Effect of C1 Inhibitor on Inflammatory and Physiologic Response Patterns in Primates Suffering from Lethal Septic Shock1

of May 7, 2014.This information is current as

Suffering from Lethal Septic ShockPhysiologic Response Patterns in Primates Effect of C1 Inhibitor on Inflammatory and

Ulrich Delvos, Fletcher B. Taylor, Jr. and C. Erik HackPatty M. Jansen, Bernd Eisele, Irma W. de Jong, Alvin Chang,

http://www.jimmunol.org/content/160/1/4751998; 160:475-484; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/160/1/475.full#ref-list-1

, 28 of which you can access for free at: cites 56 articlesThis article

Subscriptionshttp://jimmunol.org/subscriptions

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/ji/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/cgi/alerts/etocReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists All rights reserved.Copyright © 1998 by The American Association of9650 Rockville Pike, Bethesda, MD 20814-3994.The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on May 7, 2014

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on May 7, 2014

http://ww

w.jim

munol.org/

Dow

nloaded from

Effect of C1 Inhibitor on Inflammatory and PhysiologicResponse Patterns in Primates Suffering from LethalSeptic Shock1

Patty M. Jansen,* Bernd Eisele,† Irma W. de Jong,* Alvin Chang,‡ Ulrich Delvos,†

Fletcher B. Taylor, Jr.,‡ and C. Erik Hack2*§

We evaluated the effect of C1 inhibitor (C1-inh), an inhibitor of the classical pathway of complement and the contact system,on the physiologic and inflammatory response in baboons suffering from lethal Escherichia coli sepsis. Five animals pretreatedwith 500 U/kg C1-inh (treatment group; n 5 5), followed by a 9-h continuous infusion of 200 U/kg C1-inh subsequent tobacterial challenge, were compared with five controls receiving E. coli alone. Of the treatment group, one animal survived andanother lived beyond 48 h, whereas all control animals died within 27 h. In four of five treated animals, less severe pathologywas observed in various target organs. C1-inh administration did not prevent the hemodynamic or hematologic changes ob-served upon E. coli infusion. The activation of fibrinolysis and the development of disseminated intravascular coagulation wereessentially unaffected by C1-inh. However, C1-inh supplementation significantly reduced decreases in plasma levels of factor XIIand prekallikrein and abrogated the systemic appearance of C4b/c, indicating substantial inhibition of activation of the contactsystem and the classical complement pathway, respectively. Furthermore, treated animals displayed a reduced elaboration ofvarious cytokines including TNF, IL-10, IL-6, and IL-8. Thus, the administration of C1-inh may have a beneficial but modest effecton the clinical course and outcome of severe sepsis in nonhuman primates. We suggest that activated complement and/orcontact system proteases may, at least in part, contribute to the attendant manifestations of septic shock through an augmen-tation of the cytokine response. The Journal of Immunology, 1998, 160: 475–484.

Sepsis is a clinical syndrome that results from bacterial in-fection and is characterized by an extensive triggering ofmultiple endogenous mediators (1). Among the mediators

implicated are the complement system and the contact system ofintrinsic coagulation, both of which can be activated directly invitro by cell wall components of Gram-negative and Gram-posi-tive bacteria (2–4). Ample evidence has now accumulated thatactivation of these plasma cascade systems occurs in human sepsis,notably when complicated by shock and/or adult respiratory dis-tress syndrome (5–12). Activation of the complement and contactsystems results in the liberation of biologically active peptides, theanaphylatoxins and bradykinin, respectively, which may inducevasodilation and enhance the permeability of endothelial cells (13,14). It has been previously shown that the anaphylatoxins (in par-ticular C5a) and factor XIIa also stimulate neutrophils (13, 15) andinduce production of cytokines by mononuclear cells (16–22).Moreover, C5a and the terminal complex of complement, C5b-9,

may promote coagulation by evoking tissue factor expression onendothelial cells (23), and assembly of C5b-9 on the surface ofplatelets has been shown to induce the release ofa-granules, theexposure of negatively charged phospholipids, assembly of theprothrombinase complex, and release of vesicles that express pro-thrombinase activity (24–26). Hence, activation of complementand contact systems may contribute to the coagulant and inflam-matory sequelae of sepsis through several mechanisms.

Activation of both the complement and contact system is regu-lated by C1-esterase inhibitor (C1-inh).3 C1-inh, which belongs tothe superfamily of serine-proteinase inhibitors, is the only knowninhibitor of C1r and C1s, components of the classical pathway ofcomplement (27), as well as the major inhibitor of factor XII andprekallikrein of the contact system (28, 29). Although C1-inh is anacute phase protein, antigenic levels of C1-inh tend to be normalin patients with fatal septic shock, while levels of proteolyticallyinactivated C1-inh are increased, suggestive of an increased turn-over and a relative deficiency of biologically active C1-inh duringsepsis (30).

We have previously demonstrated that C1-inh substitution ther-apy in patients with septic shock may reduce the need for vaso-pressor medication and attenuate complement and contact activa-tion (31, 32). Moreover, effects of C1-inh have been observed inseveral animal models of sepsis: C1-inh supplementation abro-gated endoxin-induced disseminated intravascular coagulation and

*Central Laboratory of the Netherlands Red Cross Blood Transfusion Servicesand Laboratory for Experimental and Clinical Immunology, Academic MedicalCentre, University of Amsterdam, The Netherlands; †Behringwerke AG, Mar-burg, Germany; ‡Oklahoma Medical Research Foundation, Oklahoma City, OK73104; and §Department of Internal Medicine, Free University Hospital, Am-sterdam, The Netherlands

Received for publication April 21, 1997. Accepted for publication September23, 1997.

The costs of publication of this article were defrayed in part by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.1 Supported by Grant No. 900-512-151 from the Netherlands Organization forScientific Research (NWO). Animal experiments were supported by Behring-werke AG, Marburg, Germany.2 Address correspondence and reprint requests to Dr. C. Erik Hack, CLB, De-partment of Pathophysiology of Plasma Proteins, Plesmanlaan 125, 1066 CXAmsterdam, The Netherlands.

3 Abbreviations used in this paper: C1-inh, C1-inhibitor; NBA, normal baboonserum aged; NBS, normal baboon serum; NBS-AHG, normal baboon serum in-cubated with heat-aggregated IgG; NBP-MA-UK, normal baboon plasma incu-bated with methylamine and urokinase; PAP, plasmin-a2-antiplasmin complex;TAT, thrombin-antithrombin complex; tPA, tissue-type plasminogen activator;Fischer LSD, Fischer least significant difference.

Copyright © 1998 by The American Association of Immunologists 0022-1767/98/$02.00

by guest on May 7, 2014

http://ww

w.jim

munol.org/

Dow

nloaded from

hypotension in rabbits (33, 34) and pulmonary dysfunction in en-dotoxemic dogs (35). In this study, we evaluated the effect of i.v.administration of C1-inh on hemodynamic, coagulant, inflamma-tory, and cell injury responses in an established model of severeseptic shock in nonhuman primates. Our results indicate that in thisexperimental model, exogenous administration of C1-inh may ex-ert benificial effects, in part through modulation of cytokine re-lease, and support the notion that activation of complement and/orcontact system proteases is associated with organ injury and im-pending lethality.

Materials and MethodsC1-inh preparation

Pasteurized human C1-inh was provided by Behringwerke AG (Marburg,Germany). On SDS-PAGE, this preparation consisted of.95% nativeC1-inh.

Experimental and infusion procedures

Experiments were performed on 10 juvenile baboons (Papio anubis/Cyno-cephalus), each with a hematocrit exceeding 36% and free from tubercu-losis. The animal-handling procedures andEscherichia coli(type B) prep-aration were performed using the methodology described in previouspublications (36, 37). Briefly, baboons were fasted overnight before eachexperiment and given water ad libitum. Each animal was sedated withketamine hydrochloride (14 mg/kg, intramuscularly) on the morning of thestudy and anesthetized with sodium pentobarbital (2 mg/kg) via a percu-taneous catheter positioned in the cephalic vein. The femoral artery andboth femoral veins were cannulated aseptically and used for measuringaortic pressure, obtaining blood samples, infusing live organisms, C1-inh,and for fluid and anesthetic administration as reported elsewhere (36, 37).Gentamicin was given (9 mg/kg) as a 75-min infusion immediately afterthe E. coli infusion had been stopped, and then (4.5 mg/kg) as a 30-mininfusion at 6 and 9 h after the start of theE. coli infusion. Additionalgentamicin (4.5 mg/kg) was given as an intramuscular injection at 10 hafter the start of the infusion and twice daily for the subsequent 3 days.

