Of Proteinase-3 and Cd11b/CD18 Binds to Activated Neutrophils

The Soluble Endothelial Protein C Receptor Binds to Activated Neutrophils: Involvement of Proteinase-3 and CD11b/CD18

This information is current as Shinichiro Kurosawa, Charles T. Esmon and Deborah J. of October 1, 2021. Stearns-Kurosawa J Immunol 2000; 165:4697-4703; ; doi: 10.4049/jimmunol.165.8.4697 http://www.jimmunol.org/content/165/8/4697 Downloaded from

References This article cites 44 articles, 26 of which you can access for free at: http://www.jimmunol.org/content/165/8/4697.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on October 1, 2021

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2000 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Soluble Endothelial Protein C Receptor Binds to Activated Neutrophils: Involvement of Proteinase-3 and CD11b/CD181

Shinichiro Kurosawa,2,3* Charles T. Esmon,*† and Deborah J. Stearns-Kurosawa3*

The protein C pathway is a primary regulator of blood coagulation and a critical component of the host response to inflammatory stimuli. The most recent member of this pathway is the endothelial protein C receptor (EPCR), a type I transmembrane protein with homology to CD1d/MHC class I proteins. EPCR accelerates formation of activated protein C, a potent anticoagulant and antiinflammatory agent. The current study demonstrates that soluble EPCR binds to PMA-activated neutrophils. Using affinity chromatography, binding studies with purified components, and/or blockade with specific Abs, it was found that soluble EPCR binds to proteinase-3 (PR3), a neutrophil granule proteinase. Furthermore, soluble EPCR binding to neutrophils was partially ␤ dependent on Mac-1 (CD11b/CD18), a 2 integrin involved in neutrophil signaling, and cell-cell adhesion events. PR3 is involved in multiple diverse processes, including hemopoietic proliferation, antibacterial activity, and autoimmune-mediated vasculitis. The

␤ Downloaded from observation that soluble EPCR binds to activated neutrophils via PR3 and a 2 integrin suggests that there may be a link between the protein C anticoagulant pathway and neutrophil functions. The Journal of Immunology, 2000, 165: 4697–4703.

roteinase-3 (PR3)4 is a serine proteinase stored in neutro- In ongoing studies with the primate model of Escherichia coli- phil granules and released to the cell surface upon activa- mediated septic shock, we found that blocking the endothelial pro- P tion (1–3). PR3 is generally thought of as a soluble en- tein C receptor (EPCR) in vivo with specific mAbs transformed a zyme, in the sense of being released into the local inflammatory transient response to sublethal E. coli into a lethal response (16). http://www.jimmunol.org/ milieu upon neutrophil activation and disgorgement of granular EPCR is a type 1 transmembrane protein expressed on endothe- contents. However, several recent studies have observed that most lium that binds protein C zymogen and facilitates formation of of the neutrophil PR3 remains associated with the cell membrane activated protein C, a potent anticoagulant (17) and antiinflamma- upon activation, with very little released into the medium (4). PR3 tory protein (18, 19). A soluble form of EPCR, which exists in is possibly best known as the primary target Ag of the PR3 anti- plasma, retains full ligand-binding capability (20), and levels in- neutrophil cytoplasmic Abs (PR3-ANCA) in Wegener’s granulo- crease in patients with sepsis or systemic lupus erythematosus (21) matosis, a debilitating autoimmune disease characterized by ne- either from vascular injury or through a regulated proteolytic re- crotizing vasculitis (5, 6). PR3-ANCA are also found in patients lease of soluble receptor (22). Of particular interest in the primate with microscopic polyangiitis, a systemic vasculitic disease (7). study was the observation of a massive polymorphonuclear cell by guest on October 1, 2021 PR3 has activities that include degradation of extracellular matrix influx into specific areas of the adrenal, liver, and kidney, suggest- proteins (8), regulation of myeloid differentiation (9, 10), potenti- ing that EPCR may be involved in neutrophil interactions and leu- ation of platelet activation (11), and antibacterial action that is kocyte trafficking. independent of its enzymatic activity (12). Structurally, PR3 is In the present study, we examined whether EPCR could interact very similar to neutrophil elastase (13), but it does have unique directly with neutrophils. It was found that recombinant soluble substrates, including the membrane-bound precursors of the proin- EPCR (sEPCR) binds to activated neutrophils via PR3 and sup- ␣ ␤ ␤ flammatory TNF- and IL-1 cytokines (14, 15). Thus, PR3 ex- ported, in part, by CD11b/CD18 (Mac-1), a 2 integrin involved in pression near a vascular surface would very likely contribute to cell-cell adhesion and neutrophil-signaling events. The results are local tissue destruction and inflammation. discussed with respect to their potential importance in modulating vascular damage due to inflammatory stimuli. *Cardiovascular Biology Research, Oklahoma Medical Research Foundation, Oklahoma Materials and Methods City, OK 73104; and †Howard Hughes Medical Institute, Oklahoma City, OK 73104 Proteins Received for publication September 17, 1999. Accepted for publication July 24, 2000. The costs of publication of this article were defrayed in part by the payment of page sEPCR (3) was a recombinant soluble human EPCR and was prepared as charges. This article must therefore be hereby marked advertisement in accordance described previously (23). The construct codes for the extracellular domain of with 18 U.S.C. Section 1734 solely to indicate this fact. EPCR truncated immediately above the transmembrane domain at residue 210, with a 12-residue HPC4 epitope tag at the carboxyl terminus for calcium- 1 This research was supported by a scientist development grant from the American Heart Association (to S.K.), an Atorvastation Research Award (to S.K.), and grants dependent HPC4-immunoaffinity purification (24). sEPCR was labeled with from the National Heart, Lung, and Blood Institute of the National Institutes of Health Oregon Green 488 carboxylic acid, succinimidyl ester 5-isomer (Molecular (Grant R01 HL64787-01 to S.K.; Grants PO1 HL54804 and R37 HL30340 to C.T.E.). Probes, Eugene, OR), or Cy3 (Pharmacia-Amersham, Uppsala, Sweden) using C.T.E. is an Investigator of the Howard Hughes Medical Institute. standard methods. Human neutrophil PR3 was obtained from Athens Research 2 Address correspondence and reprint requests to Dr. Shinichiro Kurosawa, Free Rad- and Technology (Athens, GA) and biotinylated with sulfo-NHS-LC-biotin ical Biology and Aging Department, Oklahoma Medical Research Foundation, 825 (sulfosuccinimidyl 6-(biotinamido)hexanoate; Pierce, Rockford, IL). NE 13th Street, Oklahoma City, OK 73104. E-mail address: kurosawas@omrf. ouhsc.edu Monoclonal Abs 3 Current address: Free Radical Biology & Aging Department, Oklahoma Medical The preparation of and screening methods for mAbs against human EPCR Research Foundation, 825 NE 13th Street, Oklahoma City, OK 73104. have been described elsewhere (20). The PL1 mAb against neutrophil P- 4 Abbreviations used in this paper: PR3, proteinase-3; ANCA, anti-neutrophil cyto- selectin glycoprotein ligand-1 was a kind gift from Kevin Moore (Cardio- plasmic Ab; EPCR, endothelial protein C receptor; P-ANCA, perinuclear ANCA; vascular Biology Research, Oklahoma Medical Research Foundation, sEPCR, recombinant soluble EPCR. Oklahoma City, OK). The IB4 (IgG2a) hybridoma cell line was obtained

