Phospholipase activity in human mesenteric ischaemia and infarction

Currently, diagnostic tests for mesenteric ischaemia and infarction are inadequate due to poor sensitivity and specificity. In addition, many potential markers appear too late to be clinically useful. At present, definitive diagnosis can only be made at the time of surgery, which is not ideal as surgery is often to be avoided in critically ill and elderly patients. A clinically useful, minimally invasive test is likely to decrease the currently very high mortality rate and allow monitoring of 'at risk' patients during their hospital stay. A two-dimensional electrophoresis based proteomic approach was undertaken to assess plasma protein differences between patients with surgically confirmed bowel infarction and control Intensive Care patients. The major protein differences were found to be members or variants of acute phase proteins. Serum amyloid A showed the largest difference between the two patient groups, and this protein was investigated in greater depth. An analysis was performed to compare the diagnostic ability of several commonly used indicators of critical illness and bowel infarction with serum amyloid A and phospholipase A2. Although none of the variables were ideal for clinical use, plasma phospholipase A2 activity showed the best discriminatory power, as determined by Receiver Operating Characteristic curves. From a review of the literature, phospholipase AI (PLA2) appeared to be increased in the bowel as a result of ischaemia and infarction. In one patient, matched tissues were obtained, and PLA2 activity was found to be significantly higher in infarcted bowel tissue compared to ischaemic bowel tissue. PLA2 activity was significantly greater in bowel lumen than tissue, suggesting that the protein was being released, and may enter the circulation. PLA2 activity was increased in the plasma of bowel infarction patients compared with control patients, though the difference was not significant. The phospholipase activity exhibited a number of similarities to typical phospholipase A2 proteins, but also showed a number of inconsistent characteristics. For this reason, we wished to identify the protein responsible for the increased phospholipase activity in infarcted human bowel. The PLA2 activity in human bowel could not be abolished by immunoprecipitation of the PLA2 isoforms IIA (well described in bowel) and V (a closely related isoform). To investigate these proteins, a native urea protein gel devised for snake venom phospholipase A2 was modified for use with mammalian phospholipase AI. The modified gel was used to show that the protein with phospholipase activity from infarcted gut was different from normal gut PLA2 and type IIA PLA2. A number of extensions were devised for these native gels and were found to be useful both in this investigation and for venom investigations. Protein purification was undertaken to identify the protein responsible for the increased phospholipase activity in infarcted bowel. Protein was purified from infarcted human bowel using a number of techniques that exploited unusual characteristics of the protein. The purification techniques each retained the native activity of the protein and the purification could therefore be monitored with a phospholipid hydrolysis assay at each stage. The protein identified by mass spectrometry was an excellent match for cyclophilin B, an inflammatory protein that had previously been identified in rat bowel at the mRNA level (Hasel et al, 1991, Kainer & Doris, 2000). As the purification progress had been monitored throughout with a phospholipid hydrolysis assay, cyclophilin B was an unexpected identification, as it is not known to have phospholipase activity. Cyclophilin B was removed from the highly purified samples via immunoprecipitation and this process abolished all phospholipase activity. The addition of cyclosporin A, (the pharmaceutical ligand of cyclophilin B), did not effect the phospholipase activity. Cyclophilin B protein was found in normal and infarcted human bowel using Western blotting. Cyclophilin B protein also appeared to be present in the bowel lumen and plasma of several patients with bowel infarction, but not in control patients. Immunohistochemistry confirmed the ubiquitous nature of cyclophilin B that had been reported by other groups. This project has investigated the use of two dimensional gel electrophoresis based proteomics to identify proteins present in the plasma of patients with confirmed bowel infarction and control intensive care patients. The major protein classes observed were members of the acute phase proteins, which highlights the need for pre-fractionation of plasma to identify lower abundance, disease associated proteins. A series of potential plasma markers were compared using Receiver Operating Characteristic Curves. Although no ideal marker was clear from this analysis, phospholipase activity appeared to warrant further investigation. Phospholipase activity was investigated in human infarcted bowel. Protein purification identified cyclophilin B as a bowel protein that showed unusual phospholipid hydrolysing activity. Cyclophilin B is a ubiquitous protein in intestinal cell types in both normal and infarcted tissue. There appears to be release of cyclophilin B into bowel lumen and plasma under conditions of mesenteric ischaemia and infarction.