Bacterial prostatitis affects 1% of men, with increased incidence in the elderly. Acute bacterial prostatitis frequently progresses to chronicity, marked by recurrent episodes interspersed with asymptomatic periods of variable duration. Antibiotic treatment is standard of care; however, dissemination of antimicrobially resistant uropathogens threatens therapy efficacy. Thus, development of nonantibiotic-based approaches to treat chronic disease is a priority. Currently, why chronic prostatitis arises is unclear, as the immune response to prostate infection is incompletely understood. As 80% of prostatitis cases are caused by Gram-negative uropathogenic Escherichia coli (UPEC) or Gram-positive Enterococcus faecalis, we used a mouse transurethral instillation model to address the hypothesis that an innate immune response fails to develop following prostate infection with these uropathogens, leading to chronic disease. Surprisingly, infection induced robust proinflammatory cytokine expression and myeloid cell infiltration. Following a second infection, cytokine responses and innate cell infiltration were largely comparable to primary infection. Characteristic of memory responses, more lymphoid cells infiltrated the prostate in a second infection compared with a first, suggesting that adaptive immunity develops to eliminate the pathogens. Unexpectedly, bacterial burden in prostates challenged with either UPEC or E. faecalis was equal or greater than primary infection despite that a protective adaptive response to UPEC infection was evident in the bladder of the same animals. Our findings support that chronic or recurrent prostatitis develops despite strong innate immune responses and may be the result of a failure to develop immune memory to infection, pointing to actionable targets for immunotherapy.

Prostatitis is a common term encompassing the clinical conditions of acute or chronic bacterial prostatitis, chronic pelvic pain syndrome, and asymptomatic inflammatory prostatitis (1). Symptoms include urinary frequency and urgency and suprapubic, lower back, or perianal pain during micturition and sex (2). All-cause prostatitis develops in nearly 10% of men, with increased risks of chronicity after the age of 50 (3, 4). Prostatitis is more common in the elderly, similar to male urinary tract infection (UTI) (2, 58). An estimated 10% of all prostatitis cases are due to bacterial infection (2). Acute bacterial prostatitis is diagnosed by positive bacterial cultures from expressed fluids following prostate massage (9, 10). Nearly half of patients with acute infection will develop chronic bacterial prostatitis characterized by variable-length asymptomatic periods interspersed with acute symptomatic infections (2, 11, 12).

Epidemiological studies from hospitalized cohorts demonstrate that Escherichia coli or Gram-positive Enterococci spp., and in particular E. faecalis, represent up to 80% of all pathogens isolated from acute or chronic prostatitis patients (5, 1214). Similar to UTI, acute and chronic bacterial prostatitis are typically treated with analgesics to minimize pain and antibiotics, such as fluoroquinolones or trimethoprim-sulfamethoxazole, to eradicate microorganisms (5, 12, 14). Depending on the severity of acute symptoms, recommended treatment guidelines may include a single dose of oral fluoroquinolones or i.m. cephalosporin, followed by a 10-d tetracycline regimen or 2 to 4 wk of fluoroquinolones in combination with aminoglycosides (14, 15). Chronic prostatitis antibiotic regimens are similar but longer, typically ranging from 6 to 12 wk (14). These extended courses are needed to eradicate persistent or recurrent infections because of the reduced penetrance of certain antibiotics (12).

Alarmingly, unusually long regimens of up to 6 mo may lead to the loss of efficacy of fluoroquinolones in 10–30% of cases (16). These complicated antibiotic regimens highlight the difficulty in treating prostatitis and the challenge to find alternative treatment modalities. The expansion of multidrug-resistant uropathogens further compounds this problem (17). To limit development of antibiotic resistance, empirical or “best-guess” administration of antibiotics prior to microbiological analysis is recommended only in acute prostatitis (14). Despite this, most nonbacterial prostatitis cases are treated with antibiotics, as this approach, surprisingly, results in alleviation of symptoms (14). It is unclear whether this effect is ascribable to the antibiotics themselves or a placebo effect or whether bacteria are present but undetectable by current testing protocols (18). Although antibiotic treatment for prostatitis can be challenging, no alternative therapies, including immunomodulatory strategies, exist. One reason for this may be that the immune response provoked during infection is incompletely understood.

Indeed, our understanding of bacterial colonization, prostate-resident, and infiltrating immune cells and their roles in bacterial clearance or induction of specific memory in the prostate is based on a limited number of studies. For example, with respect to colonization, whereas the prostatitis-derived E. coli CP1 strain expresses fewer virulence factors than the uropathogenic E. coli (UPEC) cystitis isolate NU14, it colonizes mouse prostates to a greater extent than NU14 (19), supporting that prostate colonization is determined, in part, by the pathogen. CP1 colonizes the bladder and prostate of C57BL/6 mice equally well at 24 h postinfection (PI); however, CP1 persists longer in the prostate compared with the bladder (19). CP1 induces inflammation in infected prostates from NOD and C57BL/6 mice, but only NOD mice exhibit chronic pelvic pain (19). Adoptive transfer of IL-17A–expressing CD4 T cells from infected NOD mice into naive animals is sufficient to confer a pain phenotype and prostate-specific inflammation, suggesting that the immune response may drive chronic inflammation in the absence of infection (20). Infection of C57BL/6 mouse prostates with UPEC strain 1667 induces proinflammatory cytokine RNA expression, inflammatory cell infiltration, and tissue damage in the prostate over 2 wk (21). C3H/HeOuJ mice infected with UPEC 1667 present with persistent inflammation characterized by collagen deposition (22). When treated with a 2-wk antibiotic regimen, followed by an 8-wk washout period, these mice exhibited a reversal in collagen content and diminishment of tissue inflammation despite being infected for 28 d prior to treatment (23). When infected for 2–12 mo, C57BL/6 mice present with hyperplastic epithelia and greater numbers of macrophages and Th17 T cells in the prostate (24).

Prostate infection with Propionibacterium acnes, a Gram-positive strain associated with benign prostatic hyperplasia, induces inflammation and innate immune cell infiltration only after 1 wk of infection (25). Patient-isolated, Gram-positive uropathogens also induce pelvic pain and T cell infiltration in the NOD mouse and colonize C57BL/6 mice as well but without inducing a pain phenotype, similar to that observed in UPEC prostate infection (26). Together, these studies present a somewhat fragmented picture of the innate immune response to bacterial prostatitis and provide little insight into development of adaptive immunity to infectious prostatitis.

Recently, we postulated that bacterial prostatitis should be classified as a type of UTI, similar to cystitis or pyelonephritis (27). In addition to sharing associated risks and infection via urethral ascension, UTI and bacterial prostatitis can be caused by the same uropathogens, including UPEC and E. faecalis. In this study, we hypothesized that chronic prostatitis develops because an innate immune response to prostate infection is not induced, similar to that observed in the bladder of male mice with cystitis, in which limited immune cell infiltration leads to a failure to resolve infection (2830). To test this hypothesis, we measured colonization and the immune response in a model of acute and chronic prostatitis using a Gram-negative UPEC and Gram-positive E. faecalis strain. Surprisingly, the response to UPEC infection in the prostate, including robust neutrophil infiltration, was more similar to UTI in female animals than male mice (2830). Additionally, in UPEC and E. faecalis infection, the prostate exhibited a cytokine signature divergent from that of the bladder and, critically, unlike the bladder, failed to develop protective adaptive immunity to challenge infection, providing a possible explanation for the frequency in which chronic or recurrent prostatitis develops in men.

This study was conducted using a preclinical mouse model in controlled laboratory experiments to test the hypothesis that the immune response to infection with E. coli or E. faecalis is similar between the bladder and prostate. Animals were assigned to groups by random partition into cages. In each experiment, a minimum of three and a maximum of six animals constituted an experimental group, and all experiments were repeated 2–10 times. Data were pooled before statistical analysis. We have observed that abnormal kidneys negatively impact resolution of infection; thus, in all of our studies, we have established a priori that mice with atrophied, enlarged, or markedly pale kidneys at the time of sacrifice are to be excluded from all analyses. End points were determined before beginning experiments, and researchers were not blinded to experimental groups.

Animal experiments were conducted in accordance with the protocol number 2016-0010, approved by the Comités d’Ethique pour l’Expérimentation Animale at Institut Pasteur (the ethics committee for animal experimentation) in observation of the European Directive 2010/63/EU. In this study, C57BL/6 male mice between the ages of 6–12 wk from Charles River Laboratories France were used. Mice were anesthetized by i.p. injection of 100 mg/kg ketamine and 5 mg/kg xylazine and sacrificed by carbon dioxide inhalation.

Male mice were anesthetized as above, catheterized transurethrally, and infected with a total of 1 × 107 CFU of single or mixed bacterial inocula of indicated strains of UPEC (UTI89-GFP-ampR or UTI89-RFP-kanR) and/or E. faecalis (OG1RF, OG1RF-GFP, or OG1RF_intergenicRS00490RS00495::Tn) in 50 μl PBS as previously described (29, 3136). Antibiotic resistances and concentrations used were as follows: UTI89-GFP-ampR, 100 μg/ml ampicillin; UTI89-RFP-kanR, 50 μg/ml kanamycin; E. faecalis strain OG1RF, 250 μg/ml rifampicin; OG1RF-GFP, 250 μg/ml rifampicin; and OG1RF_intergenicRS00490RS00495::Tn, 250 μg/ml rifampicin and 7 μg/ml chloramphenicol.

Urine was collected 24 h PI and every 2–5 d thereafter. Two microliters of urine were diluted into 8 μl PBS spotted on agar plates containing antibiotics as appropriate. The qualitative presence of any bacterial growth was counted as positive for infection. The limit of detection (LOD) for this assay is 500 bacteria/ml of urine. At 25–30 d PI, we treated mice with antibiotics dependent upon the initial infection. UPEC-infected mice were given 8 mg/ml trimethoprim/sulfamethoxazole (Avemix) in the drinking water. Mice infected with E. faecalis were i.p. injected with 10 mg of carbenicillin in 100 μl PBS (37, 38). Mice coinfected with both organisms received both antibiotics. Antibiotic treatments were administered for 5 d and were followed by a washout period of 3–6 d prior to challenge infection. In the case that a majority of animals remained infected after the first washout period, a second round of antibiotics was administered. Following the washout period, monomicrobially infected mice were challenged with isogenic strains of the bacteria used for primary infection. Polymicrobially infected mice were challenged with either UPEC or E. faecalis isogenic strains.

To determine CFU in infected organs, mice were sacrificed 1, 2, 14, or 42 d PI, and bladders and prostates were removed. Organs were immediately placed in 1 ml of cold PBS, then homogenized with a Precellys 24 homogenizer. One hundred microliters from homogenized organs was serially diluted and plated on appropriate complete medium (Luria–Bertani agar for UPEC and brain–heart infusion agar for E. faecalis) with antibiotics as appropriate. The LOD for organ CFU is 20 or 40 bacteria per organ, depending upon the number and volume of dilutions plated, and is indicated by a dotted line in graphs. All sterile organs are reported at the LOD. Homogenates were clarified by microcentrifugation (17,000 × g; 4°C; 5 min), and supernatants were stored at −20°C until analysis for cytokine levels in low protein binding plates. After thawing on ice, samples were centrifuged again to remove remaining cell debris prior to analysis (200 g; 4°C; 5 min). To avoid interassay variability, whenever possible, samples were analyzed at the same time with the LEGENDplex Mouse Inflammation Panel (BioLegend). The assay was performed following the manufacturer’s instructions for measuring serum or plasma samples using a V-bottom plate with one modification: the first incubation on the shaker was extended from 2 h at room temperature to overnight at 4°C. Samples were acquired on an LSRFortessa (BD Biosciences) and analyzed using version 7.1 of the LEGENDplex Data Analysis Software (BioLegend) for Mac OS X.

Mice were sacrificed 24 h after primary or after challenge infection and the entire prostate removed. Single-cell suspensions were prepared following modification of a protocol for bladder tissue preparation previously described (30). Briefly, minced prostates were incubated in 0.34 U/ml Liberase (Roche Diagnostics) in PBS at 37°C for 45 min, with manual agitation every 15 min. Digested tissue was passed through a 100-μm filter (Miltenyi Biotec), washed, blocked with Fc Block (BD Biosciences), and immunostained with the following Abs (clone; supplier): B220 (RA3-6B2; BD Biosciences), CD11b (M1/70; BD Biosciences), CD3 (145-2C11; BD Biosciences), CD4 (RM4-5 or GK1.5; BD Biosciences), CD45 (30-F11; BD Biosciences), CD64 (X54-5/7.1; BD Biosciences), CD90.2 (53.2.1; eBioscience), CD103 (M290; BD Biosciences), F4/80 (CI:A3-1; AbD Serotec), Ly-6C/Ly-6G ([Gr-1] RB6-8C5; BD Biosciences), MHC class II (MHC II; [I-A/I-E] M5 or 114.15.2; eBioscience), NK1.1 (PK136; BD Biosciences), and Siglec F (E50-2440; BD Biosciences). Samples were acquired on an LSRFortessa (BD Biosciences) and analyzed using FlowJo Version 10 (FlowJo) for Mac OS X. Prior to cytometer acquisition, 10 μl of stained sample were added to 20 μl AccuCheck Counting Beads (Invitrogen) in 200 μl PBS to measure total cell counts in the prostate. The gating strategies used are depicted in Supplemental Fig. 3.