from American Type Culture Collection (Manassas, VA), and the TS1 sEPCR binding to activated neutrophils (IgG1) hybridoma cell line was obtained from the Developmental Studies ␮ Hybridoma Bank, Department of Biological Sciences, University of Iowa For experiments using blood, 50- l aliquots were incubated at 37°C for 10 ␮ (Iowa City, IA). Abs were purified from hybridoma supernatants by protein min with 500 nM Oregon Green-sEPCR and 0.1 g/ml PMA (Sigma). G affinity chromatography using standard methods. mAbs against CD11a Samples were placed on ice, and RBC were lysed with 1 ml FACS Lysing (clone HI111; IgG1) and CD11b [clone ICRF44 (44); IgG1] were pur- Solution (Becton Dickinson). After centrifugation, pellets were resus- chased from PharMingen (San Diego, CA). Hybridoma supernatant con- pended in 1 ml of ice-cold HHB/CaMg buffer and centrifuged. The washed taining Ab against CD11c (IgG1) was obtained from Serotec (Oxford, pellet was resuspended in ice-cold HHB/CaMg buffer and analyzed by flow U.K.). The Ab preparations against CD11b and CD11c ␣-chains were re- cytometry. Purified neutrophils were evaluated for sEPCR binding in a ␤ similar fashion, with the exception of the lysing solution. ported by their respective manufacturers to inhibit 2 integrin-mediated adhesion functions. V189 (IgG2a), an anti-human factor V mAb, served as sEPCR binding to neutrophils in the presence of Abs against an isotype control for IB4. ␤ 2 integrins Blood collection and neutrophil preparation Heparinized blood (50 ␮l) was preincubated at room temperature for 15 min with purified Ab (IB4, TS1, PL1, V189, anti-CD11a, anti-CD11b, Blood was collected by venipuncture into vacutainer tubes (Becton Dick- anti-CD11c), 10 ␮l of anti-CD11c hybridoma culture supernatant, or 10 ␮l inson, San Jose, CA) containing heparin (100 U.S.P. units), buffered so- of nonconditioned media. The final concentration of the purified Abs was dium citrate (3.8%), or EDTA (7.5% K EDTA). For anticoagulation with 3 40 ␮g/ml. Oregon Green-sEPCR (500 nM) was added in the absence or hirudin (Sigma, St. Louis, MO), blood was collected into a sterile tube presence of PMA (100 nM), and the samples were incubated at 37°C for 15 containing 50 U/ml hirudin and mixed. min. RBC were removed by hypotonic lysis, the cells were washed, and For experiments using purified neutrophils, heparinized blood was cell-associated fluorescence was determined by flow cytometry, as de- mixed with an equal volume of 6% Dextran 70 in 0.9% NaCl (McGaw, scribed above. Irvine, CA). RBC were allowed to sediment for 45–60 min at room tem-

