Shock and multiple organ failure exhibit a powerful inflammatory response with an associated activation of leukocytes and endothelial cells (1–4), as well as evidence for parenchymal cell dysfunction (5). A number of inflammatory mediators have been implicated in this process, including complement activation products (6), endotoxin (7), cytokines (7,8), various lipids (9,10), and peptides (6,11). The evidence suggests that various bioactive factors may be responsible for inducing cellular activation resulting in lethal shock.
Recently, we obtained evidence that the pancreas may be a primary tissue source of factors capable of inducing shock-like phenomena. It was further shown that these “shock” factors are generated by the pancreatic digestive enzymes (12–15). Pancreatic homogenates produced symptoms characteristic of shock, including high mortality, when injected into an animal (12,15). Pancreatic proteinases, such as trypsin and chymotrypsin, were capable of generating powerful inflammatory mediators when incubated with organs other than the pancreas, and these mediators mimicked the in vivo effects observed with the pancreatic homogenates (12). These observations suggested that the mechanism for generating potent mediators in tissue depended on a degradation (i.e., proteolytic) process. Blockage of the pancreatic serine enzymes by group-specific inhibitors markedly reduced the in vivo production of these inflammatory shock-inducing mediators and prevented the pathophysiologic response (16,17).
To initiate a more systematic analysis of the possible inflammatory mediators produced by degradative enzymes in tissue, we present here a separation of pancreatic homogenates into lipid (hydrophobic) and protein/peptide (hydrophilic) fractions. These fractions were then assessed for their ability to stimulate neutrophils (i.e., inflammatory response) and to induce systemic shock-like phenomena in the rat. This preliminary biochemical analysis is a necessary first step in fully chemical characterization of the factors, understanding the various mechanisms that produce the shock-induced symptoms, and a prerequisite in developing effective new treatments or interventions.
MATERIALS AND METHODS
Porcine pancreas was obtained fresh frozen from an abattoir (Sierra for Medical Science, Santa Fe Springs, CA). The pancreas was thawed and homogenized in approximately three volumes of normal saline at room temperature and was then coarse-filtered to remove connective tissue. The homogenates (preparations 1, 2, and 3 extracted from 370, 450, and 480 mg of wet weight of tissue per mL, respectively) were partitioned and a portion was subjected to organic extraction (18), either immediately or after incubation for 2.5 h at 37°C. The two homogenates, and fractions derived from them, are referred to here as incubated or nonincubated. The aqueous and organic phases of the chloroform/methanol /water mixture were separated and evaporated to dryness. The organic and aqueous extracts were suspended in chloroform or water, respectively, at 10% of the original homogenate volume. The aqueous fractions were filtered and then both fractions were stored at −20°C. The materials were assayed and determined to be free of endotoxin before use (19). Protein content was determined using a ninhydrin assay after acid hydrolysis (20), and phospholipid levels were estimated by phosphorous determination (21). Aliquots of the organic and aqueous extracts were taken to dryness and were dissolved in sterile saline for all animal experiments. Four separate preparations were performed in this manner.
Human neutrophils were obtained from blood samples of healthy donor volunteers under an approved Human Subjects protocol (LJIMM no. 96-293). Each donor signed an informed consent form agreeing to be an unidentified voluntary donor. Whole blood was separated using a Percoll gradient according to the procedure of Pertoft (22). The band containing neutrophils was collected, washed, and suspended in Hank's balanced salt solution (HBSS). Cell counts and viability were determined before use, and the preparations contained greater than 98% neutrophils.
Neutrophil granular release was determined by the myeloperoxidase (MPO) assay (23). Cytochalasin B-treated cells (2 × 105) in 25 μL of HBSS containing 0.25% bovine serum albumin (BSA) were incubated with 50 μL of sample in HBSS and BSA for 10 min at room temperature. Then, 50 μL of 0.2 M sodium phosphate buffer, pH 6.2, followed by 25 μL of 3.9 mM 3, 3`-dimethoxybenzidine (DMB) and 15 mM H2O2 was added to measure enzyme release. The enzyme level was detected by measuring color development as determined by reading the absorbance at 450 nm.
Oxygen free radical formation
Oxygen burst (superoxide) was determined by chemiluminescence using lucigenin as an indicator (24). Assays were performed in opaque microtiter plates. Neutrophils (4 × 105 cells/mL) in HBSS, containing 2 mg/mL BSA, were incubated at 37°C for 15 min with 0.5 μM lucigenin. Ten microliters of test material was then added to 100 μL of the cells and the emission of relative light units (RLUs) were monitored in a luminometer for 20 min with readings every minute. Data points represent the mean of three wells.