To detect E. faecalis by flow cytometry, bacteria were stained with wheat germ agglutinin conjugated to Alexa Fluor 594 (Thermo Fisher Scientific) just prior to instillation. Wheat germ agglutinin conjugate stock solution was prepared as per the manufacturer’s instructions in PBS without addition of sodium azide, aliquoted, and frozen at −20°C. Prior to staining, the stock solution was thawed, centrifuged for 10 s at 17,000 × g to eliminate protein aggregates, and the supernatant was used for staining. Wheat germ agglutinin conjugate was mixed with the bacterial inoculum at a working concentration of 10 μg/ml and incubated statically for 10 min at 37°C. The inoculum was spun at 17,000 × g for 1 min, the supernatant removed, and pellet resuspended in PBS for a total of three washes. The bacterial concentration was adjusted to 2 × 108 CFU/ml in PBS for infection.

Statistical analysis was performed in GraphPad Prism 8 (GraphPad) for Mac OS X applying the nonparametric Wilcoxon test for paired data, and the nonparametric Mann–Whitney U test (two groups) or Kruskal–Wallis test (three or more groups) for unpaired data. When more than two groups were being compared or to correct for comparisons made within an analysis or experiment, calculated p values were corrected for multiple testing with the false discovery rate (FDR) method (https://jboussier.shinyapps.io/MultipleTesting/) to determine the q value (FDR adjusted p value). All calculated q values are shown in the figures, and those that met the criteria for statistical significance (q < 0.05) are denoted with red text. Qlucore Omics Explorer 3.6 software (Qlucore) was used to calculate q values of cytokine protein expression using the nonparametric Mann–Whitney with correction for multiple testing. Calculated p values and q values are shown in Tables 1–8, in which bold text denotes p values or q values <0.05.

In the course of a previously published study, we observed that in addition to the bladder, UPEC robustly colonizes the prostate of male mice for up to 2 wk (30). As E. coli and E. faecalis are the principal uropathogens causing bacterial prostatitis, we first determined the capacity of these bacterial strains to specifically colonize the prostate, in comparison with the bladder, over time. We transurethrally infected 6–8-wk-old male C57BL/6 mice with 1 × 107 CFU of one of two representative uropathogens: UPEC strain UTI89-RFP-kanR, resistant to kanamycin and expressing red fluorescent protein, or E. faecalis strain OG1RF, resistant to rifampicin (29, 33). We quantified bacterial burden in homogenized bladders and prostates at 1, 2, and 14 d PI to model acute and chronic infection. We observed that although UPEC robustly colonized both organs in male mice, UPEC bacterial burden was significantly higher in prostates, compared with bladders, of infected mice at day 1 (Fig. 1A). Differences in UPEC colonization between prostates and bladders were no longer apparent at 2 or 14 d PI. We reported that male mouse bladders remain colonized up to 28 d following UPEC infection (30). In this study, we observed that prostates and bladders were still colonized with UPEC at 42 d PI (Supplemental Fig. 1A). Whereas UPEC colonized the prostate and bladder robustly, E. faecalis bacterial burden was lower in the bladder and in line with previous reports of E. faecalis burden in infected female mice (Fig. 1B) (3941). Notably, E. faecalis displayed a more consistent infection and pronounced tropism for the prostate over the bladder at days 1, 2, and 14 PI (Fig. 1B). Finally, as polymicrobial infections are common in humans and may lead to complicated UTI or urosepsis (4246), we measured bacterial burdens in bladders and prostates following coinfection with UPEC and E. faecalis. We instilled a total of 1 × 107 CFU in an approximate 1:1 ratio of each strain and observed that UPEC bacterial burden was significantly higher in prostates compared with the bladder at days 1 and 2 PI (Fig. 1C). In the context of this coinfection, E. faecalis colonized both organs; however, statistically significant differences in bacterial burden were no longer apparent between the bladder and the prostate because of the increased variance in E. faecalis CFU in the bladder at days 1 and 2 PI (Fig. 1D). Despite some increased variance in bacterial burden, there were no significant differences in the ability of UPEC or E. faecalis to colonize the prostate at days 1 and 2 PI in a polymicrobial infection compared with single infections with either of these organisms (Supplemental Fig. 1B). Having established mono- and polymicrobial prostate infection models with two common uropathogens, we used these models to investigate the immune response to prostatitis, focusing on early events following primary and challenge infection given the frequency in which these infections recur or become chronic.

FIGURE 1.

UPEC and E. faecalis display tropism for the prostate over the bladder. Six- to eight-week-old male C57BL/6 mice were infected with 1 × 107 CFU of (A) UPEC strain UTI89-RFP-kanR, (B) E. faecalis strain OG1RF, or (C and D) a total of 1 × 107 CFU of both bacterial strains in a 1:1 ratio. Graphs depict CFU per organ at the indicated day PI. Each circle is one mouse, filled circles represent bladder, and open circles represent prostate. Red lines indicate the median value of each group, and black dotted lines are the LOD of the assay, 20 or 40 CFU/organ. Graphs show paired data (one bladder and one prostate collected from each mouse). Data are pooled from 2 to 10 experiments, with four to seven mice per group in each experiment. CFU in the bladder and prostate were compared with each other at each time point using the nonparametric Wilcoxon test for paired data, and p values were corrected for multiple testing within each infection scenario (UPEC, E. faecalis, or polymicrobial) using the FDR method. The q values meeting the criteria for statistical significance (q < 0.05) are in red.

FIGURE 1.

UPEC and E. faecalis display tropism for the prostate over the bladder. Six- to eight-week-old male C57BL/6 mice were infected with 1 × 107 CFU of (A) UPEC strain UTI89-RFP-kanR, (B) E. faecalis strain OG1RF, or (C and D) a total of 1 × 107 CFU of both bacterial strains in a 1:1 ratio. Graphs depict CFU per organ at the indicated day PI. Each circle is one mouse, filled circles represent bladder, and open circles represent prostate. Red lines indicate the median value of each group, and black dotted lines are the LOD of the assay, 20 or 40 CFU/organ. Graphs show paired data (one bladder and one prostate collected from each mouse). Data are pooled from 2 to 10 experiments, with four to seven mice per group in each experiment. CFU in the bladder and prostate were compared with each other at each time point using the nonparametric Wilcoxon test for paired data, and p values were corrected for multiple testing within each infection scenario (UPEC, E. faecalis, or polymicrobial) using the FDR method. The q values meeting the criteria for statistical significance (q < 0.05) are in red.

Close modal

We next tested whether an inflammatory response is elicited by bacterial colonization of the prostate. As we hypothesized that the innate response to uropathogen infection in the prostate is similar to the nearly absent response observed in the bladder in male C57BL/6 mice, we included analysis of bladders for comparison. We infected male C57BL/6 mice with 1 × 107 CFU of UTI89-RFP-kanR or OG1RF or a 1:1 ratio of both strains and analyzed the expression of 13 cytokines by bead array in prostate and bladder homogenates at 1 and 2 d PI. Unexpectedly, UPEC-infected prostates expressed statistically significantly higher levels of IL-1α, IL-6, CCL2, IL-1β, IL-17A, and TNF-α compared with bladder tissue at 1 and 2 d PI (Fig. 2A, Table I). IL-23 levels were significantly elevated in the prostate over the bladder only at 2 d PI (Table I). Interestingly, IL-6 and IL-1β levels in naive prostate were significantly higher than in naive bladders, although the reasons for this are unclear (Fig. 2A, Table I). When considering the innate response within the prostate, we observed that in UPEC infection, IL-1α, IL-6, CCL2, IL-1β, IL-17A, and TNF-α were all elevated over levels in naive tissue 1 d PI (Fig. 2A, Table II). IL-1α, CCL2, and TNF-α remained significantly elevated compared with naive tissue 2 d PI; however, the overall response was short-lived, as we measured a significant reduction in IL-27, IL-23, IL-6, CCL2, IL-1β, IL-17A, TNF-α, and CSF2 from day 1 to day 2 PI (Fig. 2A, Table II). Finally, despite persistent infection, evidenced by consistent bacteriuria, cytokine levels in UPEC-infected tissues 42 d PI were not different from those measured in naive tissue, with the exception of TNF-α (Supplemental Fig. 1C, 1D).

FIGURE 2.

The cytokine response to prostate infection diverges from that observed in the bladder. Six- to eight-week-old male C57BL/6 mice were infected with 1 × 107 CFU of (A) UPEC strain UTI89-RFP-kanR, (B) E. faecalis strain OG1RF, or (C) a total of 1 × 107 CFU of both bacterial strains in a 1:1 ratio and sacrificed 1 or 2 d PI (DPI). Naive mice were included for baseline determination. Spider plots show the median values of 13 cytokines measured (picograms per milliliters) on a log scale in naive (dotted lines) or infected organs at 24 h (dashed lines) and 48 h PI (solid lines). Prostate values are depicted in orange, and bladder values are in blue. Data are pooled from two to seven experiments, with three to six mice per group in each experiment. Significance was determined using the Mann–Whitney nonparametric test with correction for multiple testing to determine the FDR adjusted p value: q < 0.05. All p and q values are listed in Tables IVI.

FIGURE 2.

The cytokine response to prostate infection diverges from that observed in the bladder. Six- to eight-week-old male C57BL/6 mice were infected with 1 × 107 CFU of (A) UPEC strain UTI89-RFP-kanR, (B) E. faecalis strain OG1RF, or (C) a total of 1 × 107 CFU of both bacterial strains in a 1:1 ratio and sacrificed 1 or 2 d PI (DPI). Naive mice were included for baseline determination. Spider plots show the median values of 13 cytokines measured (picograms per milliliters) on a log scale in naive (dotted lines) or infected organs at 24 h (dashed lines) and 48 h PI (solid lines). Prostate values are depicted in orange, and bladder values are in blue. Data are pooled from two to seven experiments, with three to six mice per group in each experiment. Significance was determined using the Mann–Whitney nonparametric test with correction for multiple testing to determine the FDR adjusted p value: q < 0.05. All p and q values are listed in Tables IVI.

Close modal
Table I.
Statistical testing of analytes defining the innate immune response to UPEC infection in the bladder versus the prostate

Bladder versus Prostate, Naive
Bladder versus Prostate, 1 d PI
Bladder versus Prostate, 2 d PI
Analyteap Valueq Valuep Valueq Valuep Valueq Value
IL-1α 2.33 × 10−1 6.54 × 10−1 1.81 × 1012 1.18 × 1011 6.84 × 103 1.48 × 102 
IL-27 7.40 × 10−1 9.77 × 10−1 3.63 × 10−1 4.29 × 10−1 4.17 × 10−1 4.93 × 10−1 
IL-23 5.84 × 10−1 9.49 × 10−1 5.36 × 10−1 5.36 × 10−1 4.43 × 103 1.44 × 102 
IL-6 1.79 × 105 2.32 × 104 1.32 × 109 4.28 × 109 4.31 × 104 5.60 × 103 
CCL2 5.69 × 10−1 9.49 × 10−1 2.39 × 1010 1.03 × 109 1.50 × 103 7.28 × 103 
IL-1β 7.10 × 103 4.62 × 102 3.87 × 105 1.01 × 104 6.84 × 103 1.48 × 102 
IFN-β 9.12 × 10−1 9.88 × 10−1 2.40 × 10−1 3.47 × 10−1 7.81 × 10−1 7.81 × 10−1 
IL-17A 7.52 × 10−1 9.77 × 10−1 1.15 × 10−2 2.49 × 102 1.13 × 102 2.10 × 102 
IL-10 2.52 × 10−1 6.54 × 10−1 5.19 × 10−1 5.36 × 10−1 1.11 × 10−1 1.80 × 10−1 
TNF-α 8.59 × 10−1 9.88 × 10−1 3.79 × 1013 4.93 × 1012 1.68 × 103 7.28 × 103 
CSF2 1.00 1.00 3.03 × 10−1 3.94 × 10−1 4.17 × 10−1 4.93 × 10−1 
IFN-γ 5.05 × 10−1 9.49 × 10−1 4.27 × 102 7.93 × 10−2 3.19 × 10−1 4.61 × 10−1 
IL-12p70 2.20 × 10−1 6.54 × 10−1 1.52 × 10−1 2.46 × 10−1 4.67 × 10−1 5.06 × 10−1 

Bladder versus Prostate, Naive
Bladder versus Prostate, 1 d PI
Bladder versus Prostate, 2 d PI
Analyteap Valueq Valuep Valueq Valuep Valueq Value
IL-1α 2.33 × 10−1 6.54 × 10−1 1.81 × 1012 1.18 × 1011 6.84 × 103 1.48 × 102 
IL-27 7.40 × 10−1 9.77 × 10−1 3.63 × 10−1 4.29 × 10−1 4.17 × 10−1 4.93 × 10−1 
IL-23 5.84 × 10−1 9.49 × 10−1 5.36 × 10−1 5.36 × 10−1 4.43 × 103 1.44 × 102 
IL-6 1.79 × 105 2.32 × 104 1.32 × 109 4.28 × 109 4.31 × 104 5.60 × 103 
CCL2 5.69 × 10−1 9.49 × 10−1 2.39 × 1010 1.03 × 109 1.50 × 103 7.28 × 103 
IL-1β 7.10 × 103 4.62 × 102 3.87 × 105 1.01 × 104 6.84 × 103 1.48 × 102 
IFN-β 9.12 × 10−1 9.88 × 10−1 2.40 × 10−1 3.47 × 10−1 7.81 × 10−1 7.81 × 10−1 
IL-17A 7.52 × 10−1 9.77 × 10−1 1.15 × 10−2 2.49 × 102 1.13 × 102 2.10 × 102 
IL-10 2.52 × 10−1 6.54 × 10−1 5.19 × 10−1 5.36 × 10−1 1.11 × 10−1 1.80 × 10−1 
TNF-α 8.59 × 10−1 9.88 × 10−1 3.79 × 1013 4.93 × 1012 1.68 × 103 7.28 × 103 
CSF2 1.00 1.00 3.03 × 10−1 3.94 × 10−1 4.17 × 10−1 4.93 × 10−1 
IFN-γ 5.05 × 10−1 9.49 × 10−1 4.27 × 102 7.93 × 10−2 3.19 × 10−1 4.61 × 10−1 
IL-12p70 2.20 × 10−1 6.54 × 10−1 1.52 × 10−1 2.46 × 10−1 4.67 × 10−1 5.06 × 10−1 
a

Values are ordered according to expression level of measured cytokines in UPEC-infected prostates at 1 d PI. The p values were determined by Mann–Whitney nonparametric test and corrected for multiple testing to determine q values (FDR adjusted p value). All p values and q values <0.05 are shown in bold.