perature. The supernatant was removed and centrifuged at 400 ϫ g (10 Microscopy Downloaded from min, 4°C). The pellet was resuspended in 25 ml of ice-cold 0.2% NaCl with mixing for 25 s, followed by 25 ml of ice-cold 1.6% NaCl. The cells were The PR3-ANCA, perinuclear ANCA (P-ANCA), and sEPCR interaction centrifuged and the pellet gently resuspended in 5 ml of HHB buffer with ethanol-fixed neutrophils was evaluated with a Leica TCS NT con- (HBSS, 10 mM HEPES, pH 7.5, 1% BSA). This was transferred to a 15-ml focal system equipped with four-laser excitation, four fluorescent detectors, conical centrifuge tube and underlayered with 5 ml of Lymphocyte Sepa- and a transmitted detector (Heidelberg, Germany). These studies were per- ration Media (density 1.077 Ϯ 0.001 g/ml at 22°C; Cellgro, Herndon, VA). formed at the Flow and Image Cytometry Laboratory, University of Okla- The sample was centrifuged at 400 ϫ g for 30 min at 4°C. The pellet homa Health Sciences Center, William K. Warren Medical Research In- containing purified neutrophils was resuspended in 5–10 ml HHB buffer. A stitute (Oklahoma City, OK). Slides with ethanol-fixed neutrophils, http://www.jimmunol.org/ typical yield was ϳ10 ϫ 106 neutrophils per preparation. P-ANCA, and FITC anti-human IgG were from The Binding Site (Bir- mingham, U.K.). PR3-ANCA was from the Varelisa Autoimmune Analy- sEPCR affinity chromatography sis kit (Pharmacia & UpJohn, Kalamazoo, MI). The cells were incubated with the autoantibodies (30 min) and washed with HHB buffer. Bound Ab Leukocytes were purified from 50 ml of citrated blood by sedimentation was detected with FITC anti-human IgG reagent according to the manu- through dextrose and hypotonic lysis, as described above. The cell pellet facturer’s directions. The slides were washed once in HHB and twice in was washed and resuspended in HBSS, 10 mM HEPES (pH 7.5), 3 mM HHB/CaMg, and then incubated with Cy3-sEPCR (1 ␮M) in HHB/CaMg CaCl2, and 0.6 mM MgCl2 and surface biotinylated with 0.5 mg/ml sulfo- buffer (30 min). The slides were washed with HHB/CaMg and mounted NHS-LC-biotin for 30 min at room temperature. The cells were washed, with a coverslip and Slow Fade reagent (Molecular Probes). The slides then lysed with 1% Nonidet P-40 in 10 mM HEPES (pH 7.5) and 0.1 mM were then visualized for FITC and Cy3 fluorescence. EDTA, and centrifuged. The supernatant was mixed with 100 ␮lof by guest on October 1, 2021 sEPCR-AffiGel 10 (Bio-Rad; ϳ1 mg sEPCR/ml resin) or Tris-inactivated Binding of PR3 and sEPCR AffiGel 10 (control) in HHB/CaMg buffer (HHB with 3 mM CaCl2, 0.6 ␮ ␮ Microtiter plate wells were coated overnight at 4°C with 4 g/ml of mAbs mM MgCl2; total volume 600 l). The samples were mixed overnight at against sEPCR (1494, 1500, or HPC4 at 50 ␮l/well). All subsequent steps 4°C, washed five times with buffer (without albumin), and eluted with were done at room temperature. The wells were washed with wash buffer HBSS, 50 mM HEPES (pH 7.5) and 5 mM EDTA. The samples were (20 mM Tris-HCl, pH 7.5, 0.1 M NaCl, 0.05% Tween 20, 3 mM CaCl , 0.6 centrifuged and the supernatant was analyzed by SDS-PAGE on a 4–20% 2 mM MgCl2), blocked, washed again, and then incubated with 100 nM gradient gel. The samples were transferred to a polyvinylidene difluoride sEPCR (50 ␮l/well) for 30 min. This approach was used to create a binding membrane, and the membrane was blocked with a 10% (w/v) nonfat dry surface because EPCR tends to be conformationally sensitive and, at least milk solution, washed, and incubated with streptavidin-alkaline phospha- for screening mAbs, immobilization of EPCR directly to plastic surfaces tase conjugate. Bound conjugate was detected with ECF substrate (Amer- destroys a majority of the available epitopes. After washing, increasing sham-Pharmacia), and image analysis was performed using a Storm 860 concentrations of biotin-PR3 were added (30 min). The plates were imager (Molecular Dynamics, Sunnyvale, CA). washed, and bound biotin-PR3 was detected with streptavidin-alkaline The procedure was scaled up so that purified neutrophils from 400 ml phosphatase (1 ␮g/ml, 30 min) and Blue Phos substrate (Kirkegaard & citrated blood were washed and lysed with 1% Nonidet P-40, 10 mM Perry Laboratories, Gaithersburg, MD). Color development was stopped HEPES (pH 7.5), and 0.1 mM EDTA. The lysate was centrifuged, and the with 2.5% EDTA, and the absorbance at 650 nm was read in a Vmax supernatant was diluted 5-fold and adjusted to 3 mM CaCl2, 0.6 mM microplate reader (Molecular Devices, Sunnyvale, CA). The background MgCl2. This was applied to an sEPCR-Affi-Gel 10 affinity column (25 ml ϳ absorbance observed in the absence of coating Ab was subtracted from all of resin coupled at 1 mg sEPCR/ml resin) equilibrated in 20 mM Tris- samples. HCl, 0.1 M NaCl, 0.2% Nonidet P-40, 3 mM CaCl2, and 0.6 mM MgCl2 Microtiter plates precoated with human PR3 (The Binding Site) were (pH 7.5), with a 10-ml precolumn of Tris-inactivated Affi-Gel 10 to reduce used to assess the ability of excess unlabeled sEPCR to inhibit binding of nonspecific interactions. The cell pellet was extracted twice more with biotin-sEPCR to purified PR3. sEPCR (0–5 ␮M) was added to the PR3- lysing buffer, and the supernatants were applied. The column was washed coated wells (30 min, room temperature), followed by 500 nM biotin- to baseline OD280 and eluted with buffer containing 50 mM HEPES (pH sEPCR (30 min). The wells were washed (HHB/CaMg), and bound biotin- 7.5) and 5 mM EDTA. The eluate fractions were pooled, concentrated, and sEPCR was detected with streptavidin-alkaline phosphatase and Blue Phos analyzed by SDS-PAGE on a 10% resolving gel. The band corresponding substrate, as described above. to the 33-kDa band observed previously was cut out and sent to the Har- vard Microchemistry Facility (Cambridge, MA) for matrix-assisted laser Normal sera and patient sera containing anti-PR3 Abs desorption/ionization mass spectrometry (MALDI-MS) analysis, trypsin digestion, and amino-terminal sequencing. Sera from five patients were kindly provided for this study by Judith James-Wood (Arthritis & Immunology Department, Oklahoma Medical Flow cytometry Research Foundation). Four of the patients (P1, P2, P3, and P5) have bi- opsy-proven Wegener’s granulomatosis, as well as anti-PR3 autoantibodies Data acquisition by flow cytometry and subsequent analysis was done with by commercial ELISA (PR3-ANCA), and were not on aggressive immu- a FACScaliber flow cytometer using CellQuest software (Becton Dickin- nosuppressive therapy at the time of collection. The fifth patient (P4) has son, San Jose, CA). Neutrophils were gated according to their relative size anti-PR3 autoantibodies, but only eye involvement and not systemic dis- (forward scatter) and relative granularity (side scatter) properties. ease, and does not fulfill the criteria for Wegener’s. These samples were The Journal of Immunology 4699 used to evaluate the influence of serum containing anti-PR3 autoantibodies on the binding of sEPCR and PR3 in vitro. Normal sera were obtained from three apparently healthy adult volunteers (N1, N2, and N3) using standard venipuncture techniques. IgG was purified for some experiments on protein G columns using standard methods. Influence of sera on sEPCR binding to PR3 Serum samples were diluted 1/200 and 1/1000 and preincubated for 30 min at room temperature on plates precoated with PR3 (The Binding Site). Biotinylated-sEPCR was added (80 nm final), and incubation continued for an additional 30 min. The plates were washed and bound biotin signal detected with streptavidin-HRP and tetramethylbenzidine substrate. The absorbance at 450 nm was read on a Vmax microplate reader. The maximal FIGURE 2. sEPCR binds to neutrophils activated in blood. Anticoagu- binding (100%) was the level of bound sEPCR after preincubation of the lated blood (as indicated) was incubated with Oregon Green-sEPCR (500 PR3 wells with buffer (10 mM HEPES, pH 7.4, 0.1 M NaCl, 0.05% Tween nM) at 37°C for 10 min in the absence (gray) or presence (black) of 0.1 20) instead of serum. ␮g/ml PMA. RBC were lysed, and the remaining washed cells were analyzed for cell-bound fluorescence (MCF) by flow cytometry. Neutrophils Results were gated according to their forward and side scatter properties. As an initial approach to evaluate the possibility of direct EPCR- neutrophil interactions, sEPCR was labeled with fluorescent Ore- gon Green and mixed with purified neutrophils, and cell-bound from the sample incubated with the sEPCR affinity resin (Fig. 3,

