INFECTION AND IMMUNITY, Apr. 2004, p. 1946–1955 Vol. 72, No. 40019-9567/04/$08.00�0 DOI: 10.1128/IAI.72.4.1946–1955.2004Copyright © 2004, American Society for Microbiology. All Rights Reserved.

N-Linked Glycosylation Is Required for C1 Inhibitor-MediatedProtection from Endotoxin Shock in Mice

Dongxu Liu, Xiaogang Gu, Jennifer Scafidi, and Alvin E. Davis III*CBR Institute for Biomedical Research, Harvard Medical School, Boston, Massachusetts 02115

Received 3 September 2003/Returned for modification 28 October 2003/Accepted 12 December 2003

C1 inhibitor (C1INH) prevents endotoxin shock in mice via a direct interaction with lipopolysaccharide(LPS). This interaction requires the heavily glycosylated amino-terminal domain of C1INH. C1INH in whichN-linked carbohydrate was removed by using N-glycosidase F was markedly less effective in protecting micefrom LPS-induced lethal septic shock. N-deglycosylated C1INH also failed to suppress fluorescein isothiocya-nate (FITC)-LPS binding to and LPS-induced tumor necrosis factor alpha mRNA expression by the murinemacrophage-like cell line, RAW 264.7, and cells in human whole blood. In an enzyme linked immunosorbentassay, the N-deglycosylated C1INH bound to LPS very poorly. In addition, C1INH was shown to bind todiphosphoryl lipid A (dLPA) but only weakly to monophosphoryl lipid A (mLPA). As with intact LPS, bindingof N-deglycosylated C1INH to dLPA and mLPA was diminished in comparison with the native protein.Removal of O-linked carbohydrate had no effect on any of these activities. Neither detoxified LPS, dLPA, normLPA had any effect on the rate or extent of C1INH complex formation with C1s or on cleavage of the reactivecenter loop by trypsin. These data demonstrate that N-linked glycosylation of C1INH is essential to mediateits interaction with the LPA moiety of LPS and to protect mice from endotoxin shock.

Septic shock caused by gram-negative bacteria is due primarilyto endotoxin lipopolysaccharide (LPS), which is a complex glyco-lipid found in the outer membrane of all gram-negative bac-teria (6). Treatment of gram-negative bacterial infections wouldbe greatly aided by substances that can effectively block pro-duction of inflammatory mediators from LPS-induced mono-nuclear phagocytes. Administration of LPS to humans or ex-perimental animals results in many of the physiological changesobserved during gram-negative bacterial infection, includingfever, hypotension, hypoglycemia, disseminated intravascularcoagulation, and shock (5). LPS is composed of two chemi-cally dissimilar structural regions: the hydrophilic repeatingpolysaccharide of the core and O-antigen structures and ahydrophobic domain known as lipid A (LPA) (40). LPA isthe toxic principle of gram-negative bacterial LPS and hasfull endotoxin activity (15, 38). Virtually all LPS-inducedbiological responses are LPA dependent (33). Therefore,recognition of LPA by cells must be the initial step in LPS-induced cellular responses. The general chemical structureof LPA from diverse gram-negative bacteria is highly con-served (40). LPA has the biological function to induce nu-clear factor-�B activation in monocytes (24) and the produc-tion of proinflammatory cytokines such as tumor necrosisfactor alpha (TNF-�) and interleukin-1 (IL-1) from macro-phages (2, 16). The acute-phase reactant LPS binding-pro-tein (LBP) binds to the LPA moiety with high affinity andfacilitates the transfer of LPS to CD14 (38, 40, 44). LPAcontains the binding domain recognized by LPS-bindingproteins, but the polysaccharide O chain and oligosaccha-ride core structure of endotoxin LPS may interfere with theinteraction of these proteins with LPA (1, 8).

Both the complement system and the contact system are im-

plicated in the pathophysiology of sepsis. C1 inhibitor (C1INH),a plasma glycoprotein that belongs to the superfamily of serineproteinase inhibitors, regulates these two systems. Levels ofproteolytically inactivated C1INH are increased in fatal septicshock, which suggests an increased turnover and a relativesecondary deficiency of biologically active C1INH (27). C1INHcan be inactivated by limited proteolytic cleavage by elastasereleased from activated neutrophils (7, 9). The inactivation ofC1INH may occur locally in inflamed tissue and thereby con-tribute to increased local complement activation (7). Both ac-tive C1INH and reactive center cleaved, inactive C1INH(iC1INH) protected mice from lethal gram-negative endotoxinshock (22). C1INH blocked the binding of Salmonella entericaserovar Typhimurium LPS to the murine macrophage cell line,RAW 264.7, and to human blood cells. It directly interactedwith LPS and suppressed LPS-induced TNF-� mRNA expres-sion. Deletion of the amino-terminal 97 amino acid residuesabolished the ability of C1INH to bind to LPS. Therefore,C1INH, in addition to its function as a serine protease inhib-itor, has a novel anti-inflammatory function mediated via theheavily glycosylated amino-terminal nonserpin domain (22).

We provide here evidence that N deglycosylation signifi-cantly reduced C1INH-mediated protection of mice from le-thal LPS-induced shock. The data also demonstrate that N-deglycosylated C1INH did not bind to LPS, could not inhibitthe binding of LPS to RAW 264.7 cells or to human bloodcells, and could not prevent the activation of these cells by LPS.Furthermore, additional experiments demonstrate that C1INHbinding to LPS was mediated via LPA and that this bindingalso was reduced by removal of N-linked carbohydrate fromC1INH.

MATERIALS AND METHODS

Reagents. C1INH and C1s were from Advanced Research Technologies (SanDiego, Calif.). Moloney murine leukemia virus (M-MuLV) reverse transcriptase(RT), N-glycosidase F (from Flavobacterium meningosepticum), and neuramini-dase (from Arthrobacter ureafaciens) were purchased from New England BioLabs

* Corresponding author. Mailing address: CBR Institute, 800 Hun-tington Ave., Boston, MA 02115. Phone: (617) 278-3379. Fax: (617)278-3490. E-mail: aldavis@cbr.med.harvard.edu.