Each animal was i.v. challenged with a lethal dose ofE. coli (4 3 1010

CFU/kg of body weight), given as a 2-h infusion. The time point at whichthe infusion was started is further indicated as T1 0, a time point ofnhours thereafter referred to as T1 n h. Time points before the start of thechallenge are indicated as T2 n h.

Two experimentalE. coli groups were studied: one group consisted offive baboons that received an initial dose of 500 U/kg C1-inh (1 U is theamount of C1-inh present in 1 ml of normal human plasma, which is equalto 2.50mM/L or 270 mg/L; Ref. 30) administered i.v. over a 30-min periodbefore the start of the lethalE. coli challenge, followed by a continuousinfusion of 200 U/kg for 9 h (treatment group). C1-inh doses were basedon studies in human volunteers (38) in which human native C1-inh wascleared from the circulation with a fractional catabolic rate of 2.5% of theplasma pool per hour and has an apparentt1/2 of ;28 h; thus, presumablythis scheme of administration would maintain concentrations of C1-inh at;10 times the level observed in normal human plasma during the first 9 hof the experiment. The second group consisted of five animals that receivedsaline according to a similar scheme of administration before and afterlethal E. coli infusion (control group).

All animals were maintained under anesthesia and monitored for 10 h.They were observed continuously for an additional 36 h and daily for amaximum of 7 days. Blood samples were collected at given time points forhematology, clinical chemistry, and C1-inh determinations. Additionalsamples were collected on EDTA/soy bean trypsin inhibitor (final concen-trations, 10 mM and 0.1 mg/ml, respectively) before (T1 0), and at 0.5,1, 2, 3, 4, 6, 8, and 10 h afterE. coli challenge for plasma determinationof cytokines, complement activation products, contact system proteins,neutrophil degranulation products, and coagulation/(anti)fibrinolytic pa-rameters. Baboons surviving for 7 days were considered permanent survi-vors and were subsequently killed with sodium pentobarbital for necropsyon the eighth day.

Assays

All assays used in this study were developed with mono- or polyclonal Absraised against human proteins. When possible, standards were preparedwith baboon plasma or serum to correct for differences in affinities of theAbs for baboon vs human proteins. When human standards had to be used,we established that dilution curves of these standards were parallel to thoseobtained with baboon plasma samples. Notably, the lower affinity of the

Abs for baboon proteins may have led to an underestimation of the baboonproteins.Antigenic C1-inh. Plasma levels of exogenously administered C1-inhwere measured by nephelometer (Behringwerke Nephelometer Analyzer,Behringwerke AG) and expressed as mg/L. Functional C1-inh levels weremeasured by RIA as described previously (30).Complement activation products.C3b/c and C4b/c in baboon plasmawere quantified by RIA as reported elsewhere (39–41). C3b/c was ex-pressed as a percentage of the amount present in normal baboon serumaged (NBA), i.e., normal baboon serum (NBS) incubated for 7 days at37°C in the presence of 0.02% (w/v) NaN3. Results for C4b/c were ex-pressed as a percentage of the amount generated in NBS by incubation withheat-aggregated IgG (NBS-AHG). Levels of the terminal complex of com-plement (C5b-9) were measured by ELISA according to the instructions ofthe manufacturer (Behringwerke AG, Marburg, Germany), and were ex-pressed asmg/L with reference to a serially diluted standard of humanC5b-9.Cytokines. Plasma concentrations of TNF-a, IL-6, IL-8, and IL-10 weremeasured by ELISA as previously described (42–44).Coagulation and (anti)fibrinolytic parameters.Levels of tissue-typeplasminogen activator (t-PA), plasminogen activator inhibitor type 1 (PAI-1), and thrombin-antithrombin III (TAT) complexes were determined byELISA as described previously (41, 45, 46). Values were expressed asng/ml. Plasmin-a2-antiplasmin (PAP) complexes were measured by RIA asdescribed (46). PAP complex levels were expressed as percentage of thelevel present in normal baboon plasma in which a maximal amount ofcomplexes was generated by incubation with an equal volume of urokinase(50 mg/ml) in the presence of 0.2 mol/L methylamine (final concentration)to inactivatea2-macroglobulin, further referred to as NBP-MA-UK.Measurement of prekallikrein and factor XII in plasma.Plasma prekal-likrein and factor XII were determined by a sandwich-type ELISA. Flat-bottom microtiter (96-well) plates (Dynatech, Plochingen, Germany) werecoated overnight at room temperature with 100ml of 2.5 mg/ml anti-humanprekallikrein mAb K15, or mAb OT-2 against human factor XII, in car-bonate buffer, pH 9.5, and blocked for 30 min with 150ml PBS containing2% (v/v) cow’s milk. All subsequent incubations were in 100-ml volumesat room temperature, and plates were washed after each incubation withPBS/0.02% (w/v) Tween-20. The plates were then incubated for 2 h withbaboon plasma samples diluted in 100ml high performance ELISA (HPE)buffer (CLB, Amsterdam, The Netherlands). Bound prekallikrein and fac-tor XII were detected by subsequent 1-h incubation with HPE buffer con-taining 1mg/ml of biotinylated mAbs 13G11 (kindly provided by Dr. R. W.Colman, Temple University, Philadelphia, PA) and F3, respectively, fol-lowed by a 1:10,000 dilution of streptavidin-polymerized horseradish per-oxidase (polyHRP; CLB) in PBS/2% (v/v) cow’s milk for 30 min. Theplates were developed with a solution of 100mg/ml of 3,5,39,59-tetram-ethylbenzidin (Merck, Darmstadt, Germany) with 0.003% (v/v) H2O2 in0.11 mol/L sodium acetate, pH 5.5. The reaction was stopped by the ad-dition of an equal volume of 2 mol/L H2SO4 to the wells. Serial dilutionsof normal pooled baboon plasma was used as a standard. Values wereexpressed as percentage of the amount present beforeE. coli infusion(T 1 0).Neutrophil degranulation products.Elastase-a1-protease inhibitor com-plexes were determined with a RIA that has been described in detail else-where (47). Results were expressed as nanograms of elastase per milliliterby reference to a standard curve that consisted of normal baboon plasma towhich human neutrophil elastase (Elastin Products Co., Pacific, MO) wasadded at a final concentration of 2mg/ml. In this standard,.95% of theelastase is complexed toa1-antitrypsin.

Statistical analysis

Results are expressed as mean6 SEM. Statistical analysis was performedusing a commercial statistical package (StatView; Abacus Concepts, Inc.,Berkeley, CA). Comparisons between groups during the course of the ob-servation period were performed using repeated measures analysis of vari-ance (ANOVA). Data were analyzed by two-tailed ANOVA to determinethe significance of differences in means between groups at given times.Within one group, differences from baseline levels were determined withANOVA using Fischer’s least significant difference (Fischer LSD). Statis-tical significance was designated at the 95% confidence level.

ResultsRecovery of C1-inh

C1-inh was infused into baboons to achieve a concentration of;10 times the level observed in normal human plasma (i.e., 270mg/L) just before the start of theE. coli infusion (T 1 0). Figure

476 EFFECT OF C1-inh IN LETHAL PRIMATE SEPSIS

by guest on May 7, 2014

http://ww

w.jim

munol.org/

Dow

nloaded from

1 shows the course of antigenic levels of C1-inh in the treatmentgroup. Baseline levels of endogenous C1-inh 30 min beforeE. colichallenge (T-0.5 h) were,200 mg/L, which probably resultedfrom poor cross-reactivity of the employed anti-human C1-inh Abswith PapioC1-inh. C1-inh levels at T1 0 were 22436 86 mg/L(range: 2016–2644), and remained elevated for at least 10 h. Dur-ing this entire observation period,.95% of circulating humanC1-inh in both groups consisted of native, uncleaved C1-inh, asdetermined by RIA, based on the binding capacity of functionalC1-inh to C1s (Ref. 30; not shown). In the control group, levels ofantigenic C1-inh remained below the limit of detection (notshown).

