Endogenous endotoxin participates in causing a panenteric inflammatory ileus after colonic surgery

Abstract

Objective: To investigate muscularis inflammation and endogenous endotoxin as causes of postoperative ileus.

Background: Postoperative inflammatory ileus of the colon is associated with a significant delay in gastrointestinal transit. We investigated whether these changes are caused by the downstream obstructive barrier of the surgically altered colon or by small intestinal muscularis inflammation itself. Furthermore, we evaluated the mechanistic role of gut derived endotoxin in the development of postoperative intestinal dysfunction.

Methods: Rats underwent surgical manipulation of the colon. Isolated gastrointestinal transit was analyzed in animals with ileostomy. The perioperative emigration of intracolonic particles was investigated by colonic luminal injection of fluorescently labeled LPS and microspheres. Mediator mRNA induction was quantified by real-time RT-PCR. Muscularis leukocytic infiltrates were characterized. In vitro circular muscle contractility was assessed in a standard organ bath.

Results: Ileostomy rats presented with a significant delay in small intestinal transit after colonic manipulation. This was associated with leukocyte recruitment and inflammatory mediator mRNA induction within the small intestinal muscularis. Colonic manipulation caused the transference of intracolonic LPS and microspheres into the intestinal muscularis. Postoperative in vitro small intestinal circular muscle contractility was impaired by 42% compared with controls. Gut decontamination and TLR-4 deletion significantly alleviated the small intestinal muscularis inflammation and prevented intestinal muscle dysfunction.

Conclusions: Selective colonic manipulation initiates a distant inflammatory response in the small intestinal muscularis that contributes to postoperative ileus. The data provide evidence that gut-derived bacterial products are mechanistically involved in the initiation of this remote inflammatory cascade.

FIGURE 1. Transit histograms for the distribution of nonabsorbable fluorescein labeled dextran along the intestine of control rats (A) and rats with ileostomy (B) 90 minutes after oral administration (Stom, stomach; SB, small bowel; n = 5 each). A, Colonic manipulation results in a delay in upper gastrointestinal transit. B, In sham-operated animals (light gray bars), the transit marker accumulated in the terminal intestine and the feces. Colonic manipulation (dark gray bars) caused a significant delay in intestinal transit.

FIGURE 2. Bethanechol dose-response curves demonstrating impairment in small intestinal circular smooth muscle contractility 24 hours after colonic manipulation (n = 5 each). The contractile activity in response to the muscarinic agonist bethanechol (1, 3, 10, 100, and 300 μmol/L) was quantified in mid jejunal muscle strips from control, sham-operated, and colon-manipulated animals. Data are mean ± SEM.

FIGURE 3. Mechanical trace of spontaneous contractile activity recorded from small intestinal circular smooth muscle strips 24 hours after colonic manipulation. The muscle was subjected to a combined DFU (5 mg/L) and l-NIL (30 μmol/L) perfusion. Initial and final portions of the trace are shown to the bottom on an expanded time scale, demonstrating changes in contractile activity. The jejunal muscle contractions from a colonically manipulated animal shows a marked increase in spontaneous activity in presence of DFU and l-NIL.

FIGURE 4. Upper panel: Myeloperoxidase staining and ED1-immunohistochemistry of small intestinal muscularis whole-mounts demonstrating leukocyte recruitment after colonic manipulation. A and B, Occasionally a few MPO-positive cells and ED-1-positive resident macrophages were observed within control muscularis whole mounts. C and D, Infiltrating MPO- and ED-1-positive cells 24 hours after colonic manipulation (A–D, original magnification ×200). Lower panel: E, Histogram showing a significant increase in the number of infiltrating PMNs and monocytes after colonic manipulation compared with controls and shams (n = 5 each). Leukocytes were counted at a magnification of ×200. Data are mean ± SEM.

FIGURE 5. Histograms showing the results of real-time RT-PCR analysis of inflammatory mediator mRNAs in the small intestinal muscularis after sham operation (at 3 hours) and colonic manipulation (3, 6, and 18 hours postoperatively), compared with control expression (CM, colonic manipulation; n = 4–6). IL-6-, ICAM-1-, MCP-1- and iNOS-mRNA exhibited a single expression peak 3 hours after colonic manipulation, whereas COX-2 peaked at 3 hours and again at 18 hours postoperatively. Note that all mediators, except iNOS, were up-regulated in sham operations, which included laparotomy with cecal eventration (F). Data are mean ± SEM.

FIGURE 6. Fluorescence histology in colonic (A–C) and jejunal (D, E) muscularis whole mounts after intracolonic injection of LPS-Alexa 488 and colonic manipulation. No fluorescence was seen in preparations from sham-operated rats (A, colon, D, jejunum). Colonic manipulation resulted in a distinct uptake of the fluorescently labeled LPS in stellate shaped cells within the colonic muscularis 3 hours postoperatively (B). Using ED1-immunohistochemistry, these cells were identified as macrophages (C). Intestinal muscularis whole mounts exhibited only weak fluorescence after colonic manipulation (E) (A–E, original magnification ×200).

FIGURE 7. Appearance of intraluminally administered fluorescently labeled microspheres in the colonic (A–C) and jejunal muscularis (D–F) 3 hours after colonic manipulation. There were only a few microspheres in the colonic (A) and small intestinal muscularis (D) of sham-operated animals. Colonic manipulation caused a massive uptake of green colored microspheres (B; 0.05 μm in diameter) and a moderate accumulation of the larger, orange microspheres (C; 0.5 μm in diameter) in the colonic muscularis. Microspheres of the same sizes, indeed with less quantity, were also seen in the small intestinal muscularis after colonic manipulation (original magnifications: A, B, D–F: ×200; C: ×100).

FIGURE 8. Bethanechol dose-response curves demonstrating the involvement of bacterial products in the impairment of small intestinal circular smooth muscle contractility 24 hours after colonic manipulation. A, Pretreatment with antibiotics significantly prevents small intestinal smooth muscle dysfunction after colonic manipulation (n = 5 each). The dose-response curves in B depict a significant improvement of postoperative small intestinal muscle contractility after colonic manipulation of Toll-like receptor 4 knockout mice compared with manipulated wild-type mice (n = 6–8). Data are mean ± SEM.