1946

on February 14, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

(Beverly, Mass.). O-Glycosidase (from Diplococcus pneumoniae) was fromRoche Diagnostics GmbH (Mannheim, Germany). �-Phenylenediamine dihy-drochloride substrates, LPS, and fluorescein isothiocyanate (FITC)-conjugatedLPS from Salmonella enterica serovar Typhimurium were obtained from SigmaChemical Co. (St. Louis, Mo.), as were serovar Typhimurium LPS-derived mono-phosphoryl lipid A (mLPA), diphosphoryl lipid A (dLPA), and detoxified lipo-polysaccharide (dLPS). Rabbit anti-human C1INH antibody was purchased fromDako (Glostrup, Denmark). Goat anti-rabbit immunoglobulin G (H�L) conju-

gated with horseradish peroxidase was purchased from Pierce Biotechnology(Rockford, Ill.).

Enzymatic deglycosylation of C1INH. Deglycosylation was performed as de-scribed previously (12). N-linked carbohydrate was removed from plasma-derivedC1INH by treatment with 50 U of N-glycosidase F/ml for 2 h at 37°C. O-linkedcarbohydrate was removed by incubation with 0.1 U of O-glycosidase and 0.1 Uof neuraminidase/ml overnight at 37°C. To remove both N- and O-linked car-bohydrates, C1INH (10 �g/ml) was treated with 50 U of N-glycosidase F, 0.1 Uof O-glycosidase, and 0.1 U of neuraminidase/ml overnight at 37°C. Each of thesereactions was performed in sodium phosphate buffer (50 mM, pH 7.5).

Mouse endotoxin shock model. C57BL/6J mice (male, 6 to 8 weeks, 18 to 22 g)(Charles River Laboratories, Wilmington, Mass.) were injected intraperitoneally(i.p.) with a lethal dose (20 mg/kg) of serovar Typhimurium LPS mixed witheither normal intact, N-deglycosylated, O-deglycosylated, or both N- and O-deglycosylated C1INH (200 �g/mouse). Control mice were injected (i.p.) withLPS (20 mg/kg) alone, with a mixture of LPS (20 mg/kg) and the buffer used fordeglycosylation, with the deglycosylation buffer alone, with a mixture of LPS andthe glycosidase enzymes, or with the glycosidases alone. Mice were monitored for5 days. Kaplan-Meier survival analysis was performed by using GraphPad Prism4.00 (GraphPad Software, Inc., San Diego, Calif.). Survival curves of the differentgroups were compared by using the log-rank test (GraphPad Prism 4.00). Allexperiments were performed in compliance with relevant laws and institutionalguidelines and were approved by the CBR Institute for Biomedical ResearchAnimal Care and Use Committee.

Cell culture. The murine RAW 264.7 macrophage cell line (American TypeCulture Collection [ATCC], Manassas, Va.) was cultured in Dulbecco minimalessential medium (DMEM; ATCC) supplemented with 10% fetal bovine serum(FBS; Invitrogen, Carlsbad, Calif.) at 37°C in 5% CO2. Confluent macrophageswere detached by washing with phosphate-buffered saline (PBS; pH 7.4).

FACS. The murine RAW 264.7 macrophages were incubated with FITC-conjugated LPS (175 ng/ml) in the absence or presence of the various deglyco-sylated C1INH (150 �g/ml) preparations in DMEM containing 10% FBS for 15min at 37°C. In other experiments, human peripheral venous blood from anormal volunteer was collected into a tube containing EDTA (1 mg/ml of wholeblood). Aliquots of the whole blood were treated with LPS at a final concentra-tion of 175 ng/ml in the absence or presence of added N-deglycosylated C1INH(150 �g/ml), O-deglycosylated C1INH (150 �g/ml), or both N- and O-deglyco-

FIG. 1. SDS-PAGE analysis of deglycosylated C1INH. C1INH (10�g) was incubated with N-glycosidase F, O-glycosidase, and neuramin-idase as described in Materials and Methods. The various forms ofC1INH were analyzed by SDS-PAGE and stained with Coomassiebrilliant blue.

FIG. 2. Effect of deglycosylated C1INH on survival of mice in gram-negative endotoxin LPS-induced lethal endotoxemia. C57BL/6J mice wereinjected i.p. with a mixture of LPS (20 mg/kg) with native, plasma-derived C1INH (‚, n � 20), with N-deglycosylated C1INH (Œ, n � 10), withO-deglycosylated C1INH (E, n � 10), or with both N- and O-deglycosylated C1INH (F, n � 10) (200 �g/per mouse). Control mice were injected(i.p.) with LPS (20 mg/kg) alone (■, n � 21) or a mixture of LPS (20 mg/kg) with the glycosidase buffer (�, n � 5). The indicated P values arefor each group in comparison with the group treated with native, plasma-derived C1INH.

VOL. 72, 2004 C1 INHIBITOR-MEDIATED PROTECTION FROM ENDOTOXIN SHOCK 1947

on February 14, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

sylated C1INH (150 �g/ml) for 15 min at 37°C. Cells were fixed with fluores-cence-activated cell sorter (FACS) solution after being washed with PBS threetimes. The binding of FITC-conjugated LPS was analyzed on a FACSCaliburapparatus (Becton Dickinson, San Jose, Calif.) by using CellQuest software.

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).C1INH was incubated with C1s (ratio, 1:1) for 20 min at 37°C after the additionof mLPA, dLPA, or dLPS from serovar Typhimurium LPS. Reactions were sub-jected to electrophoresis with a 10% SDS-Tris-glycine polyacrylamide gel (Invitro-gen). The gels were stained with Coomassie blue and then dried for 1 h at 80°C.