fluorescence was monitored by flow cytometry. As shown in Fig. lane 2). More bands were observed with longer exposure times, Downloaded from 1, the fluorescently labeled sEPCR bound to PMA-activated neu- but correlated with those in the control sample (lane 1). trophils. Without activation (no PMA), the cell-associated fluores- The procedure was scaled up, and the corresponding band from cence was only slightly higher than the background cell autofluo- the EDTA eluate of an sEPCR-affinity column was cut from the gel rescence (neutrophils alone). In contrast, the fluorescence and sent to the Harvard Microchemistry Facility for MALDI-MS associated with the PMA-activated neutrophils (solid black) was at analysis, trypsin digestion, and microsequencing. From one pep-

least an order of magnitude higher than the controls. tide of the digest, the sequence identified was LVNVVLGAH- http://www.jimmunol.org/ To test whether the above phenomenon could be observed in the NVR, which was used in a search of the Swiss-Protein database. presence of all the components of blood, fluorescently labeled The sequence was a unique match to residues of Leu70-Arg81 of sEPCR was added to blood samples anticoagulated with citrate, PR3 (molecular mass ϳ30 kDa), a serine proteinase stored in the EDTA, heparin, or hirudin, and binding was evaluated by flow secretory vesicles and primary and azurophilic granules of neutro- cytometry after lysis of RBC (Fig. 2). Essentially the same results phils. This sequence of PR3 is distinct from the homologous region of were obtained, and activation of the neutrophils again greatly en- neutrophil elastase, azurocidin, or cathepsin G, which are closely re- hanced sEPCR binding. An EDTA buffer system supported about lated structurally and also stored in azurophilic granules (25). half the sEPCR binding relative to the others, the difference with To test directly whether sEPCR was binding to PR3, in vitro citrate probably attributable to the stronger chelating properties of studies were done using purified components. To create a binding by guest on October 1, 2021 EDTA. These experiments were repeated multiple times with surface, microtiter plate wells were coated with mAbs to EPCR, blood from three different donors with essentially identical results. and the Ab surface was then saturated with sEPCR. This approach To search for the neutrophil-binding protein(s) on an analytical was used to create a binding surface because direct immobilization scale, surface-biotinylated leukocytes were lysed and incubated of EPCR on plastic surfaces was found to destroy the conformation with sEPCR immobilized on a resin. As a control, an equal volume of EPCR. Purified biotin-PR3 was then added and, in each case, of lysate was incubated with Tris-inactivated resin (no sEPCR) and the Ab-sEPCR surface supported biotin-PR3 binding in a saturable treated identically to the affinity resin sample. After extensive manner (Fig. 4A). These are nonoverlapping Abs that either block washing, the resins were eluted with buffer containing EDTA. The protein C/activated protein C binding to EPCR (1494) (17, 26 ), do eluates were electrophoresed on 4–20% gradient gels, transferred not block ligand binding (1500), or bind to the carboxyl-terminal to a membrane, blocked, and then processed for biotin signal. A single major band was observed at about 33–35 kDa on the gel

FIGURE 3. sEPCR affinity chromatography. The lysate of surface-biotinylated leukocytes was incubated with sEPCR-affinity resin (lane 2)or Tris-inactivated control resin (lane 1). EDTA eluates were processed for FIGURE 1. sEPCR binds to activated neutrophils. Purified neutrophils biotin signal by SDS-PAGE, transfer to polyvinylidene difluoride mem- were incubated with Oregon Green-sEPCR (500 nM) at 37°C for 10 min in branes, and incubation with streptavidin-alkaline phosphatase and sub- the absence (no PMA) or presence (solid black) of 0.1 ␮g/ml PMA. The strate. A single major band of ϳ33 kDa is observed in the eluate from the autofluorescence of neutrophils alone is indicated. sEPCR affinity resin. 4700 SOLUBLE EPCR BINDS TO ACTIVATED NEUTROPHILS

tag (HPC4). None of the anti-sEPCR mAbs screened to date ap- pear to inhibit the sEPCR-PR3 interaction. In addition, increasing concentrations of unlabeled sEPCR decreased biotin-sEPCR binding to PR3-coated wells (Fig. 4B). Thus, the sEPCR-PR3-binding interaction was saturable and dissociable with excess, unlabeled ligand using purified components. ANCA directed against PR3 (PR3-ANCA) are found in patients with some autoimmune vasculitides, including Wegener’s granulomatosis and microscopic polyangiitis. ANCA react with neutrophil granular proteins, and two common ANCA types are distin- guished by indirect immunofluorescence on ethanol-fixed neutrophils. One type, P-ANCA, produces an artifactual perinuclear staining and results from Ab binding primarily to myeloperoxidase, and occasionally to other cationic granule proteins, which diffuses during fixation to the negatively charged nucleus. The other type, PR3-ANCA, produces a diffuse, cytoplasmic staining directed toward PR3. The PR3-ANCA pattern is highly specific for Wegener’s granulomatosis, and typically Ͼ80% of Wegener’s patients test positive for PR3-ANCA by both immunofluorescence Downloaded from and ELISA against purified PR3 (5, 27, 28). This technique was used in two-color confocal microscopy colocalization studies with ethanol-fixed neutrophils (Fig. 5). sEPCR FIGURE 4. sEPCR binds to purified PR3. A, Wells coated with anti- was labeled with Cy3, a fluorescent probe that emits a red color. sEPCR mAb 1494 (F), 1500 (Ⅺ), or HPC4 (‚) were blocked and then The PR3-ANCA (or P-ANCA) was detected with FITC anti- saturated with sEPCR. Biotin-PR3 was added as indicated, and bound human IgG that emits a green color. When the two signals overlap, http://www.jimmunol.org/ probe was detected with streptavidin-alkaline phosphatase and Blue Phos a yellow color is observed. Cy3-sEPCR staining was diffuse and substrate (absorbance at 650 nm). B, PR3-coated wells were preincubated distributed throughout the cytoplasm (Fig. 5, A and E), and the with increasing concentrations of unlabeled sEPCR. Biotin-sEPCR (500 PR3-ANCA produced the typical green cytoplasmic staining (Fig. nM) was added, and bound biotin was detected with streptavidin-alkaline 5B). When the images were overlaid, extensive areas of diffuse, phosphatase and Blue Phos substrate. yellow color were observed in the cell cytoplasms (Fig. 5C). In by guest on October 1, 2021