Aqueous and lipid fractions were titrated in both assays to determine relative potency and endpoint concentrations. Costimulation studies were performed by incubating the cells for 15 min with a substimulatory level of a known neutrophil activator (i.e., formyl-methionyl-leucyl-phenylalanine [FMLP] or platelet-activating factor [PAF]), followed by the addition of a substimulatory level of the hydrophobic or hydrophilic extract.
Precast Novex Tris-glycine gels (Invitrogen, Carlsbad, CA) were used for the SDS-PAGE separation of the aqueous extracts. The samples were dissolved in the pH 8.6 Tris-glycine SDS sample buffer and were heated to 85°C for 2 min. Ten microliters of sample was loaded into each well. Electrophoresis was carried out at 125 V for 90 min. The gel was stained used the SilverQuest staining kit (Invitrogen). Molecular weights were estimated using the Novex Mark 12 unstained standard mix.
Animals and surgical preparation
Animal experiments were reviewed and approved by the University of California at San Diego Animal Use and Care Committee. All surgical manipulations were performed in adherence to the National Institutes of Health Guidelines on the Use of Laboratory Animals.
Male Wistar rats (350–450 g; Charles River Laboratories, Wilmington, MA) were housed in a controlled environment and were maintained on a standard pellet diet for at least 3 days before initiation of experimental procedures. Anesthesia was administered i.p. using 50 mg/kg sodium pentobarbital (Abbott Laboratories, North Chicago, IL). The right jugular vein was cannulated with PE50 tubing (Clay Adams, Parsippany, NJ) to provide additional anesthesia as necessary as determined by toe pinch reflex. The left femoral artery was cannulated using PE50 tubing (Clay Adams) to assess mean arterial blood pressure and heart rate throughout the course of the experiment.
Homogenates and extracted fractions were injected into the systemic circulation of the animals via the jugular vein. To normalize the injection volume, each animal received 0.5 mL of the experimental fraction per 100 g of animal weight. Cardiovascular overload was avoided by introducing the fraction over 45 s. The animals were then observed for 60 min. Blood pressure and heart rate were continuously recorded using a Statham Pressure Transducer (Gould Medical Products, Oxnard, CA) amplified by a CTH-2000 amplifier (CB Sciences, St. Dover, NH). The results were recorded onto a G3 Macintosh computer using the MacLab 8E interface (AD Instruments, Mountain View, CA). Animals surviving the entire observation period were sacrificed with an intravenous overdose of sodium pentobarbital (100 mg/rat; Abbott Laboratories).
Whole pancreatic homogenates, and the lipid and aqueous extracts derived from them, were evaluated for their ability to activate neutrophils. Figure 1 shows the ability of these preparations to cause the release of MPO from human neutrophils. Both nonincubated and incubated homogenates induced the release of MPO, however, the incubated homogenate was more potent than the nonincubated homogenate. A similar result was obtained when the lipid extracts from the incubated and nonincubated homogenates were compared. The hydrophobic fraction from the incubated homogenate gave measurably higher levels of enzyme release than the corresponding fraction from nonincubated homogenate. No MPO release was observed from cells exposed to corresponding levels of the aqueous extracts, indicating that the material in these fractions failed to induce neutrophil activation. The MPO release induced by 5 μM FMLP is shown for comparison.
Both incubated and nonincubated pancreatic homogenates were capable of inducing an oxygen free radical (superoxide) burst in neutrophils, another independent indicator of cellular activation (Fig. 2). The incubated homogenate appears to be approximately as potent as micromolar FMLP, whereas the nonincubated homogenate is again less active. It should be noted that even the nonincubated homogenate came from ischemic tissue, and therefore one would expect that some level of enzyme digestion (i.e., mediator generation) had already taken place while harvesting the pancreas.
Cellular coactivation studies
The lipid fraction from the incubated homogenate stimulates the free radical burst in the neutrophils (Fig. 3) with an efficacy comparable with that of the whole homogenate or micromolar FMLP (see Fig. 2). As in the previous MPO experiment, there was no neutrophil activation by the aqueous fractions from either the incubated or nonincubated homogenates (data not shown). These data indicate that the ability of pancreatic homogenate to activate neutrophils is exhibited by only the lipid extracts and not the aqueous fractions. Neither the 10-fold diluted lipid extract from the incubated homogenate nor the undiluted lipid extract from the nonincubated homogenate induced an oxygen burst in these cells. These results suggest that the active lipid factors were enzymatically generated over time.
