The intravenous injection of LPS rapidly evokes fever. We have hypothesized that its onset is mediated by prostaglandin (PG)E2 quickly released by Kupffer cells (Kc). LPS, however, does not stimulate PGE2 production by Kc as rapidly as it induces fever; but complement (C) activated by LPS could be the exciting agent. To test this hypothesis, we injected LPS (2 or 8 μg/kg) or cobra venom factor (CVF, an immediate activator of the C cascade that depletes its substrate, ultimately causing hypocomplementemia; 25 U/animal) into the portal vein of anesthetized guinea pigs and measured the appearance of PGE2, TNF-α, IL-1β, and IL-6 in the inferior vena cava (IVC) over the following 60 min. LPS (at both doses) and CVF induced similar rises in PGE2 within the first 5 min after treatment; the rises in PGE2 due to CVF returned to control in 15 min, whereas PGE2 rises due to LPS increased further, then stabilized. LPS given 3 h after CVF to the same animals also elevated PGE2, but after a 30- to 45-min delay. CVF per se did not alter basal PGE2 and cytokine levels and their responses to LPS. These in vivo effects were substantiated by the in vitro responses of primary Kc from guinea pigs to C (0.116 U/ml) and LPS (200 ng/ml). These results indicate that LPS-activated C rather than LPS itself triggers the early release of PGE2 by Kc.
- portal vein cannulation
- cobra venom factor
- pyrogenic cytokines
fever develops quickly after the intravenous bolus administration of a pyrogenic dose of LPS to conscious guinea pigs, rats, and other species, but the afferent mechanism that induces this response is controversial. It is generally thought that it is mediated by pyrogenic cytokines produced secondarily in response to the LPS challenge, rather than by its direct action. Tumor necrosis factor-α (TNF-α), IL-1β, and IL-6 are the major cytokines implicated in this response (9). There is, however, a temporal disconnect between the first appearance of these cytokines and the onset of the febrile response to intravenous LPS; that is, fever appears within 10–15 min (13, 51) whereas TNF-α, the first cytokine to appear, is not detectable until 30 min after LPS treatment (18, 23, 24, 39). This temporal discrepancy is less evident after the intraperitoneal administration of low to moderate doses of LPS, when the latency of fever onset is ∼60 min but becomes evident also when higher doses are administered, when the onset latency approaches that after intravenous LPS (7).
We have shown previously that the onsets of the febrile responses to intravenous and intraperitoneal LPS are correlated with the appearance of LPS in the liver's Kupffer cells (Kc) (31), the body's principal clearinghouse of LPS (16, 34, 45) and source of pyrogenic cytokines (10). As cytokines are not constitutively expressed by Kc and their de novo production occurs after some delay (32), we and others have proposed, to account for the promptness of the febrile response to intravenous LPS, that the peripheral pyrogenic signal could be transmitted to the preoptic-anterior hypothalamic area (POA, the presumptive locus of the febrigenic controller) via a neuronal rather than a humoral mechanism (6). Thus evidence was adduced by us and others (12, 49, 59) that the vagus and its hepatic branch, in particular (52), may convey the pyrogenic signal to the brain. Indeed, IL-1β injected into the portal vein of anesthetized rats had been shown earlier to increase the electrical activity of the vagus (36). But because this cytokine lags behind the onset of fever induced by intravenous LPS, it also seems unlikely that it could be the direct, peripheral mediator responsible for vagal activation.
We (6, 50) and others (12, 44) have suggested that, alternatively, PGE2 could be that mediator. It is synthesized by all macrophages, including Kc, in response to LPS (46), and its levels in plasma are correlated with the febrile course (35). PGE2 receptors are also present on peripheral sensory neurons, e.g., in kidneys (26), lungs (27), stomach (54), and jejunum (19), as well as on nodose ganglion vagal sensory neurons receiving information from thoracic and abdominal compartments (12), and on cervical vagal afferents (25). It has also been proposed that, alternatively, in lieu of stimulating vagal terminals, peripherally synthesized PGE2 could trigger fever by being transported to the brain as an albumin-bound complex (22, 43). Paradoxically, however, the elevation of plasma PGE2 after LPS challenge is faster in vivo (22, 35) than the ability of macrophages to synthesize PGE2 in vitro (33, 37). It would seem, therefore, that LPS per se could not be the stimulus for the rapid production of PGE2 in vivo. Hence, presumably, another factor rapidly evoked by LPS should drive this response. The complement (C) cascade is immediately activated in vivo by the presence of LPS (58), and we have recently reported that the anaphylatoxic complement component 5a (C5a) is an essential mediator of the febrile response to LPS (5, 30). It was shown earlier that the in vitro production of PGE2 by Kc is stimulated by C5a, whereas C depletion limits this production (15, 41, 48); Kc express abundant C5a receptors (14, 48, 61).