The effect of C1-inh administration on physiologic responsepatterns to E. coli

Table I shows the conditions, the number of organisms infused andfound circulating in the blood at T1 2 h, and survival times ofanimals in control and treatment groups. The meanE. coli dosageof C1-inh-treated and control groups was similar, i.e., 9.453 1010

and 8.703 1010 CFU/kg, respectively (p . 0.05). Moreover, nosignificant difference was noted in the number of organisms cir-culating at T1 2 h (4.643 107 and 4.773 107 CFU/ml in treatedand control animals, respectively), indicating that C1-inh admin-istration did not interfere with bacterial clearance. Administrationof C1-inh rescued one of five baboons in the treatment group.Moreover, one animal receiving C1-inh survived for 62.5 h, anunusual occurrence in this model of sepsis. None of the excipientcontrol animals survived beyond 27 h, and the mean survival timein this group was 19.4 h. However, possibly due to the limitednumber of animals included in this study, comparison of the sur-vival curves using the likelihood ratio test failed to indicate a sig-nificant difference in survival time of treated vs control groups( p . 0.05).

The influence of C1-inh on clinical and hematologic responsepatterns to lethalE. coli challenge is summarized in Table II. In-fusion of E. coli produced a severe hypotensive shock in all ba-boons, which was unaffected by supplementation with C1-inh. Inboth control and treatment groups, a significant decline in meansystemic arterial pressure was noted at 2 h, remaining below base-line values during the 10-h observation period. Mean heart rate andtemperature were similarly elevated in all animals.E. coli-infusionwith or without C1-inh induced a prompt fall in leukocyte count to40% of control at T1 1 h, a steady decline in platelet count to 30%of control at T1 10 h, a small rise in hematocrit from T1 6 h andon, and a progressive decrease in fibrinogen levels over the entireobservation period. Changes in mean systemic arterial pressure,

heart rate, temperature, white blood cell, platelet counts, and he-matocrit were not different between the nonsurviving C1-inh-treated animals and the animal surviving treatment (not shown).However, the rate of fibrinogen consumption appeared to be mod-estly decreased in the surviving C1-inh-supplemented animal(Table II).

To evaluate whether the administration of C1-inh affected bio-chemical changes related to organ damage, we compared severalmarkers of cell injury at T1 0 and T1 10 h in C1-inh-treated andcontrol groups. Table III shows that levels of blood urea nitrogen,creatinine, uric acid, lactate dehydrogenase (LDH), serum glu-tamic oxaloacetic transaminase (SGOT), and serum glutamic pyru-vate transaminase (SGPT) were increased in all animals at 10 h.The magnitude of the increases, however, tended to be lower in theC1-inh treatment group, especially in the treated baboon that wasrescued from lethal challenge, and in the animal surviving for.60h after C1-inh supplementation. At 10 h postchallenge, these an-imals exhibited only moderate changes in uric acid, LDH, SGOT,and SGPT, while a pronounced increase in the levels of theseparameters was observed in the treated and control animals thatdied within 14 to 27 h.

Postmortem examinations were conducted in all baboons. Tis-sues were removed for analysis within minutes after death, therebyavoiding autolytic changes. Kidneys and adrenals removed fromall baboons in the control group showed evidence of wide-spreadmicrovascular thrombosis with extensive infarction and hemor-rhage. In each case, alveolar capillary congestion, edema, and ag-gregation of neutrophils were observed in the lungs. There wassevere vascular congestion and accumulation of neutrophils in thevascular spaces of the liver. The spleen showed lymphoid follic-ular necrosis and medullary congestion. In contrast, in the surviv-ing animal (No. 9) in the C1-inh treatment group, all organs ap-peared unaffected, while in the treated animal surviving for.60 h(No. 8), the only significant changes were limited to the lungs,which showed moderate edema without signs of thrombosis. Inaddition, in one nonsurviving C1-inh-treated animal (No. 5), C1-inh treatment protected adrenals and kidneys, in which no pathol-ogy was observed, while in another animal (No. 10), the lungs,showing only mild congestion and leukocyte influx, were essentiallyspared. Thus, less severe damage to various organs was observed infour of five C1-inh-treated animals when compared with controls.

Effect of high dose C1-inh supplementation on inflammatoryresponse patterns in lethal E. coli sepsis

Complement activation.Plasma levels of C4b/c, C3b/c, andC5b-9 were measured to evaluate the effect of C1-inh supplemen-tation on E. coli-induced complement activation (Fig. 2,A-C). Inthe control group, circulating levels of C4b/c and C3b/c continuedto rise during the entire observation period, reaching maximal lev-els of 13.36 3.1% and 10.96 2.8%, respectively, of fully acti-vated NBS at T1 10 h (Fig. 2,A andB). Administration of highdoses of C1-inh almost completely abrogated the appearance ofC4b/c (p , 0.0001) in all animals thus treated, indicating efficientinhibition of PapioC1 by human C1-inh. Moreover, C1-inh mark-edly attenuated the appearance of C3b/c at all time points (p ,0.01), suggesting that at least part of the C3 activation had oc-curred via the classical pathway. Plasma levels of the terminalcomplex of complement, i.e., C5b-9, also rapidly increased uponE. coli challenge (Fig. 2C). In control animals, peak levels of36916 221mg/L were noted at T1 2 h, remaining elevated untilthe end of the observation period. C1-inh treatment only modestlyaffected concentrations of C5b-9 (p 5 0.05), and peak levels of30616 289 mg/L were measured at 2 h after theE. coli infusionwas started. A uniform response was observed for all activation

FIGURE 1. Recovery of C1-inh in baboons with lethal sepsis.Mean 6 SEM levels of antigenic C1-inh in the treatment group receiv-ing 500 U/kg 1 200 U/kg/9 h of human C1-inh.

477The Journal of Immunology

by guest on May 7, 2014

http://ww

w.jim

munol.org/

Dow

nloaded from

products in the treatment group, and this response appeared unre-lated to survival and/or organ damage.Fibrinolytic response. In agreement with previous observations(41, 46), a pronounced activation of (anti)fibrinolysis was ob-served with lethal E. coli challenge (Fig. 3,A–C). In both controland treatment groups, a protracted increase of t-PA concentrationswas noted from T1 1 h and on, to reach a maximum level of77.86 16.1 and 117.66 15.2 ng/ml at T1 10 h in control andC1-inh treatment groups, respectively (Fig. 3A). Concentrations of

circulating PAP complexes, which reflect the generation of plas-min, the key enzyme of the fibrinolytic system, transiently in-creased uponE. coli infusion, peak levels being 10.36 3.5% and15.76 4.2% of maximally activated plasma at T1 2 h in controlsand treated animals, respectively (Fig. 3B). This difference in timecourse of t-PA and PAP complexes was not statistically significantbetween both groups (p . 0.05). In contrast, C1-inh administra-tion markedly attenuated the appearance of PAI into the circula-tion, especially at and beyond T1 6 h (Fig. 3C; p , 0.0001). In

Table I. Weight, sex, E. coli dose, and E. coli concentration in blood at the end of the infusion (T 1 2 hr), and survival in control and C1-inhtreatment groupsa

No.Weight

(kg) SexE. coli Dose

(31010 CFU/kg)

E. coli Dose inBlood at T 1 2(3107 CFU/kg) Survival (h)

Control group7 5.5 F 9.48 15.00 19.5

11 5.0 F 10.02 3.65 273 6.6 M 7.68 1.72 174 6.1 M 8.16 1.28 15

7b 6.8 M 8.15 2.22 18.5mean 6 SEM 6.0 60.3 8.70 6 0.45 4.77 6 2.59 19.4 6 2.0

C1-inh group5 6.4 F 9.00 7.90 14.56 5.0 M 9.72 1.84 178 5.7 F 10.26 4.40 62.59 6.8 F 8.34 3.85 .168

10 5.9 F 9.92 5.20 25.5mean 6 SEM 5.9 6 0.3 9.45 6 0.35 4.64 6 0.99 .57.5 6 28.9

a No significant difference was noted by two-tailed ANOVA.

