plates (Nunc Life Technologies, Rochester, NY) overnight at 4 °C. Plates were blocked with 1% bovine serum albumin containing 5% goat serum in PBS -tween. Vaginal lavage (50 µl) in a twofold serial dilution were added to each well. Horseradish peroxidase -conjugated goat -anti-mouse IgG and IgA isotypes (50 µl) (Southern Biotechnology Associates, Inc., Birmingham, AL) were added to each plate and incubated for one hour and developed with 2,2′-azino-bis (3-ethylbenzthiazoline-6-sulfonic acid) (ABTS). The optical density was measured at 490 nm on a microplate reader and the results were generated with a standard curve. Data sets are displayed using the corresponding absorbance values as mean concentrations (ng/ml) ± SD and represent the mean values from triplicate experiments.
Fertility assay. Three weeks after reinfection, female uninfected control, mice infected at ZT3 and ZT15 were placed in cages with proven fertile male C57Bl/6J mice (Jackson Laboratory, Bar Harbor, MA), at either three females or two females to one male mouse. The female mice were weighed every three days after one week until they have gained approximately 10 grams to confirm pregnancy. Once pregnancy has been confirmed, mice were sacrificed and dissected to determine the number of pups 73,74 . Statistical analysis. Nonparametric one-way analysis of variance (ANOVA) was used to analyse the statistical differences in immune response and fertility rate between the treatment groups. Two-way ANOVA was used to determine the difference in infectivity between the treatment groups. In addition, we also did a post hoc test after the one way and two-way ANOVA, to determine the actual statistical relationship between the treatment groups. Chi-square test was used to determine the difference in pregnancy rate between the different treatment groups. Statistical significance was determined at P < 0.05. GraphPad Prism (La Jolla, CA) is the statistical package that was used for analysing the data.

Data Availability
All data and results have been added to this manuscript and the Supplementary Material section.