DNases improve effectiveness of antibiotic treatment in murine polymicrobial sepsis

Abstract

Introduction: Neutrophil extracellular traps (NETs) have various beneficial and detrimental effects in the body. It has been reported that some bacteria may evade the immune system when entangled in NETs. Thus, the aim of the current study was to evaluate the effects of a combined DNase and antibiotic therapy in a murine model of abdominal sepsis.

Methods: C57BL/6 mice underwent a cecum-ligation-and-puncture procedure. We used wild-type and knockout mice with the same genetic background (PAD4-KO and DNase1-KO). Mice were treated with (I) antibiotics (Metronidazol/Cefuroxime), (II) DNAse1, or (III) with the combination of both; mock-treated mice served as controls. We employed a streak plate procedure and 16s-RNA analysis to evaluate bacterial translocation and quantified NETs formation by ELISA and immune fluorescence. Western blot and proteomics analysis were used to determine inflammation.

Results: A total of n=73 mice were used. Mice that were genetically unable to produce extended NETs or were treated with DNases displayed superior survival and bacterial clearance and reduced inflammation. DNase1 treatment significantly improved clearance of Gram-negative bacteria and survival rates. Importantly, the combination of DNase1 and antibiotics reduced tissue damage, neutrophil activation, and NETs formation in the affected intestinal tissue.

Conclusion: The combination of antibiotics with DNase1 ameliorates abdominal sepsis. Gram-negative bacteria are cleared better when NETs are cleaved by DNase1. Future studies on antibiotic therapy should be combined with anti-NETs therapies.

Keywords: NETs; abdominal sepsis; appendicitis; biomarker; extracellular traps; neutrophils; prospective.

Figure 1

Macroscopic differences and colocalization of MPO/H3cit in immunofluorescence staining. (A) Open situs of a control animal. Note the swelling compared to (C). (B) Open situs of a D1 knockout, treated with antibiotics. Note the vascular injections, adhesions, and swelling compared to (C). (C) Open situs of a PAD4 knockout, treated with antibiotics. (D) Wide field image of IF staining of the lung, obtained from control animal (MPO green, H3cit red, DAPI blue), and assessment of colocalization (only places where both signals are present are colored). (E) Wide field image of IF staining of intestine, obtained from PAD4-KO treated with antibiotics (MPO green, H3cit red, and DAPI blue) and assessment of colocalization (only places where both signals are present are colored). (F) Evaluation of morphometry and colocalization of MPO/H3cit in the lung, intestine, and liver. Highest level of colocalization in the lung tissue was observed in wild-type mice treated with antibiotics. Significantly less colocalization was observed in the intestine of PAD4-mice treated with antibiotics compared to control animals. Highest level of colocalization in the intestine was observed in wild-type mice treated with antibiotics. Significantly less colocalization was observed in the liver tissue in PAD4-mice treated with antibiotics compared to control animals. Data shown as mean ± SD. Statistics: for comparison, one-way ANOVA with Dunnett´s correction.

Figure 2

Assessment of survival, sepsis severity, and bacterial clearance. (A) Survival in wild-type mice, shown by the number of mice in the cohorts in the figure legend: treatment with D1 did not affect survival; however, antibiotic treatment improved the outcome. Mice treated with antibiotics and DNAse1 48 h after CLP procedure showed the best survival. Logrank test for trend showed no significant difference. (B) Survival in genetically modified mice, shown by the number of mice in the cohorts in the figure legend: PAD4-knockout mice treated with antibiotics showed significantly improved survival in the logrank test for trend (p=0,0294) compared to DNAse1-knockout mice. (C) Murine sepsis score in wild-type mice 96 h after CLP procedure: treatment only with D1 or antibiotics did not affect the outcome; however, combined treatment showed significantly lower sepsis score. (D) Murine sepsis score in genetically modified mice 96 h after CLP procedure: mice treated with antibiotics with reduced amount and size of NETs (PAD4-knockout) showed the lowest sepsis score compared to the other groups. (E) Assessment of nucleosomes: found significantly less NETs degradation products (nucleosomes/histones) in wild-type mice, when treated with D1 or antibiotics, particularly in mice, treated with the combination of antibiotics and D1. (F) Assessment of bacterial colonization in liver and lung tissues in wild-type mice: bacterial clearance was significantly improved in mice, treated with D1 or antibiotics. Best bacterial clearance was observed in mice treated with the combination of D1 and antibiotics. (G) Assessment of in-sample biodiversity via 16s-rRNA-gene sequencing analysis: bacterial biodiversity was significantly reduced in mice receiving the combination of D1 and antibiotics. Data shown as mean ± SD. Statistics: for comparison, one-way ANOVA with Dunnett’s correction or Kruskal–Wallis test with Dunn’s correction.

Figure 3

Spectral analysis of bacterial colonization in (A) intestine, (B) liver, (C) lung, and (D) Spleen. Not all bacteria seem to be equally affected by anti-NETs therapy. However Gram-negative bacteria are cleared significantly better when NETs are dissolved by D1 and treated with antibiotics. Data shown as mean ± SD. Statistics: for comparison, one-way ANOVA with Dunnett’s correction. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.