Table II.
Statistical testing of analytes defining the innate immune response to UPEC infection in the prostate over time

Naive versus 1 d PI
Naive versus 2 d PI
1 versus 2 d PI
aAnalytep Valueq Valuep Valueq Valuep Valueq Value
IL-1α 1.31 × 10−9 8.50 × 10−9 1.76 × 10−5 2.29 × 10−4 7.88 × 10−1 7.88 × 10−1 
IL-27 4.94 × 10−1 5.36 × 10−1 2.54 × 10−2 8.25 × 10−2 2.90 × 10−3 1.88 × 10−2 
IL-23 9.83 × 10−2 1.65 × 10−1 3.56 × 10−1 4.63 × 10−1 2.83 × 10−2 4.58 × 10−2 
IL-6 4.31 × 10−3 1.12 × 10−2 8.32 × 10−1 9.42 × 10−1 7.87 × 10−3 2.56 × 10−2 
CCL2 1.31 × 10−9 8.50 × 10−9 9.28 × 10−5 4.02 × 10−4 3.02 × 10−2 4.58 × 10−2 
IL-1β 6.39 × 10−4 2.08 × 10−3 8.69 × 10−1 9.42 × 10−1 1.65 × 10−2 3.58 × 10−2 
IFN-β 4.78 × 10−1 5.36 × 10−1 8.96 × 10−2 1.46 × 10−1 1.08 × 10−1 1.28 × 10−1 
IL-17A 1.45 × 10−2 3.14 × 10−2 9.81 × 10−1 9.81 × 10−1 6.72 × 10−3 2.56 × 10−2 
IL-10 9.93 × 10−1 9.93 × 10−1 4.52 × 10−2 1.18 × 10−1 5.46 × 10−2 7.10 × 10−2 
TNF-α 3.90 × 10−9 1.69 × 10−8 8.62 × 10−5 4.02 × 10−4 1.07 × 10−2 2.78 × 10−2 
CSF2 4.84 × 10−1 5.36 × 10−1 8.28 × 10−2 1.46 × 10−1 3.17 × 10−2 4.58 × 10−2 
IFN-γ 1.12 × 10−1 1.65 × 10−1 7.57 × 10−2 1.46 × 10−1 4.48 × 10−1 4.85 × 10−1 
IL-12p70 1.14 × 10−1 1.65 × 10−1 2.66 × 10−1 3.85 × 10−1 2.59 × 10−3 1.88 × 10−2 

Naive versus 1 d PI
Naive versus 2 d PI
1 versus 2 d PI
aAnalytep Valueq Valuep Valueq Valuep Valueq Value
IL-1α 1.31 × 10−9 8.50 × 10−9 1.76 × 10−5 2.29 × 10−4 7.88 × 10−1 7.88 × 10−1 
IL-27 4.94 × 10−1 5.36 × 10−1 2.54 × 10−2 8.25 × 10−2 2.90 × 10−3 1.88 × 10−2 
IL-23 9.83 × 10−2 1.65 × 10−1 3.56 × 10−1 4.63 × 10−1 2.83 × 10−2 4.58 × 10−2 
IL-6 4.31 × 10−3 1.12 × 10−2 8.32 × 10−1 9.42 × 10−1 7.87 × 10−3 2.56 × 10−2 
CCL2 1.31 × 10−9 8.50 × 10−9 9.28 × 10−5 4.02 × 10−4 3.02 × 10−2 4.58 × 10−2 
IL-1β 6.39 × 10−4 2.08 × 10−3 8.69 × 10−1 9.42 × 10−1 1.65 × 10−2 3.58 × 10−2 
IFN-β 4.78 × 10−1 5.36 × 10−1 8.96 × 10−2 1.46 × 10−1 1.08 × 10−1 1.28 × 10−1 
IL-17A 1.45 × 10−2 3.14 × 10−2 9.81 × 10−1 9.81 × 10−1 6.72 × 10−3 2.56 × 10−2 
IL-10 9.93 × 10−1 9.93 × 10−1 4.52 × 10−2 1.18 × 10−1 5.46 × 10−2 7.10 × 10−2 
TNF-α 3.90 × 10−9 1.69 × 10−8 8.62 × 10−5 4.02 × 10−4 1.07 × 10−2 2.78 × 10−2 
CSF2 4.84 × 10−1 5.36 × 10−1 8.28 × 10−2 1.46 × 10−1 3.17 × 10−2 4.58 × 10−2 
IFN-γ 1.12 × 10−1 1.65 × 10−1 7.57 × 10−2 1.46 × 10−1 4.48 × 10−1 4.85 × 10−1 
IL-12p70 1.14 × 10−1 1.65 × 10−1 2.66 × 10−1 3.85 × 10−1 2.59 × 10−3 1.88 × 10−2 
a

Values are ordered according to expression level of measured cytokines in UPEC-infected prostates at 1 d PI. The p values were determined by Mann–Whitney nonparametric test and corrected for multiple testing to determine q values (FDR adjusted p value). All p values and q values <0.05 are shown in bold.

In E. faecalis–infected mice, IL-1α, IL-6, CCL2, IL-1β, and TNF-α were expressed at levels significantly higher in the prostate than in the bladder at 1 d PI, and IL-1α, IL-6, CCL2, and IL-1β remained elevated at 2 d PI (Fig. 2B, Table III). Notably, no cytokines were significantly elevated over naive tissue levels in the prostate at 1 d PI with E. faecalis after correction for multiple testing (Fig. 2B, Table IV). At 2 d PI with E. faecalis, IL-1α, CCL2, and IFN-γ were elevated over naive tissue levels, and similar to that seen in UPEC infection, the response contracted from 1 to 2 d PI, with a reduction in IL-23 and IL-6. By contrast, TNF-α and IFN-γ were expressed at significantly higher levels at 2 d PI compared with day 1 PI (Fig. 2B, Table IV). Globally, cytokine levels induced after E. faecalis infection were lower than those measured in UPEC-infected bladders and prostates, closely mirroring that measured in naive tissue (Fig. 2A, 2B).

Table III.
Statistical testing of analytes defining the innate immune response to E. faecalis infection in the bladder versus the prostate

Bladder versus Prostate, 1 d PI
Bladder versus Prostate, 2 d PI
aAnalytep Valueq Valuep Valueq Value
IL-1α 3.73 × 10−3 1.21 × 10−2 4.82 × 10−4 2.09 × 10−3 
IL-27 9.25 × 10−2 1.72 × 10−1 9.77 × 10−1 9.77 × 10−1 
IL-23 2.32 × 10−1 3.36 × 10−1 3.78 × 10−1 5.46 × 10−1 
IL-6 9.24 × 10−12 1.20 × 10−10 3.60 × 10−5 3.22 × 10−4 
CCL2 2.54 × 10−3 1.12 × 10−2 4.96 × 10−5 3.22 × 10−4 
IL-1β 2.58 × 10−3 1.12 × 10−2 1 × 10−2 3.26 × 10−2 
IFN-β 7.13 × 10−1 7.13 × 10−1 2.66 × 10−1 4.32 × 10−1 
IL-17A 4.88 × 10−1 5.76 × 10−1 8.43 × 10−1 9.13 × 10−1 
IL-10 1.60 × 10−1 2.60 × 10−1 7.99 × 10−1 9.13 × 10−1 
TNF-α 1.08 × 10−2 2.78 × 10−2 2.19 × 10−1 4.07 × 10−1 
CSF2 5.86 × 10−1 6.35 × 10−1 5.51 × 10−1 7.17 × 10−1 
IFN-γ 3.05 × 10−1 3.97 × 10−1 9.40 × 10−2 2.44 × 10−1 
IL-12p70 1.28 × 10−2 2.78 × 10−2 1.81 × 10−1 3.91 × 10−1 

Bladder versus Prostate, 1 d PI
Bladder versus Prostate, 2 d PI
aAnalytep Valueq Valuep Valueq Value
IL-1α 3.73 × 10−3 1.21 × 10−2 4.82 × 10−4 2.09 × 10−3 
IL-27 9.25 × 10−2 1.72 × 10−1 9.77 × 10−1 9.77 × 10−1 
IL-23 2.32 × 10−1 3.36 × 10−1 3.78 × 10−1 5.46 × 10−1 
IL-6 9.24 × 10−12 1.20 × 10−10 3.60 × 10−5 3.22 × 10−4 
CCL2 2.54 × 10−3 1.12 × 10−2 4.96 × 10−5 3.22 × 10−4 
IL-1β 2.58 × 10−3 1.12 × 10−2 1 × 10−2 3.26 × 10−2 
IFN-β 7.13 × 10−1 7.13 × 10−1 2.66 × 10−1 4.32 × 10−1 
IL-17A 4.88 × 10−1 5.76 × 10−1 8.43 × 10−1 9.13 × 10−1 
IL-10 1.60 × 10−1 2.60 × 10−1 7.99 × 10−1 9.13 × 10−1 
TNF-α 1.08 × 10−2 2.78 × 10−2 2.19 × 10−1 4.07 × 10−1 
CSF2 5.86 × 10−1 6.35 × 10−1 5.51 × 10−1 7.17 × 10−1 
IFN-γ 3.05 × 10−1 3.97 × 10−1 9.40 × 10−2 2.44 × 10−1 
IL-12p70 1.28 × 10−2 2.78 × 10−2 1.81 × 10−1 3.91 × 10−1 
a

Values are ordered according to expression level of measured cytokines in UPEC-infected prostates at 1 d PI. The p values were determined by Mann–Whitney nonparametric test and corrected for multiple testing to determine q values (FDR adjusted p value). All p values and q values <0.05 are shown in bold.

Table IV.
Statistical testing of analytes defining the innate immune response to E. faecalis infection in the prostate over time

Naive versus 1 d PI
Naive versus 2 d PI
1 versus 2 d PI
aAnalytep Valueq Valuep Valueq Valuep Valueq Value
IL-1α 3.29 × 10−1 6.31 × 10−1 2.28 × 10−3 1.48 × 10−2 4.90 × 10−2 9.11 × 10−2 
IL-27 9.70 × 10−1 9.70 × 10−1 1.00 1.00 7.32 × 10−1 7.62 × 10−1 
IL-23 1.09 × 10−2 6.88 × 10−2 4.65 × 10−1 8.63 × 10−1 1.06 × 10−2 3.43 × 10−2 
IL-6 1.59 × 10−2 6.88 × 10−2 7.87 × 10−1 9.68 × 10−1 1.74 × 10−2 4.52 × 10−2 
CCL2 1.31 × 10−2 6.88 × 10−2 6.53 × 10−4 8.49 × 10−3 8.32 × 10−2 1.35 × 10−1 
IL-1β 4.37 × 10−1 6.31 × 10−1 8.19 × 10−1 9.68 × 10−1 3.08 × 10−1 4.45 × 10−1 
IFN-β 2.90 × 10−1 6.31 × 10−1 6.57 × 10−1 9.48 × 10−1 4.90 × 10−2 9.11 × 10−2 
IL-17A 7.21 × 10−1 9.13 × 10−1 3.85 × 10−1 8.35 × 10−1 4.68 × 10−1 6.08 × 10−1 
IL-10 3.98 × 10−1 6.31 × 10−1 6.32 × 10−1 9.48 × 10−1 5.82 × 10−1 6.87 × 10−1 
TNF-α 3.67 × 10−1 6.31 × 10−1 1.87 × 10−2 6.09 × 10−2 8.00 × 10−3 3.43 × 10−2 
CSF2 9.27 × 10−1 9.70 × 10−1 9.49 × 10−1 1.00 7.62 × 10−1 7.62 × 10−1 
IFN-γ 7.72 × 10−1 9.13 × 10−1 3.95 × 10−3 1.71 × 10−2 9.71 × 10−4 6.31 × 10−3 
IL-12p70 3.58 × 10−2 1.16 × 10−1 2.94 × 10−1 7.63 × 10−1 1.02 × 10−4 1.33 × 10−3 

