Cutaneous Burn Injury Modulates Urinary Antimicrobial Peptide Responses and the Urinary Microbiome

Abstract

Objectives: Characterization of urinary bacterial microbiome and antimicrobial peptides after burn injury to identify potential mechanisms leading to urinary tract infections and associated morbidities in burn patients.

Design: Retrospective cohort study using human urine from control and burn subjects.

Setting: University research laboratory.

Patients: Burn patients.

Interventions: None.

Measurements and main results: Urine samples from catheterized burn patients were collected hourly for up to 40 hours. Control urine was collected from "healthy" volunteers. The urinary bacterial microbiome and antimicrobial peptide levels and activity were compared with patient outcomes. We observed a significant increase in urinary microbial diversity in burn patients versus controls, which positively correlated with a larger percent burn and with the development of urinary tract infection and sepsis postadmission, regardless of age or gender. Urinary psoriasin and β-defensin antimicrobial peptide levels were significantly reduced in burn patients at 1 and 40 hours postadmission. We observed a shift in antimicrobial peptide hydrophobicity and activity between control and burn patients when urinary fractions were tested against Escherichia coli and Enterococcus faecalis urinary tract infection isolates. Furthermore, the antimicrobial peptide activity in burn patients was more effective against E. coli than E. faecalis. Urinary tract infection-positive burn patients with altered urinary antimicrobial peptide activity developed either an E. faecalis or Pseudomonas aeruginosa urinary tract infection, suggesting a role for urinary antimicrobial peptides in susceptibility to select uropathogens.

Conclusions: Our data reveal potential links for urinary tract infection development and several morbidities in burn patients through alterations in the urinary microbiome and antimicrobial peptides. Overall, this study supports the concept that early assessment of urinary antimicrobial peptide responses and the bacterial microbiome may be used to predict susceptibility to urinary tract infections and sepsis in burn patients.

Figure 1. Burn injury significantly alters the diversity of the urinary microbiome

A) NMDS using Bray-Curtis Dissimilarity analysis comparing the bacterial diversity at the Genus level within the urinary microbiome of controls (black squares) and burn patients (red circles). The letter represents the subject identification, while the number represents the time (hours) post-admission for each subject. The patients within the black dotted circle represent control patients, which cluster away from burn patients within the red dotted circle at all time-points. B) The most abundant bacterial genera in controls and burn patients are indicated in horizontal bar graphs. Controls are indicated by blue bars. Burn subjects are indicated by red bars. All genera shown are significantly different between the 2 cohorts (p<0.05).

Figure 2. Urinary protein levels of AMPs after burn injury

A) Psoriasin levels are significantly decreased after burn injury at 1 and 40 hours post-admission. B) Lactoferrin levels are not significantly different after burn injury. C) hBD1 levels are significantly decreased after burn injury at hour 1 and 40 hours post-admission. D) hBD2 levels are significantly decreased at 40 hours post-admission. E) HMGB-1 levels are not statistically different at either time-point. F) Hepcidin levels are not statistically different at either time-point. * = p<0.05 and # = p<0.001 with Mann Whitney test, n = 8-24.

Figure 3. Bacterial growth inhibition of E. coli with urine fractions from control patients compared to the same fractions of burned patients at 1 hour and 40 hours post-admission

A) Areas of clearing in E. coli bacterial lawns in fractions 7-12 from urine samples of a representative control and burn patient at 1 and 40 hours post-admission. B) Antimicrobial activity against E. coli demonstrates notable changes in multiple fractions, primarily in fractions 2, 10-11, and 17-20. Arrow to the right indicates increasing peptide hydrophobicity. The median is shown for all groups: open circles=controls; grey circles=burn patients at 1 hour post-admission; black circles=burn patients at 40 hours post-admission. Fractions were assessed in duplicate with wells containing 1μl of sterile water and 1 μl LL-37 (100 μM; GeneScript) as negative and positive controls, respectively.

Figure 4. Bacterial growth inhibition of E. faecalis with urine fractions from control patients compared to the same fractions of burned patients at 1 hour and 40 hours post-admission

A) Zones of bacterial growth inhibition of E. faecalis present in fractions 7-12 withcontrol urine and fractions from burn patients 1 and 40 hours post-admission. B) Antimicrobial activity against E. faecalis is altered following burn injury in the majority of urine fractions and antimicrobial activity against E. faecalis demonstrates notable changes in multiple fractions. Arrow to the right indicates increasing peptide hydrophobicity. The median is shown for all groups: open circles: controls; grey circles: burn at 1 hour post-admission; black circles: burn at 40 hours post-admission. Fractions were assessed in duplicate with wells containing 1μl of sterile water and 1 μl LL-37 (100 μM; GeneScript) as negative and positive controls, respectively.