We have hypothesized, therefore, that LPS-activated C5a may trigger PGE2 production by Kc. The purpose of this study was to test this hypothesis, using two different approaches: the first was to determine whether the rapid production of PGE2 by the liver induced by LPS in vivo is indeed C-dependent; cobra venom factor (CVF) was used to address this question because it immediately activates the C cascade, thereby adding all the C components to the circulation, and because, furthermore, it ultimately depletes the C substrate for the cascade, thereby causing hypocomplementemia (8, 15, 56, 57; see also Discussion). The second was to verify whether Kc, in fact, secretes PGE2 more rapidly in response to C than to LPS in vivo and in vitro. A corollary was to substantiate that cytokines are not liberated as early as PGE2 in response to LPS.
MATERIALS AND METHODS
Male, pathogen-free, Hartley guinea pigs were used in these experiments. The guinea pigs (550–650 g; Charles River Laboratories, Wilmington, MA) were quarantined for 1 wk, three to a cage, in the vivarium of our Department of Comparative Medicine before any experimental use. Tap water and food (Agway Prolab guinea pig) were available ad libitum. The ambient temperature (Ta) in the animal rooms was 23 ± 1°C; light and darkness alternated, with light on from 0600 to 1800. All animal protocols were approved by the University of Tennessee Health Science Center Animal Care and Use Committee and fully conform with the standards established by the U. S. Animal Welfare Act and by the document entitled “Guiding Principles for Research Involving Animals and Human Beings” (3).
For the in vivo guinea pig studies, the solvent of most drugs was sterile, nonpyrogenic, isotonic (0.9%) NaCl solution (pyrogen-free saline, PFS; Abbott Laboratories, Chicago, IL); for the in vitro studies, it was its PBS analog. Heparin was purchased from Elkins-Sinn (Cherry Hill, NJ). LPS was Salmonella enteritidis LPS B (lot #651628; Difco Laboratories, Detroit, MI), suspended in PFS. Cobra venom factor (CVF; cat #233552) was from Calbiochem-Novabiochem (San Diego, CA). Murine serum C was purchased from Sigma-Aldrich (cat. #S3269), and recombinant murine (rm)IL-1β was from R&D Systems (cat. #401-ML, lot #BN021021; Minneapolis, MN).
Cannulation of the Portal Vein and Inferior Vena Cava of Guinea Pigs
The guinea pigs were anesthetized with ketamine (35 mg/kg im) and xylazine (5 mg/kg im) and prepared for surgery. Under aseptic conditions, a midline laparotomy upward from the umbilicus, ∼3 cm long, was performed. The ileum and mesentery were exposed and the vascular arcades in the mesentery visualized. Two connecting tributary mesenteric veins were identified and back-ligated to stop inflow to these veins. The junction just proximal to the two veins was carefully incised, and a sterilized, RenaSil silicon cannula (ID 0.012 in., OD 0.025 in.; Braintree Scientific, Braintree, MA) prefilled with heparinized (50 IU/ml) PFS was inserted and directed toward the portal vein. Another two ligatures were placed around the proximal end of the cannula to secure it in place. The laparotomy incision was closed.
Immediately thereafter, a sterilized, HelixMark silicon cannula (ID 0.020 in., OD 0.037 in.; Baxter Healthcare, McGraw Park, IL) prefilled with heparinized PFS was inserted into the left jugular vein and guided toward the inferior vena cava (IVC). The incision was sutured. The animals remained under surgical anesthesia and were maintained on heating pads set at 30°C for the experimental tests.