Table II. Hemodynamic parameters, vital signs, and hematologic response patterns in baboons after infusion of lethal E. coli alone (n 5 5) andafter lethal challenge supplemented with C1-inh (n 5 5)a

Time (h)

T 1 0 T 1 1 T 1 2 T 1 3 T 1 4 T 1 5 T 1 6 T 1 8 T 1 10

MSAPControl 119.0 6 6.7 109.0 6 2.9 74.4 6 7.4* 65.0 6 7.0* 61.6 6 7.8* 59.2 6 7.3* 72.2 6 4.4* 72.6 6 10.6* 68.6 6 10.9*C1-inh 113.6 6 4.1 110.0 6 1.6 66.0 6 6.4* 59.0 6 1.9* 58.0 6 4.1* 66.0 6 4.3* 73.0 6 4.6* 73.0 6 3.4* 72.6 6 6.3*

HRControl 136.0 6 11.2 156.0 6 12.9* 176.0 6 9.3* 192.0 6 5.8* 193.0 6 8.9* 194.0 6 4.0* 190.0 6 10.0* 192.0 6 9.7* 195.0 6 8.7*C1-inh 133.0 6 12.2 168.0 6 9.7* 189.0 6 8.4* 193.0 6 4.4* 186.0 6 7.5* 188.0 6 3.7* 186.0 6 5.1* 189.0 6 6.4* 186.0 6 5.1*

TempControl 36.3 6 0.5 35.9 6 0.3 36.4 6 0.4 37.0 6 0.5* 37.1 6 0.4* 37.2 6 0.4* 37.1 6 0.4* 37.3 6 0.6* 36.9 6 0.4*C1-inh 36.6 6 0.2 36.7 6 0.2 37.3 6 0.3* 37.5 6 0.3* 37.6 6 0.4* 37.4 6 0.6* 37.5 6 0.6* 37.2 6 0.3* 37.4 6 0.3*

WBCControl 5.8 6 0.7 2.3 6 0.2* 1.4 6 0.2* ND 1.4 6 0.2* ND 1.6 6 0.3* ND 2.9 6 0.3*C1-inh 3.5 6 0.2 1.3 6 0.1* 0.8 6 0.1* ND 1.0 6 0.1* ND 1.1 6 0.2* ND 1.9 6 0.4*

PlateletsControl 323.4 6 26.0 255.0 6 17.7* 215.2 6 17.4* ND 174.8 6 15.8* ND 137.5 6 7.9* ND 104.8 6 21.4*C1-inh 273.8 6 39.8 249.8 6 37.9* 184.2 6 26.7* ND 143.6 6 19.7* ND 122.2 6 13.1* ND 86.0 6 15.8*

HCTControl 37.2 6 1.1 37.8 6 1.3 37.4 6 1.3 ND 37.8 6 1.8 ND 40.5 6 2.0* ND 42.5 6 0.6*C1-inh 34.4 6 0.7 37.4 6 0.5 35.2 6 0.5 ND 35.8 6 0.6 ND 38.8 6 0.8* ND 40.8 6 1.0*

FibrinogenControl 100 86.4 6 7.8* 78.0 6 5.2* ND 13.2 6 8.2* ND 4.2 6 3.2* ND 2.0 6 0.8*C1-inh 100 87.4 6 4.2* 46.8 6 8.4* ND 13.0 6 4.1* ND 4.2 6 3.2* ND 4.6 6 3.1*

IndividualNo. 5 100 76 27 ND 12 ND ,1 ND ,1No. 6 100 89 47 ND 13 ND ,1 ND 3No. 8 100 102 33 ND ,1 ND ,1 ND ,1No. 9b 100 86 75 ND 27 ND 17 ND 17No. 10 100 84 52 ND 12 ND ,1 ND ,1a Values represent mean 6 SEM changes in mean systemic arterial pressure (MSAP, in mmHg), heart rate (HR, in beats/min), temperature (°C), white blood cell

counts (WBC, in 103/mm3), hematocrit (HCT, in %), and fibrinogen (% of baseline) after start of the infusion (T 1 0). Individual fibrinogen levels in the C1-inh treatmentgroup are given. The course of all parameters was not different between both groups by ANOVA-repeated measurements.

bAnimals surviving for .7 days.*p , 0.05 vs baseline by two-tailed ANOVA using Fischer PLSD.

478 EFFECT OF C1-inh IN LETHAL PRIMATE SEPSIS

by guest on May 7, 2014

http://ww

w.jim

munol.org/

Dow

nloaded from

both groups, peak levels of PAI were noted at 8 h postchallenge,and were 51806 877 and 23796 446 ng/ml in controls andC1-inh-supplemented animals, respectively.Clotting response.TAT complexes were measured in control andC1-inh groups to determine the extent of thrombin generation (Ta-ble IV). Consistent with a pronounced decline in fibrinogen andplatelets observed upon lethalE. coli administration (Table II),indicating the onset of diffuse intravascular coagulation (DIC), cir-culating TAT complexes sharply increased from T1 1 h and on,and maximal levels of 22366 314 and 16686 452 ng/ml weremeasured at 6 h postchallenge in control and treatment groups,respectively. Although the course of TAT complexes was not sta-tistically different between both groups, the magnitude of increasewas markedly lower in the surviving C1-inh-treated animal thatalso showed a retarded fibrinogen consumption (animal No. 9;Tables II and IV).Cytokine response patterns.The administration ofE. coli wasassociated with a transient increase of plasma TNF concentrations(Fig. 4A). Consistent with our previous studies (48, 49), peak TNFlevels of 16.36 1.9 ng/ml were observed in control animals at T12 h, i.e., at the end of theE. coli infusion. Administration of lethalE. coli supplemented with C1-inh elicited a similar, but signifi-cantly lower TNF response, with peak plasma concentrations of11.06 1.7 ng/ml (p , 0.05). Notably, the lowest TNF peak wasobserved in the treated animal survivingE. coli challenge (No. 9;4.1 ng/ml at T1 2 h).

C1-inh administration markedly modulated the appearance ofIL-10, especially during the later stages of the septic proces (Fig.4B). In excipient control baboons, IL-10 levels sharply increasedto reach a maximum of 35646 772 pg/ml 3 h after the start of theexperiment and were still elevated at T1 10 h. Upon C1-inhtreatment, peak levels of IL-10 were noted at T1 3 h (2, 2666767 pg/ml), to become near baseline values at the end of the ob-servation period (p , 0.05). In addition, administration of C1-inhsignificantly attenuated the sustained release of IL-6 and IL-8 intothe circulation, and maximal levels of 279.66 74.9 and 167.4619.1 ng/ml were noted at 8 and 4 h, respectively, i.e., a two- tofivefold reduction compared with peak control levels (Fig. 4,C andD; p , 0.001).

FIGURE 2. Complement activation after lethal E. coli infusion.Mean 6 SEM plasma levels of C3b/c (A), C4b/c (B), and C5b-9 (C) incontrol (v) and C1-inh (V) groups. C3b/c and C4b/c are expressed aspercentage of NBA and NBS-AHG, respectively. Differences betweenboth group by ANOVA repeated measurements were significant at p ,0.05.

Table III. Comparison of the effect of C1-inh treatment on markers of cell injury in baboons infused with lethal E. coli a

Time (h)

BUN Creatinine Uric acid LDH SGOT SGPT

T 1 0 T 1 10 T 1 0 T 1 10 T 1 0 T 1 10 T 1 0 T 1 10 T 1 0 T 1 10 T 1 0 T 1 10

ControlMean 22 39 0.6 2.5 ,0.1 1.3 278 2582 42 687 42 2836 SEM 63 62 60.1 60.2 60.1 619 6420 65 6183 610 6130

C1-inhNo. 5 18 41 0.6 2.3 ,0.1 0.6 277 2369 33 511 60 213No. 6 19 45 0.5 2.4 ,0.1 1.3 363 2780 43 657 41 445No. 8b 19 38 0.7 2.0 ,0.1 0.4 228 1010 26 134 23 38No. 9c 13 25 0.7 1.9 ,0.1 0.5 229 1200 18 166 27 50No. 10 12 23 0.5 1.8 ,0.1 0.9 253 1574 33 672 38 247

Mean 16 34 0.6 2.1 ,0.1 0.7 270 1787 31 428 38 1996SEM 62 64 60.04 60.1 60.2* 625 6340 64 6117 66 675

a Values represent mean 6 SEM of blood urea nitrogen (BUN, in mg/dl), creatinine (mg/dl), uric acid (mg/dl), lactate dehydrogenase (LDH, in U/L), serum glutamicoxaloacetic transaminase (SGOT, in U/L), and serum glutamic pyruvate transaminase (SGPT, in U/L). Individual values of the animals in the C1-inh treatment groupare given.

bAnimals surviving for 62.5 h.c Animals surviving for .7 days.*p , 0.05 by Mann-Whitney U comparison.