FIGURE 5. Coincubation of sEPCR and PR3-ANCA on ethanol-fixed neutrophils. Neutrophils were incubated with PR3-ANCA- or P-ANCA-positive serum, and bound autoantibody was detected with FITC anti-human IgG. The diffuse granular cytoplasmic staining (A) usually corresponds to Ab against PR3. The perinuclear pattern (D) results from Ab against primarily myeloperoxidase. Cy3-sEPCR produced a diffuse cytoplasmic staining (B and E). Incubation with both PR3-ANCA and sEPCR produced extensive areas of yellow in the cytoplasm, indicating colocalization of the fluorescent signals (C). Little overlap of signals was observed when P-ANCA and sEPCR were incubated with the cells (F). The Journal of Immunology 4701 contrast, similar experiments with P-ANCA (Fig. 5D) and Cy3- surfaces. To address this, sEPCR binding to neutrophils activated sEPCR (Fig. 5E) did not show colocalization (Fig. 5F). Thus, the in whole blood was evaluated in the presence of buffer or mAbs to ␤ sEPCR localized to neutrophil cytoplasmic sites in parallel with the 2 integrins (Fig. 7). IB4 Ab (IgG2a) binds to the common ␤ ␤ ␤ the PR3-ANCA, whose target Ag is PR3. -chain of the 2 integrin family (CD18) and blocks 2 integrin- Since the PR3-ANCA and sEPCR colocalized on the cells, the mediated adhesion (33). This Ab reduced sEPCR binding to PMA- effect of patient sera containing PR3-ANCA on sEPCR binding to activated neutrophils by 81% ( p Ͻ 0.0001). Interestingly, TS-1 Ab purified PR3 was tested (Fig. 6). The PR3 was immobilized on the (IgG1) had either no effect, or slightly increased, sEPCR binding surface of microtiter plate wells and preincubated with buffer to the cells. TS-1 also binds to CD18, but to a different epitope, and (100%), normal serum samples (N1, N2, and N3), or patient sera does not inhibit neutrophil adhesion to matrix proteins. As con- containing PR3-ANCA (P1-P5). Although the normal serum sam- trols, PL-1 (IgG1), which binds to neutrophil P-selectin glycopro- ples did not affect sEPCR binding to PR3, preincubation with three tein ligand-1 (34), had no effect on sEPCR binding, nor did an of the patient sera diluted 1/200 resulted in reduced sEPCR bind- irrelevant isotype-matched Ab (V189; IgG2a). Essentially identi- ing (P2, P3, and P4). Further dilution of these sera to 1/1000 re- cal results were observed when IB4 was added to purified neutro- sulted in sEPCR binding similar to the normal samples. Incubation phils and sEPCR binding to PMA-activated neutrophils was de- with the P1 and P5 sera did not appreciably alter sEPCR binding termined (data not shown). ␤ to the PR3-coated wells. The relative ability to reduce sEPCR To address which 2 integrin supports sEPCR binding, the neu- binding did not correlate with the PR3-ANCA titer of the patient trophils were preincubated with mAbs against CD11a, CD11b, or samples (data not shown) and is more likely related to the PR3 hybridoma culture supernatant containing Ab against CD11c. Only epitope specificity of the autoantibodies. A similar result was ob- the Ab against CD11b inhibited sEPCR binding to the activated served using confocal microscopy in which pretreatment of the neutrophils ( p Ͻ 0.0001) (Fig. 7). This indicates that at least Downloaded from ethanol-fixed neutrophils with PR3-ANCA, but not P-ANCA (pos- Mac-1 is involved in the neutrophil receptor complex that supports itive and negative control sera with the commercial PR3 ELISA), sEPCR binding, probably as a PR3/Mac-1 heterocomplex. reduced the signal observed from Oregon Green-labeled sEPCR The question then arose as to whether IB4, the anti- binding by ϳ50%, relative to untreated cells (data not shown). CD18-blocking Ab, could alter PR3-ANCA IgG binding to acti- Other potential modulators of this system were tested as well, vated neutrophils. Neutrophils were preincubated with IB4 (10 ␮g/ including the known EPCR ligands, protein C, and activated pro- ml) in the absence or presence of each patient PR3-ANCA IgG http://www.jimmunol.org/ tein C. Our preliminary data indicate that neither protein C, acti- (100 ␮g/ml) and activated with cytochalasin B/fMLP, and bound vated protein C, nor ␣1-antitrypsin (a natural inhibitor of PR3 Ab was detected with fluorescein-anti-human IgG by flow cytom- protease activity) significantly alters the sEPCR-PR3 interaction etry. There was no difference in the amount of patient IgG bound (data not shown). in the presence of IB4 Ab (data not shown). The simplest inter- The ability of serum containing PR3-ANCA to reduce sEPCR pretation is that the ability of IB4 to block sEPCR binding to the binding to PR3 raised the possibility that sEPCR may modulate the cells (and presumably to PR3) is a steric effect, rather than a direct protein-protein interactions in this system. Although it is generally inhibition of a ligand interaction. Interestingly, IgG from P1 and accepted that PR3-ANCA binds to neutrophil PR3, it is less widely P5, which did not reduce sEPCR binding to activated neutrophils, by guest on October 1, 2021 appreciated that accessory molecules may be involved as well. A also bound very little to the activated neutrophils relative to the previous study indicated that elastase and azurocidin, and possibly other three autoantibodies. ␤ PR3, are ligands for the 2 integrin, Mac-1 (CD11a/CD18), on the ␤ neutrophil surface (29). Mac-1 is a member of the 2 integrin family, Discussion consisting of LFA-1 (CD11a/CD18), Mac-1 (CD11b/CD18), p150,95 The current data demonstrate that sEPCR binds to activated neu- ␣ (CD11c/CD18), and a newly described fourth member ( d/CD18) trophils, and PR3, the Wegener’s autoantigen, was identified as a (30). These are heterodimeric complexes with a common ␤-chain component of the binding site based on sEPCR affinity chroma- ␣ ␣ (CD18) and unique -chains (CD11a, b, c, and d), and are involved tography results and in vitro binding studies using the purified in neutrophil adhesion and signaling (30–32). These observations raised the possibility that sEPCR may be ␤ binding to a preassembled PR3- 2 integrin complex on the cell