Isolation of Guinea Pig Kupffer Cells
This method was modified from Do et al. (11). Briefly, under deep ketamine (50 mg/kg ip) xylazine (50 mg/kg ip) anesthesia, the guinea pigs were laparotomized, their portal vein isolated, and their liver perfused in situ through the portal vein with, successively, 15–20 ml of HBSS and 30 ml of liver perfusion medium (Gibco/Invitrogen, Carlsbad, CA); drainage was through the IVC. The liver was removed and submerged in ice-cold serum-free RPMI 1640 medium. To obtain Kc, the gall bladder was removed and the liver was minced with a sterile razor. The minces were digested in 50 ml of liver digest medium (Gibco/Invitrogen) at 37°C, with occasional shaking. The resulting cell suspension was filtered through a 75-μm mesh, and the filtrate was centrifuged at 50 g for 5 min, to pellet the hepatocytes. The supernatant was removed and pelleted by centrifugation at 600 g for 10 min. The cells were next washed once with HBSS, layered over a 1,038 mg/ml solution of Percoll (Pharmacia-Amersham, Piscataway, NJ), and centrifuged at 400 g for 20 min. The resulting debris-containing upper layer was removed. The lower layer was diluted threefold with PBS and centrifuged at 600 g for 10 min, to pellet the cells. The washed cell pellets from the Percoll centrifugation were suspended in PBS, 0.5% BSA, and 2 mM EDTA. They were then isolated by anti-CD-11b-conjugated magnetic beads cell sorting over MS + MiniMACS separation columns (Miltenyi, Auburn, CA), according to the instructions provided by the manufacturer. Approximately 2–4 × 106 cells were collected from the livers of four animals.
LPS in the plasma of guinea pigs was evaluated using a chromogenic Limulus amebocyte lysate assay (Pyrochrome; Associates of Cape Cod, Woods Hole, MA), according to the supplier's instructions. The detection limit of this assay was 0.005 endotoxin unit (EU). The First International Standard for Endotoxin (84/650, World Health Organization) was used as the reference. Endotoxin concentrations are expressed as international EUs per milliliter.
PGE2 in guinea pig plasma was analyzed using an enzyme immunoassay (EIA) kit (high sensitivity prostaglandin E2 EIA Kit #931–001; Assay Designs, Ann Arbor, MI), according to the manufacturer's instructions; the prostaglandin synthetase inhibitor indomethacin (10 μg/ml) was added to all the blood samples immediately after collection. All of the samples were diluted before analysis in the assay buffer supplied, according to the manufacturer's instructions. The detection limit of this assay was 8.26 pg/ml.
The levels of PGE2 in the incubation media of guinea pigs Kc were determined by our radioimmunoassay (RIA), as previously described (4). Briefly, it is based on the competition of PGE2 in the test samples with 3H-labeled PGE2 for binding to anti-PGE2 antibody. A 100-μl aliquot of culture medium was added to the RIA assay buffer (0.1 mM phosphate buffer, pH 7.4, containing 0.9% sodium chloride, 0.1% sodium azide, and 0.1% gelatin), mixed with appropriate amounts of labeled PGE2 and reconstituted antiserum, and incubated overnight at 4°C. The assay tubes were then placed on ice, and 1.0 ml of cold charcoal-dextran suspension was added. Fifteen minutes later, the tubes were centrifuged at 2,200 g for 10 min at 4°C, and the supernatants decanted into scintillation vials. Radioactivity was determined by scintillation spectrometry (Packard Tricarb 200CA). Percent binding was compared with a standard curve and the amounts of PGE2 in the samples calculated.
The determination of TNF-α in the plasma of guinea pigs was performed by a bioassay based on the cytotoxic effect of TNF-α on the mouse fibrosarcoma cell line WEHI 164 subclone 13(c). The assay was performed using sterile, 96-well microtiter plates. Serial dilutions of biological samples or different concentration of TNF-α standard (code 88/532, National Institute for Biological Standards and Control, South Mimms, UK) were incubated for 24 h in wells that had been seeded with 50,000 actinomycin D-treated WEHI 164 cells. The number of surviving cells after 24 h was measured by use of the dimethylthiazol-diphenyl tetrazolium bromide (MTT) colorimetric assay. Plasma IL-1β and IL-6 were determined by bioassays based on the dose-dependent growth stimulation of the D10 and B9 hybridoma cell lines, respectively. These assays were performed also using sterile, 96-well microtiter plates. In each well, 5,000 D10 or B9 cells were incubated for 72 h with serial dilutions of biological samples or with different concentrations of IL-1β or IL-6 standards (code 86/680 and 89/548, National Institute for Biological Standards and Control). The number of cells in each well was measured using the MTT assay.