The possibility that the lipid fraction could act in concert with other neutrophil stimuli was examined. Based on the titration data obtained from our previous studies, neutrophils were exposed to substimulatory (1:10) levels of the lipid fractions. After 15 min, a substimulatory dose (0.1 μM) of either PAF or FMLP was added to the cells and the production of superoxide was monitored. Neither the primary nor secondary agonists alone could induce superoxide production (Fig. 3). However, when the lipid-primed cells were subsequently exposed to the same concentration of PAF or FMLP, measurable superoxide release was observed. These results indicate that the lipid fraction can act in concert, either additively or cooperatively, with other stimuli to activate neutrophils. Preincubation of the cells with the aqueous fraction followed by 0.1 μM PAF or FMLP resulted in no increase in O2− release (data not shown). This suggests that the aqueous fraction (unlike the lipid fraction) does not act as an enhancing or priming agent for neutrophils.
In vivo studies
Intravenous injection of clarified pancreatic homogenate into an animal has previously been shown to induce severe shock-like symptoms in the rat (12). The question that we addressed with the first series of animal experiments was whether the pancreatic homogenate could induce symptoms of shock and did the response intensify after an incubation period. Rats were injected with nonincubated and incubated pancreatic homogenate (0.5 mL per 100 g body weight) and were monitored for changes in heart rate and blood pressure. Figure 4A shows a representative experiment. Nonincubated homogenate induced an apparent decrease in blood pressure and heart rate immediately postinjection, followed by a gradual return to normal levels after about 60 min. Injection of an equal volume of the incubated homogenate resulted in a profound drop in both blood pressure and heart rate that resulted in the animal's death (Fig. 4B). No mortality was observed (0/3) in animals that received the nonincubated homogenates, whereas three of four animals administered the incubated homogenate died and all exhibited extensive decreases in both blood pressure and heart rate. These in vivo results are summarized in Table 1.
Lipid fractions were emulsified in sterile saline and the phospholipid content was determined (21). Because the cell activation studies indicated that the lipid extracts from the incubated homogenates were more potent in stimulating neutrophils than were extracts from the nonincubated homogenates, it was decided to evaluate only the lipid extracts from incubated homogenates in vivo. Animals were injected with approximately 200 μM phospholipid and the hemodynamic parameters were observed. Three of four lipid preparations resulted in dramatic depression of both blood pressure and heart rate and subsequent mortality. The fourth animal exhibited a marked decrease in these parameters but survived. Figure 5 shows a typical rat experiment where injection of the lipid extract causes a rapid lethal response. These results further indicate that there are bioactive lipids in the pancreatic homogenate capable of inducing both cell activation and lethal shock.
Aqueous fractions were administered to the animals as described in the lipid extract studies. Protein analysis indicated that the animals received 80 to 100 mg of protein/peptide per injection. The concentrated aqueous fraction from a nonincubated homogenate was tested and the results are shown in Figure 6A. Shock symptoms (i.e., decrease in heart rate and blood pressure) were induced immediately and the animal died approximately 8 min after injection. Aqueous material (80–100 mg) from an incubated pancreas homogenate (Fig. 6B) induced a dramatic decrease in blood pressure and heart rate and even more rapid mortality. All aqueous fractions tested from incubated homogenates were lethal (n = 4). This result clearly indicates that there are nonlipid (i.e., hydrophilic) moieties in the pancreatic homogenate capable of inducing lethal shock phenomena. Table 1 summarizes the results of these in vivo studies.
Characterization of the aqueous extracts
Figure 7 shows the electrophoresis patterns obtained for a number of aqueous extracts from various pancreatic homogenates. The 1–4 I fractions represent four separate aqueous extracts from incubated homogenates. The 1–4 N fractions represent four separate aqueous extracts from nonincubated homogenates. The patterns observed were very similar, having a number of visible bands at less that 20 kDa. The most densely staining bands are less than 6 kDa, and peptides less than 3 kDa may have diffused out of the gel during fixation and staining. It appears that there may be more material in the incubated than in the nonincubated samples, especially in preparations 1 and 3. Variations in the patterns of peptidic material observed for the various preparations is not surprising, even though homogenates were performed on equal wet weight portions of different pancreas. However, each of the aqueous extracts tested were lethal in the rat (see Table 1).