RT-PCR. Total RNA was isolated from the RAW 264.7 macrophages inducedwith LPS (175 ng/ml) in the presence or absence of N-deglycosylated C1INH,O-deglycosylated C1INH, or both N- and O-deglycosylated C1INH and thenreverse transcribed by using M-MuLV RT with oligo(dT)20 primers (Invitrogen)for 1 h at 37°C. PCR primers were designed to generate mouse TNF-� and�-actin fragments, each with lengths of 200 bp (TNF-�, sense [5�-ATGAGCACAGAAAGCATGATCC-3�] and antisense [5�-GAGGCCATTTGGGAACTTCTC-3�]; �-actin, sense [5�-TGGATGACGATATCGCTGC-3�] and antisense [5�-AGGGTCAGGATACCTCTCTT-3�]). PCR products were analyzed on 1.2%(wt/vol) agarose gels containing 0.5 �g of ethidium bromide/ml and visualizedunder UV light. Band density was analyzed and quantified by using ImageQuantsoftware (Molecular Dynamics, Sunnyvale, Calif.). In addition, human periph-eral venous blood was collected from a normal volunteer and anticoagulated withEDTA (1 mg/ml of whole blood). Aliquots of the whole blood were treated withLPS at a final concentration of 175 ng/ml with or without added C1INH (5 to 150�g/ml) for 15 min at 37°C. In other experiments, total RNA was isolated from theblood leukocytes and was reverse transcribed by using M-MuLV RT with oligo(dT)20 primers. PCR primers were designed for human TNF-� (sense [5�-ATGAGCACTGAAAGCATGATCCGGGACGTG-3�] and antisense [5�-AGGTCCCTGGGGAACTCTTCCCTCTG-3�]) and human �-actin (sense [5�-ATGGATGATGATATCGCCGCGCTCGTCGTC-3�] and antisense [5�-AGGGTGAGGATGCCTCTCTTGCTCTG-3�]).

Enzyme-linked immunosorbent assay (ELISA). Plates (polyvinyl chloride, 96U-bottom wells; Becton Dickinson, Franklin Lakes, N.J.) were coated withmLPA, dLPA, dLPS, and LPS at room temperature overnight. C1INH (150�g/ml) was incubated with mLPA-, dLPA-, dLPS-, and LPS-coated plates for 1 hat room temperature, respectively. In other experiments, deglycosylated C1INH(150 �g/ml) was treated for 2 h at 37°C or overnight at 37°C and was stopped bychilling, followed by incubation in LPS (175 ng/ml)-coated plates for 1 h at roomtemperature. Rabbit anti-human C1INH antibody (1:1,000) was incubated for1 h at room temperature. After being washed, plates were incubated with Im-munoPure goat anti-rabbit immunoglobulin G (H�L) conjugated with horse-radish peroxidase (1:100,000). Color reactions were developed for 3 min at roomtemperature, and reactions were terminated with 3 N HCl. The absorbance ofeach well was measured at 490 nm by using Revelation Microsoft in an MRXMicroplate reader (Dynex Technologies, Chantilly, Va.). Standard samples weredetected with the different concentrations of human C1INH binding to rabbitanti-human C1INH antibody.

RESULTS

Analysis of deglycosylated C1INH. Treatment of plasma-derived intact C1INH with neuraminidase reduced the molec-ular mass, as judged by SDS-PAGE, from 106 to 96 kDa (Fig.1). Removal of N- and O-linked carbohydrate resulted in de-creases to 89 and 83 kDa, respectively, while the combinationof N- and O-glycosidase treatment reduced the molecular massto 75 kDa. These apparent size differentials are consistent withpreviously published data (30).

Removal of N-linked carbohydrate abrogates the ability ofC1INH to protect mice from lethal endotoxin shock. C1INHprevented the adverse biologic effects of LPS via a mechanismunrelated to protease inhibition and this protection appearedto be secondary to a direct interaction with endotoxin (22).Deletion of the amino-terminal 97 amino acid residues abol-ished the ability of C1INH to bind to LPS in vitro. However,the available quantities of this recombinant truncated C1INHare insufficient for testing in animals. We previously demon-strated, by using the lowest dose of LPS (20 mg/kg), that

FIG. 3. Effect of deglycosylated C1INH on the binding of FITC-conjugated LPS to the murine macrophage cell line, RAW 264.7. LPSbinding, thick line; control, shaded field; mean fluorescence intensitiesfor the control and treated cells are indicated in the upper left andupper right corners, respectively, in each panel. (A) RAW 264.7 mac-rophages were incubated with FITC-LPS (175 ng/ml) in the absence orpresence of C1INH, N-deglycosylated C1INH, O-deglycosylated C1INH,or both N- and O-deglycosylated C1INH (each at 150 �g/ml) in DMEMcontaining 10% FBS for 15 min at 37°C. (B) RAW 264.7 macrophageswere incubated with FITC-LPS (175 ng/ml) in the presence of differentdoses of O-deglycosylated C1INH or untreated native C1INH (150, 75,37.5, 10, or 5 �g/ml) in DMEM containing 10% FBS for 15 min at 37°C.(C) RAW 264.7 macrophages were incubated with FITC-LPS (175 ng/ml)in the presence of iC1INH (150 �g/ml) lacking N-, O-, or both N- andO-linked glycosylation. Cells were fixed with FACS solution after beingwashed with PBS three times and were analyzed by FACS.