Five microliters of plasma were added to wells placed in agarose gel containing standard sheep erythrocytes sensitized with hemolysin (Kit RC001; The Binding Site, San Diego, CA). Plates were incubated for 18 h at 4°C and then for 1 h at 37°C. The diameters of the zones of hemolysis around each well (radial immunodiffusion) were measured. These values were converted to the percentage activity of complement in the samples by interpolating from calibration curves plotted using the manufacturer's standard, diluted according to the manufacturer's directions. The detection limit of this assay was 32 CH100 units/ml. The results are expressed as the percentage of C activity remaining after CVF treatment compared with the pretreatment level.
To determine initially the efficacy of the clearance of LPS by the liver as directly as possible and to correlate this with the production of PGE2 by the liver, we injected two doses of LPS (2 and 8 μg/kg) directly into the portal vein (iportal) of anesthetized guinea pigs and measured their appearance in the IVC at 0, 1, 2.5, 5, 10, 15, 30, 45, and 60 min after their administration. The volume of each blood collection was 0.4 ml; it was immediately replaced by an equal volume of PFS. PFS (0.2 ml/kg) was also the control solution for this experiment.
To verify whether the rapid production by the liver of LPS-induced PGE2 is C-dependent, we injected CVF (25 U/animal) into the portal vein of anesthetized guinea pigs. CVF causes the immediate activation of the alternative pathway of the C cascade (8, 56, 57); the consequently formed C components then interact with their cognate receptors on cells and activate them. PFS (0.2 ml/kg) was the control solution. Blood samples were collected from the IVC at 0, 1, 2.5, 5, 10, 15, 30, 45, and 60 min after this treatment and analyzed for their PGE2, TNF-α, IL-1β, and IL-6 levels.
The activation of C induced by CVF, however, continues unabated, so that, ultimately, C becomes significantly reduced (8, 15, 56, 57; see also Discussion). To determine, therefore, the effect of hypocomplementation on LPS-induced PGE2 and cytokine production, we injected into the portal vein of these CVF-treated animals a second dose of LPS (2 μg/kg) or PFS (0.2 ml/kg) 180 min after CVF. The guinea pigs previously treated with PFS received LPS (2 μg/kg). A PFS+PFS control was considered redundant in this design and was, therefore, not performed. Plasma samples were collected from the IVC and analyzed as before. The level of hypocomplementemia was determined by analyzing plasma C levels before and 180 min after CVF administration.
To compare the effects of C and LPS on Kc PGE2 production directly, the Kc from guinea pigs were seeded at 1 × 105 cells/well in 24-well tissue culture microplates (0.5 ml/well) in RPMI 1640 medium with l-glutamine containing 20% heat-inactivated FBS and incubated overnight at 37°C. The next morning, the medium was replaced with fresh RPMI 1640 medium containing 1% FBS, and the following test agents were added: none (NA; 0.5 ml of medium), murine serum C (0.116 U/ml), S. enteritidis LPS (200 ng/ml), rmIL-1β (1 ng/ml), and C + these factors. These doses were based on preliminary dose-response studies. Samples of culture media were collected at 0, 2.5, 10, 30, and 60 min after these treatments, and their PGE2 contents analyzed.
The results are reported here as means ± SE. The data were evaluated by a repeated-measures analysis of variance model, where factor 1 was the between-group factor (the experimental treatment) and factor 2 the within-subject factor (the different sampling periods). The analyses were performed using Instat 3 (GraphPad software; Instant Biostatistics, San Diego, CA). Each variable was considered to be independent. The 5% level of probability was accepted as statistically significant.
LPS at both doses (2 and 8 μg/kg) injected into the portal vein appeared in IVC plasma virtually immediately (1 min); the level of the lower dose was reduced by half in 5 min, and then stabilized over the remainder of the 60-min experimental period. The plasma level of the higher dose, however, remained at its high, initial level over the duration of this experiment (Fig. 1A). The LPS level of the PFS-treated controls remained stable throughout the hour, although it was slightly elevated (i.e., above 0) due presumably to the surgery-associated manipulation of the animals' intestinal tract (60).
IVC plasma PGE2 levels were not significantly elevated by the administration of PFS, but they rose within 2.5 min to approximately half of their maximal value after the administration of both LPS doses, then continued to rise more slowly over the next 25 min. They then both stabilized at ∼200 pg/ml above their basal level for the remainder of the hour, irrespective of the LPS dose (Fig. 1B).
IVC plasma PGE2 levels increased within 2.5 min, peaked in 5 min, and then returned to control levels in 15 min following the administration of CVF (Fig. 2A.). PFS administration had no significant effect on PGE2 levels.
