Expression and Significance of the HIP/PAP and RegIIIγ Antimicrobial Peptides during Mammalian Urinary Tract Infection

Abstract

Recent evidence indicates that antimicrobial peptides (AMPs) serve key roles in defending the urinary tract against invading uropathogens. To date, the individual contribution of AMPs to urinary tract host defense is not well defined. In this study, we identified Regenerating islet-derived 3 gamma (RegIIIγ) as the most transcriptionally up-regulated AMP in murine bladder transcriptomes following uropathogenic Escherichia coli (UPEC) infection. We confirmed induction of RegIIIγ mRNA during cystitis and pyelonephritis by quantitative RT-PCR. Immunoblotting demonstrates increased bladder and urinary RegIIIγ protein levels following UPEC infection. Immunostaining localizes RegIIIγ protein to urothelial cells of infected bladders and kidneys. Human patients with UTI have increased urine concentrations of the orthologous Hepatocarcinoma-Intestine-Pancreas / Pancreatitis Associated Protein (HIP/PAP) compared to healthy controls. Recombinant RegIIIγ protein does not demonstrate bactericidal activity toward UPEC in vitro, but does kill Staphylococcus saprophyticus in a dose-dependent manner. Kidney and bladder tissue from RegIIIγ knockout mice and wild-type mice contain comparable bacterial burden following UPEC and Gram-positive UTI. Our results demonstrate that RegIIIγ and HIP/PAP expression is induced during human and murine UTI. However, their specific function in the urinary tract remains uncertain.

Fig 1. Confirmation of AMP mRNA regulation during experimental cystitis by qRT-PCR.

The fold-change in AMP mRNA expression 6 and 24 hpi, relative to a pool of uninfected bladders, is shown. * P < 0.05, Mann-Whitney U test versus uninfected bladders (n = 4 bladders/group, uninfected and 6 hpi; n = 5 bladders/group, 24 hpi).

Fig 2. RegIIIγ mRNA and RegIIIγ protein expression are induced during UPEC cystitis, and RegIIIγ is secreted into the urinary stream following UPEC infection.

(A) Fold-change in bladder RegIIIγ mRNA levels in response to intravesical inoculation of 10⁸ CFU of UPEC strain UTI89. * P = 0.0286, Mann-Whitney U test versus uninfected bladders (n = 4 bladders/group). (B) Relative RegIIIγ mRNA levels increase 24 hpi following inoculation with 10⁸ CFU of UPEC, but not in response to Gram-positive uropathogens. * P = 0.0286, Mann-Whitney U test, UPEC versus uninfected bladders (n = 4 bladders/group). (C) RegIIIγ protein (arrow) is induced in wild-type (WT) bladder 16 hpi with UPEC, but expression is absent at earlier times, and as well as in RegIIIγ ^-/- bladders. Bladder RegIIIγ co-migrates with recombinant (r) RegIIIγ protein. A non-specific reactive band (n.s., arrowhead) migrates more rapidly and is present in WT and RegIIIγ ^-/- bladder lysates. (D) RegIIIγ protein is undetectable in sterile WT urine, but secreted 24 hpi with UPEC. Gram-positive inocula, E. faecalis (EF) and S. saprophyticus (SS), fail to induce RegIIIγ secretion, and RegIIIγ immunoreactivity is absent in RegIIIγ ^-/- urine 24 hpi with UPEC.

Fig 3. RegIIIγ protein localizes to urothelium following UPEC infection.

RegIIIγ immunoreactivity is absent in uninfected bladders. Non-specific staining of UPEC is seen 2 and 6 hpi. Beginning 16 hpi, patchy urothelial immunoreactivity is observed. Urothelial RegIIIγ expression is most prominent 24 hpi and persists to a lesser extent 48 hpi. All figures are 20x magnification.

Fig 4. Induction of RegIIIγ mRNA and RegIIIγ protein by UPEC in C3H/HeOuJ kidneys.

(A) Significant RegIIIγ mRNA induction by UPEC in C3H/HeOuJ kidneys, with progressive accumulation over time. (* P = 0.0286, Mann-Whitney U test versus uninfected C3H/HeOuJ kidneys, n = 4 kidneys/time). (B) Absent RegIIIγ immunostaining in uninfected C3H/HeOuJ urothelium, which exhibits uniform apical expression of Uroplakin 3a (Upk3a; red). U: Urinary space. (C) RegIIIγ protein (green) localizes to the hypertrophied urothelium of C3H/HeOuJ kidneys 28 dpi, with reduced expression of Upk3a. Urothelial border is indicated by dashed line. Both images are 20x magnification.

Fig 5. The human RegIIIγ orthologue, HIP/PAP, is secreted in infected human urine.

(A) HIP/PAP levels were quantified by ELISA and adjusted for urine creatinine to control for differences in urine concentration. The resulting PAP/Cr ratio is significantly elevated in urine from patients with positive urine cultures, compared to sterile urine (n = 26 urines/group; * P < 0.0001, Mann-Whitney U test). The horizontal line indicates the mean. (B) Western blotting demonstrates absent of HIP/PAP in sterile (S) urine, whereas infected (I) urine from two children with UTI exhibits immunoreactivity.

Fig 6. Recombinant RegIIIγ protein exhibits bactericidal activity toward S. saprophyticus but not UPEC, and RegIIIγ deletion does not affect susceptibility to experimental UTI.

(A) Recombinant RegIIIγ kills S. saprophyticus not UPEC in a dose-dependent manner. 10⁵ CFU bacteria were incubated with indicated concentrations of recombinant C-terminally cleaved RegIIIγ for 2 hours at 37°C in 10 mM MES, pH 5.5 with 25 mM NaCl, and colonies were enumerated. (B-C) Experimental UTI. Comparable bacterial burden in wildtype (WT) and RegIIIγ ^-/- urinary tracts following inoculation with 10⁸ CFU of UPEC strain UTI89 (B) or S. saprophyticus (C). There was no significant difference between genotypes by Mann-Whitney U test (p > 0.05). Horizontal lines indicate geometric means.