Naive versus 1 d PI
Naive versus 2 d PI
1 versus 2 d PI
aAnalytep Valueq Valuep Valueq Valuep Valueq Value
IL-1α 3.29 × 10−1 6.31 × 10−1 2.28 × 10−3 1.48 × 10−2 4.90 × 10−2 9.11 × 10−2 
IL-27 9.70 × 10−1 9.70 × 10−1 1.00 1.00 7.32 × 10−1 7.62 × 10−1 
IL-23 1.09 × 10−2 6.88 × 10−2 4.65 × 10−1 8.63 × 10−1 1.06 × 10−2 3.43 × 10−2 
IL-6 1.59 × 10−2 6.88 × 10−2 7.87 × 10−1 9.68 × 10−1 1.74 × 10−2 4.52 × 10−2 
CCL2 1.31 × 10−2 6.88 × 10−2 6.53 × 10−4 8.49 × 10−3 8.32 × 10−2 1.35 × 10−1 
IL-1β 4.37 × 10−1 6.31 × 10−1 8.19 × 10−1 9.68 × 10−1 3.08 × 10−1 4.45 × 10−1 
IFN-β 2.90 × 10−1 6.31 × 10−1 6.57 × 10−1 9.48 × 10−1 4.90 × 10−2 9.11 × 10−2 
IL-17A 7.21 × 10−1 9.13 × 10−1 3.85 × 10−1 8.35 × 10−1 4.68 × 10−1 6.08 × 10−1 
IL-10 3.98 × 10−1 6.31 × 10−1 6.32 × 10−1 9.48 × 10−1 5.82 × 10−1 6.87 × 10−1 
TNF-α 3.67 × 10−1 6.31 × 10−1 1.87 × 10−2 6.09 × 10−2 8.00 × 10−3 3.43 × 10−2 
CSF2 9.27 × 10−1 9.70 × 10−1 9.49 × 10−1 1.00 7.62 × 10−1 7.62 × 10−1 
IFN-γ 7.72 × 10−1 9.13 × 10−1 3.95 × 10−3 1.71 × 10−2 9.71 × 10−4 6.31 × 10−3 
IL-12p70 3.58 × 10−2 1.16 × 10−1 2.94 × 10−1 7.63 × 10−1 1.02 × 10−4 1.33 × 10−3 
a

Values are ordered according to expression level of measured cytokines in UPEC-infected prostates at 1 d PI. The p values were determined by Mann–Whitney nonparametric test and corrected for multiple testing to determine q values (FDR adjusted p value). All p values and q values <0.05 are shown in bold.

In the coinfection scenario, more cytokines were significantly elevated in the prostate compared with the bladder than in either monomicrobial infection at day 1 PI, including IL-1α, IL-27, IL-6, CCL2, IL-1β, IL-17A, IL-10, TNF-α, CSF2, and IFN-γ (Fig. 2C, Table V). IL-1α, IL-6, IL-1β, and TNF-α levels in the prostate remained elevated over bladder-associated levels at 2 d PI (Fig. 2C, Table V). With the exception of IL-27 and IFN-β, all cytokines measured were significantly increased at 1 d PI compared with naive levels (Fig. 2C, Table VI). Although IL-1α, IL-6, CCL2, IL-1β, IL-17A, and TNF-α remained elevated compared with naive tissue at 2 d PI, no cytokines were significantly different between 1 and 2 d PI in the polymicrobial prostate infection when the two organisms colonize the prostate similarly to single infections with either uropathogen (Fig. 2C, Table VI). Finally, although IL-12p70 levels were statistically significantly different in several scenarios, such as between day 1 and day 2 PI in UPEC and E. faecalis infection, expression levels were near or at the LOD in all infection scenarios.

Table V.
Statistical testing of analytes defining the innate immune response to polymicrobial infection in the bladder versus the prostate

Bladder versus Prostate, 1 d PI
Bladder versus Prostate, 2 d PI
aAnalytep Valueq Valuep Valueq Value
IL-1α 4.62 × 10−12 3.00 × 10−11 1.55 × 10−4 2.02 × 10−3 
IL-27 3.11 × 10−3 4.49 × 10−3 1.00 1.00 
IL-23 4.20 × 10−1 4.20 × 10−1 6.50 × 10−2 1.41 × 10−1 
IL-6 1.07 × 10−6 3.49 × 10−6 1.33 × 10−2 4.33 × 10−2 
CCL2 2.08 × 10−6 5.41 × 10−6 4.99 × 10−2 1.30 × 10−1 
IL-1β 4.86 × 10−13 6.32 × 10−12 6.99 × 10−3 3.03 × 10−2 
IFN-β 4.97 × 10−2 5.87 × 10−2 7.01 × 10−1 9.12 × 10−1 
IL-17A 2.80 × 10−5 6.07 × 10−5 4.42 × 10−1 7.18 × 10−1 
IL-10 8.73 × 10−4 1.42 × 10−3 5.05 × 10−1 7.30 × 10−1 
TNF-α 6.39 × 10−9 2.77 × 10−8 2.68 × 10−3 1.74 × 10−2 
CSF2 2.57 × 10−2 3.35 × 10−2 9.59 × 10−1 1.00 
IFN-γ 6.86 × 10−5 1.27 × 10−4 1.00 1.00 
IL-12p70 8.68 × 10−2 9.40 × 10−2 8.47 × 10−2 1.57 × 10−1 

Bladder versus Prostate, 1 d PI
Bladder versus Prostate, 2 d PI
aAnalytep Valueq Valuep Valueq Value
IL-1α 4.62 × 10−12 3.00 × 10−11 1.55 × 10−4 2.02 × 10−3 
IL-27 3.11 × 10−3 4.49 × 10−3 1.00 1.00 
IL-23 4.20 × 10−1 4.20 × 10−1 6.50 × 10−2 1.41 × 10−1 
IL-6 1.07 × 10−6 3.49 × 10−6 1.33 × 10−2 4.33 × 10−2 
CCL2 2.08 × 10−6 5.41 × 10−6 4.99 × 10−2 1.30 × 10−1 
IL-1β 4.86 × 10−13 6.32 × 10−12 6.99 × 10−3 3.03 × 10−2 
IFN-β 4.97 × 10−2 5.87 × 10−2 7.01 × 10−1 9.12 × 10−1 
IL-17A 2.80 × 10−5 6.07 × 10−5 4.42 × 10−1 7.18 × 10−1 
IL-10 8.73 × 10−4 1.42 × 10−3 5.05 × 10−1 7.30 × 10−1 
TNF-α 6.39 × 10−9 2.77 × 10−8 2.68 × 10−3 1.74 × 10−2 
CSF2 2.57 × 10−2 3.35 × 10−2 9.59 × 10−1 1.00 
IFN-γ 6.86 × 10−5 1.27 × 10−4 1.00 1.00 
IL-12p70 8.68 × 10−2 9.40 × 10−2 8.47 × 10−2 1.57 × 10−1 
a

Values are ordered according to expression level of measured cytokines in UPEC-infected prostates at 1 d PI. The p values were determined by Mann–Whitney nonparametric test and corrected for multiple testing to determine q values (FDR adjusted p value). All p values and q values <0.05 are shown in bold.

Table VI.
Statistical testing of analytes defining the innate immune response to polymicrobial infection in the prostate over time

Naive versus 1 d PI
Naive versus 2 d PI
1 versus 2 d PI
aAnalytep Valueq Valuep Valueq Valuep Valueq Value
IL-1α 5.77 × 10−8 1.50 × 10−7 7.04 × 10−5 2.23 × 10−4 7.07 × 10−1 8.02 × 10−1 
IL-27 1.28 × 10−1 1.38 × 10−1 6.77 × 10−1 7.33 × 10−1 3.81 × 10−2 9.91 × 10−2 
IL-23 1.76 × 10−4 3.27 × 10−4 4.70 × 10−2 7.64 × 10−2 7.40 × 10−1 8.02 × 10−1 
IL-6 6.63 × 10−10 4.31 × 10−9 8.96 × 10−6 1.16 × 10−4 2.75 × 10−1 3.97 × 10−1 
CCL2 5.79 × 10−8 1.50 × 10−7 7.08 × 10−5 2.23 × 10−4 2.32 × 10−2 7.54 × 10−2 
IL-1β 6.93 × 10−11 9.00 × 10−10 8.58 × 10−5 2.23 × 10−4 1.78 × 10−2 7.54 × 10−2 
IFN-β 7.33 × 10−1 7.33 × 10−1 3.03 × 10−1 3.94 × 10−1 5.20 × 10−2 1.13 × 10−1 
IL-17A 2.12 × 10−6 4.58 × 10−6 3.53 × 10−3 7.65 × 10−3 1.21 × 10−1 1.97 × 10−1 
IL-10 4.04 × 10−2 4.78 × 10−2 1.44 × 10−1 2.07 × 10−1 5.42 × 10−1 7.05 × 10−1 
TNF-α 5.54 × 10−8 1.50 × 10−7 8.14 × 10−5 2.23 × 10−4 1.17 × 10−2 7.54 × 10−2 
CSF2 4.01 × 10−3 5.22 × 10−3 4.20 × 10−1 4.97 × 10−1 9.47 × 10−2 1.76 × 10−1 
IFN-γ 4.01 × 10−3 5.22 × 10−3 8.24 × 10−1 8.24 × 10−1 2.26 × 10−2 7.54 × 10−2 
IL-12p70 4.12 × 10−4 6.69 × 10−4 3.42 × 10−2 6.35 × 10−2 8.60 × 10−1 8.60 × 10−1 

Naive versus 1 d PI
Naive versus 2 d PI
1 versus 2 d PI
aAnalytep Valueq Valuep Valueq Valuep Valueq Value
IL-1α 5.77 × 10−8 1.50 × 10−7 7.04 × 10−5 2.23 × 10−4 7.07 × 10−1 8.02 × 10−1 
IL-27 1.28 × 10−1 1.38 × 10−1 6.77 × 10−1 7.33 × 10−1 3.81 × 10−2 9.91 × 10−2 
IL-23 1.76 × 10−4 3.27 × 10−4 4.70 × 10−2 7.64 × 10−2 7.40 × 10−1 8.02 × 10−1 
IL-6 6.63 × 10−10 4.31 × 10−9 8.96 × 10−6 1.16 × 10−4 2.75 × 10−1 3.97 × 10−1 
CCL2 5.79 × 10−8 1.50 × 10−7 7.08 × 10−5 2.23 × 10−4 2.32 × 10−2 7.54 × 10−2 
IL-1β 6.93 × 10−11 9.00 × 10−10 8.58 × 10−5 2.23 × 10−4 1.78 × 10−2 7.54 × 10−2 
IFN-β 7.33 × 10−1 7.33 × 10−1 3.03 × 10−1 3.94 × 10−1 5.20 × 10−2 1.13 × 10−1 
IL-17A 2.12 × 10−6 4.58 × 10−6 3.53 × 10−3 7.65 × 10−3 1.21 × 10−1 1.97 × 10−1 
IL-10 4.04 × 10−2 4.78 × 10−2 1.44 × 10−1 2.07 × 10−1 5.42 × 10−1 7.05 × 10−1 
TNF-α 5.54 × 10−8 1.50 × 10−7 8.14 × 10−5 2.23 × 10−4 1.17 × 10−2 7.54 × 10−2 
CSF2 4.01 × 10−3 5.22 × 10−3 4.20 × 10−1 4.97 × 10−1 9.47 × 10−2 1.76 × 10−1 
IFN-γ 4.01 × 10−3 5.22 × 10−3 8.24 × 10−1 8.24 × 10−1 2.26 × 10−2 7.54 × 10−2 
IL-12p70 4.12 × 10−4 6.69 × 10−4 3.42 × 10−2 6.35 × 10−2 8.60 × 10−1 8.60 × 10−1 
a

Values are ordered according to expression level of measured cytokines in UPEC-infected prostates at 1 d PI. The p values were determined by Mann–Whitney nonparametric test and corrected for multiple testing to determine q values (FDR adjusted p value). All p values and q values <0.05 are shown in bold.

Notably, in UPEC infection, although prostates remained colonized up to 42 d, cytokine levels quickly returned to naive levels, suggesting that persistent infection might render animals refractory to induction of inflammatory mediators in the case of a second or challenge infection. To test this hypothesis, we measured cytokine concentrations in the supernatant from tissue homogenates of challenged prostates. We infected cohorts of male C57BL/6 mice with 1 × 107 CFU of UPEC or OG1RF or a 1:1 ratio of both strains and monitored bacterial clearance by culturing urine samples every 2–5 d over the course of 1 mo. Following antibiotic treatment to resolve infection and a short washout period, mice with sterile urine were challenged with a second infection. Mice that received UPEC or E. faecalis were challenged with an isogenic strain of the same species used for primary infection. Animals receiving a primary polymicrobial infection were monomicrobially challenged with either UPEC or E. faecalis. Cytokine concentrations of IL-1α, IL-17A, TNF-α, and IFN-γ were significantly elevated 1 d PI in UPEC-challenged prostates that had previously been exposed to a primary UPEC infection, whereas IL-27 and IFN-β were significantly decreased following challenge infection (Fig. 3A, Table VII). By contrast, E. faecalis challenge infection induced increased expression only of IL-1α and CCL2 over levels observed following primary infection (Fig. 3B, Table VII). IL-12p70 also was significantly different between primary and challenge infection with UPEC or E. faecalis; however, as in primary infections, expression levels of this cytokine were very low. Surprisingly, total cytokine expression was significantly reduced only following challenge with either UPEC or E. faecalis compared with the response induced by primary polymicrobial infection (Fig. 3C, Table VIII). Indeed, IL-1α, IL-6, CCL2, IL-1β, IL-17A, IL-10, TNF-α, and CSF2 were all significantly reduced after UPEC infection, and all cytokines measured except for IL-23 were reduced following E. faecalis infection (Fig. 3C, Table VIII). Notably, we did not observe this pattern in the bladder of these same animals (Supplemental Fig. 2).