The iportal injection of LPS 3 h after PFS induced the same pattern of IVC PGE2 elevation as was observed when injected at time 0, albeit that PGE2 levels increased only approximately half as much (PFS+LPS, Fig. 1B). But LPS given 3 h after CVF pretreatment, when plasma C levels were reduced to ∼9% relative to their basal levels (the basal level of C was 393 ± 68 CH100 units/ml; 3 h after CVF treatment, it was reduced to 34 ± 14 CH100 units/ml; data from 6 animals), did not significantly raise PGE2 levels until 30–45 min after its injection (CVF+LPS, Fig. 2B). The injection of PFS into these hypocomplementemic guinea pigs (CVF+PFS, Fig. 2B) had no effect on PGE2 levels; these were not significantly different from those of PFS injected into untreated animals (Fig. 2A).
TNF-α (Fig. 3A), IL-1β (Fig. 3C), and IL-6 (Fig. 3E) became detectable in IVC plasma 30 and 45 min after LPS treatment, respectively. They were not elevated by acute CVF or PFS per se. Hypocomplementemia (Figs. 3B, D, F) also did not affect the appearance, rate of rise, or magnitude of the LPS-induced elevations of these cytokines.
IL-1β alone induced the release of PGE2 by primary, freshly isolated Kc from guinea pigs between 30 and 60 min after their addition, but LPS was without effect throughout the 60-min incubation period. On the other hand, C alone and C+LPS or IL-1β very quickly (2.5–10 min) triggered PGE2 increases of similar, apparently maximal magnitudes (Fig. 4).
The present results show that IVC PGE2 rose rapidly and to similar levels within 5 min after the iportal administration of CVF or LPS. But, whereas the elevation caused by CVF was reversed over the following 10 min, that due to LPS continued to a higher level over the same interval, then plateaued (Figs. 1B and 2A). On the other hand, 3 h after CVF treatment, when ∼91% of the original C level was depleted, PGE2 levels did not increase significantly until 45 min after the administration of LPS (Fig. 2B). Both CVF and LPS cause the immediate activation of the alternative pathway of the C cascade, but that due to CVF continues unabated, reducing C and, hence, limiting its effects (8, 15, 56, 57; see also below). Moreover, whereas no PGE2 was detectable within 60 min after the addition of LPS to freshly isolated Kc, the addition of C alone or in combination with it induced the generation of PGE2 between 2.5 and 10 min (Fig. 4). Pyrogenic cytokines were not detectable in vivo until 30–45 min after the iportal injection of LPS (Fig. 3); neither the activation of C nor its reduction affected these in vivo responses.
We infer from these data that the almost immediate appearance of PGE2 in IVC plasma after the iportal injection of LPS supports the notion, advanced previously by us (5, 6, 28, 30), that it may be mediated by C5a rapidly induced by LPS. On the other hand, its continued production in these animals, as manifested by its further, slower rise to a higher plateau at 15 min, its delayed elevation in the C-insufficient, LPS-treated guinea pigs (Fig. 2B), as well as the late production of cytokines in all the animals under all the experimental conditions (Fig. 3) are probably all the result of the direct, LPS binding protein (LBP)-cluster designation 14 (CD14)-Toll-like receptor 4 (TLR4)-myeloid differentiation protein 2 (MD2)-mediated activation of Kc by LPS (53).
Like LPS, CVF activates virtually immediately C3. The activation of C3 by LPS yields a derivative, C3b, which forms a complex with factor B, C3bBb, a convertase that cleaves the α chain of C3 and thereby enables the further production of its downstream components (58); C3bBb is very labile (t1/2 = 1.5 min at 37°C). By contrast, CVF itself forms a complex with factor B, CVFBb, that is functionally analogous to C3bBb, but much more stable (t1/2 = 7 h) due to its resistance to the control mechanisms that limit the activity of the normal C3 convertase. Consequently, C3 activation continues unabated, depleting C3 and all its downstream products (8, 56). The present finding that C triggered the immediate production of PGE2 by Kc, whereas its depletion limited this production thus substantiates similar, earlier findings by others (15, 40, 41, 47, 48). But our further in vivo finding that the similarly quick release of PGE2 evoked by LPS at time 0 was abrogated when C was reduced 3 h later, though anticipated, is novel. It confirms that the initial, very early, LPS-induced PGE2 rise was indeed C- rather than LPS-mediated. On the basis of the duration of the PGE2 response to the acute effect of CVF, that is, the rapid activation, then reduction of C3, it was therefore brief. The similarities in the onset, magnitude, and duration of the PGE2 response to LPS, irrespective of the dose administered (Fig. 1B), further reinforce the notion that the initial, quick, LPS-induced release of PGE2 is C-mediated. On the other hand, the subsequent rise of IVC PGE2 to higher levels at 15 min and the 15-min delay in its appearance when C was reduced indicate that a secondary, slower, C-independent mechanism also underlay the LPS-induced Kc production of PGE2; this mechanism was, presumably, the LPS-TLR4 signaling pathway (38, 42). To our best knowledge, this is the first in vivo demonstration of a two-part modulation of the PGE2 response to the iportal injection of LPS.