1948 LIU ET AL. INFECT. IMMUN.

on February 14, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

resulted in 100% mortality of C57BL/6J mice within 48 h thata single dose of C1INH (200 �g) improved survival to 45 and50% when administered via the i.p. and intravenous routes,respectively. A mixture of LPS (20 mg/kg) and C1INH (200�g/per mouse) given i.p. increased survival to 65% (22). Inorder to evaluate the role of carbohydrate in C1INH-mediatedprotection from endotoxin shock, mice were injected i.p. withLPS mixed with either native C1INH, N-deglycosylated C1INH,O-deglycosylated C1INH, or both N- and O-deglycosylated

C1INH (200 �g/mouse). Treatment with native C1INH andO-deglycosylated C1INH resulted in 55 and 70% survival, re-spectively (Fig. 2). These survival curves were not significantlydifferent (P � 0.6322). N-deglycosylated C1INH and both N-and O-deglycosylated C1INH, on the other hand, resulted insurvival rates of only 20 and 10%, respectively, at 72 h. Thesurvival of each of these groups was significantly less than thatof mice treated with native C1INH (P � 0.0062 and 0.0013,respectively). However, although only 2 of 10 and 1 of 10 mice

FIG. 3—Continued.

VOL. 72, 2004 C1 INHIBITOR-MEDIATED PROTECTION FROM ENDOTOXIN SHOCK 1949

on February 14, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

survived in the groups treated with N-deglycosylated C1INHand with N- and O-deglycosylated C1INH, respectively, theirsurvival rates were significantly different from mice treatedwith LPS alone (no mice survived of 21 given LPS alone; P �0.0087 and 0.0151). Therefore, N-deglycosylated C1INH andN- and O-deglycosylated C1INH may provide some protectionfrom endotoxin shock, but the level of protection is much less

than that with either native or O-deglycosylated C1INH. Con-trol mice treated with either N- or O-glycosidase alone allsurvived, whereas none of five mice treated with N-glycosidaseplus LPS and one of five mice treated with both N- and O-glycosidase plus LPS survived (data not shown). These data

FIG. 4. Effect of deglycosylated C1INH on the binding of FITC-conjugated LPS to human blood cells. LPS binding, thick line; control,shaded field. Mean fluorescence intensities for the control and treatedcells are indicated in the upper left and upper right corners, respec-tively, in each panel. The human blood cells were incubated for 15 minat 37°C with FITC-LPS (175 ng/ml) in the presence of C1INH (150�g/ml) lacking N-, O-, or both N- and O-linked glycosylation. Cellswere fixed with FACS solution after being washed with PBS threetimes and were analyzed by FACS.

FIG. 3—Continued.

1950 LIU ET AL. INFECT. IMMUN.

on February 14, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

indicate that N-linked glycosylation of C1INH is essential toprotect mice from lethal LPS-induced endotoxin shock.

C1INH lacking N glycosylation failed to suppress LPS bind-ing to the murine macrophage cell line, RAW 264.7, and tohuman blood cells. The finding that C1INH has the ability toblock the binding of serovar Typhimurium LPS to macro-phages prompted us to further investigate whether N- or O-linked glycosylation participates in this suppression of LPSbinding. We tested by FACS the ability of deglycosylatedC1INH to inhibit LPS binding to RAW 264.7 macrophages inthe presence of 10% FBS. N-deglycosylated C1INH (150 �g/ml) lost the ability to suppress FITC-LPS binding to RAW264.7 cells (Fig. 3A). However, removal of O-linked carbohy-drate had no effect on the inhibition of LPS binding (Fig. 3B).As described previously, C1INH in which the reactive centerloop has been cleaved with trypsin (iC1INH) also inhibits thebinding of LPS to RAW 264.7 cells. As with intact C1INH,

removal of N-linked carbohydrate eliminates this activity,whereas O-deglycosylation has no effect (Fig. 3C). Neither theN-glycosidase nor the O-glycosidase interfered with FITC-LPSbinding to RAW 264.7 cells (data not shown). In addition,whole human blood was incubated with LPS (175 ng/ml) in thepresence of C1INH (150 �g/ml) and the three different formsof deglycosylated C1INH. N deglycosylation eliminated theability of C1INH to block the binding of LPS to human bloodcells, whereas C1INH lacking O-linked glycosylation retainedthe ability to suppress LPS binding (Fig. 4).

In addition, TNF-� mRNA expression induced by LPS wasdetected in RAW 264.7 cells by RT-PCR. LPS-mediated up-regulation of TNF-� mRNA was completely suppressed aftertreatment with native plasma-derived C1INH (150 �g/ml) orwith O-deglycosylated C1INH (150 �g/ml) (Fig. 5A). Similarly,C1INH with O-linked carbohydrate removed (150 �g/ml) alsosuppressed LPS-induced TNF-� mRNA expression by cells in

FIG. 5. Effect of deglycosylated C1INH on LPS-induced TNF-�mRNA expression in the murine macrophage cell line, RAW 264.7 andhuman blood cells. (A) total RNA from RAW 264.7 macrophages wasisolated after treatment with LPS (175 ng/ml) in the presence ofC1INH or with N-, O-, or both N- and O-deglycosylated C1INH (all at150 �g/ml) for 30 min at 37°C. RT-PCR was performed with mouseTNF-� cDNA and �-actin cDNA primers. The band intensity wasquantitated by using ImageQuant software (Molecular Dynamics), wasnormalized to the � actin level, and is expressed relative to the amountof PCR product present in the samples treated with LPS alone. (B) To-tal RNA from human blood cells was isolated after treatment of wholehuman blood with LPS (175 ng/ml) in the presence of C1INH (5, 10,37.5, 75, or 150 �g/ml) or with N-deglycosylated C1INH (150 �g/ml),O-deglycosylated C1INH (5, 10, 37.5, 75, or 150 �g/ml), or N- andO-deglycosylated C1INH (150 �g/ml) (upper panel) for 30 min at37°C. RT-PCR was performed with human TNF-� cDNA and �-actincDNA primers. PCR products were quantitated as in panel A.

VOL. 72, 2004 C1 INHIBITOR-MEDIATED PROTECTION FROM ENDOTOXIN SHOCK 1951

on February 14, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

whole human blood (Fig. 5B). However, removal of N-linkedcarbohydrate completely destroyed the ability of C1INH to in-hibit LPS-mediated upregulation of TNF-� by both RAW 264.7cells and by human peripheral blood cells (Fig. 5A and B).