479The Journal of Immunology

by guest on May 7, 2014

http://ww

w.jim

munol.org/

Dow

nloaded from

Contact system proteases.Mean plasma levels of factor XII andprekallikrein declined by;20 to 25% after 1 h in theuntreatedcontrol group, which was statistically different from baseline val-ues (Fig. 5,A andB; T 1 1 to T 1 10 h vs baseline:p , 0.05 byFischer LSD). In contrast, in the animals receiving C1-inh treat-ment, factor XII levels remained relatively stable until the end ofthe observation period and were only significantly reduced at T110 h (Fig. 5A). Similarly, plasma prekallikrein did not decline untilafter 6 h, although this decrease was not statistically different frominitial levels (Fig. 5B). Comparison of the untreated and treatedgroups indicated a significant difference in the 10-h course of fac-tor XII and prekallikrein (p , 0.05).Neutrophil degranulation. Elastase-a1-antitrypsin complexeswere assayed in the plasma of both groups to study the effects ofC1-inh administration onE. coli-induced neutrophil degranulation.In control animals, levels of these complexes steeply increasedshortly after start of the bacterial challenge, reaching plateau levelsfrom T 1 3 h and on, with maximal concentrations of 14916 82ng/ml at T1 10 h. C1-inh treatment neither affected kinetics norlevels of elastase complexes, and the highest concentrations werenoted at 6 h (14486 109 ng/ml; not shown).

DiscussionThe activation of complement and contact system proteases in hu-man and animal septic shock has been well documented (1, 5–12,41, 50, 51). In this study, we aimed to restrict activation of theseplasma cascade systems in baboons challenged with lethalE. coliby exogenous administration of C1-inh, to evaluate whether inhi-bition of these enzymes modulated the pathophysiologic responseobserved.

Metabolic studies with radiolabeled C1-inh in human volunteershave yielded a fractional catabolic rate of 2.5% of the plasma poolof C1-inh per hour (38). Based on this catabolic rate, we calculatedthat a continuous infusion of 200 U/kg over 9 h would be neces-sary to sustain the increase in circulating C1-inh induced by aloading dose of 500 U of C1-inh/kg of body weight. Constantlevels after the loading dose were indeed observed (Fig. 1), indi-cating that the fractional catabolic rate in humans can be used tocalculate dosing of C1-inh to baboons.

A rise in activation products of various components of the com-plement system was observed immediately after starting theE. coliinfusion. C1-inh supplementation abrogated the increase of plasmaC4b, indicating efficient inhibition of classical pathway activation.By contrast, the appearance of C3b and C5b-9 was incompletelyblocked by C1-inh, which indicates that circulating organisms mayhave directly activated the alternative pathway and/or that only asmall percentage of classical pathway zymogens need to be acti-vated to cleave their substrates in a catalytic manner, resulting inless efficient inhibition of activation downstream the cascade. Inaddition, the more pronounced effect of C1-inh administration onthe generation of C3b as compared with C5b-9 generation mayalso support a bypass mechanism of activation of C5, which ismediated by reactive oxygen species with the formation of a novelterminal complex containing oxidized C5 rather than C5b (52, 53).Notably, reduced activation of the complement cascade had noeffect on the clearance of the infusedE. coli bacteria, since circu-lating numbers of these organisms at 2 h postchallenge were sim-ilar in treatment and control groups. Thus, although complete in-hibition of complement at the level of C3 may impair bacterialclearance (54), inhibition at the level of C1 apparently does not.

C5a is generally regarded to be the most powerful anaphyla-toxin: i.v. administration of purified C5a to animals can inducehypotension (55), and pretreatment with anti-C5a Abs in a primatemodel of sepsis results in a recovery in mean arterial pressure (56).Moreover, it potently stimulates neutrophils to generate toxic ox-ygen radicals, to degranulate, and to aggregate (13). Our data showthat downstream complement activation was only modestlyblocked by C1-inh supplementation; effects of C1-inh are thereforenot likely related to inhibition of C5a generation.

Infusion of E. coli was associated with a protracted drop inantigenic levels of factor XII and prekallikrein within 1 h of start-ing the experiment, which was impeded by supplementation withC1-inh (Fig. 5). Reduced levels of factor XII and prekallikrein inthe control group likely reflected activation of the contact system(51). Despite the apparent inhibition of contact system activationin the treatment group, C1-inh administration was unable to pre-vent the severe hypotension observed uponE. coli infusion. Thisfinding does not agree with a study by Pixley et al. showing thatpretreatment with a mAb that inhibits activation of factor XII ab-rogated the secondary decline in arterial pressure observed in thismodel (51). Moreover, in further contrast to the data presentedhere, we demonstrated that blockade of the contact system by useof this anti-factor XII mAb modestly reduced the release ofneutrophil elastase and t-PA and inhibited the generation of

FIGURE 3. (Anti-)fibrinolytic parameters in baboons upon lethal E.coli challenge. Mean 6 SEM plasma levels of t-PA (A), plasmin-a2-antiplasmin complexes (B), and PAI-1 (C) in control (v) and C1-inh(V) groups. PAP complexes are expressed as percentage of NBP-MA-UK. The difference between both groups by ANOVA repeated mea-surements were significant for PAI-1. NS, not significant.

480 EFFECT OF C1-inh IN LETHAL PRIMATE SEPSIS

by guest on May 7, 2014

http://ww

w.jim

munol.org/

Dow

nloaded from

PAP complexes, suggestive of the involvement of the contactsystem in neutrophil activation and fibrinolysis (41). The find-ings described here need to be reconciled with those data aswell as with reported hemodynamic effects of C1-inh in septicpatients (31, 32) and endotoxemic rabbits (34). In the presentstudy, some degree of factor XII and prekallikrein activation islikely to have occurred at the C1-inh concentrations used (assuggested by slightly reduced factor XII and prekallikrein lev-els at the end of the observation period; Fig. 5). In addition,although bacteria or their products are likely candidates to haveinitiated activation of the contact system shortly after start ofthe challenge, negatively charged surfaces such as cell mem-branes and extracullar matrix exposed as a consequence of sep-sis-induced tissue damage may have contributed to contact sys-tem activation during later stages of the septic process. We havepreviously reported that in vitro, various glycosaminoglycans

may alter the relative contribution of serpins to inactivation ofcontact proteases, and a twofold protection of inhibition ofa-factor XIIa andb-factor XIIa by C1-inh could be demonstratedin the presence of dextran sulfate (57). Moreover, activator-bound factor XIIa may not be accessible by C1-inh, whereas itcan still be inhibited by Abs (58). The inactivation rate of C1-inh may therefore depend on the identity, localization, andphase (fluid or solid) of the in vivo activator, and parameterssuch as severity and distribution of organ damage and/or themodel of sepsis employed may influence the relative efficacy ofC1-inh supplementation. Thus, some activation of the contactsystem and the generation of bradykinin may have escaped in-hibition by C1-inh, but not by anti-factor XII mAb, in this ba-boon model. We suggest that this may explain observed differ-ences between the effects of C1-inh and anti-factor XIImAb (51).

FIGURE 4. Mean 6 SEM plasma levels of TNF (A), IL-10 (B), IL-6 (C), and IL-8 (D) in control (v) and C1-inh (V) groups after a lethal dose oflive E. coli. Differences between groups were significant ( p , 0.05) by ANOVA repeated measurements.

Table IV. Thrombin generation in baboons after lethal E. coli infusion with or without supplementation with C1-inha

Time (h)

T 1 0 T 1 1 T 1 2 T 1 3 T 1 4 T 1 6 T 1 8 T 1 10

Control 10 6 5 50 6 13 223 6 35 964 6 918 2088 6 396 2236 6 314 2020 6 505 1292 6 274C1-inh 3 6 0 26 6 1 240 6 67 918 6 267 1554 6 378 1668 6 452 1140 6 270 575 6 131

No. 5 4 27 113 1005 2230 2391 1606 860No. 6 3 24 343 1261 2125 2771 1163 803No. 8 2 23 453 1681 2040 1798 1031 439No. 9b 1 25 130 251 280 232 190 139No. 10 3 31 162 390 1097 1146 1709 636

a Values represent mean 6 SEM changes in thrombin-antithrombin III (TAT) complexes (in ng/ml) after start of the challenge (T 1 0). Individual TAT levels are givenfor the C1-inh treatment group. No significant difference was observed by ANOVA-repeated measurements.

b Surviving C1-inh-treated animals.

