The development of quantitative in vitro methods for the analysis of various mediators of the inflammatory response has led to a greater understanding of the functions of such mediators. Indeed, recent investigations have suggested that the complement (C) ~ system plays a central role as an effector of a number of events in the inflammatory process (1, 2). A point which has emerged from various in vitro studies, however, is that there are a multiplicity of mediators for any given biological function. For example, fragmentation products from each of the complement, clotting, and kinin systems have been reported to increase vascular permeability, contract smooth muscle, and chemotactically attract polymorphonuclear leukocytes (PMN) (3). Important questions for future investigations will be aimed at determining the relative quantitative and qualitative importance in vivo of each of the known biological effectors in mediating any given aspect of the inflammatory process.A cardinal event in the acute inflammatory response is the accumulation of PMN at the inflammatory site. While we have previously discussed the importance of the fifth component of C (C5) in the generation of PMN, as well as mononuclear leukocyte chemotactic activity in vitro (4, 5), information on the role C5 plays in mediating the local accumulation of leukocytes in vivo is quite sparse. In this report, data will be presented which suggest an important role for C5 in the early accumulation of PMN in inflammatory exudates. Materials and MethodsAnimals.--Hartley-strain albino male guinea pigs were obtained from the Animll Resources Branch of the National Institutes of Health. Male mice of the lines B10D2/SN "new" (C5-normal) and B10D2/SN "old" (C5-deficient) (6) were purchased from Jackson Laboratories, Bar Harbor, Maine.Inflammatory Stimulants.--SheUfish glycogen (Mann Research Labs. Inc., New York) was used at a final concentration of 0.5% (w/v) in pyrogen-free saline. Endotoxin derived from I Abbreviations used in this paper: BSA, bovine serum albumin; C, complement; CS-deficient, B10D2/SN "old"; C5-normal, B10D2/SN "new"; GVB, gelatin Veron~l buffer; PBS, phosphate-buffered saline; PMN, polymorphonuclear leukocytes.

Endotoxic lipopolysaccharides (LPS) isolated from Serratia marcescens, Veillonella alcalescens, and Salmonella typhosa were potent in their ability to induce fixation of complement (C') in normal guinea pig, rabbit, mouse, and human serum. The C'-fixing ability of LPS was pronounced even when assays were performed in undiluted serum, and was lost after each of four chemical modifications which resulted in loss of biological toxicities. The detoxification procedures had in common the cleavage of ester-bound, long-chain carboxylic acids. The ability of biologically active LPS to fix C' in normal guinea pig serum was reflected chiefly in dramatic uptake of classical C'3 (C'3t); fixation of C'1, C'4, and C'2 was virtually undetectable. Hence, it was the capacity for fixation of C'3t which was lost most overtly during detoxification. Addition of immune serum to the assay mixtures resulted in detectable fixation of C'1 and C'4. Biologically active LPS also fixed more of these components than did detoxified LPS. Immune serum restored the ability of detoxified LPS to fix C'3t, but whether this is by the original pathway is not yet clear. We concluded that the loss of certain biological activities and the loss of ability to fix C'3t in normal serum after LPS detoxification involved loss or rearrangement of substrates on LPS which either initiated or supported, or both, its interaction with the complement system. It was apparent that the ability to fix C' can serve as a valuable in vitro indicator of the integrity of the toxic conformation of biologically active LPS membrane fragments. These experiments supported the hypothesis that certain of the biological activities induced by endotoxins are mediated via the complement system.

A number of polymorphonuclear leukocyte (PMN) chemotactic factors derived from complement components have been described. We sought to determine which of these factors accounted for the majority of PMN chemotactic activity in rabbit serum and guinea pig serum treated with preformed immune complexes. Normal rabbit sera, guinea pig sera, or rabbit sera deficient in C6 were treated with homologous antibovine serum-albumin-bovine serum-albumin complexes. Sera so treated contained one major PMN chemotactic factor which was heat-stable (56 C for 30 min), had a molecular weight of approximately 15,000, and did not require the presence of C6 for its generation. This factor appears to be analogous to C5a.