N-linked glycosylation of C1INH is crucial for its interac-tion with LPS. LPS is composed of two structural regions: thehydrophilic repeating polysaccharide of the core and O-anti-gen structures and the hydrophobic LPA domain (40). Virtu-ally all of the biological responses induced by LPS are depen-dent upon LPA (15, 33, 38). As expected from the experimentsdescribed above, deletion of N-linked carbohydrate or of N-and O-linked carbohydrate from C1INH completely elimi-nated its ability to bind to LPS, but the binding of O-deglyco-

sylated C1INH (150 �g/ml) was unaltered in comparison withnative C1INH (Fig. 6A). To further investigate the mechanismof interaction of C1INH with endotoxin, intact C1INH and thedifferent forms of deglycosylated C1INH were analyzed forbinding to immobilized mLPA, dLPA, or dLPS by using anELISA. Analysis of C1INH binding to mLPA (0.5, 1.0, and 5.0ng/ml), dLPA (0.5, 1.0, and 5.0 ng/ml), and dLPS (0.5, 1.0, and5.0 ng/ml) showed binding to both dLPA and mLPA but verylittle binding to dLPS (Fig. 6B). Similarly, O-deglycosylatedC1INH bound to both mLPA and dLPA (Fig. 6C). N-degly-cosylated (and N/O-deglycosylated) C1INH showed very littlebinding to mLPA. Both, however, did bind to dLPA, althoughnot as well as did native or O-deglycosylated C1INH. Thesedata suggested that N-linked glycosylation is essential to theinteraction of C1INH with LPS. As with most other proteinsthat bind to LPS, LPA likely represents the primary structurein the LPS molecule that is recognized by C1INH.

Neither LPA nor dLPS alter the ability of C1INH to complexwith C1s. To examine whether mLPA, dLPA, and dLPS had aneffect on the formation of C1INH-C1s complexes, native, ac-tive C1INH (10 �g) was incubated with mLPA, dLPA, anddLPS, respectively, followed by the addition of C1s (10 �g).The C1INH treated with mLPA, dLPA, and dLPS retained theability to form C1INH-C1s complexes (Fig. 7A). In addition,mLPA, dLPA, and dLPS did not interfere with the susceptibilityof the reactive center loop to cleavage with trypsin (Fig. 7B).

FIG. 6. Analysis of the binding of deglycosylated C1INH to LPSand LPA. The interaction of C1INH or deglycosylated C1INH withimmobilized LPS, mLPA, dLPA, and dLPS was analyzed by ELISA.(A) Binding of deglycosylated C1INH (150 �g/ml) to LPS (50, 100, or175 ng/ml) in comparison with native C1INH (150 �g/ml); (B) bindingof C1INH (150 �g/ml) to mLPA (0.5, 1.0, or 5.0 �g/ml), dLPA (0.5,1.0, or 5.0 �g/ml), and dLPS (0.5, 1.0, and 5.0 �g/ml); (C) binding ofdeglycosylated C1INH (150 �g/ml) to mLPA (1.0 �g/ml), dLPA (1.0�g/ml), or dLPS (1.0 �g/ml).

1952 LIU ET AL. INFECT. IMMUN.

on February 14, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

FIG. 7. Effects of mLPA, dLPA, and dLPS on the formation of C1INH-C1s complexes and C1INH cleavage by trypsin. (A) mLPA (0.1, 0.2, 0.5, or1.0 �g), dLPA (0.1, 0.2, 0.5, or 1.0 �g), and dLPS (0.1, 0.2, 0.5, or 1.0 �g) have no effect on the rate or extent of C1INH-C1s (10 �g: 10 �g) complexformation, as assessed by SDS-PAGE stained with Coomassie brilliant blue. (B) mLPA (0.5, 25, or 50 ng), dLPA (0.1, 0.5, 25, or 50 ng), and dLPS (0.1,0.5, 25, or 50 ng) have no effect on the formation of cleaved C1INH (10 �g) by trypsin, as assessed by SDS-PAGE stained with Coomassie brilliant blue.

VOL. 72, 2004 C1 INHIBITOR-MEDIATED PROTECTION FROM ENDOTOXIN SHOCK 1953

on February 14, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

DISCUSSION

The data presented here demonstrate that N-linked glyco-sylation of C1INH is required for the interaction of C1INHwith endotoxin. Removal of N-linked carbohydrate, but not theremoval of O-linked carbohydrate, resulted in a significantreduction in the ability of C1INH to protect mice from LPS-mediated endotoxin shock (Fig. 2). Consistent with this obser-vation, N-linked carbohydrate also was required for inhibitionof LPS binding to RAW 264.7 cells (Fig. 3) and to humanblood cells (Fig. 4) and for the induction of TNF-� synthesis bythese cells (Fig. 5). Lastly, in an ELISA, the direct binding ofC1INH to LPS was dependent on N glycosylation (Fig. 6A).Previous data demonstrated that iC1INH, which has no pro-tease inhibitory activity, retained the ability to bind to LPS butthat binding required the presence of the amino-terminal non-serpin domain (22). A recombinant C1INH variant with dele-tion of the amino-terminal 97 amino acids retained the abilityto inactivate proteases but did not bind to LPS (11, 22). Thisdomain consists of 120 amino acids and is extremely heavilyglycosylated with three N-linked and at least seven O-linkedcarbohydrate groups (4). Three additional N-linked groups arewithin the serpin domain. Deglycosylation of C1INH with N-glycanase, O-glycanase, or both has no significant effect onprotease inhibitory function (30). The combination of the cur-rent and previous findings, therefore, indicate that the threeN-linked carbohydrates within the amino-terminal domain atAsn residues 3, 47, and 59 and not those within the serpin do-main are required for the reactivity of C1INH with endotoxin.