481The Journal of Immunology

by guest on May 7, 2014

http://ww

w.jim

munol.org/

Dow

nloaded from

Administration of high-dose C1-inh reduced circulating levelsof TNF, IL-6, IL-8, and IL-10 elicited by lethal infusion withE.coli. Differences in cytokine release were not due to variations inbacterial challenge, since the number of organisms in the infusionfluid was similar for both treatment and control groups. Rather, ourdata suggest the possibility of links between activated complementand/or contact proteases and the cytokine response during severesepsis. In vitro, the anaphylatoxins as well as bradykinin and factorXIIa can stimulate the synthesis and release of early response cy-tokines such as TNF, IL-1 and IL-6 (16–22). Recently, engage-ment of monocyte receptors for the fixed C3 fragments iC3b andC3b has been shown to induce IL-1 synthesis or synthesis andsecretion, respectively (59, 60). Our results do not allow conclu-sions regarding the mechanism of reduced cytokine release upon1-inh administration. However, considering the mild effects of thistreatment on the activation of C3 and C5 (see Fig. 2), we favor theexplanation that the attenuated cytokine release resulted from re-duced generation of contact activation products.

Many symptoms of septic shock can be reproduced by directinfusion of TNF into animals, and anti-TNF Abs reduces LPS-induced mortality and abrogates many of the attendant manifesta-tions of sepsis (61, 62). Among other effects, TNF has been shownto increase endothelial procoagulant activity (63) and induce thesecretion of pro- and antiinflammatory cytokines such as IL-10,IL-6, and IL-8 (64–66). We show here that lowest TNF levelswere measured in a treated surviving animal displaying retardedfibrinogen consumption and reduced thrombin formation. Survivalbenefit in this animal may therefore be interpreted to result indi-rectly from an impediment of systemic TNF release. On the otherhand, we have previously reported that the progressive elaborationof IL-6 and IL-8 during the later stages in this model is associatedwith sepsis-induced tissue damage and death, which may occur

despite the absence of circulating immunoreactive TNF (43, 49).Accordingly, the ability of C1-inh to modulate ongoing productionof IL-6 and IL-8 may go beyond a reduction of TNF release, andprotective effects onE. coli-induced organ injury during C1-inhtreatment may be linked to a direct interruption of these protago-nists in the proinflammatory merge of cascades. Modulatory ef-fects of C1-inh treatment on the cytokine response may also ex-plain observed effect on PAI-1 release (Fig. 2C). Synthesis andrelease of PAI-1 from endothelial cells and hepatocytes is inducedby IL-1, IL-6, and TNF (67–69). PAI-1 belongs to the serpin fam-ily and acts as a pseudosubstrate for t-PA and urokinase-type plas-minogen activator, forming inactive t-PA/PAI or urokinase-typeplasminogen activator/PAI complexes, respectively (70). How-ever, since t-PA levels were unaffected by C1-inh supplementa-tion, and PAP concentrations peaked before inhibitory effects onPAI became apparent, i.e., not until 4 h postchallenge, the atten-uating effects of C1-inh on (anti)fibrinolysis are likely of second-ary importance in thisE. coli model.

Hematologic and biochemical response profiles revealed that thelethal effects ofE. coli were related to the occurrence of dissem-inated intravascular coagulation and organ damage. None of thecontrol animals lived beyond 27 h, while in the treatment group,one animal survived the challenge and another lived beyond 48 h.Moreover, pathologic examination revealed less severe damage tovarious organs in four of the five animals receiving C1-inh, con-sistent with reduced organ dysfunction during the later stages (Ta-ble III). Notably, we administered a twofold lower dose of C1-inhto two additional animals, one of which survived (data not shown).These findings, together with the observed effects on the elabora-tion of cytokines, support the notion that C1-inh does interferewith reactions that occur in the microenvironment of the plasma/target cell interface and show that C1-inh supplementation has abeneficial, although mild effect on the inflammatory and physio-logic sequelae of lethalE. coli challenge.

In conclusion, we demonstrate here that administration of C1-inh to baboons suffering from lethal sepsis blocked classical com-plement activation and reduced the decrease in plasma levels offactor XII and prekallikrein. We suggest that, in this model, acti-vated complement and/or contact system proteases may promoteE. coli-induced organ injury and lethality, at least in part, by aug-mentating the cytokine response.

References1. Hack, C. E., and L. G. Thijs. 1991. The orchestra of mediators in the pathogenesis

of septic shock: a review. InUpdate in Intensive Care and Emergency Medicine.J. L. Vincent, ed. Springer-Verlag, Berlin, Germany, p. 232.

2. Morrison, D. C., and R. J. Ulevitch. 1978. The effect of bacterial endotoxins onhost mediator systems.Am. J. Pathol. 93:527.

3. Kalter, E. S., W. C. van Dijk, A. Timmermans, J. Verhoef, and B. N. Bouma.1983. Activation of purified human plasma prekallikrein triggered by cell wallfractins ofEscherichia coliandStaphylococcus aureus. J. Infect. Dis. 148:692.

4. Morrison, D. C., and L. F. Kline. 1977. Activation of classical and properdinpathways of complement by bacterial lipopolysaccharides (LPS).J. Immunol.118:362.

5. Nuijens, J. H., C. C. M. Huijbregts, A. J. M. Eerenberg-Belmer, J. J. Abbink,R. J. M. Strack van Schijndel, R. J. F. Felt-Bersma, L. G. Thijs, and C. E. Hack.1988. Quantification of plasma factor XIIa-C1-inhibitor and kallikrein-C1-inhib-itor complexes in sepsis.Blood 72:1841.

6. Mason, J. W., U. Kleeberg, P. Dolan, and R. W. Colman. 1970. Plasma kallikreinand Hageman factor in Gram-negative bacteremia.Ann. Intern. Med. 73:545.

7. Hack, C. E., J. H. Nuijens, R. J. F. Felt-Bersma, W. O. Schreuder,A. J. M. Eerenberg-Belmer, J. Paardekooper, W. Bronsveld, and L. G. Thijs.1989. Elevated plasma levels of the anaphylatoxins C3a and C4a are associatedwith a fatal outcome in sepsis.Am. J. Med. 86:20.

8. McCabe, W. R. 1973. Serum complement levels in bacteremia due to Gram-negative organisms.N. Engl. J. Med. 288:21.

9. Fearon, D. T., S. Ruddy, P. H. Schur, and W. R. McCabe. 1975. Activation of theproperdin pathway of complement in patients with Gram-negative bacteremia.N. Engl. J. Med. 292:937.

10. Hammerschmidt, D. E., L. D. Hudson, L. J. Weaver, P. R. Craddock, andH. S. Jacob. Association of complement activation and elevated plasma-C5a with

FIGURE 5. Mean 6 SEM plasma levels of factor XII (A) and prekal-likrein (B) in baboons receiving a lethal E. coli infusion (v) or a lethalE. coli dose supplemented with C1-inh (V). Comparison by ANOVArepeated measurements resulted in a significant difference ( p , 0.05)between both groups. *, p , 0.05 compared with baseline levels (T 10) by Fischer LSD.

482 EFFECT OF C1-inh IN LETHAL PRIMATE SEPSIS

by guest on May 7, 2014

http://ww

w.jim

munol.org/

Dow

nloaded from

adult respiratory distress syndrome: pathophysiological relevance and possibleprognostic value.Lancet 1:947.

11. Kalter, E. S., M. R. Daha, J. W. Ten Cate, J. Verhoef, and B. N. Bouma. 1985.Activation and inhibition of Hageman factor-dependent pathways and the com-plement system in uncomplicated bacteremia or bacterial shock.J. Infect. Dis.151:1019.

12. Groenevelt, A. B. J., W. Bronsveld, and L. G. Thijs. 1986. Haemodynamic de-terminants of mortality in human septic shock.Surgery 99:140.

13. Vogt, W. 1986. Anaphylatoxins: possible roles in disease.Complement 3:177.14. Mason, D. T., and K. L. Melmon. 1965. Effects of bradykinin on forearm venous

tone and vascular resistance in man.Circ. Res. 17:106.15. Wachtfogel, Y. T., R. A. Pixley, U. Kucich, W. Abrams, G. Weinbaum,

M. Schapira, and R. Colman. 1986. Purified plasma factor XIIa aggregates humanneutrophils and causes degranulation.Blood 67:1731.

