Host-pathogen interactions mediating pain of urinary tract infection

Abstract

Background: Pelvic pain is a major component of the morbidity associated with urinary tract infection (UTI), yet the molecular mechanisms underlying UTI-induced pain remain unknown. UTI pain mechanisms probably contrast with the clinical condition of asymptomatic bacteriuria (ASB), characterized by significant bacterial loads without lack symptoms.

Methods: A murine UTI model was used to compare pelvic pain behavior elicited by infection with uropathogenic Escherichia coli strain NU14 and ASB strain 83972.

Results: NU14-infected mice exhibited pelvic pain, whereas mice infected with 83972 did not exhibit pain, similar to patients infected with 83972. NU14-induced pain was not dependent on mast cells, not correlated with bacterial colonization or urinary neutrophils. UTI pain was not influenced by expression of type 1 pili, the bacterial adhesive appendages that induce urothelial apoptosis. However, purified NU14 lipopolysaccharide (LPS) induced Toll-like receptor 4 (TLR4)-dependent pain, whereas 83972 LPS induced no pain. Indeed, 83972 LPS attenuated the pain of NU14 infection, suggesting therapeutic potential.

Conclusions: These data suggest a novel mechanism of infection-associated pain that is dependent on TLR4 yet independent of inflammation. Clinically, these findings also provide the rational for probiotic therapies that would minimize the symptoms of infection without reliance on empirical therapies that contribute to antimicrobial resistance.

Figure 1

NU14 induces pelvic pain in female mice. Referred visceral hyperalgesia was measured as responses to mechanical stimulation of the pelvic region using von Frey filaments of 5 calibrated forces. Data are reported as the mean percentage of positive response ± SEM before instillation of bacteria (baseline) and at PID 1, 2, 3, 4, 5, 6, 7, 10, and 14. A) Responses to pelvic stimulation of saline-infected female C57BL/6J mice (B6, n=15). B) Responses to pelvic stimulation of female B6 mice infected with 83972 (n=10). C) Responses to pelvic stimulation of female B6 mice infected with NU14 (n=15). D) Responses to pelvic stimulation of female Kit^W-sh/Kit^W-sh mice infected with NU14 (n=9). ANOVA indicated a significant increase in response frequency from baseline at all filaments tested in NU14-treated mice at PIDs 1–10 (p<0.05), with no significant differences in baseline between saline and NU14 treated mice. NU14-treated mice exhibited a significant increase in response frequency from baseline in the largest 3 filament only at PID14 (P<0.05). E) Percent response at each PID were calculated as total responses to all fibers relative to base line responses. The symbol key shown in (A) applies to panels A–D.

Figure 2

UTI-associated pelvic pain is attenuated by organ crosstalk. Modulation of referred visceral hyperalgesia by colonic lidocaine was measured as responses to mechanical stimulation of the pelvic region with von Frey filaments of 5 intensities. Responsiveness was characterized prior to infection (baseline), PID 1 following bacterial infection, and 45 min. following colonic administration of 2% lidocaine on PID1. Instilling 50 μl lidocaine into the colon of B6 mice (B, n=15) or mast cell deficient Kit^W-sh mice (C, n=5) reduced pelvic pain responses (P<0.05), whereas lidocaine had no significant effect on mice infected with 83972 (A, n=10). Data are reported as the mean percentage of positive responses ± SEM.

Figure 3

NU14-induced pelvic pain is not correlated with bacterial colonization of the bladder. Wild-type mice were infected with either 83972 or NU14, and bacterial colonization was measured 24 hours (n=10 for 83972, n=20 for NU14) or 14 days (n=10 for 83972, n=15 for NU14) after infection. Both 83972 and NU14 colonized the bladder, however NU14 colonization was significantly greater than 83972 at 24 hours (A) and 14 days (C) after infection (**P<0.01). Mean colonization levels are indicated by solid lines, bladder homogenates with out detectable bacterial colonization appear on the x-axis. Pelvic pain was not correlated with bladder colonization at 24 hours (B, r= −0.20) or 14 days (D, r= −0.17) after infection.