Attempts to isolate or identify factors capable of activating neutrophils from plasma or shock animals have failed because there was too little material available for isolation (25). However, factors exhibiting activities similar to the plasma factors have been demonstrated in organ homogenates, and the most abundant source of these factors is the pancreas (12). The present study confirms our earlier observations that pancreatic homogenate is an abundant source of factors capable of activating neutrophils and inducing acute shock-like phenomena in rats (12,15). The observations presented here indicate that the levels or potency of these active factors increased when the homogenates were incubated. The earlier result suggested that the active factors were generated as a result of degradative processes (i.e., proteolysis) and not just the release of preformed mediators already sequestered in the pancreatic tissue. The differences in activities observed in incubated versus nonincubated homogenates of pancreatic tissue, as presented in this study, further supports this hypothesis.
Physical separation of the lipids (i.e., hydrophobic fraction) from the hydrophilic or proteinaceous material permitted a further characterization of these tissue-derived shock factors.
The lipids in the hydrophobic fractions induced both neutrophil activation and a lethal shock reaction similar to the functions and effects known for PAF and other bioactive lipids. Others have demonstrated that the pancreatic homogenates not only activate circulating leukocytes, but also contain some of the key inflammatory elements in shock (26) that depress parenchymal cell function and survival (15,25). At present, the bioactive component(s) in the lipid fraction of these homogenates have not been identified, but are likely known mediators such as the various arachidonate products that exhibit many of the same biological effects. Preliminary studies in which we separated the lipids by thin-layer chromatography indicated that material recovered from the regions containing the neutral lipids and lyso-lipid components, but not the region containing PAF, activated the neutrophils (data not shown). Lysophosphatidylcholine (27) and eicosanoids (10), which should be contained in these two active lipid fractions, could be released from pancreatic tissue by the action of phospholipase A2 (28).
The priming effect of tissue lipid fractions on human neutrophils has previously been observed for lipids derived from mesenteric lymph (29). These priming effects may in part explain the apparent cooperation between tissue-derived lipid shock factors and products released from activated neutrophil such as PAF and LTB4.
The aqueous extracts, unlike the pancreatic homogenates, were proteinase free (data not shown) and failed to activate neutrophils based on the results observed in this study. Even without an ability to activate neutrophils, the hydrophilic material was effective in inducing a lethal shock-like response. These results demonstrated that neutrophil activation may be only one of several essential components involved in shock. This observation is in line with the fact that whole pancreatic homogenate induced direct parenchymal cell apoptosis in the microcirculation irrespective of whether leukocytes were present or not (12). Furthermore, evidence has suggested that neutrophils do play an important role in shock-induced lung injury by gut-derived inflammatory factors (17,29). Therefore, our results show that there may be multiple pathways leading to this lethal systemic event.
Our observation showing that the aqueous fraction did not act as a priming signal in human neutrophils was consistent with results obtained using aqueous extracts of mesenteric lymph from shock animals (29). There were similarities in effects observed between the lymph studies and the pancreatic homogenate studies. The aqueous fractions in each study failed to activate the human neutrophil, whereas the lipid extracts from both sources activated and primed the neutrophils. It was proposed that phospholipase A2 in the gut may be responsible for releasing the bioactive lipids in the mesenteric lymph (28).
Preliminary characterization of the aqueous fraction of pancreatic homogenate indicated that there was both high- and low-molecular-weight material in the mixture (14). The unexpected result seen in Figure 7 indicated that there were distinct bands of material in the mixture and that the majority of the material is less than a molecule weight of 20 kDa. Although it required milligrams of the crude aqueous extract to produce a lethal response in rats, the quantity of active component(s) needed to induce shock is probably much less, as will be determined as the purification of this mixture progresses. Based on the total activity in the homogenates and the relative activity expressed by the lipid and aqueous extracts, the potency of these two separated fractions appears to be comparable for inducing a lethal response (i.e., shock).
The current results support our basic hypothesis that the combination of factors released as a result of proteolytic digestion of pancreatic or other tissues produce both cell activation and organ dysfunction. Both our data and published evidence have demonstrated that powerful mediators exist in pancreatic homogenates and that the levels or potency increased as the tissue was incubated (12,13). Organ homogenates, other than the pancreas, incubated with proteinases also produced powerful inflammatory mediators, but not in the absence of the enzymes (12). Proteinases appear to provide the initial degradative action that exposes cellular elements such as lipids to the subsequent action of lipases. The consequence of these actions generates potent bioactive lipids, as well as protein fragments, that can each induce a shock response, but via different mechanisms. The lipids may act indirectly by stimulating neutrophils to release various mediators capable of attacking the vasculature, whereas the bioactive protein/peptides appear capable of attacking the vasculature directly. Many vasoactive peptides have been studied in the context of shock, including bradykinin (11), myocardial depressant factor (30), and tachykinins (31). There are also studies using crude mixtures of protein digests known as peptone that have been reported to induce shock (32,33). These observations suggest that peptide agonists may induce shock via their ability to interact with specific receptors, such as the NK-X (34) or PAR-X (35) receptors. Each of these receptors is known to participate in the inflammatory process.