FIGURE 3.

The cytokine response is largely unchanged following challenge infection. Six- to eight-week-old male C57BL/6 mice were infected with 1 × 107 CFU of (A) UPEC strain UTI89-RFP-kanR, (B) E. faecalis strain OG1RF, or (C) a total of 1 × 107 CFU of both bacterial strains in a 1:1 ratio. At 25–30 d PI (DPI), all animals were treated with one to two cycles of antibiotics, as described in 2Materials and Methods for 5 d, followed by a 3–5-d washout period. Mice with sterile urine were challenged with 1 × 107 CFU of (A) UPEC strain UTI89-GFP-ampR, (B) E. faecalis strain OG1RF_intergenicRS00490RS00495::Tn, or (C) in primary polymicrobially infected mice; half of the cohort was challenged with UTI89-GFP-ampR, and the other half was challenged with OG1RF_intergenicRS00490RS00495::Tn. Spider plots show the median values of 13 cytokines measured (picograms per milliliter) on a log scale at 24 h after primary (dotted lines) or challenge infection (solid lines). Data are pooled from two to seven experiments, with two to six mice per group in each experiment. (A and B) Values for prostate cytokine expression in naive and primary infected prostate are replotted (dotted lines) from Fig. 2 for ease in comparing primary to challenge infection. Significance was determined using the Mann–Whitney nonparametric test with correction for multiple testing to calculate the FDR adjusted p value: q < 0.05. All p and q values are listed in Tables VII and VIII.

FIGURE 3.

The cytokine response is largely unchanged following challenge infection. Six- to eight-week-old male C57BL/6 mice were infected with 1 × 107 CFU of (A) UPEC strain UTI89-RFP-kanR, (B) E. faecalis strain OG1RF, or (C) a total of 1 × 107 CFU of both bacterial strains in a 1:1 ratio. At 25–30 d PI (DPI), all animals were treated with one to two cycles of antibiotics, as described in 2Materials and Methods for 5 d, followed by a 3–5-d washout period. Mice with sterile urine were challenged with 1 × 107 CFU of (A) UPEC strain UTI89-GFP-ampR, (B) E. faecalis strain OG1RF_intergenicRS00490RS00495::Tn, or (C) in primary polymicrobially infected mice; half of the cohort was challenged with UTI89-GFP-ampR, and the other half was challenged with OG1RF_intergenicRS00490RS00495::Tn. Spider plots show the median values of 13 cytokines measured (picograms per milliliter) on a log scale at 24 h after primary (dotted lines) or challenge infection (solid lines). Data are pooled from two to seven experiments, with two to six mice per group in each experiment. (A and B) Values for prostate cytokine expression in naive and primary infected prostate are replotted (dotted lines) from Fig. 2 for ease in comparing primary to challenge infection. Significance was determined using the Mann–Whitney nonparametric test with correction for multiple testing to calculate the FDR adjusted p value: q < 0.05. All p and q values are listed in Tables VII and VIII.

Close modal
Table VII.
Statistical testing of analytes expressed during monomicrobial primary or challenge infection with UPEC or E. faecalis

Primary versus Challenge, UPEC
Primary versus Challenge, E. faecalis
aAnalytep Valueq Valuep Valueq Value
IL-1α 3.47 × 10−4 3.83 × 10−3 8.69 × 10−4 5.65 × 10−3 
IL-27 5.54 × 10−3 1.44 × 10−2 2.44 × 10−1 3.53 × 10−1 
IL-23 5.78 × 10−1 7.51 × 10−1 7.03 × 10−1 8.31 × 10−1 
IL-6 1.35 × 10−1 1.96 × 10−1 7.68 × 10−1 8.32 × 10−1 
CCL2 9.75 × 10−1 9.75 × 10−1 6.46 × 10−3 2.80 × 10−2 
IL-1β 3.32 × 10−2 5.39 × 10−2 1.73 × 10−1 2.81 × 10−1 
IFN-β 8.54 × 10−3 1.59 × 10−2 2.78 × 10−2 9.04 × 10−2 
IL-17A 6.94 × 10−3 1.50 × 10−2 1.23 × 10−1 2.29 × 10−1 
IL-10 7.93 × 10−1 8.59 × 10−1 9.74 × 10−1 9.74 × 10−1 
TNF-α 8.84 × 10−4 3.83 × 10−3 4.98 × 10−2 1.08 × 10−1 
CSF2 7.66 × 10−1 8.59 × 10−1 5.12 × 10−1 6.65 × 10−1 
IFN-γ 3.03 × 10−3 9.84 × 10−3 4.91 × 10−2 1.08 × 10−1 
IL-12p70 7.96 × 10−4 3.83 × 10−3 5.20 × 10−4 5.65 × 10−3 

Primary versus Challenge, UPEC
Primary versus Challenge, E. faecalis
aAnalytep Valueq Valuep Valueq Value
IL-1α 3.47 × 10−4 3.83 × 10−3 8.69 × 10−4 5.65 × 10−3 
IL-27 5.54 × 10−3 1.44 × 10−2 2.44 × 10−1 3.53 × 10−1 
IL-23 5.78 × 10−1 7.51 × 10−1 7.03 × 10−1 8.31 × 10−1 
IL-6 1.35 × 10−1 1.96 × 10−1 7.68 × 10−1 8.32 × 10−1 
CCL2 9.75 × 10−1 9.75 × 10−1 6.46 × 10−3 2.80 × 10−2 
IL-1β 3.32 × 10−2 5.39 × 10−2 1.73 × 10−1 2.81 × 10−1 
IFN-β 8.54 × 10−3 1.59 × 10−2 2.78 × 10−2 9.04 × 10−2 
IL-17A 6.94 × 10−3 1.50 × 10−2 1.23 × 10−1 2.29 × 10−1 
IL-10 7.93 × 10−1 8.59 × 10−1 9.74 × 10−1 9.74 × 10−1 
TNF-α 8.84 × 10−4 3.83 × 10−3 4.98 × 10−2 1.08 × 10−1 
CSF2 7.66 × 10−1 8.59 × 10−1 5.12 × 10−1 6.65 × 10−1 
IFN-γ 3.03 × 10−3 9.84 × 10−3 4.91 × 10−2 1.08 × 10−1 
IL-12p70 7.96 × 10−4 3.83 × 10−3 5.20 × 10−4 5.65 × 10−3 
a

Values are ordered according to expression level of measured cytokines in UPEC-infected prostates at 1 d PI. The p values were determined by Mann–Whitney nonparametric test and corrected for multiple testing to determine q values (FDR adjusted p value). All p values and q values <0.05 are shown in bold.

Table VIII.
Statistical testing of analytes expressed during primary polymicrobial infection or challenge infection with UPEC or E. faecalis

Polymicrobial Primary versus UPEC Challenge
Polymicrobial Primary versus E. faecalis Challenge
aAnalytep Valueq Valuep Valueq Value
IL-1α 6.64 × 10−5 2.88 × 10−4 2.10 × 10−10 9.10 × 10−10 
IL-27 1.52 × 10−1 1.62 × 10−1 3.04 × 10−2 3.59 × 10−2 
IL-23 8.79 × 10−2 1.04 × 10−1 9.73 × 10−2 9.73 × 10−2 
IL-6 2.71 × 10−4 7.04 × 10−4 4.44 × 10−5 9.62 × 10−5 
CCL2 2.33 × 10−5 1.51 × 10−4 4.38 × 10−13 5.69 × 10−12 
IL-1β 3.96 × 10−6 5.15 × 10−5 2.21 × 10−12 1.44 × 10−11 
IFN-β 1.62 × 10−1 1.62 × 10−1 4.24 × 10−2 4.60 × 10−2 
IL-17A 2.10 × 10−3 3.90 × 10−3 3.32 × 10−7 8.63 × 10−7 
IL-10 2.13 × 10−2 3.08 × 10−2 1.38 × 10−4 2.57 × 10−4 
TNF-α 1.23 × 10−4 3.99 × 10−4 2.87 × 10−10 9.34 × 10−10 
CSF2 8.35 × 10−3 1.36 × 10−2 2.44 × 10−4 3.96 × 10−4 
IFN-γ 4.54 × 10−2 5.90 × 10−2 4.02 × 10−4 5.80 × 10−4 
IL-12p70 5.60 × 10−4 1.21 × 10−3 2.75 × 10−3 3.58 × 10−3 

Polymicrobial Primary versus UPEC Challenge
Polymicrobial Primary versus E. faecalis Challenge
aAnalytep Valueq Valuep Valueq Value
IL-1α 6.64 × 10−5 2.88 × 10−4 2.10 × 10−10 9.10 × 10−10 
IL-27 1.52 × 10−1 1.62 × 10−1 3.04 × 10−2 3.59 × 10−2 
IL-23 8.79 × 10−2 1.04 × 10−1 9.73 × 10−2 9.73 × 10−2 
IL-6 2.71 × 10−4 7.04 × 10−4 4.44 × 10−5 9.62 × 10−5 
CCL2 2.33 × 10−5 1.51 × 10−4 4.38 × 10−13 5.69 × 10−12 
IL-1β 3.96 × 10−6 5.15 × 10−5 2.21 × 10−12 1.44 × 10−11 
IFN-β 1.62 × 10−1 1.62 × 10−1 4.24 × 10−2 4.60 × 10−2 
IL-17A 2.10 × 10−3 3.90 × 10−3 3.32 × 10−7 8.63 × 10−7 
IL-10 2.13 × 10−2 3.08 × 10−2 1.38 × 10−4 2.57 × 10−4 
TNF-α 1.23 × 10−4 3.99 × 10−4 2.87 × 10−10 9.34 × 10−10 
CSF2 8.35 × 10−3 1.36 × 10−2 2.44 × 10−4 3.96 × 10−4 
IFN-γ 4.54 × 10−2 5.90 × 10−2 4.02 × 10−4 5.80 × 10−4 
IL-12p70 5.60 × 10−4 1.21 × 10−3 2.75 × 10−3 3.58 × 10−3 
a

Values are ordered according to expression level of measured cytokines in UPEC-infected prostates at 1 d PI. The p values were determined by Mann–Whitney nonparametric test and corrected for multiple testing to determine q values (FDR adjusted p value). All p values and q values <0.05 are shown in bold.

Infection induced a stronger cytokine response in the prostate than predicted, suggesting that a robust cellular response might develop in response to prostatitis. To assess immune cell infiltration following bacterial infection, we instilled UPEC or E. faecalis into male mice and analyzed digested prostate tissue by flow cytometry at 24 h PI (gating strategy in Supplemental Fig. 3). Following infection with either strain, the total number of cells in the prostate was unchanged compared with naive animals (Fig. 4A, compare orange to black dots). CD45+ immune cells were significantly higher only in UPEC-infected prostates, and not in E. faecalis–infected tissue, compared with naive prostates (Fig. 4B). In contrast to UPEC infection of the bladder (30), eosinophils only modestly infiltrated the prostate in primary UPEC infection and were not different from the naive state in E. faecalis infection (Fig. 4C). Similar to the female UTI model, UPEC infection induced robust neutrophil and monocyte-derived cell infiltration, which was significantly elevated in the infected prostate above the numbers of these cells in naive tissue (Fig. 4D). Monocyte-derived cells were both MHC II and MHC II+, likely representing recently infiltrated cells (MHC II) and those undergoing maturation to macrophages (MHC II+) (29). In E. faecalis infection, no populations were significantly elevated over naive levels after correction for multiple testing. Finally, APCs, such as resident macrophages, CD11b+ dendritic cells (DCs), and CD103+ DCs, were not different from naive levels at 24 h PI in either infection (Fig. 4E).

FIGURE 4.

UPEC infection induces a greater immune cell infiltration in the prostate than E. faecalis. Six- to eight-week-old male C57BL/6 mice were infected with 1 × 107 CFU of either red fluorescent UPEC strain UTI89-RFP-kanR or E. faecalis OG1RF stained with wheat germ agglutinin conjugated to Alexa Fluor 594. At 24 h after primary or after challenge infection, mice were sacrificed, and prostates were analyzed by flow cytometry. Naive mice were included for baseline determination. Supplemental Fig. 3 depicts the gating strategy used for these experiments. Graphs depict (A) total prostate cells, (B) total CD45+ immune cells, and (CE) total specified immune cell populations. Data are pooled from two to three experiments, with five to seven mice per group in each experiment. Each dot represents one mouse, lines are medians, black dots depict naive mice, orange dots denote values from primary infection, and green dots represent data from challenge infection. Statistical significance was determined using the nonparametric Kruskal–Wallis test, and p values were corrected for multiple testing using the FDR method. Statistically significant q values (q < 0.05) are in red. MD, monocyte-derived.

FIGURE 4.