FIGURE 7. Mac-1 (CD11b/CD18) supports sEPCR binding to neutrophils. Heparinized blood (50 ␮l) was incubated with buffer (no mAb), 40 ␮g/ml Abs (IB4, TS1, PL1, anti-CD11a, anti-CD11b, isotype controls), or FIGURE 6. PR3-ANCA inhibits the sEPCR-PR3 interaction. PR3- 10 ␮l anti-CD11c hybridoma supernatant. Oregon Green-sEPCR was coated wells were pretreated for 30 min with normal human serum (N1, added (500 nM) in the absence or presence of activation with PMA at 37°C N2, and N3) or PR3-ANCA-positive serum from patients (P1-P5) at dilu- for 15 min. RBC were removed by hypotonic lysis, and binding of fluo- tions of 1/200 (black) and 1/1000 (white). Biotin-sEPCR (80 nM) was rescent sEPCR to the washed cells was analyzed by flow cytometry. The added, and bound biotin signal was detected after 30 min with streptavidin- IB4 Ab against CD18 and the anti-CD11b Ab significantly inhibited p Ͻ 0.0001 compared with no ,ء .HRP and substrate. The OD at 450 nm was read. Maximal binding (100%) sEPCR binding to activated neutrophils refers to bound biotin-sEPCR in buffer-treated wells. mAb or nonconditioned media controls (control for anti-CD11c). 4702 SOLUBLE EPCR BINDS TO ACTIVATED NEUTROPHILS

␤ enzyme. Furthermore, the 2 integrin Mac-1 was identified as con- binding complex does provide insight into the partial metal depen- tributing to sEPCR binding to the activated neutrophils. These ob- dence observed for sEPCR binding to the activated neutrophils ␤ servations, in combination with an earlier study indicating that since 2 integrin function is metal ion dependent (45). PR3 forms heterocomplexes with Mac-1 on neutrophils (29), led The current studies were done with sEPCR, raising the question us to our working model in which sEPCR binds to PR3 and PR3/ of whether EPCR anchored in the endothelial membrane shares Mac-1 complexes on activated neutrophils. this ability to bind PR3. In this regard, EPCR mRNA levels in- Although a causative role for the PR3-ANCA autoantibodies in crease rapidly in response to endotoxin challenge in a rat model of the etiology of autoimmune vasculitis remains controversial, many septic shock (22). Although the tissue EPCR Ag levels do not rise in vitro studies have demonstrated that they are capable of prop- appreciably, there is a substantial increase in sEPCR levels in the agating an inflammatory response by cross-linking Fc␥ receptors plasma, suggesting a regulated pathway for EPCR synthesis and Ј as well as by PR3 binding to F(ab )2 regions (35–40). In Wege- sEPCR release, possibly through thrombin stimulation of the en- ner’s patients, PR3 is expressed on the surface of circulating neu- dothelium and subsequent metalloproteinase activity (46). This trophils (41, 42), and PR3-ANCA binding to PR3 on the surface of concept is supported by observations that significantly elevated TNF-␣-primed neutrophils sets off a full-blown activation re- levels of circulating soluble EPCR are observed in patients with sponse, including degranulation and production of oxygen radicals sepsis or systemic lupus erythematosus (21). Regulated release of (35). However, mechanisms that may modulate this process of soluble EPCR may play a role in Wegener’s patients in whom vasculitis are poorly understood. The current study suggests that selective microvascular beds in the nose, sinuses, lungs, and kid- sEPCR may modulate the ability of PR3-ANCA to bind to PR3 on neys are the initial targets of the vasculitis (7), sites in which mem- neutrophils, potentially regulating the inflammatory response and brane-bound EPCR expression is relatively poor (47). However, vasculitic injury. However, this potential regulatory role is spec- unlike its membrane-bound parent, soluble EPCR is not restricted Downloaded from ulative and it must be recognized that there is no in vivo data to a particular vascular bed and could be recruited for binding to defining a pathogenic role for ANCA, and the possibility remains neutrophils. that the presence of ANCA in the vasculitis patients is an These observations represent novel additions toward under- epiphenomenona. standing the fundamental mechanisms of inflammation and the One prerequisite for the current model of vascular injury is local molecular basis for the vasculitis in Wegener’s granulomatosis and accumulation of primed neutrophils that will eventually adhere to related vasculitides. Additional studies are required to further char- http://www.jimmunol.org/ the endothelium to provide a local source of inflammatory medi- acterize the effect of PR3-ANCA and other modulators on the pair- ␤ ators. The 2 integrins, which include Mac-1 (CD11b/CD18), are wise interactions between PR3, soluble EPCR, and CD11b/CD18 expressed on granulocytes and monocytes and are involved in cell- on activated neutrophils. cell adhesion and signaling events. Integrin expression is induc- ible, and circulating leukocytes from patients with active Wegen- Acknowledgments er’s granulomatosis show significantly increased surface We thank Dr. Rodger McEver (University of Oklahoma), for support and expression of Mac-1 that declines to normal levels upon treatment helpful discussions. We gratefully acknowledge the assistance of Jim (43). Expression of two other ␤ integrins (LFA-1 and p150,95) Henthorn with the confocal microscopy and image analysis and Noriko