Figure 4

Type 1 pilus status does not influence UTI-associated pelvic pain. Referred visceral hyperalgesia was measured as responses to mechanical stimulation of the pelvic region with von Frey filaments of 5 intensities. Responsiveness was characterized at baseline and 24 hours after bacterial infection. A) Responses to pelvic stimulation of female B6 mice infected with 83972 that does not have the fimB-fimD gene cluster (no type 1 pili, n=10). B) Responses to pelvic stimulation of female B6 mice infected with 83972:pREG153 that does not have the fimB-fimD gene cluster (no type 1 pili, n=10). C) Responses to pelvic stimulation of female wild-type mice infected with 83972:pGB4 that does have the fimB-fimD gene cluster (expresses type 1 pili, n=10). D) Responses to pelvic stimulation of wild-type mice infected with NU14 that does have fimH (expresses type 1 pili, n=20). E) Responses to pelvic stimulation of wild-type mice infected with NU14-1 that does not have fimH (greatly reduced expression of type 1 pili, n=10). Data are reported as the mean percentage of positive responses ± SEM. ANOVA indicated a significant increase in responses from baseline at all filaments tested in NU14 and NU14-1-treated mice at 24 hours after infection (p<0.05), with no significant differences in baseline between any of the five groups of infected mice. F) Percent responses 24 hours after infection were calculated as total responses to all fibers relative to baseline responses for 83972 (while bar), 83972:pREG153 without (light gray bar), 83972:pGB4 (dark gray bar) and NU14 with or with out fimH (black bars).

Figure 5

NU14-induced pelvic pain is not correlated with urine neutrophil myeloperoxidase. Female B6 mice were infected with either 83972, 83972:pGB4, NU14 or NU14-1 and neutrophil myeloperoxidase (MPO, a mesure of inflammation) was quantified 6 (n=28, 10, 23 and 10 respectively) or 24 hours after infection (n=30, 8, 15 and 10 respectively). NU14-1 MPO levels are significantly lower than both NU14 and 83972:pGB4 (p<0.05) 6 hours after infection with no other statistically significant differences between any of groups (A). NU14-1 MPO levels are significantly lower than 83972:pGB4 and NU14 MPO levels are significantly higher than 83972 (p<0.05) 24 hours after infection with no other statistically significant differences between any of groups (C). Mean MPO levels are indicated by solid lines. Pelvic pain was not correlated with MPO levels at 6 (B, r= −0.01) or 24 hours (D, r= 0.39) after infection. Histograms of CD80⁺ cells 4 hours (E) and 8 hours (F) following LPS stimulation (gray solid line with fill is medium, black dashed line is isotype control, black dotted line in 83972 and black solid line is NU14). G) Both NU14 and 83972 LPS significantly increased numbers of CD80⁺ macrophages compared to the medium (ANOVA, P<0.05; * indicates a significant increase compared to medium at 4 hour after LPS stimulation). H) Both NU14 and 83972 LPS induce significantly increased levels of IL-6 (ANOVA, P<0.05; * indicates a significant increase compared to medium at 8 hours after LPS stimulation).

Figure 6

Lipopolysacharride induces pelvic pain and represents a therapeutic target for pain relief. Referred visceral hyperalgesia was measured as responses to mechanical stimulation of the pelvic region with von Frey filaments of 5 intensities. Responsiveness was characterized at baseline, 1, 4, 24, 48, 72 and 96 hours after 83972 or NU14 lipoploysacharride (LPS) instillation. A) Responses to pelvic stimulation of female B6 mice instilled with NU14 LPS. Data are reported as the mean percentage of positive responses ± SEM (n=8 for 83972 and NU14). ANOVA indicated a significant increase in response frequency from baseline at all filaments tested in NU14 LPS-treated mice at 1, 4 and 24 hours after instillation (p<0.05), with no significant differences in baseline between 83972 and NU14 treated mice. B) Percent responses 1, 4, 24, 48, 72 and 96 hours after instillation were calculated as total responses to all fibers relative to baseline responses for 83972 or NU14 LPS instilled mice. C) LPS-induced MPO levels in 83972 and NU14 LPS instilled mice. D) C3H-HeJ mice (n=10) instilled with NU14 LPS exhibited a significant reduction in pelvic pain compared with C3H-HeOuJ mice (n=10) instilled in NU14 LPS (P<0.05). Percent responses 4 hours after instillation were calculated as total responses to all fibers relative to baseline responses. E) Left panel is a cartoon and experimental timeline (intervention was saline or 83972 LPS). Right panel is a graph of the percent responses at PID 1, 2, 3, 4, 5 and 6 days after NU14 instillation were calculated as total responses to all fibers relative to baseline responses for all groups of mice (NU14 then saline, n=8; NU14 then 83972 LPS, n=9).