LPS interacts with a number of glycoproteins, including thethree components of the cell membrane receptor for LPS,CD14, MD-2 and Toll-like receptor 4 (TLR4) (12, 19, 36, 37,42, 43). LPS is transported by the plasma protein, LPS-bindingprotein (LBP), and is transferred to the LPS receptor on thecell surface. Engagement of this receptor ultimately results intriggering of innate immune responses, in particular, increasedexpression of TNF-� and a variety of other cytokines. MD-2has two N glycosylation sites at Asn26 and Asn114, whereasTLR4 has nine N-linked sites in its amino-terminal extracellu-lar domain (12). Binding of LPS to the LPS receptor requiresN-linked glycosylation of both MD-2 and TLR4 (12) but not ofCD14 (36). It is, however, not known whether the glycosylationof TLR4 is required to directly mediate the binding of LPS orif it is necessary for stabilization of the complex receptor (12).In the case of MD-2, however, the binding site for LPS hasbeen shown to consist of a positively charged domain that issimilar to the binding domains described in a number of otherLPS-binding proteins (23, 29). Therefore, with MD-2, as mightbe expected, the role of glycosylation must be to maintain theconformation of the protein rather than to participate directlyin binding. Although a number of residues within CD14 havebeen shown to be required for binding of LPS, the precisebiochemical basis of this binding also remains to be defined(19, 36, 37, 42, 43).

LPS is released from the outer membrane of gram-negativebacteria and is largely responsible for the symptoms of septicshock. The active component of endotoxin, LPA, is a phospho-glycolipid with an acylated and phosphorylated dihexosamineheadgroup. The polysaccharide component contains antigenicdeterminants but does not contribute to endotoxin activity (32,

33). The removal of an acid labile phosphate group and normalfatty acid groups from dLPA diminishes endotoxic activity.Although mLPA lacks many of the endotoxic properties ofLPS (2), in vitro it induces the production of proinflammatorycytokines from macrophages (16) and gamma interferon andIL-2 from lymphocytes (10). However, mLPA is markedly lessactive than is LPS in the induction of these cytokines. Inaddition, mLPA and LPS may differentially regulate the pro-duction of some cytokines. For example, mLPA induced IL-10to a greater extent than did LPS (34). In addition, mLPAactivates human dendritic cells and T cells and enhances thegeneration of both Th1- and Th2-specific immune responses inmice (13, 18). Previously, we demonstrated that C1INH inhib-its the interaction of LPS with LBP and, ultimately, with mac-rophages (22). The data presented here show that C1INHinteracts with the LPA moiety within LPS to prevent this in-teraction. C1INH at a fixed concentration (150 �g/ml) boundto immobilized dLPA but did not bind as efficiently to mLPA,and bound extremely poorly to dLPS (Fig. 6B). As with bindingto LPS, the binding of C1INH to LPA also was reduced by Ndeglycosylation but not by removal of O-linked carbohydrate(Fig. 6C). However, the reduction in binding of N-deglycosy-lated versus native C1INH to dLPA (Fig. 6C) was not asstriking as the reduction in binding of N-deglycosylated C1INHto LPS (Fig. 6A). This could be a result of variable and incom-plete deglycosylation in the two experiments. However, a morelikely explanation is that, analogous to the apparent situationwith MD-2, the binding site on the C1INH amino-terminaldomain does not reside on the N-linked carbohydrate itself,but is on the peptide backbone. Removal of carbohydrate mayresult in alteration of the conformation of this domain andpartially mask the binding site. The smaller dLPA moleculemay be able to more readily gain access to the binding site,whereas the larger LPS molecule would be unable to interactwith the site. Binding of LPS (or dLPA) to many LPS-bindingproteins is via interaction of the phosphate groups on the LPAwith specific positively charged residues within the LPS-bind-ing protein (1, 3, 14, 17, 20, 21, 25, 26, 28, 31, 35, 39, 41).C1INH has one Arg (at position 18) and 3 Lys residues (atpositions 22, 30, and 55) within the amino-terminal domain.Mutagenesis studies are in progress to test this hypothesis andto define which of the three carbohydrate groups within theamino-terminal domain are required for binding to endotoxin.

In summary, the binding of C1INH to LPS prevents theinteraction of LPS with LBP which in turn prevents the deliv-ery of LPS to cells that express the LPS receptor complex. Thecharacteristics of this binding are that it is mediated by theLPA moiety of LPS, it does not require an intact C1INHreactive center loop (and therefore is not dependent on pro-tease inhibitory activity), and it requires the C1INH amino-terminal nonserpin domain and intact N-linked carbohydrate.

ACKNOWLEDGMENTS

This study was supported by Public Health Service grants HD22082and HD33727 from the National Institute of Child Health and HumanDevelopment.

REFERENCES

1. Appelmelk, B. J., Y.-Q. An, M. Geerts, B. G. Thijs, H. A. De Boer, D. M.MacLaren, J. de Graaff, and J. H. Juijens. 1994. Lactoferrin is a lipidA-binding proten. Infect. Immun. 62:2628–2632.

1954 LIU ET AL. INFECT. IMMUN.

on February 14, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

2. Astiz, M. E., R. C. Rackow, J. G. Still, S. T. Howell, A. Cato, K. B. VonEschen, J. T. Ulrich, J. A. Rudbach, G. McMahon, R. Vargas, and W. Stern.1995. Pretreatment of normal humans with monophosphoryl lipid A inducestolerance to endotoxin: a prospective, double-blind, randomized, controlledtrial. Crit. Care Med. 23:9–17.

3. Battafarano, R. J., P. S. Dahlberg, C. A. Ratz, J. W. Johnston, B. H. Gray,J. R. Haseman, K. H. Mayo, and D. L. Dunn. 1995. Peptide derivatives ofthree distinct lipopolysaccharide binding proteins inhibit lipopolysaccharide-induced tumor necrosis factor-alpha secretion in vitro. Surgery 118:318–324.