2 by guest on October 1, 2021 was normal in the Wegener’s patients, as was Mac-1 expression on Hidari and Kandice Swindle for technical assistance. cells from patients with other systemic diseases (systemic lupus erythematosus, myeloperoxidase-positive systemic vasculitis, sep- References sis). An additional study found that CD18 on TNF-␣-primed neu- 1. Jenne, D. E., J. Tschopp, J. Ludemann, B. Utecht, and W. L. Gross. 1990. We- gener’s autoantigen decoded. Nature 346:520. trophils was required for the respiratory burst activation induced 2. Labbaye, C., P. Musette, and Y. E. Cayre. 1991. Wegener autoantigen and my- by an anti-PR3 mAb (40). These observations are consistent with eloblastin are encoded by a single mRNA. Proc. Natl. Acad. Sci. USA 88:9253. the current data demonstrating Mac-1 participation in the sEPCR 3. Csernok, E., M. Ernst, W. Schmitt, D. F. Bainton, and W. L. Gross. 1994. Ac- tivated neutrophils express proteinase 3 on their plasma membrane in vitro and in binding site on neutrophils, possibly as a PR3/Mac-1 heterocom- vivo. Clin. Exp. Immunol. 95:244. plex (29). A recent study further demonstrated that PR3-ANCA 4. Witko-Sarsat, V., E. M. Cramer, C. Hieblot, J. Guichard, P. Nusbaum, S. Lopez, P. Lesavre, and L. Halbwachs-Mecarelli. 1999. Presence of proteinase 3 in se- was capable of a transient activation of rolling neutrophils under cretory vesicles: evidence of a novel, highly mobilizable intracellular pool dis- flow conditions with resultant firm adhesion to a platelet selectin tinct from azurophil granules. Blood 94:2487. surface (44). This activation and adhesion was Mac-1 mediated 5. Bajema, I. M., E. C. Hagen, F. J. van der Woude, and J. A. Bruijn. 1997. We- ␥ gener’s granulomatosis: a meta-analysis of 349 literary case reports. J. Lab. Clin. and dependent on interaction of the ANCA with Fc receptors. Med. 129:17. One prediction from our working model is that sEPCR may mod- 6. Hoffman, G. S. 1998. Wegener’s granulomatosis. In Rheumatology. ulate Mac-1 integrin-mediated events, such as outside-in signaling J. Klippel and P. Dieppe, eds. Mosby International, London, pp. 22.1–22.10. 7. Jennette, J. C. 1998. Renal involvement in systemic vasculitis. In Heptinstall’s or adhesion, and these studies are in progress. Based on current Pathology of the Kidney. J. C. Jennette, J. L. Olson, M. M. Schwartz, and models, neutrophils play a role in the early stages of vasculitic F. G. Silva, eds. Lippincott-Raven Publishers, Philadelphia, pp. 1059–1095. 8. Rao, N. V., N. G. Wehner, B. C. Marshall, W. R. Gray, B. H. Gray, and injury, and the later fulminant lesions typically consist of lympho- J. R. Hoidal. 1991. Characterization of proteinase-3 (PR-3), a neutrophil serine cytes and macrophages. proteinase: structural and functional properties. J. Biol. Chem. 266:9540. The current studies demonstrating Mac-1 participation in the 9. Bories, D., M. C. Raynal, D. H. Solomon, Z. Darzynkiewicz, and Y. E. Cayre. 1989. Down-regulation of a serine protease, myeloblastin, causes growth arrest binding complex for sEPCR are uniquely limited by the epitope and differentiation of promyelocytic leukemia cells. Cell 59:959. specificity of the Abs used. Mac-1 has a diverse ligand repertoire 10. Skold, S., B. Rosberg, U. Gullberg, and T. Olofsson. 1999. A secreted proform (ICAM-1, fibrinogen, complement iC3b, coagulation factor X, of neutrophil proteinase 3 regulates the proliferation of granulopoietic progenitor cells. Blood 93:849. neutrophil elastase, azurocidin, PR3), and there is no particular 11. Renesto, P., L. Halbwachs-Mecarelli, P. Nusbaum, P. Lesavre, and M. Chignard. reason to believe that PR3 has a singular preference for binding to 1994. Proteinase 3: a neutrophil proteinase with activity on platelets. J. Immunol. only one ␤ integrin. It is quite possible that the epitopes of the 152:4612. 2 12. Campanelli, D., P. A. Detmers, C. F. Nathan, and J. E. Gabay. 1998. Azurocidin Abs against CD11a and CD11c may not have overlapped the li- and a homologous serine protease from neutrophils: differential antimicrobial and gand binding site. Additional studies with other Abs are ongoing to proteolytic properties. J. Clin. Invest. 85:904. 13. Fujinaga, M., M. M. Chernaia, R. Halenbeck, K. Koths, and M. N. James. 1996. further evaluate LFA-1 or p150,95 as potential partners in the The crystal structure of PR3, a neutrophil serine proteinase antigen of Wegener’s sEPCR-binding complex. Participation of Mac-1 in the sEPCR- granulomatosis antibodies. J. Mol. Biol. 261:267. The Journal of Immunology 4703