UPEC infection induces a greater immune cell infiltration in the prostate than E. faecalis. Six- to eight-week-old male C57BL/6 mice were infected with 1 × 107 CFU of either red fluorescent UPEC strain UTI89-RFP-kanR or E. faecalis OG1RF stained with wheat germ agglutinin conjugated to Alexa Fluor 594. At 24 h after primary or after challenge infection, mice were sacrificed, and prostates were analyzed by flow cytometry. Naive mice were included for baseline determination. Supplemental Fig. 3 depicts the gating strategy used for these experiments. Graphs depict (A) total prostate cells, (B) total CD45+ immune cells, and (CE) total specified immune cell populations. Data are pooled from two to three experiments, with five to seven mice per group in each experiment. Each dot represents one mouse, lines are medians, black dots depict naive mice, orange dots denote values from primary infection, and green dots represent data from challenge infection. Statistical significance was determined using the nonparametric Kruskal–Wallis test, and p values were corrected for multiple testing using the FDR method. Statistically significant q values (q < 0.05) are in red. MD, monocyte-derived.

Close modal

Given that cytokine responses were similar or elevated in response to challenge infection in monomicrobial infections, we measured immune cell infiltration into the prostate 24 h after challenge infection. We infected mice with 1 × 107 CFU of UPEC or OG1RF and monitored infection via urine collection. One month PI following antibiotic treatment and a washout period, mice with sterile urine were challenged with isogenic strains of UPEC or E. faecalis. Prostates were collected, digested, and stained for flow cytometric analysis 24 h PI as above. Whereas the total numbers of prostate cells and total CD45+ cells were not different between UPEC primary and challenge infection, E. faecalis–challenged prostates had more total cells and greater CD45+ cell infiltration compared with the primary infection (Fig. 4A, 4B, compare orange to green dots for each bacterial strain). Eosinophil infiltration was not different in comparison with primary infection with either strain, and given how little this population changed overall, these cells may not play an important role in the response to prostatitis (Fig. 4C). In UPEC infection, although neutrophils and both subsets of monocyte-derived macrophages were elevated in a primary infection compared with naive levels, these populations were not further increased in UPEC challenge compared with the primary infection (Fig. 4D). Conversely, neutrophil and MHC II+ monocyte-derived macrophage infiltration was significantly elevated in E. faecalis–challenged animals compared with the numbers observed 24 h after primary infection (Fig. 4D). Resident macrophages and CD103+ DC populations increased in mice challenged with UPEC or E. faecalis, whereas CD11b+ DCs increased only in E. faecalis–challenged mice (Fig. 4E). Overall, these results demonstrate that in contrast to that observed in male mouse bladders (30), myeloid cell recruitment is robust in response to both a primary and challenge UPEC. By contrast, a primary E. faecalis infection is largely silent, and challenge infection by this Gram-positive organism is required for robust myeloid cell infiltration.

Dependent upon the infecting organism, significant innate immune cell infiltration was evident after primary or challenge infection. We next determined whether recruited phagocytes efficiently took up bacteria during primary or challenge infection. In infected organs, UTI89-RFP-kanR is easily detectable by flow cytometry (29, 30); however, we did not have a red fluorescent strain of E. faecalis. In testing a GFP-expressing E. faecalis strain, we encountered challenges similar to those with using GFP-expressing UPEC in the bladder in that tissue autofluorescence masked the GFP signal (Supplemental Fig. 4). Therefore, we stained E. faecalis bacterial cell walls with wheat germ agglutinin conjugated to Alexa Fluor 594 to visualize immune cells that had taken up E. faecalis. Overall, in primary and challenge infection, we observed greater numbers of UPEC-containing immune cells in the prostate in comparison with the number of E. faecalis–containing cells (Fig. 5A–C). When we considered bacteria-containing phagocytes between primary and challenge infection, we observed that there were no differences in the number of UPEC+ cells in any of the populations we measured (Fig. 5A–C). However, in E. faecalis infection, we observed that the total number of CD45+ cells containing bacteria was increased in the challenge infection compared with primary infection (Fig. 5A). Among specific phagocyte populations, significant increases in the number of resident macrophages and mature (MHC II+) monocyte-derived cells containing E. faecalis were observed 24 h after challenge infection (Fig. 5B, 5C).

FIGURE 5.

Bacterial uptake is decreased during challenge infection. Six- to eight-week-old male C57BL/6 mice were infected with 1 × 107 CFU of either red fluorescent UPEC strain UTI89-RFP-kanR or E. faecalis OG1RF stained with wheat germ agglutinin conjugated to Alexa Fluor 594. At 24 h after primary or after challenge infection, mice were sacrificed and prostates analyzed by flow cytometry. Graphs show the (AC) absolute number or (DF) percentage of cells positive for red fluorescent bacteria within the specified immune cell populations depicted. Supplemental Fig. 3 shows the gating strategy used for these experiments. Data are pooled from two to three experiments, with five to seven mice per group in each experiment. Each dot represents one mouse, lines are medians, black dots depict naive mice, orange dots denote values from primary infection, and green dots represent data from challenge infection. Statistical significance was determined using the nonparametric Kruskal–Wallis test, and p values were corrected for multiple testing using the FDR method. The q values meeting the criteria for statistical significance (q < 0.05) are depicted in red. MD, monocyte-derived.

FIGURE 5.

Bacterial uptake is decreased during challenge infection. Six- to eight-week-old male C57BL/6 mice were infected with 1 × 107 CFU of either red fluorescent UPEC strain UTI89-RFP-kanR or E. faecalis OG1RF stained with wheat germ agglutinin conjugated to Alexa Fluor 594. At 24 h after primary or after challenge infection, mice were sacrificed and prostates analyzed by flow cytometry. Graphs show the (AC) absolute number or (DF) percentage of cells positive for red fluorescent bacteria within the specified immune cell populations depicted. Supplemental Fig. 3 shows the gating strategy used for these experiments. Data are pooled from two to three experiments, with five to seven mice per group in each experiment. Each dot represents one mouse, lines are medians, black dots depict naive mice, orange dots denote values from primary infection, and green dots represent data from challenge infection. Statistical significance was determined using the nonparametric Kruskal–Wallis test, and p values were corrected for multiple testing using the FDR method. The q values meeting the criteria for statistical significance (q < 0.05) are depicted in red. MD, monocyte-derived.

Close modal

We then calculated the percentage of cells containing bacteria between primary and challenge infection for the indicated immune cell populations in the prostate. We observed that the percentage of CD45+ immune cells containing bacteria was decreased in UPEC-challenged mice, whereas it was increased in E. faecalis–challenged mice (Fig. 5D). A greater proportion of resident immune cell populations, such as macrophages and CD103+ DCs, contained E. faecalis following challenge infection (Fig. 5E) Although these resident populations were not different in UPEC-challenged mice compared with primary infection, a significant decrease in the percentage of neutrophils containing bacteria was evident (Fig. 5F). These results suggest that primary infection may have altered neutrophils or their precursors possibly by impairing their capacity to take up bacteria, improving their ability to digest bacteria or making them more susceptible to death in UPEC challenge infection, thereby potentially contributing to a failure to resolve the infection and to the development of chronicity.

We originally hypothesized that a lack of innate immunity led to chronic prostate infection; however, both infections induced significant levels of cytokine expression and myeloid cell infiltration. Therefore, we asked whether lymphoid cell infiltration was induced during UPEC or E. faecalis primary and challenge infection. Using flow cytometry, we analyzed the numbers of T cells, NK cells, NKT cells, and innate lymphoid cells present in prostates of naive, primary, and challenge-infected mice. In a primary infection, T cells, comprised almost entirely of CD4+ T cells were significantly elevated over naive levels in the prostate following UPEC, but not E. faecalis infection (Fig. 6A, 6B, compare orange to black dots). The global CD3+ T cell population, and specifically the CD4+ T cell pool, was significantly increased upon challenge infection with either bacterial strain over their respective primary infection levels, suggesting that a memory response was induced (Fig. 6A, 6B, compare green to orange dots). NK cells were also increased in challenge infection compared with primary infection with either organism (Fig. 6C). NKT cells were significantly elevated in prostates following primary UPEC infection and E. faecalis challenge (Fig. 6C). Finally, the total number of innate lymphocytes was unchanged at all timepoints, suggesting that these cells do not play a role in prostatitis as no infiltration was induced (Fig. 6C).

FIGURE 6.

Effector immune cell infiltration is increased in prostates challenged with UPEC or E. faecalis. Six- to eight-week-old male C57BL/6 mice were infected with 1 × 107 CFU of red fluorescent UPEC strain UTI89-RFP-kanR or E. faecalis OG1RF stained with wheat germ agglutinin conjugated to Alexa Fluor 594. At 24 h after primary or after challenge infection, prostates were analyzed by flow cytometry. Naive mice were included for baseline determination. See Supplemental Fig. 3, gating strategy for these experiments. Graphs show (A) CD3+ T cells, (B) CD3+ CD4+ T cells, and (C) total specified immune cell populations. Data are pooled from two to three experiments, three to seven mice per group in each experiment. Each dot is one mouse, lines are medians, black dots depict naive mice, orange dots denote values from primary infection, and green dots represent data from challenge infection Statistical significance was determined using the nonparametric Kruskal–Wallis test, and p values were corrected for multiple testing using the FDR method. Statistically significant q values (q < 0.05) are in red.

FIGURE 6.

Effector immune cell infiltration is increased in prostates challenged with UPEC or E. faecalis. Six- to eight-week-old male C57BL/6 mice were infected with 1 × 107 CFU of red fluorescent UPEC strain UTI89-RFP-kanR or E. faecalis OG1RF stained with wheat germ agglutinin conjugated to Alexa Fluor 594. At 24 h after primary or after challenge infection, prostates were analyzed by flow cytometry. Naive mice were included for baseline determination. See Supplemental Fig. 3, gating strategy for these experiments. Graphs show (A) CD3+ T cells, (B) CD3+ CD4+ T cells, and (C) total specified immune cell populations. Data are pooled from two to three experiments, three to seven mice per group in each experiment. Each dot is one mouse, lines are medians, black dots depict naive mice, orange dots denote values from primary infection, and green dots represent data from challenge infection Statistical significance was determined using the nonparametric Kruskal–Wallis test, and p values were corrected for multiple testing using the FDR method. Statistically significant q values (q < 0.05) are in red.

Close modal

Given the pronounced effector cell infiltration we observed during UPEC and E. faecalis challenge infection, we hypothesized that this strong adaptive immune response would mediate protection against subsequent bacterial colonization. In female and male bladder infection, UPEC induces a protective immune response, although this response is not sterilizing, and animals challenged with UPEC are still colonized but at significantly lower levels compared with a primary infection (29, 30). Given the robust innate response in the prostate to UPEC infection compared with the near lack of response in E. faecalis infection, we investigated whether a protective adaptive immune response arises following infection with either uropathogen. We infected cohorts of male C57BL/6 mice with 1 × 107 CFU UPEC or OG1RF or with a 1:1 ratio of both strains. We determined the bacterial burden in the prostate and bladder 24 h PI in a subset of animals after a primary infection and monitored bacterial clearance over the course of 1 mo in the remaining mice. Mice that received UPEC or E. faecalis were challenged with an isogenic strain of the same bacteria used for primary infection. Animals receiving a primary polymicrobial infection were challenged with either UPEC or E. faecalis to evaluate the impact of primary coinfection on development of an adaptive immune response to either bacterial strain. In the bladder, 24 h after challenge infection, the bacterial burden was significantly decreased compared with that in primary UPEC-infected bladders, as we previously reported (Fig. 7A) (30). Surprisingly, prostate UPEC burdens from the same mice were nearly 10 times higher in challenged organs compared with the bacterial burden observed in prostates following the primary infection (Fig. 7A). Bacterial CFU in prostates challenged with E. faecalis were similar to primary infection levels, whereas quite surprisingly, the bladders from these same animals harbored nearly 100-fold greater bacterial burden than in primary infection (Fig. 7B). In the polymicrobial infection scenario, strikingly, protection against UPEC was lost in the bladders of these mice (Fig. 7C). In addition, protection against UPEC or E. faecalis was not observed in the prostates of mice that experienced a primary polymicrobial infection (Fig. 7C, 7D). Taken together, these data suggest that a protective adaptive response specifically does not develop to prostatitis despite that a nonsterilizing adaptive immunity is mounted in the bladder. Furthermore, these data support that E. faecalis plays a role in the suppression of adaptive immune responses, particularly in the bladder.

FIGURE 7.

The prostate fails to mount a protective immune response against bacterial reinfection. Six- to eight-week-old male C57BL/6 mice were infected with 1 × 107 CFU of (A) UPEC strain UTI89-RFP-kanR, (B) E. faecalis strain OG1RF, or (C) 1 × 107 CFU of both bacterial strains in a 1:1 ratio. One half of each group was sacrificed at 24 h after primary (1°) infection. At 25–30 d PI, remaining animals were treated with one to two cycles of antibiotics for 5 d, followed by a 3–5-d washout period. Mice with sterile urine were challenged with 1 × 107 CFU of (A) UPEC strain UTI89-GFP-ampR, (B) E. faecalis strain OG1RF_intergenicRS00490RS00495::Tn, or in primary polymicrobially infected mice; half of the cohort was challenged with (C) UTI89-GFP-ampR, and the other half was challenged (D) with OG1RF_intergenicRS00490RS00495::Tn. Mice were sacrificed 24 h after challenge infection (2°). Graphs depict CFU per organ. Each circle is one mouse, filled circles are bladder CFU, and open circles are prostate CFU. Red lines are the median value of each group, and black dotted lines are the LOD of the assay, 20 or 40 CFU/organ. Data are pooled from two to four experiments, each with two to six mice per group. CFU from 1 and 2° infection in bladder or prostate were compared using the nonparametric Mann–Whitney U test. Within each infection scenario (UPEC, E. faecalis, and polymicrobial), p values were corrected for multiple testing using the FDR method. Statistically significant q values (q < 0.05) are in red.