In conclusion, two classes of mediators have been isolated from pancreatic homogenate and each is capable of inducing shock. It appears that prolonged digestion of the tissue enhances the level of these mediators in the homogenate. One class is hydrophobic and contains a mixture of lipids, and the other is hydrophilic and contains a complex mixture of protein fragments. Only the lipid mediators activated neutrophils, and both fractions proved lethal in the rat model. We hypothesize that the bioactive protein-derived factors act directly on the vasculature without the involvement of, or requirement for, activated neutrophils. Consequently, the action of these peptidic agents may occur at the level of the endothelium or the smooth muscle cells. Identification of an aqueous peptide fraction that contains potentially lethal shock factors, separate from the bioactive lipid fraction, is the first step in isolating and further characterizing specific protein-derived mediators, as well as determining their cellular targets.
The authors thank Mr. Sean LaPerruque and Mrs. Marleen Kawahara for their expert assistance. The assistance of Mr. Michael Oubre in the preparation of the manuscript is gratefully acknowledged.
1. Barry MC, Wang JH, Kelly CJ, Sheehan SJ, Redmond HP, Bouchier-Hayes DJ: Plasma factors augment neutrophil and endothelial cell activation during aortic surgery. Eur J Vasc Endovasc Surg 13:81–87, 1997.
2. Simpson R, Alon R, Kobzik L, Valeri CR, Shepro D, Hechtman HB: Neutrophil and non-neutrophil-mediated injury in intestinal ischemia-reperfusion. Ann Surg 218( 4):444–453, 1993.
3. Abello PA, Buchman TG, Bulkley GB: Shock and multiple organ failure. Adv Exp Med Biol 366:253–268, 1994.
4. Barroso-Aranda J, Zweifach BW, Mathison JC, Schmid-Schoenbein GW: Neutrophil activation, tumor necrosis factor, and survival after endotoxic and hemorrhagic shock. J Cardiovasc Pharmacol 25(Suppl 2):S23–S29, 1995.
5. Lefer AM: Interaction between myocardial depressant factor and vasoactive mediators with ischemia and shock. Am J Physiol 252(2 Pt 2):R193–R205, 1987.
6. Xiao F, Eppihimer MJ, Willis BH, Carden DL: Complement-mediated lung injury and neutrophil retention after intestinal ischemia-reperfusion. J Appl Physiol 82:1459–1465, 1997.
7. Endo S, Inada K, Yamada Y, Takakuwa T, Kasai T, Nakae H, Yoshida M, Ceska M: Plasma endotoxin and cytokine concentrations in patients with hemorrhagic shock. Crit Care Med 22:949–955, 1994.
8. Takakuwa T, Endo S, Nakae H, Kikichi M, Suzuki T, Inada K, Yoshida M: Plasma levels of TNF-α, endothelin-1 and thrombomodulin in patients with sepsis. Res Commun Chem Pathol Pharmacol 84:261–269, 1994.
9. Mozes T, Braquet P, Filep J: Platelet-activating factor: an endogenous mediator of mesenteric ischemia-reperfusion-induced shock. Am J Physiol 257(4 Pt 2):R872–R877, 1989.
10. Lefer AM: Eicosinoids: their role in splenic ischemia and shock. In Marston A, Buckley GB, Fiddian-Green RG, Haglund UH (ed):Splanchnic Arterial Occlusion Shock.
The C.V. Mosby Co., St. Louis, MO, 1989, pp 213–227.
11. Hoffmann TF, Steinbauer M, Waldner H, Messmer K: Exogenous bradykinin enhances ischemia/reperfusion injury of pancreas in rats. J Surg Res. 62: 144–151, 1996.
12. Kistler E, Hugli T, Schmid-Schönbein GW: The pancreas as a source of cell activating factors. Microcirculation 7:183–192, 2000.
13. Traverso LW, Gomez RR: Hemodynamic characterization of a canine pancreatic shock factor. Proc Soc Exp Biol Med 168:245–253, 1981.