14. Coeshott, C., C. Ohnemus, A. Pilyavskaya, S. Ross, M. Wieczorek, H. Kroona, 30. Carlos, T. M., and J. M. Harlan. 1994. Leukocyte-endothelial adhesion mole- A. H. Leimer, and J. Cheronis. 1999. Converting enzyme-independent release of cules. Blood 84:2068. tumor necrosis factor ␣ and IL-1␤ from a stimulated human monocytic cell line 31. Clark, E. A., and J. S. Brugge. 1995. Integrins and signal transduction pathways: in the presence of activated neutrophils or purified proteinase 3. Proc. Natl. Acad. the road taken. Science 268:233. Sci. USA 96:6261. 32. Van der Vieren, M., H. Le Trong, C. L. Wood, P. F. Moore, T. St.John, and 15. Robache-Gallea, S., V. Morand, J. M. Bruneau, B. Schoot, E. Tagat, E. Realo, D. E. G. W. M. Staunton. 1995. A novel leukointegrin, ␣-d ␤-2, binds prefer- S. Chouaib, and S. Roman-Roman. 1995. In vitro processing of human tumor entially to ICAM-3. Immunity 683. necrosis factor-␣. J. Biol. Chem. 270:23688. 33. Wright, S. D., P. E. Rao, W. C. Van Voorhis, L. S. Craigmyle, K. Iida, 16. Taylor, F. B., Jr., D. J. Stearns-Kurosawa, S. Kurosawa, G. Ferrell, A. Chang, Z. M. A. Talle, E. F. Westberg, G. Goldstein, and S. C. Silverstein. 1983. Identifi- Laszik, G. Peer, and C. T. Esmon. 2000. The endothelial protein C receptor aids cation of the C3bi receptor of human monocytes and macrophages by using in host defense against E. coli sepsis. Blood 95:1680. monoclonal antibodies. Proc. Natl. Acad. Sci. USA 80:5699. 17. Stearns-Kurosawa, D. J., S. Kurosawa, J. S. Mollica, G. L. Ferrell, and 34. Moore, K. L., K. D. Patel, R. E. Bruehl, F. Li, D. A. Johnson, H. S. Lichenstein, C. T. Esmon. 1996. The endothelial cell protein C receptor augments protein C R. D. Cummings, D. F. Bainton, and R. P. McEver. 1995. P-selectin glycoprotein activation by the thrombin-thrombomodulin complex. Proc. Natl. Acad. Sci. USA ligand-1 mediates rolling of human neutrophils on P-selectin. J. Cell Biol. 128: 93:10212. 661. 18. Fijnvandraat, K., B. Derkx, M. Peters, R. Bijlmer, A. Sturk, M. H. Prins, 35. Falk, R. J., R. S. Terrell, L. A. Charles, and J. C. Jennette. 1990. Anti-neutrophil S. J. H. van Deventer, and J. Wouter ten Cate. 1995. Coagulation activation and cytoplasmic autoantibodies induce neutrophils to degranulate and produce oxy- tissue necrosis in meningococcal septic shock: severely reduced protein C levels gen radicals in vitro. Proc. Natl. Acad. Sci. USA 87:4115. predict a high mortality. Thromb. Haemostasis 73:15. 36. Kettritz, R., J. C. Jennette, and R. J. Falk. 1997. Crosslinking of ANCA-antigens 19. Smith, O. P., and B. White. 1999. Infectious pupura fulminans: diagnosis and stimulates superoxide release by human neutrophils. J. Am. Soc. Nephrol. 8:386. treatment. Br. J. Haematol. 104:202. 37. Ralston, D. R., C. B. Marsh, M. P. Lowe, and M. D. Wewers. 1997. Antineu- 20. Kurosawa, S., D. J. Stearns-Kurosawa, N. Hidari, and C. T. Esmon. 1997. Iden- trophil cytoplasmic antibodies induce monocyte IL-8 release: role of surface pro- tification of functional endothelial protein C receptor in human plasma. J. Clin. teinase-3, ␣1-antitrypsin, and Fc␥ receptors. J. Clin. Invest. 100:1416. Invest. 100:411. 38. Tomer, Y., O. Lider, B. Gilburd, R. Hershkoviz, P. L. Meroni, A. Wiik, and 21. Kurosawa, S., D. J. Stearns-Kurosawa, C. W. Carson, A. D’Angelo, Y. Shoenfeld. 1997. Anti-neutrophil cytoplasmic antibody-enriched IgG induces P. Della Valle, and C. T. Esmon. 1998. Plasma levels of endothelial cell protein adhesion of human T lymphocytes to extracellular matrix proteins. Clin. Immu- C receptor are elevated in patients with sepsis and systemic lupus erythematosus: nol. Immunopathol. 83:245. Downloaded from lack of correlation with thrombomodulin suggests involvement of different patho- 39. Mayet, W. J., and z. B. K. Meyer. 1993. Antibodies to proteinase 3 increase logical processes. Blood 91:725. adhesion of neutrophils to human endothelial cells. Clin. Exp. Immunol. 94:440. 22. Gu, J.-M., Y. Katsuura, G. L. Ferrell, P. Grammas, and C. T. Esmon. 1999. 40. Reumaux, D., P. J. Vossebeld, D. Roos, and A. J. Verhoeven. 1995. Effect of Endotoxin and thrombin elevate rodent endothelial cell protein C receptor mRNA tumor necrosis factor-induced integrin activation on Fc␥ receptor II-mediated levels and increase receptor shedding in vivo. Blood 95:1687. signal transduction: relevance for activation of neutrophils by anti-proteinase 3 or 23. Fukudome, K., S. Kurosawa, D. J. Stearns-Kurosawa, X. He, A. R. Rezaie, and anti-myeloperoxidase antibodies. Blood 86:3189. C. T. Esmon. 1996. The endothelial cell protein C receptor: cell surface expres- 41. Csernok, E., W. H. Schmitt, M. Ernst, D. F. Bainton, and W. L. Gross. 1993.

sion and direct ligand binding by the soluble receptor. J. Biol. Chem. 271:17491. Membrane surface proteinase 3 expression and intracytoplasmic immunoglobulin http://www.jimmunol.org/ 24. Rezaie, A. R., M. M. Fiore, P. F. Neuenschwander, C. T. Esmon, and on neutrophils from patients with ANCA-associated vasculitides. Adv. Exp. Med. J. H. Morrissey. 1992. Expression and purification of a soluble tissue factor Biol. 336:45. fusion protein with an epitope for an unusual calcium-dependent antibody. Pro- 42. Muller Kobold, A. C., C. G. Kallenberg, and J. W. Tervaert. 1998. Leukocyte tein Expression Purif. 3:453. membrane expression of proteinase 3 correlates with disease activity in patients 25. Jenne, D. E. 1994. Structure of the azurocidin, proteinase 3, and neutrophil elas- with Wegener’s granulomatosis. Br. J. Rheumatol. 37:901. tase genes: implications for inflammation and vasculitis. Am. J. Respir. Crit. Care 43. Haller, H., J. Eichhorn, K. Pieper, U. Gobel, and F. C. Luft. 1996. Circulating Med. 150:S147. leukocyte integrin expression in Wegener’s granulomatosis. J. Am. Soc. Nephrol. 26. Fukudome, K., and C. T. Esmon. 1994. Identification, cloning and regulation of 7:40. a novel endothelial cell protein C/activated protein C receptor. J. Biol. Chem. 44. Radford, D. J., C. O. Savage, and G. B. Nash. 2000. Treatment of rolling neu- 269:26486. trophils with antineutrophil cytoplasmic antibodies causes conversion to firm 27. Schultz, D. R., and E. C. Tozman. 1995. Antineutrophil cytoplasmic antibodies: integrin-mediated adhesion. Arthritis Rheum. 43:1337.

major autoantigens, pathophysiology, and disease associations. Semin. Arthritis 45. Bohnsack, J. F., and X. N. Zhou. 1992. Divalent cation substitution reveals by guest on October 1, 2021 Rheum. 25:143. CD18- and very late antigen-dependent pathways that mediate human neutrophil 28. Wong, R. C., R. A. Silvestrini, J. A. Savige, D. A. Fulcher, and E. M. Benson. adherence to ﬁbronectin. J. Immunol. 149:1340. 1999. Diagnostic value of classical and atypical antineutrophil cytoplasmic an- 46. Xu, J., D. Qu, N. L. Esmon, and C. T. Esmon. 2000. Metalloproteolytic release tibody (ANCA) immunoﬂuorescence patterns. J. Clin. Pathol. 52:124. of endothelial protein C receptor. J. Biol. Chem. 275:6038. 29. Cai, T. Q., and S. D. Wright. 1996. Human leukocyte elastase is an endogenous 47. Laszik, Z., A. Mitro, F. B. Taylor Jr., G. Ferrell, and C. T. Esmon. 1997. Human ␣ ␤ ligand for the integrin CR3 (CD11b/CD18, Mac-1, M 2) and modulates poly- protein C receptor is present primarily on endothelium of large blood vessels: morphonuclear leukocyte adhesion. J. Exp. Med. 184:1213. implications for the control of the protein C pathway. Circulation 96:3633.