FIGURE 7.

The prostate fails to mount a protective immune response against bacterial reinfection. Six- to eight-week-old male C57BL/6 mice were infected with 1 × 107 CFU of (A) UPEC strain UTI89-RFP-kanR, (B) E. faecalis strain OG1RF, or (C) 1 × 107 CFU of both bacterial strains in a 1:1 ratio. One half of each group was sacrificed at 24 h after primary (1°) infection. At 25–30 d PI, remaining animals were treated with one to two cycles of antibiotics for 5 d, followed by a 3–5-d washout period. Mice with sterile urine were challenged with 1 × 107 CFU of (A) UPEC strain UTI89-GFP-ampR, (B) E. faecalis strain OG1RF_intergenicRS00490RS00495::Tn, or in primary polymicrobially infected mice; half of the cohort was challenged with (C) UTI89-GFP-ampR, and the other half was challenged (D) with OG1RF_intergenicRS00490RS00495::Tn. Mice were sacrificed 24 h after challenge infection (2°). Graphs depict CFU per organ. Each circle is one mouse, filled circles are bladder CFU, and open circles are prostate CFU. Red lines are the median value of each group, and black dotted lines are the LOD of the assay, 20 or 40 CFU/organ. Data are pooled from two to four experiments, each with two to six mice per group. CFU from 1 and 2° infection in bladder or prostate were compared using the nonparametric Mann–Whitney U test. Within each infection scenario (UPEC, E. faecalis, and polymicrobial), p values were corrected for multiple testing using the FDR method. Statistically significant q values (q < 0.05) are in red.

Close modal

One percent of all men will experience bacterial prostatitis in their lifetime, and half of them will have chronic relapses in the months following infection for unknown reasons. The only treatment option is antibiotics with regimens that may last from a few weeks to several months, increasing the risk of development of antibiotic resistance. Although the symptoms and microbiological causes of bacterial prostatitis are reasonably well investigated, little is known about the prostate immune compartment and its role in resolving or preventing infection. In this study, we used a mouse model to determine whether the immune response in the prostate following catheter-mediated transurethral instillation of UPEC or E. faecalis is sufficient to prevent chronic infection or reinfection. Mouse prostates remained colonized for at least 4 wk by UPEC or E. faecalis despite an acute proinflammatory response, including robust neutrophil and monocyte-derived cell infiltration, in the case of UPEC infection. Notably, the cytokine and cellular immune response was the same or greater upon challenge infection with both strains, although surprisingly, the prostate was not protected from colonization following a challenge infection regardless of a pronounced lymphocytic infiltration. These results suggest that a failure to develop memory following a primary infection, which may include B or T cell–mediated adaptive immunity or, intriguingly, innate immune memory, may be responsible for the development of chronic or recurrent infection. The underlying mechanisms for a lack of protection are still unknown but may include that despite robust T cell infiltration, T cells are not appropriately polarized, activated, or specific for uropathogen Ags. It may also be that inappropriate activation of innate immune cells leads to suppression of effector cells. Finally, particularly in the case of E. faecalis infection, it may be that the bacteria inhibit effector cell function, in which case, targeting the suppressive mechanism of the bacteria would be expected to improve protection against a second infection.

Both UPEC and E. faecalis displayed tropism for the prostate over the bladder at 24 h PI, but E. faecalis persisted at higher levels in the prostate compared with the bladder up to 2 wk PI. Despite persistent infection, E. faecalis did not elicit as strong an inflammatory response compared with UPEC. Differences in initial bacterial burden between UPEC and E. faecalis may be only partially responsible for the divergent cytokine response to the two organisms, as E. faecalis can suppress NF-κB–mediated proinflammatory responses in myeloid cells, dampening proinflammatory cytokine pathways (41). Additionally, E. faecalis–mediated immunosuppression may promote colonization or fitness of other bacteria, such as UPEC (46, 47).

The greater cytokine response in the prostate over the bladder in the first 2 d PI was surprising, as we recently showed that cytokine levels are globally suppressed in the bladder following UPEC infection in male mice compared with female mice (30). These findings suggest that during infection, cytokine expression in the prostate remains localized, and mechanisms that result in inadequate cytokine expression in the bladder are restricted to this tissue. Additionally, we observed different baseline levels of IL-6 and IL-1β between prostate and bladder, which may lead to divergent innate immune, and possibly adaptive, responses, as these tissues may be differently primed to respond to infection. However, insights into the prostate-specific role of IL-6 and IL-1β are needed to verify this hypothesis. In fact, these cytokines underpin early stage inflammatory processes, such as neutrophil recruitment, and adaptive immune processes, including Th17 polarization (4850). Albeit often synergistically correlated, IL-6 and IL-1β may also function independently (51). As an added layer of complexity, the inflammatory role of IL-6 is diverse. For example, IL-6−/− mice infected with E. coli, Candida albicans, or Mycobacterium tuberculosis have significantly reduced neutrophil numbers and survival, with a poorer inflammatory response (5254). However, aerosol exposure or i.p. injection of endotoxin into IL-6−/− mice induces local and systemic increases in TNF-α, CXCL2, and neutrophil infiltration in the lungs at significantly higher levels than in wild-type mice in which IL-6 production is observed (55). Therefore, the role of IL-6 in immunity of the prostate may be quite complex. Furthermore, despite differences in the magnitude of cytokine response between UPEC and E. faecalis infection, IL-1α, IL-6, CCL2, and IL-1β were significantly increased in the prostate over the bladder for 2 d PI with either bacteria, suggesting that common antibacterial innate immune pathways may be elicited in the prostate in which these cytokines are equally pivotal.

Compared with the bladder (30), the acute immune response to UPEC infection in the prostate was composed almost entirely of neutrophils and monocytes. The greater neutrophil involvement may be due to the copious secretion of IL-1α in the prostate, which recruits neutrophils in other tissues (56). IL-1α was five times higher than any other cytokine in UPEC infection and notably absent in E. faecalis infection. The role of IL-1α in immunity is less clear than IL-1β, even though they both signal through IL-R–MyD88-dependent inflammatory pathways (57). Constitutively active and normally sequestered in the cell cytoplasm, IL-1α is translocated to the plasma membrane during stress conditions and released by necrotic cells, inducing tissue inflammation (58). It is rapidly inducible in cells of hematopoietic origin, such as macrophages or DCs, and expressed by stromal cells such as keratinocytes or endothelial cells, as well (59). IL-1α may also induce local inflammation in an alarmin-like fashion. Notably, a sterile inflammatory response to necrotic HMGB1-deficient cells leads to IL-1α–dependent acute neutrophil infiltration in two different in vivo models (56). Similar mechanisms may be at play in UPEC-infected prostates, in which the release of IL-1α from necrotic cells and tissue damage provoked by UPEC invasion mediates immune cell infiltration. Thus, IL-1α–activated stromal cells would secrete proinflammatory cytokines in an autocrine inflammatory loop to enhance neutrophil and monocyte recruitment via the release of chemotactic mediators. Of great clinical interest, the use of anakinra, an IL-1R antagonist, greatly reduced infection-associated pathology in a mouse model of cystitis (60), supporting that mechanisms targeting the immune response may provide relief to patients suffering from prostatitis.

We observed a pronounced lack of protection against recurrent infection in prostates of UPEC- and E. faecalis–infected mice at 24 h PI despite that protective immunity was observed in the bladder of the same mice at the same timepoint following UPEC infection. This supports that the memory response to prostate infection is impaired. Potentially, lymphocytes in the two organs respond differently to challenge infection because of the divergent cytokine environment, lymphocytes specific for bacterial Ags do not infiltrate the prostate, may do so at a later timepoint, or that bacteria in the prostate are simply inaccessible to responding immune cells. Previously, we showed that the bladder adaptive immune response upon challenge infection is significantly improved when macrophages are depleted prior to primary infection (29). This subversive role of macrophages may also occur in the prostate. For example, in UPEC or E. faecalis infection, a vast majority of bacteria engulfing cells were resident or monocyte-derived macrophages, and few bacteria could be found in tissue DCs.

In summary, we found that mouse prostates remained chronically infected for at least 4 wk by UPEC or E. faecalis. Although infection induced strong soluble or cellular innate responses, ultimately, the prostate was not protected against challenge infection with either bacteria. This finding may underlie the high rate of chronic prostatitis observed in men. Overall, these results support that the prostate is reactive to bacterial colonization but fails to develop lasting protection in the face of recurrent infection. Understanding mechanisms underlying this ineffectual response include determining the Ag specificity of cellular or humoral responses that arise; however, the tools necessary for these investigations will need to be developed. These and other future studies will then ultimately contribute to the identification of targetable pathways for immunotherapeutic approaches in prostatitis.

We thank insightful discussions and support from Livia Lacerda Mariano, Camila Rosat Consiglio, Jérémy Boussier, Kao Hsien-Neng, Choo Pei Yi, Giovanna Barba, Gerald Spaeth, and Maria Luisa Pisanelli. We thank Dr. Kimberly Kline for sharing E. faecalis strains and critical feedback on the project and manuscript. Dr. Karen Sfanos and Dr. Darragh Duffy provided much-appreciated constructive critiques on the manuscript as well. We also thank the University of Glasgow, including Dr. Simon Milling and Dr. Robert Nibbs, for the academic opportunity offered through the integrated master’s program with work placement in life sciences.

This work was supported by funding from Agence Nationale de la Recherché ANR-17-CE17-0014.

The online version of this article contains supplemental material.

Abbreviations used in this article:

DC

dendritic cell

FDR

false discovery rate

LOD

limit of detection

MHC II

MHC class II

PI

postinfection

UPEC

uropathogenic E. coli

UTI

urinary tract infection.

