Oral Polyphosphate Suppresses Bacterial Collagenase Production and Prevents Anastomotic Leak Due to Serratia marcescens and Pseudomonas aeruginosa

Abstract

Objective: The objective of this study was to determine the effect of polyphosphate on intestinal bacterial collagenase production and anastomotic leak in mice undergoing colon surgery.

Background: We have previously shown that anastomotic leak can be caused by intestinal pathogens that produce collagenase. Because bacteria harbor sensory systems to detect the extracellular concentration of phosphate which controls their virulence, we tested whether local phosphate administration in the form of polyphosphate could attenuate pathogen virulence and prevent leak without affecting bacterial growth.

Methods: Groups of mice underwent a colorectal anastomosis which was then exposed to collagenolytic strains of either Serratia marcescens or Pseudomonas aeruginosa via enema. Mice were then randomly assigned to drink water or water supplemented with a 6-mer of polyphosphate (PPi-6). All mice were sacrificed on postoperative day 10 and anastomoses assessed for leakage, the presence of collagenolytic bacteria, and anastomotic PPi-6 concentration.

Results: PPi-6 markedly attenuated collagenase and biofilm production, and also swimming and swarming motility in both S. marcescens and P. aeruginosa while supporting their normal growth. Mice drinking PPi-6 demonstrated increased levels of PPi-6 and decreased colonization of S. marcescens and P. aeruginosa, and collagenase activity at anastomotic tissues. PPi-6 prevented anastomotic abscess formation and leak in mice after anastomotic exposure to S. marcescens and P. aeruginosa.

Conclusions: Polyphosphate administration may be an alternative approach to prevent anastomotic leak induced by collagenolytic bacteria with the advantage of preserving the intestinal microbiome and its colonization resistance.

Fig. 1. Polyphosphate suppresses the collagenolytic activity, biofilm production, and motility of S2-S. marcescens

(A), Dynamics of gelatin-degrading activity in S. marcescens grown in TSB. n=6. (B), PPi-6 suppresses the collagenolytic activity of S. marcescens (blue color curve) in a concentration dependent manner (n=6/group, *p< 0.01 between groups of PPi-6=0 and PPi-6 ≥0.05 mM) without affecting its growth (black color curve). (C) Gelatin and collagen I but not collagen IV are degraded by S. marcescens. n=6/group. (C), (D), PPi-6 does not affect the gelatin-degrading activity of the S. marcescens supernatant. (E), PPi-6 suppresses biofilm production by S. marcescens (blue color curve) in a concentration dependent manner (n=6/group, *p< 0.01 between groups of PPi-6=0 and PPi-6 ≥10 mM) without affecting its growth. n=4/group, *p< 0.01. (F), PPi-6 suppresses swimming and swarming motilities in a concentration dependent manner.

Fig. 2. Mouse model of S2- S. marcescens –induced anastomotic leak

(A), Anastomosis healing score (AHS): 0= normal healing, 1= loose adhesions, 2 = dense adhesions and inflammation, 3= abscess with dense adhesions/inflammation and 4= gross leak. Arrows indicate the line of the anastomosis. (B), Introduction of S2-S. marcescens to anastomotic tissues and antibiotic exposure induce anastomotic leak. n= 5 for anastomosis only, n=10 for anastomosis+S1, n=4 for anastomosis+S2, n=10 for anastomosis+S2+antibiotic, *p<0.01 for anastomosis+S2 +antibiotic (Chi-square test). (C), Antibiotic treatment promotes higher colonization of anastomotic tissues by S. marcescens. n=4 for anastomosis+S2, n=5 for anastomosis+S2+antibiotic, *p<0.01 (Student t-test). (D) Schematic presentation of the model and experimental protocol. On POD0, a distal colonic transection and anastomosis is performed. Antibiotics are administered twice daily, starting one day prior to surgery and continued until POD3. S2-S. marcescens is delivered by enema at POD1, POD2, and POD3, twice a day, 100 µl of S. marcescens solution in 10% glycerol, OD600 nm = 1.0.

Fig. 3. PPi-6 promotes healing in mouse model of S2- S. marcescens –induced anastomotic leak

(A–C), AHS is improved by oral PPi-6 as seen by the absence of gross leaks (AHS3 and AHS4) (A), significant decrease of the mean AHS, n=10, p<0.05 (B), no abscesses in healed tissues (C). (D), PPi-6 concentration is increased at the site of anastomosis in mice drinking PPi-6 n=5 for anastomosis only, n=4 for S2+ antibiotic, n=4 for S2=antibiotic +PPi-6, *p<0.05 (Student t-test) (E), colonization of anastomotic tissues by S. marcescens is attenuated by oral PPi-6, n=5/group, *p<0.01 (Student t-test) (F), total collagenase activity at anastomotic tissues is decreased by oral PPi-6, n=5, *p<0.01 (Student t-test).

Fig. 4. PPi-6 promotes healing in P. aeruginosa–induced anastomotic leak

(A), PPi-6 suppresses the gelatin-degrading activity of P. aeruginosa MPAO1-P2 grown in TSB. n=6. (B), PPi-6 suppresses biofilm production by P. aeruginosa MPAO1-P2. (C, D), PPi-6 suppresses swarming (C) and swimming (D) motility in MPAO1-P2. (E–G), oral PPi-6 causes a significant decrease in the mean AHS, n=10, p<0.05 (E); attenuation of P. aeruginosa colonization at anastomotic tissues, n=5, *p<0.01 (F) decrease of total collagenase activity at anastomotic tissues, n=5, *p<0.01 (G).