4. Bock, S. C., K. Skriver, E. Nielsen, H. C. Thogersen, B. Wiman, V. H.Donaldson, R. L. Eddy, J. Marrinan, E. Radziejewska, R. Huber, T. B.Shows, and S. Magnussen. 1986. Human C1 inhibitor: primary structure,cDNA cloning, and chromosomal localization. Biochemistry 25:4292–4301.

5. Bone, R. C. 1993. gram-negative sepsis: a dilemma of modern medicine. Clin.Microbiol. Rev. 6:57–68.

6. Bone, R. C. 1991. The pathogenesis of sepsis. Ann. Intern. Med. 115:457–469.

7. Brower, M. S., and P. C. Harpel. 1982. Proteolytic cleavage and inactivationof �2 plasmin inhibitor and C1 inactivator by human polymorphonuclearleukocyte elastase. J. Biol. Chem. 257:9849–9854.

8. Caccavo, D., A. Afeltra, S. Pece, G. Giuliani, M. Freudenberg, C. Galanos,and E. Jirillo. 1999. Lactoferrin-lipid A-lipopolysaccharide interaction: in-hibition by anti-human lactoferrin monoclonal antibody AGM 10.14. Infect.Immun. 67:4668–4672.

9. Caliezi, C., W. A. Wuillemin, S. Zeerleder, M. Redondo, B. Eisele, and C. E.Hack. 2001. C1-esterase inhibitor: an anti-inflammatory agent and its poten-tial use in the treatment of diseases other than hereditary angioedema.Pharmacol. Rev. 52:91–112.

10. Carozzi, S., M. Salit, A. Cantaluppi, M. G. Nasini, S. Barocci, S. Cantarella,and S. Lamperi. 1989. Effect of monophosphoryl lipid A on the in vitrofunction of peritoneal leukocytes from uremic patients on continuous am-bulatory peritoneal dialysis. J. Clin. Microbiol. 27:1748–1753.

11. Coutinho, M., K. S. Aulak, and A. E. Davis III. 1994. Functional analysis ofthe serpin domain of C1 inhibitor. J. Immunol. 153:3648–3654.

12. da Silva Correia, J., and R. J. Ulevitch. 2002. MD-2 and TLR4 N-linkedglycoproteins are important for a functional lipopolysaccharide receptor.J. Biol. Chem. 277:1845–1854.

13. De Becker, G., V. Moulin, B. Pajak, C. Bruck, M. Francotte, C. Thiriart, J.Urbain, and M. Moser. 2000. The adjuvant monophosphoryl lipid A in-creases the function of antigen-presenting cells. Int. Immunol. 12:807–815.

14. Ferguson, A. D., W. Welte, E. Hofmann, B. Lindner, O. Holst, J. W. Coulton,and K. Diederichs. 2000. A conserved structural motif for lipopolysacchariderecognition by procaryotic and eucaryotic proteins. Structure 8:585–592.

15. Grabarek, J., G. Her, V. N. Reinhold, and J. Hawiger. 1990. Endotoxic lipidA interaction with human platelets: structure-function analysis of lipid Ahomologs obtained from Salmonella Minnesota Re595 lipopolysaccharide.J. Biol. Chem. 265:8117–8121.

16. Henricson, B. E., C. L. Mnathey, P. Y. Perera, T. A. Hamilton, and S. N.Vogel. 1993. Dissociation of lipopolysaccharide (LPS)-inducible gene expres-sion in murine macrophages pretreated with smooth LPS versus monophos-phoryl lipid A. Infect. Immun. 61:2325–2333.

17. Hoess, A., S. Watson, G. R. Siber, and R. Liddington. 1993. Crystal structureof an endotoxin-neutralizing protein from the horseshoe crab, Limulus anti-LPS factor, at 1.5 A resolution. EMBO J. 12:3351–3356.

18. Ismaili, J., J. Rennesson, E. Aksoy, J. Vekemans, B. Vincart, Z. Amraoui, F.Van Laethem, M. Goldman, and P. M. Dubois. 2002. Monophosphoryl lipidA activates both human dendritic cells and T cells. J. Immunol. 168:926–932.

19. Juan, T. S., E. Hailman, M. J. Kelley, L. A. Busse, E. Davy, C. J. Empig, L. O.Narhi, S. D. Wright, and H. S. Lichenstein. 1995. Identification of a lipo-polysaccharide binding domain in CD14 between amino acids 57 and 64.J. Biol. Chem. 270:5219–5224.

20. Lamping, N., A. Hoess, B. Yu, T. C. Park, C.-J. Kirschning, C. Pfeil, D.Reuter, S. D. Wright, F. Herrman, and R. R. Schumann. 1996. Effects ofsite-directed mutagenesis of basic residues (Arg 94, Lys 95, Lys 99) ofLPS-binding protein on binding and transfer of LPS and subsequent immunecell activation. J. Immunol. 157:4648–4656.

21. Little, R. G., D. N. Kelner, E. Lim, D. J. Burke, and P. J. Conlon. 1994.Functional domains of recombinant bactericidal/permeability increasing pro-tein (rBPI23). J. Biol. Chem. 269:1865–1872.

22. Liu, D., S. Cai, X. Gu, J. Scafidi, X. Wu, and A. E. Davis III. 2003. C1inhibitor prevents endotoxin shock via a direct interaction with lipopolysac-charide. J. Immunol. 171:2594–2601.

23. Mancek, M., P. Pristovsek, and R. Jerala. 2002. Identification of LPS-binding peptide fragment of MD-2, a Toll-receptor accessory protein. Bio-chem. Biophys. Res. Commun. 292:880–885.

24. Mansell, A., A. Reinicke, D. Marharet Worral, and L. A. J. O’Neill. 2001.The serine protease inhibitor antithrombin III LPS-mediated NF-�B activa-tion by TLR-4. FEBS Lett. 508:313–317.