16. Goodman, M. G., D. E. Chenoweth, and W. O. Weigle. 1982. Induction of in-terleukin 1 secretion and enhancement of humoral imunity by binding of humanC5a to macrophage surface C5a receptors.J. Exp. Med. 1156:912.

17. Haeffner-Cavaillon, N., J.-M. Cavaillon, M. Laude, and M. D. Kazatchkine.1987. C3a (C3adesArg) induces production and release of interleukin 1 by culturedhuman monocytes.J. Immunol. 139:794.

18. Okusawa, S., K. B. Yancey, J. W. M. van der Meer, S. Endres, G. Lonnermann,K. Hefter, M. M. Frank, J. F. Burke, C. A. Dinarello, and J. A. Gelfand. 1988.C5a stimulates secretion of tumor necrosis factor from human mononuclear cellsin vitro: comparison with secretion of interleukin 1b and interleukin 1a. J. Exp.Med. 168:443.

19. Scholz, W., M. R. McClurg, G. J. Cardenas, M. Smith, D. J. Noonan, T. E. Hugli,and E. L. Morgan. 1990. C5a-mediated release of interleukin 6 by human mono-cytes.Clin. Immunol. Immunopathol. 57:297.

20. Toossi, Z., J. R. Sedor, M. A. Mettler, B. Everson, T. Young, and O. D. Ratnoff.1992. Induction of expression of monocyte interleukin 1 by Hageman factor(factor XII). Proc. Natl. Acad. Sci. USA 89:11969.

21. Ratnoff, O. D., J. A. Voytus, and Z. Toossi. 1995. Inhibition of expression ofmonocyte interleukin-1 by inhibitors of Hageman factor (factor XII).J. Lab. Clin.Med. 125:247.

22. Pan, Z. K., B. L. Zuraw, C. C. Lung, E. R. Prossnitz, D. D. Browning, andR. D. Ye. 1996. Bradykinin stimulates NF-kB activation and interleukin 1b geneexpression in cultured human fibroblasts.J. Clin. Invest. 98:2042.

23. Carson, S. D., and D. R. Johnson. 1990. Consecutive enzyme cascades: comple-ment activation at the cell surface triggers increased tissue factor activity.Blood762:361.

24. Sims, P. J., E. M. Faioni, T. Wiedmer, and S. J. Shattil. 1988. Complementproteins C5b-9 cause release of membrane vesicles from the platelet surface thatare enriched in the membrane receptor for coagulation factor Va and expressprothrombinase activity.J. Biol. Chem. 263:18205.

25. Wiedmer, T., C. T. Esmon, and P. J. Sims. 1988. On the mechanism by whichcomplement proteins C5b-9 increase platelet prothrombinase activity.J. Biol.Chem. 261:14587.

26. Wiedmer, T., C. T. Esmon, and P. J. Sims. 1986. Complement proteins C5b-9stimulate procoagulant activity through platelet prothrombinase.Blood 68:875.

27. Sim, R. B., A. Reboul, G. J. Arlaud, C. L. Villiers, and M. G. Colomb. 1979.Interaction of125I-labelled complement subcomponents C1r and C1s with pro-tease inhibitors in plasma.FEBS Lett. 97:111.

28. De Agostini, A., H. R. Lijnen, R. A. Pixley, R. W. Colman, and M. Schapira.1984. Inactivation of factor-XII active fragment in normal plasma: predominantrole of C1-inhibitor.J. Clin. Invest. 73:1542.

29. Schapira, M., C. F. Scott, and R. W. Colman. 1982. Contribution of plasmaprotease inhibitors to the inactivation of kallikrein in plasma.J. Clin. Invest.69:462.

30. Nuijens, J. H., A. J. M. Eerenberg-Belmer, C. C. M. Huijbregts, W. O. Schreuder,R. J. F. Felt-Bersma, J. J. Abbink, L. G. Thijs, and C. E. Hack. 1989. Proteolyticinactivation of plasma C1-Inhibitor in sepsis.J. Clin. Invest. 84:443.

31. Hack, C. E., A. C. Ogilvie, B. Eisele, P. M. Jansen, J. Wagstaff, and L. G. Thijs.1994. Initial studies on the administration of C1-esterase inhibitor to patients withseptic shock or with a vascular leak syndrome induced by interleukin-2 therapy.In Bacterial Endotoxins: Basic Science to Anti-Sepsis Treatment. J. Levin,A. Sturk, T. van der Poll, and S. J. H. van Deventer SJH, eds. Wiley-Liss, Inc.,New York, p. 335.

32. Hack, C. E., H. J. Voerman, B. Eisele, H.-O. Keinecke, J. H. Nuijens,A. J. M. Eerenberg, A. Ogilvie, R. J. M. Strack van Schijndel, U. Delvos, andL. G. Thijs. 1992. C1-esterase inhibitor substitution in sepsis.Lancet 339:378.

33. Triantaphyllopoulos, D. C., and M. S. Cho. 1986. Effect of injection of C1-inactivator on the platelet count and blood coagulation in rabbits infused withendotoxin.Thromb. Haemost. 55:293.

34. Scherer, R. U., R. M. Giebler, U. Schmidt, D. Paar, and W. J. Kox. 1996. Theinfluence of C1-esterase inhibitor substitution on coagulation and cardiorespira-tory parameters in an endotoxin-induced rabbit model of hypercoagulability.Se-min. Thromb. Hemost. 22:357.

35. Guerrero, R., F. Velasco, M. Rodriguez, A. Lopez, R. Rojas, M. A. Alvarez,R. Villalba, V. Rubio, A. Torres, and D. del Castillo. 1993. Endotoxin-inducedpulmonary dysfunction is prevented by C1-esterase inhibitor.J. Clin. Invest. 91:2754.

36. Taylor, F. B., Jr., A. Chang, C. T. Esmon, A. D’Angelo, S. Vigano-D’Angelo,and K. E. Blick. 1987. Protein C prevents the coagulopathic and lethal effects ofEscherichia coliinfusion in baboons.J. Clin. Invest. 79:918.

37. Taylor, F. B., Jr., T. E. Emerson, Jr., R. Jordan, A. K. Chang, and K. E. Blick.1988. Antithrombin-III prevents the lethal effects ofEscherichia coliinfusion inbaboons.Circ. Shock 26:227.

38. Quastel, M., R. Harrison, M. Cicardi, C. A. Alper, and F. S. Rosen. 1983. Be-havior in vivo of normal and dysfunctional C1 inhibitor in normal subjects andpatients with hereditary angioneurotic edema.J. Clin. Invest. 71:1041.

39. Hack, C. E., J. Paardekooper, R. J. T. Smeenk, J. J. Abbink, A. J. M. Eerenberg,and J. H. Nuijens. 1988. Disruption of the internal thioester bond in the thirdcomponent of complement (C3) results in the exposure of neodeterminants alsopresent on activation products of C3: an analysis with monoclonal antibodies.J. Immunol. 141:1602.

40. Wolbink, G. J., J. Bollen, J. W. Baars, R. J. M. ten Berge, A. J. G. Swaak,J. Paardekooper, and C. E. Hack. 1993. Application of a monoclonal antibodyagainst a neoepitope on activated C4 in an ELISA for the quantification of com-plement activation via the classical pathway.J. Immunol. Methods 163:67.

41. Jansen, P. M., R. A. Pixley, M. Brouwer, I. W. de Jong, A. C. K. Chang,C. E. Hack, F. B. Taylor, Jr., and R. W. Colman. 1996. Inhibition of factor XIIin septic baboons attenuates the activation of complement and fibrinolytic sys-tems and reduces the release of interleukin-6 and neutrophil elastase.Blood 87:2337.

42. Helle, M., L. Boeije, E. R. de Groot, A. de Vos, and L. A. Aarden. 1991. SensitiveELISA for interleukin-6: detection of IL-6 in biological fluids: synovial fluids andsera.J. Immunol. Methods 38:47.

43. Jansen, P. M., J. van Damme, W. Put, I. W. de Jong, F. B. Taylor, Jr., andC. E. Hack. 1995. Monocyte chemotactic protein 1 (MCP-1) is released duringlethal and sublethal bacteremia in baboons.J. Infect. Dis. 171:1640.

44. Van der Pouw Kraan, T. C. T. M., L. C. M. Boeije, R. J. T. Smeenk, J. Wijdenes,and L. A. Aarden. 1995. Prostaglandin-E2 is a potent inhibitor of human inter-leukin-12 production.J. Exp. Med. 181:775.