Although not specifically demonstrated in the in vivo portions of this study, the in vitro results of experiment 3 strongly suggest that the targets of C5a, the anaphylatoxin specifically implicated in the febrile responses to intravenous and intraperitoneal LPS (28, 30), are in all probability the Kc. Although hepatic stellate, sinusoidal endothelial, and mast cells also express its principal receptor, C5aR1, it is most abundant in Kc (48). These cells are critically linked to the onset of the febrile response to LPS (31). Mast cells are not involved (5), and there is no evidence that the other cell types are implicated. C5aR1 is a G-protein-linked receptor that acts by increasing intracellular inositol-1,4,5-triphosphate and Ca2+ (47), rapidly activating cyclooxygenase (COX)-1-catalyzed PGE2 production (2, 40, 46, 47). COX-1 is constitutive, functionally coupled mainly with cytosolic PGE2 synthase (PGES) and therefore prepared to quickly synthesize PGE2 (55). On the other hand, it is generally agreed that the production of PGE2 induced by LPS per se is initiated by the LBP-mediated transfer of LPS to the receptor complex CD14/TLR4/MD2 and is associated with the upregulation of, specifically, COX-2 and microsomal PGES-1 (1, 20); in rats, both enzymes are induced in the liver ∼30 min after LPS challenge (21). This delay in their biosynthesis presumably accounts for the lateness of the secondary appearance of PGE2 observed in the present study. The cytokines, then also present in the blood, may further contribute to this late rise (Fig. 4). Indeed, as in previous studies (17, 18, 23, 24), these became evident in IVC plasma around 30–45 min after the iportal administration of LPS (Fig. 3), a delay that reaffirms that they probably have no role in the initiation of the febrile response to iv LPS. It is generally agreed that the LPS-TLR4 complex is the system that induces the production of pyrogenic cytokines (53). But, because the guinea pigs were anesthetized and maintained on heating pads during the present experiments (anesthesia impairs thermoregulatory responses; their body temperatures were monitored therefore only to insure their stability), the concurrence of the observed cytokines and PGE2 responses with the normal febrile course of LPS-treated animals could not be verified. They do concord, however, with those reported in other, previous studies in conscious animals (5, 9, 18, 22–24, 29, 51).
In summary, because the very early appearance of PGE2 in IVC plasma coincides temporally with the onset of fever, these findings would support the notion that PGE2 quickly elaborated by Kc stimulated by LPS-activated C could be the factor that stimulates vagal terminals in the liver or circulates to the brain and, hence, may be responsible for the prompt initiation of the febrile response to intravenous LPS, as postulated previously (6, 22, 43, 49, 50). This interpretation may also help to clarify the correlation between its elevation in plasma and the febrile course (35).
In conclusion, the present data suggest that LPS injected into the portal vein (and, by inference, circulating LPS arriving in the liver) causes the very rapid appearance of PGE2 in the blood via a two-part effect, one very rapid, but brief, exerted on C5aR1-expressing Kc consequent to the virtually immediate activation by LPS of the alternative pathway of the C cascade, and another slower, but more prolonged, induced by the recognition of LPS by the TLR4 signaling complex. These results are compatible with the hypothesis that LPS-activated C rather than LPS itself rapidly triggers the release of PGE2 by Kc.
This work was supported by National Institutes of Health Grant NS-34857 to C. M. Blatteis.
We gratefully acknowledge Dr. Steve Hopkins (North Western Injury Research Collaboration, MRC Trauma Group, Hope Hospital, Salford, UK) for the generous gift of the cell lines used for the cytokine bioassays in this study. We would also like to thank Gregg Short and Daniel Morse for their graphic art support.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 the American Physiological Society