1
Krieger
,
J. N.
,
L.
Nyberg
,
J. C.
Nickel
.
1999
.
NIH consensus definition and classification of prostatitis.
JAMA
282
:
236
237
.
2
Khan
,
F. U.
,
A. U.
Ihsan
,
H. U.
Khan
,
R.
Jana
,
J.
Wazir
,
P.
Khongorzul
,
M.
Waqar
,
X.
Zhou
.
2017
.
Comprehensive overview of prostatitis.
Biomed. Pharmacother.
94
:
1064
1076
.
3
Krieger
,
J. N.
,
S. W.
Lee
,
J.
Jeon
,
P. Y.
Cheah
,
M. L.
Liong
,
D. E.
Riley
.
2008
.
Epidemiology of prostatitis.
Int. J. Antimicrob. Agents
31
(
Suppl. 1
):
S85
S90
.
4
Schaeffer
,
A. J.
2003
.
Epidemiology and demographics of prostatitis.
Eur. Urol. Suppl.
2
:
5
10
.
5
Lipsky
,
B. A.
1999
.
Prostatitis and urinary tract infection in men: what’s new; What’s true?
Am. J. Med.
106
:
327
334
.
6
Foxman
,
B.
2010
.
The epidemiology of urinary tract infection.
Nat. Rev. Urol.
7
:
653
660
.
7
Ingersoll
,
M. A.
2017
.
Sex differences shape the response to infectious diseases.
PLoS Pathog.
13
: e1006688.
8
Rowe
,
T. A.
,
M.
Juthani-Mehta
.
2013
.
Urinary tract infection in older adults.
Aging Health
9
:
519
528
.
9
Meares
,
E. M.
,
T. A.
Stamey
.
1972
.
The diagnosis and management of bacterial prostatitis.
Br. J. Urol.
44
:
175
179
.
10
Nickel
,
J. C.
,
D.
Shoskes
,
Y.
Wang
,
R. B.
Alexander
,
J. E.
Fowler
Jr.
,
S.
Zeitlin
,
M. P.
O’Leary
,
M. A.
Pontari
,
A. J.
Schaeffer
,
J. R.
Landis
, et al
.
2006
.
How does the pre-massage and post-massage 2-glass test compare to the Meares-Stamey 4-glass test in men with chronic prostatitis/chronic pelvic pain syndrome?
J. Urol.
176
:
119
124
.
11
Bowen
,
D. K.
,
E.
Dielubanza
,
A. J.
Schaeffer
,
2015
.
Chronic bacterial prostatitis and chronic pelvic pain syndrome.
BMJ Clin. Evid.
,
2015
:
1802
.
12
Gill
,
B. C.
,
D. A.
Shoskes
.
2016
.
Bacterial prostatitis.
Curr. Opin. Infect. Dis.
29
:
86
91
.
13
Domingue
,
G. J.
,
W. J. G.
Hellstrom
.
1998
.
Prostatitis.
Clin. Microbiol. Rev.
11
:
604
613
.
14
Lipsky
,
B. A.
,
I.
Byren
,
C. T.
Hoey
.
2010
.
Treatment of bacterial prostatitis.
Clin. Infect. Dis.
50
:
1641
1652
.
15
Coker
,
T. J.
,
D. M.
Dierfeldt
.
2016
.
Acute bacterial prostatitis: diagnosis and management.
Am. Fam. Physician
93
:
114
120
.
16
Naber
,
K. G.
,
K.
Roscher
,
H.
Botto
,
V.
Schaefer
.
2008
.
Oral levofloxacin 500 mg once daily in the treatment of chronic bacterial prostatitis.
Int. J. Antimicrob. Agents
32
:
145
153
.
17
Stamatiou
,
K.
,
N.
Pierris
.
2017
.
Mounting resistance of uropathogens to antimicrobial agents: a retrospective study in patients with chronic bacterial prostatitis relapse.
Investig. Clin. Urol.
58
:
271
280
.
18
Delarosette
,
J.
,
M. R.
Hubregtse
,
E. J. H.
Meuleman
,
M. V. M.
Stolkengelaar
,
F. M. J.
Debruyne
.
1993
.
Diagnosis and treatment OF 409 patients with prostatitis syndromes.
Urology
41
:
301
307
.
19
Rudick
,
C. N.
,
R. E.
Berry
,
J. R.
Johnson
,
B.
Johnston
,
D. J.
Klumpp
,
A. J.
Schaeffer
,
P.
Thumbikat
.
2011
.
Uropathogenic Escherichia coli induces chronic pelvic pain.
Infect. Immun.
79
:
628
635
.
20
Quick
,
M. L.
,
L.
Wong
,
S.
Mukherjee
,
J. D.
Done
,
A. J.
Schaeffer
,
P.
Thumbikat
.
2013
.
Th1-Th17 cells contribute to the development of uropathogenic Escherichia coli-induced chronic pelvic pain.
PLoS One
8
: e60987.
21
Boehm
,
B. J.
,
S. A.
Colopy
,
T. J.
Jerde
,
C. J.
Loftus
,
W.
Bushman
.
2012
.
Acute bacterial inflammation of the mouse prostate.
Prostate
72
:
307
317
.
22
Wong
,
L.
,
P. R.
Hutson
,
W.
Bushman
.
2014
.
Prostatic inflammation induces fibrosis in a mouse model of chronic bacterial infection.
PLoS One
9
: e100770.
23
Wong
,
L.
,
P. R.
Hutson
,
W.
Bushman
.
2015
.
Resolution of chronic bacterial-induced prostatic inflammation reverses established fibrosis.
Prostate
75
:
23
32
.
24
Simons
,
B. W.
,
N. M.
Durham
,
T. C.
Bruno
,
J. F.
Grosso
,
A. J.
Schaeffer
,
A. E.
Ross
,
P. J.
Hurley
,
D. M.
Berman
,
C. G.
Drake
,
P.
Thumbikat
,
E. M.
Schaeffer
.
2015
.
A human prostatic bacterial isolate alters the prostatic microenvironment and accelerates prostate cancer progression.
J. Pathol.
235
:
478
489
.
25
Shinohara
,
D. B.
,
A. M.
Vaghasia
,
S. H.
Yu
,
T. N.
Mak
,
H.
Bruggemann
,
W. G.
Nelson
,
A. M.
De Marzo
,
S.
Yegnasubramanian
,
K. S.
Sfanos
.
2013
.
A mouse model of chronic prostatic inflammation using a human prostate cancer-derived isolate of Propionibacterium acnes.
Prostate
73
:
1007
1015
.
26
Murphy
,
S. F.
,
J. F.
Anker
,
D. J.
Mazur
,
C.
Hall
,
A. J.
Schaeffer
,
P.
Thumbikat
.
2019
.
Role of gram-positive bacteria in chronic pelvic pain syndrome (CPPS).
Prostate
79
:
160
167
.
27
Lupo
,
F.
,
M. A.
Ingersoll
.
2019
.
Is bacterial prostatitis a urinary tract infection?
Nat. Rev. Urol.
16
:
203
204
.
28
Ingersoll
,
M. A.
,
K. A.
Kline
,
H. V.
Nielsen
,
S. J.
Hultgren
.
2008
.
G-CSF induction early in uropathogenic Escherichia coli infection of the urinary tract modulates host immunity.
Cell. Microbiol.
10
:
2568
2578
.
29
Mora-Bau
,
G.
,
A. M.
Platt
,
N.
van Rooijen
,
G. J.
Randolph
,
M. L.
Albert
,
M. A.
Ingersoll
.
2015
.
Macrophages subvert adaptive immunity to urinary tract infection.
PLoS Pathog.
11
: e1005044.
30
Zychlinsky Scharff
,
A.
,
M.
Rousseau
,
L.
Lacerda Mariano
,
T.
Canton
,
C. R.
Consiglio
,
M. L.
Albert
,
M.
Fontes
,
D.
Duffy
,
M. A.
Ingersoll
.
2019
.
Sex differences in IL-17 contribute to chronicity in male versus female urinary tract infection.
JCI Insight
5
: e122998.
31
Hung
,
C. S.
,
K. W.
Dodson
,
S. J.
Hultgren
.
2009
.
A murine model of urinary tract infection.
Nat. Protoc.
4
:
1230
1243
.
32
Zychlinsky Scharff
,
A.
,
M. L.
Albert
,
M. A.
Ingersoll
,
2017
.
Urinary tract infection in a small animal model: transurethral catheterization of male and female mice.
J. Vis. Exp.
130
:
54432
.
33
Dunny
,
G. M.
,
B. L.
Brown
,
D. B.
Clewell
.
1978
.
Induced cell aggregation and mating in Streptococcus faecalis: evidence for a bacterial sex pheromone.
Proc. Natl. Acad. Sci. USA
75
:
3479
3483
.
34
Debroy
,
S.
,
R.
van der Hoeven
,
K. V.
Singh
,
P.
Gao
,
B. R.
Harvey
,
B. E.
Murray
,
D. A.
Garsin
.
2012
.
Development of a genomic site for gene integration and expression in Enterococcus faecalis.
J. Microbiol. Methods
90
:
1
8
.
35
Kristich
,
C. J.
,
V. T.
Nguyen
,
T.
Le
,
A. M.
Barnes
,
S.
Grindle
,
G. M.
Dunny
.
2008
.
Development and use of an efficient system for random mariner transposon mutagenesis to identify novel genetic determinants of biofilm formation in the core Enterococcus faecalis genome.
Appl. Environ. Microbiol.
74
:
3377
3386
.
36
Dale
,
J. L.
,
K. B.
Beckman
,
J. L. E.
Willett
,
J. L.
Nilson
,
N. P.
Palani
,
J. A.
Baller
,
A.
Hauge
,
D. M.
Gohl
,
R.
Erickson
,
D. A.
Manias
, et al
.
2018
.
Comprehensive functional analysis of the Enterococcus faecalis core genome using an ordered, sequence-defined collection of insertional mutations in strain OG1RF.
mSystems
3
: e00062-18.
37
Libke
,
R. D.
,
C.
Regamey
,
J. T.
Clarke
,
W. M. M.
Kirby
.
1973
.
Synergism of carbenicillin and gentamicin against enterococci.
Antimicrob. Agents Chemother.
4
:
564
568
.
38
Baker
,
D. A.
,
V. T.
Androle
.
1973
.
The treatment of difficult urinary-tract infections with carbenicillin indanyl sodium.
J. Infect. Dis.
127
(
Suppl. 2
):
S136
S142
.
39
Kau
,
A. L.
,
S. M.
Martin
,
W.
Lyon
,
E.
Hayes
,
M. G.
Caparon
,
S. J.
Hultgren
.
2005
.
Enterococcus faecalis tropism for the kidneys in the urinary tract of C57BL/6J mice.
Infect. Immun.
73
:
2461
2468
.
40
Singh
,
K. V.
,
S. R.
Nallapareddy
,
B. E.
Murray
.
2007
.
Importance of the ebp (endocarditis- and biofilm-associated pilus) locus in the pathogenesis of Enterococcus faecalis ascending urinary tract infection.
J. Infect. Dis.
195
:
1671
1677
.
41
Tien
,
B. Y. Q.
,
H. M. S.
Goh
,
K. K. L.
Chong
,
S.
Bhaduri-Tagore
,
S.
Holec
,
R.
Dress
,
F.
Ginhoux
,
M. A.
Ingersoll
,
R. B. H.
Williams
,
K. A.
Kline
.
2017
.
Enterococcus faecalis promotes innate immune suppression and polymicrobial catheter-associated urinary tract infection.
Infect. Immun.
85
: e00378-17.
42
Brogden
,
K. A.
,
J. M.
Guthmiller
,
C. E.
Taylor
.
2004
.
Human polymicrobial infections.
Lancet
365
:
253
255
.
43
Siegmanigra
,
Y.
,
T.
Kulka
,
D.
Schwartz
,
N.
Konforti
.
1993
.
The significance of polymicrobial growth in urine: contamination or true infection.
Scand. J. Infect. Dis.
25
:
85
91
.
44
Siegman-Igra
,
Y.
,
T.
Kulka
,
D.
Schwartz
,
N.
Konforti
.
1994
.
Polymicrobial and monomicrobial bacteraemic urinary tract infection.
J. Hosp. Infect.
28
:
49
56
.
45
Kline
,
K. A.
,
A. L.
Lewis
.
2016
.
Gram-positive uropathogens, polymicrobial urinary tract infection, and the emerging microbiota of the urinary tract.
Microbiol. Spectr.
DOI: 10.1128/microbiolspec.UTI-0012-2012.
46
Tay
,
W. H.
,
K. K. L.
Chong
,
K. A.
Kline
.
2016
.
Polymicrobial-Host interactions during infection.
J. Mol. Biol.
428
:
3355
3371
.
47
Kao
,
P. H. N.
,
K. A.
Kline
.
2019
.
Dr. Jekyll and mr. Hide: how Enterococcus faecalis subverts the host immune response to cause infection.
J. Mol. Biol.
431
:
2932
2945
.
48
Netea
,
M. G.
,
A.
Simon
,
F.
van de Veerdonk
,
B.-J.
Kullberg
,
J. W. M.
Van der Meer
,
L. A. B.
Joosten
.
2010
.
IL-1β processing in host defense: beyond the inflammasomes.
PLoS Pathog.
6
: e1000661.
49
Rincon
,
M.
2012
.
Interleukin-6: from an inflammatory marker to a target for inflammatory diseases.
Trends Immunol.
33
:
571
577
.
50
Bettelli
,
E.
,
Y. J.
Carrier
,
W. D.
Gao
,
T.
Korn
,
T. B.
Strom
,
M.
Oukka
,
H. L.
Weiner
,
V. K.
Kuchroo
.
2006
.
Reciprocal developmental pathways for the generation of pathogenic effector TH17 and regulatory T cells.
Nature
441
:
235
238
.
51
McGeough
,
M. D.
,
C. A.
Pena
,
J. L.
Mueller
,
D. A.
Pociask
,
L.
Broderick
,
H. M.
Hoffman
,
S. D.
Brydges
.
2012
.
Cutting edge: IL-6 is a marker of inflammation with No direct role in inflammasome-mediated mouse models.
J. Immunol.
189
:
2707
2711
.
52
Dalrymple
,
S. A.
,
R.
Slattery
,
D. M.
Aud
,
M.
Krishna
,
L. A.
Lucian
,
R.
Murray
.
1996
.
Interleukin-6 is required for a protective immune response to systemic Escherichia coli infection.
Infect. Immun.
64
:
3231
3235
.
53
Romani
,
L.
,
A.
Mencacci
,
E.
Cenci
,
R.
Spaccapelo
,
C.
Toniatti
,
P.
Puccetti
,
F.
Bistoni
,
V.
Poli
.
1996
.
Impaired neutrophil response and CD4(+) T helper cell 1 development in interleukin 6-deficient mice infected with Candida albicans.
J. Exp. Med.
183
:
1345
1355
.
54
Ladel
,
C. H.
,
C.
Blum
,
A.
Dreher
,
K.
Reifenberg
,
M.
Kopf
,
S. H. E.
Kaufmann
.
1997
.
Lethal tuberculosis in interleukin-6-deficient mutant mice.
Infect. Immun.
65
:
4843
4849
.
55
Xing
,
Z.
,
J.
Gauldie
,
G.
Cox
,
H.
Baumann
,
M.
Jordana
,
X. F.
Lei
,
M. K.
Achong
.
1998
.
IL-6 is an antiinflammatory cytokine required for controlling local or systemic acute inflammatory responses.
J. Clin. Invest.
101
:
311
320
.
56
Chen
,
C.-J.
,
H.
Kono
,
D.
Golenbock
,
G.
Reed
,
S.
Akira
,
K. L.
Rock
.
2007
.
Identification of a key pathway required for the sterile inflammatory response triggered by dying cells.
Nat. Med.
13
:
851
856
.
57
Dinarello
,
C. A.
2009
.
Immunological and inflammatory functions of the interleukin-1 family.
Annu. Rev. Immunol.
27
:
519
550
.
58
Garlanda
,
C.
,
C. A.
Dinarello
,
A.
Mantovani
.
2013
.
The interleukin-1 family: back to the future.
Immunity
39
:
1003
1018
.
59
Di Paolo
,
N. C.
,
D. M.
Shayakhmetov
.
2016
.
Interleukin 1 alpha and the inflammatory process.
Nat. Immunol.
17
:
906
913
.
60
Ambite
,
I.
,
M.
Puthia
,
K.
Nagy
,
C.
Cafaro
,
A.
Nadeem
,
D. S. C.
Butler
,
G.
Rydstrom
,
N. A.
Filenko
,
B.
Wullt
,
T.
Miethke
,
C.
Svanborg
.
2016
.
Molecular basis of acute cystitis reveals susceptibility genes and immunotherapeutic targets.
PLoS Pathog.
12
: e1005848.

The authors have no financial conflicts of interest.

Supplementary data