45. Boermeester, M. A., P. A. M. van Leeuwen, S. M. Coyle, G. J. Wolbink,C. E. Hack, and S. F. Lowry. 1995. Interleukin-1 blockade attenuates mediatorrelease and dysregulation of the hemostatic mechanism during human sepsis.Arch. Surg. 130:739.

46. De Boer, J. P., A. A. Creasey, A. Chang, D. Roem, M. C. Brouwer,A. J. M. Eerenberg, C. E. Hack, and F. B. Taylor, Jr. 1993. Activation patternsof coagulation and fibrinolysis in baboons following infusion with lethal or sub-lethal dose ofEscherichia coli. Circ. Shock 39:59.

47. Nuijens, J. H., J. J. Abbink, Y. T. Wachtfogel, R. W. Colman, A. J. M. Eerenberg,D. Dors, A. J. M. Kamp, R. J. M. Strack van Schijndel, L. G. Thijs, andC. E. Hack. 1992. Plasma elastase-a1-antitrypsin and lactoferrin in sepsis: evi-dence for neutrophils as mediators in fatal sepsis.J. Lab. Clin. Med. 119:159.

48. Jansen, P. M., T. C. T. M. van der Pouw Kraan, I. W. de Jong, G. van Mierlo,J. Wijdenes, A. A. Chang, L. A. Aarden, F. B. Taylor, Jr., and C. E. Hack. 1996.Release of interleukin-12 in experimentalEscherichia coliseptic shock in ba-boons: relation to plasma levels of interleukin-10 and interferon-g. Blood 87:5144.

49. Jansen, P. M., I. W. de Jong, M. Hart, K. J. Kim, L. A. Aarden, L. B. Hinshaw,F. B. Taylor, Jr., and C. E. Hack. 1996. The release of leukemia inhibitory factorin primate sepsis. Analysis of the role of TNF-a. J. Immunol. 156:4401.

50. Pixley, R. A., R. de la Cadena, J. D. Page, N. Kaufman, E. G. Wyshock,R. W. Colman, A. Chang, and F. B. Taylor, Jr. 1992. Activation of the contactsystem in lethal hypotensive bacteremia in a baboon model.Am. J. Pathol. 140:897.

51. Pixley, R. A., R. de la Cadena, J. D. Page, N. Kaufman, E. G. Wyshock,A. Chang, F. B. Taylor, Jr., and R. W. Colman. 1993. The contact system con-tributes to hypotension but not disseminated intravascular coagulation in lethalbacteremia: in vivo use of a monoclonal anti-factor XII antibody to block contactactivation in baboons.J. Clin. Invest. 91:61.

52. Finn, A., B. P. Morgan, N. Rebuck, N. Klein, C. A. Rogers, M. Hibbs, M. Elliott,D. F. Shore, T. W. Evans, S. Strobel, and N. Moat. 1996. Effects of inhibition ofcomplement activation using recombinant soluble complement receptor 1 on neu-trophil CD11b/CD18 and L-selectin expression and release of interleukin-8 andelastase in simulated cardiopulmonary bypass.J. Thorac. Cardiovasc. Surg. 111:451.

53. Vogt, W., B. Damerau, I. von Zabern, R. Nolte, and D. Brunahl. 1989. Non-enzymatic activation of the fifth component of human complement by oxygenradicals: some properties of the activation product, C5b-like C5.Mol. Immunol.26:1133.

54. Quezado, Z. M. N., W. D. Hoffman, J. A. Winkelstein, I. Yatsiv, C. A. Koev,L. C. Cork, R. J. Elin, P. Q. Eichacker, and C. Natanson. 1994. The third com-ponent of complement protects againstEscherichia coliendotoxin-induced shockand multiple organ failure.J. Exp. Med. 179:569.

55. Lundberg, C., F. Marceau, and T. E. Hugli. 1987. C5a-induced hemodynamic andhematologic changes in the rabbit.Am. J. Pathol. 128:471.

56. Stevens, J. H., P. O’Hanley, J. M. Shapiro, F. G. Mihm, P. S. Satoh, J. A. Collins,and T. A. Raffin. 1986. Effects of anti-C5a antibodies on the adult respiratorydistress syndrome in septic primates.J. Clin. Invest. 77:1812.

57. Wuillemin, W. A., E. Eldering, F. Citarella, C. P. de Ruig, H. ten Cate, andC. E. Hack. 1996. Modulation of contact system proteases by glycosaminogly-cans: selective enhancement of the inhibition of factor XIa.J. Biol. Chem. 271:12913.

58. Dors, D. M., J. H. Nuijens, C. C. M. Huijbregts, and C. E. Hack. 1992. A novelsensitive assay for functional factor XII based on the generation of kallikrein-C1-inhibitor complexes in factor XII-deficient plasma by glass-bound factor XII.Thromb. Haemost. 67:644.

59. Couturier, C., N. Haeffner-Cavaillon, L. Weiss, E. Fischer, andM. D. Kazatchkine. 1990. Induction of cell-associated interleukin 1 through stim-ulation of the adhesion-promoting proteins LFA-1 (CD11a/CD18) and CR3(CD11b/CD18) 0f human monocytes.Eur. J. Immunol. 20:999.

483The Journal of Immunology

by guest on May 7, 2014

http://ww

w.jim

munol.org/

Dow

nloaded from

60. Bacle, F., N. Haeffner-Cavaillon, M. Laude, C. Couturier, andM. D. Kazatchkine. 1990. Induction of IL-1 release through stimulation of theC3b/C4b complement receptor type one (CR1, CD35) on human monocytes.J. Immunol. 144:147.

61. Remick, D. G., R. G. Kunkel, J. W. Larrick, and S. L. Kunkel. 1987. Acute invivo effects of human recombinant tumor necrosis factor.Lab. Invest. 56:583.

62. Beutler, B., W. Milsark, and A. Cerami. 1985. Passive immunization againstcachectin/tumor necrosis factor protects mice from lethal effects of endotoxin.Science 229:869.

63. Nawroth, P. P., and D. M. Stern. 1986. Modulation of endothelial cell hemostaticproperties by tumor necrosis factor.J. Exp. Med. 163:740.

64. Van der Poll, T., J. Jansen, M. Levi, H. ten Cate, J. W. ten Cate, andS. J. H. van Deventer. 1994. Regulation of interleukin-10 release by tumor ne-crosis factor in humans and chimpanzees.J. Exp. Med. 180:1985.

65. Fong, Y., K. J. Tracey, L. L. Moldawer, D. G. Hesse, K. R. Manogue,J. S. Kenney, A. T. Lee, G. C. Kuo, A. C. Allison, S. F. Lowry, and A. Cerami.1989. Antibodies to cachectin/tumor necrosis factor reduce interleukin 1b andinterleukin 6 appearance during lethal bacteremia.J. Exp. Med. 170:1627.

66. Van Deventer, S. J. H., M. Hart, T. van der Poll, C. E. Hack, and L. A. Aarden.1993. Endotoxin and tumor necrosis factor-alpha-induced interleukin-8 release inhumans.J. Infect. Dis. 167:461.

67. Van Hinsbergh, V. W. M., T. Kooistra, E. A. van den Berg, H. M. G. Princen,W. Fiers, and J. J. Emeis. 1988. tumor necrosis factor increases the production ofplasminogen activator inhibitor in human endothelial cells in vitro and rats invivo. Blood 72:1467.

68. Schleef, R. R., M. P. Bevilacqua, M. Sawdey, M. A. Gimbrone, Jr., andJ. J. Loskutoff. 1988. Cytokine activation of vascular endothelium.J. Biol. Chem.263:5797.

69. De Boer, J. P., J. J. Abbink, M. C. Brouwer, C. Meijer, D. Roem, G. P. Voorn,J. W. J. Lambers, J. A. van Mourik, and C. E. Hack. 1991. PAI-1 synthesis in thehuman hepatoma cell line Hep G2 is increased by cytokines: evidence that theliver contributes to acute phase behaviour of PAI-1.Thromb. Haemost. 65:181.

70. Travis, J. J., and G. S. Salvesen. 1983. Human plasma proteinase inhibitors.Annu. Rev. Biochem. 254:655.

484 EFFECT OF C1-inh IN LETHAL PRIMATE SEPSIS

by guest on May 7, 2014

http://ww

w.jim

munol.org/

Dow

nloaded from