14. Schmid-Schönbein GW, Kistler E, Hugli TE: Mechanisms for cell activation and its consequences for biorheology and microcirculation: multi-organ failure in shock. Biorheology 38:185–201, 2001.
15. Kistler E, Lefer A, Hugli T, Schmid-Schönbein GW: Plasma activation during splenic arterial occlusion shock. Shock 14:30–34, 2000.
16. Mitsuoka H, Kistler EB, Schmid-Schoenbein GW: Generation of in vivo activating factors in the ischemic intestine by pancreatic enzymes. Proc Natl Acad Sci USA 97:1772–1777, 2000.
17. Deitch EA, Shi HP, Lu Q, Feketeova, E, Xu DZ: Serine proteases are involved in the pathogenesis of trauma-hemorrhagic shock-induced gut and lung injury. Shock 19:452–456, 2003.
18. Bligh EG, Dyer WJ: A rapid method of total lipid extraction and purification. Can J Biochem Physiol 37:911–917, 1959.
19. E-TOXATE Technical Bulletin No. 210: Limulus Amebocyte Lysate Test, Sigma Chemical Co., St. Louis, MO. Revised April 2000.
20. Starcher B: A ninhydrin-based assay to quantitate the total protein content of tissue samples. Anal Biochem 292:125–129, 2001.
21. Fiske CM, Subbarow Y: The colorimetric determination of phosphorous. J Biol Chem 66:375–400, 1925.
22. Pertoft H: Fractionation of cells and subcellular particles with Percoll. J Biochem Biophys Methods 44:1–30, 2000.
23. Hasegawa H, Suzuki K, Nakaji S, Sugawara K: Analysis and assessment of the capacity of neutrophils to produce reactive oxygen species in a 96-well microplate format using lucigenin- and luminol-dependant chemiluminescence. J Immunol Methods 210:1–10, 1997.
24. Webster RO, Henson P: Rapid micromeasurement of neutrophil exocytosis. Inflammation 3:129–135, 1978.
25. Kramp WJ, Schmid-Shoenbein GW, Hoyt DB, Hugli TE: Pancreatic inflammatory mediators related to neutrophil activation and shock. FASEB J 16:A1194, 2002.
26. Schmid-Schoenbein GW, Hugli TE, Mitsuoka H, Kistler EB: Mechanisms for microvascular cell activation: pancreatic digestive enzyme derived inflammatory mediators. In Schmid-Schoenbein GW, Granger DN (ed):Molecular Basis for Microcirculatory Disorders.
Springer-Verlag, Paris, 2003, pp 255–268.
27. Ryborg AK, Deleuran B, Sogaard H, Kragballe K: Intracutaneous injection of lysophosphatidylcholine induces skin inflammation and accumulation of leukocytes. Acta Derm Venereol 80:242–246, 2000.
28. Gonzalez RJ, Moore EE, Ciesla DJ, Meng X, Biffi WL, Silliman CC: Post-hemorrhagic shock mesenteric lymph lipids prime neutrophils for enhanced cytotoxicity via phospholipase A2. Shock 16:218–222, 2001.
29. Dayal SD, Hauser CJ, Feketeova E, Fekete Z, Adams JM, Lu Q, Xu DZ, Zaets S, Deitch EA: Shock mesenteric lymph-induced rat polymorphonuclear neutrophil activation and endothelial cell injury is mediated by aqueous factors. J Trauma 52:1048–1055, 2002.
30. Leffler JN, Litvin Y, Barenholz Y, Lefer AM: Proteolysis in formation of a myocardial depressant factor during shock. Am J Physiol 224:824–831, 1973.
31. Severini C, Improta G, Falconieri-Erspamer G, Salvadori S, Erspamer V: The tachykinin peptide family. Pharmacol Rev 54:285–322, 2002.
32. Dragstedt CA, Mead FB: Further studies on the mechanism of peptone shock. J Pharmacol 63:400–406, 1937.
33. Feldberg W, O'Connor WJ: The liberation of histamine from the perfused lung by peptone. J Physiol (Lond) 90:288–295, 1937.
34. Patacchini R, Maggi CA: Peripheral tachykinin receptors as targets for new drugs. Eur J Pharmacol. 429: 13–21, 2001.
35. Dery O, Corvera CU, Steinhoff M, Bunnett NW: Proteinase-activated receptors: novel mechanisms of signaling by serine proteases. Am J Physiol 274:C1429–C1452, 1998.