25. Nagaoka, I., S. Hirota, F. Niyonsaba, M. Hirata, Y. Adachi, H. Tamura, andD. Heumann. 2001. Cathelicidin family of antibacterial peptides CAP18 andCAP11 inhibit the expression of TNF-� by blocking the binding of LPS toCD14� cells. J. Immunol. 167:3329–3338.

26. Nagaoka, I., S. Hirota, F. Niyonsaba, M. Hirata, Y. Adachi, H. Tamura, S.Tanaka, and D. Heumann. 2002. Augmentation of the lipopolysaccharide-neutralizing activities of human cathelicidin CAP18/LL-37-derived antimi-crobial peptides by replacement with hydrophobic and cationic amino acidresidues. Clin. Diagn. Lab. Immunol. 9:972–982.

27. 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. Proteolytic inactivation of plasma C1 inhibitor in sepsis. J. Clin. In-vestig. 84:443–450.

28. Odell, E. W., R. Sarra, M. Foxworthy, D. S. Chapple, and R. W. Evans. 1996.Antibacterial activity of peptides homologous to a loop region in humanlactoferrin. FEBS Lett. 382:175–178.

29. Re, F., and J. L. Strominger. 2003. Separate functional domains of humanMD-2 mediate Toll-like receptor 4-binding and lipopolysaccharide respon-siveness. J. Immunol. 171:5272–5276.

30. Reboul, A., M. Prandini, and M. Colomb. 1987. Proteolysis and deglycosy-lation of human C1 inhibitor: effect on functional properties. Biochem. J.24:117–121.

31. Reyes, O., M. G. Vallespi, H. E. Garay, L. J. Cruz, L. J. Gonzalez, G. Chinea,W. Buurman, and M. J. Arana. 2002. Identification of single amino acidresidues essential for the binding of lipopolysaccharide (LPS) to LPS bindingprotein (LBP) residues 86–89 using an Ala-scanning library. J. Peptide Sci.8:144–150.

32. Ribi, E. 1986. Structure-function relationship of bacterial adjuvants, p. 35. InR. M. Nervig, P. M. Gough, M. L. Kaeverle, and C. A. Whetstone (ed.),Advances in carries and adjuvants for veterinary biologicals. Iowa StateUniversity Press, Ames.

33. Rietschel, E. T., T. Kirikae, F. U. Schade, U. Mamat, G. Schmidt, H. Lop-pnow, A. J. Ulmer, U. Zahringer, U. Seydel, F. di Padova, M. Schreier, andH. Brade. 1994. Bacterial endotoxin: molecular relationships of structure toactivity and function. FASEB J. 8:217–225.

34. Salkowski, C. A., G. R. Detore, and S. N. Vogel. 1997. Lipopolysaccharideand monophosphoryl lipid A differentially regulate interleukin-12, gammainterferon, and interleukin-10 mRNA production in murine macrophages.Infect. Immun. 65:3239–3247.

35. Scott, M. G., A. C. E. Vreugdenhil, W. A. Buurman, R. E. W. Hancock, andM. R. Gold. 2000. Cationic antimicrobial peptides block the binding oflipopolysaccharide (LPS) to LPS binding protein. J. Immunol. 164:549–553.

36. Stelter, F., M. Bernheiden, R. Menzel, R. S. Jack, S. Witt, X. Fan, M. Pfister,and C. Schutt. 1997. Mutation of amino acids 39–44 of human CD14 abro-gates binding of lipopolysaccharide and Escherichia coli. Eur. J. Biochem.243:100–109.

37. Stelter, F., H. Loppnow, R. Menzel, U. Grunwald, M. Bernheiden, R. S. Jack,A. J. Ulmer, and C. Schutt. 1999. Differential impact of substitution of aminoacids 9–13 and 91–101 of human CD14 on soluble CD14-dependent activa-tion of cells by lipopolysaccharide. J. Immunol. 163:6035–6044.

38. Takada, H., and S. Kotani. 1989. Structural requirements of lipid A forendotoxicity and other biological activities. CRC Crit. Rev. Microbiol. 16:423–477.

39. Taylor, A. H., G. Heavner, M. Nedelman, D. Sherris, E. Brunt, D. Knight,and J. Ghrayeb. 1995. Lipopolysaccharide neutralizing peptides reveal a lipidA binding site of LPS binding protein. J. Biol. Chem. 270:17934–17938.

40. Ulevitch, R. J., and P. S. Tobias. 1995. Receptor-dependent mechanisms ofcell stimulation by bacterial endotoxin. Annu. Rev. Immunol. 13:437–457.

41. van Berkel, P. H., M. E. Geerts, H. A. van Veen, M. Mericskay, H. A. de Boer,and J. H. Nuijens. 1997. N-terminal stretch Arg2, Arg3, Arg4 and Arg5 ofhuman lactoferrin is essential for binding to heparin, bacterial lipopolysac-charide, human lysozyme and DNA. Biochem. J. 328:145–151.

42. Viriyakosol, S., and T. N. Kirkland. 1995. A region of human CD14 requiredfor lipopolysaccharide binding. J. Biol. Chem. 270:361–368.

43. Viriyakosol, S., J. C. Mathison, P. S. Tobias, and T. N. Kirkland. 2000.Structure-function analysis of CD14 as a soluble receptor for lipopolysac-charide. J. Biol. Chem. 275:3144–3149.

44. Wilde, C. G., J. J. Seilhamer, M. McGrogan, N. Ashton, J. L. Snable, J. C.Lane, S. R. Leong, M. B. Thornton, K. L. Miller, R. W. Scott, and M. N.Marra. 1994. Bactericidal/permeability-increasing protein and lipopolysac-charide (LPS)-binding protein. LPS binding properties and effects on LPS-mediated cell activation. J. Biol. Chem. 269:17411–17416.

Editor: A. D. O’Brien

VOL. 72, 2004 C1 INHIBITOR-MEDIATED PROTECTION FROM ENDOTOXIN SHOCK 1955

on February 14, 2018 by guest

http://iai.asm.org/

Dow

nloaded from