(EGF-TM7) Receptor CD97 the Epidermal Growth Factor-Seven

Home , Chemokine receptor, EMR2, EMR3

The Epidermal Growth Factor-Seven Transmembrane (EGF-TM7) Receptor CD97 Is Required for Neutrophil Migration and Host Defense This information is current as of September 23, 2021. Jaklien C. Leemans, Anje A. te Velde, Sandrine Florquin, Roelof J. Bennink, Kora de Bruin, René A. W. van Lier, Tom van der Poll and Jörg Hamann J Immunol 2004; 172:1125-1131; ; doi: 10.4049/jimmunol.172.2.1125 Downloaded from http://www.jimmunol.org/content/172/2/1125

References This article cites 52 articles, 19 of which you can access for free at: http://www.jimmunol.org/ http://www.jimmunol.org/content/172/2/1125.full#ref-list-1

Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists by guest on September 23, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

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 © 2004 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

The Epidermal Growth Factor-Seven Transmembrane (EGF-TM7) Receptor CD97 Is Required for Neutrophil Migration and Host Defense1

Jaklien C. Leemans,* Anje A. te Velde,* Sandrine Florquin,† Roelof J. Bennink,‡ Kora de Bruin,‡ Rene´A. W. van Lier,§ Tom van der Poll,* and Jo¨rg Hamann2§

The epidermal growth factor-seven transmembrane (EGF-TM7) family is a group of seven-span transmembrane receptors pre- dominantly expressed by cells of the immune system. Family members CD97, EGF module-containing mucin-like receptor (EMR) 1, EMR2, EMR3, EMR4, and EGF-TM7-latrophilin-related protein are characterized by an extended extracellular region with a variable number of N-terminal EGF-like domains. EGF-TM7 receptors bind cellular ligands as demonstrated by the interaction of CD97 with decay accelerating factor (CD55) and dermatan sulfate. Investigating the effect of newly generated mAb on the Downloaded from migration of neutrophilic granulocytes, we here report for the first time in vivo data on the function of CD97. In dextran sulfate sodium-induced experimental colitis, we show that homing of adoptively transferred neutrophils to the colon was significantly delayed when cells were preincubated with CD97 mAb. The consequences of this defect in neutrophil migration for host defense are demonstrated in a murine model of Streptococcus pneumoniae-induced pneumonia. Mice treated with CD97 mAb to EGF domain 1 (1B2) and EGF domain 3 (1C5) displayed a reduced granulocytic inflammatory infiltrate at 20 h after inoculation. This was associated with a significantly enhanced outgrowth of bacteria in the lungs at 44 h and a strongly diminished survival. http://www.jimmunol.org/ Together, these findings indicate an essential role for CD97 in the migration of neutrophils. The Journal of Immunology, 2004, 172: 1125–1131.

D97 (1–7) is a defining member of the epidermal growth joined by a noncovalent linkage with the TM7/cytoplasmic ␤ factor seven transmembrane (EGF-TM7)3 family (8). subunit. C Molecules of this family, comprising also EGF module- Due to alternative RNA splicing, the number of EGF domains in containing mucin-like related protein (EMR)1 (F4/80) (9–11), EGF-TM7 receptors is variable (8). For human CD97 (hCD97), iso- EMR2 (12), EMR3 (13), EMR4 (FIRE) (14–16), and EGF-TM7- forms with three, four, and five EGF domains have been identified (4), by guest on September 23, 2021 latrophilin-related protein (17, 18), are characterized by an ex- designated hereafter as hCD97(EGF1,2,5), hCD97(EGF1,2,3,5), and tended extracellular region. They possess N-terminal EGF-like do- hCD97(EGF1,2,3,4,5). Mouse CD97 (mCD97) also exists in three mains, which are connected by a stalk region to TM7 segments isoforms (5, 6). Next to isoforms with three and four EGF domains, homologous to those found in peptide hormone-binding class B G indicated as mCD97(EGF1,2,4) and mCD97(EGF1,2,3,4), a third iso- protein-coupled receptors (12, 13, 17, 19). First discovered for form mCD97(EGF1,2,X,3,4) was detected. This isoform has a se- CD97 (4), there is growing evidence that most if not all EGF-TM7 quence of 45 aa between the second and third EGF domain that does receptors are expressed at the cell surface as heterodimers (5, 15, not correspond to known protein modules. EGF domains 3 and 4 in 17, 20, 21). After cleavage of the polypeptide at a G protein-cou- mCD97 are the homologues of EGF domains 4 and 5 in hCD97, pled receptor-proteolytic site (22) immediately proximal to the first respectively (6). Schematic structures of CD97 in humans and mice ␣ transmembrane segment, the large extracellular subunit is re- are depicted in Fig. 1. Receptors of the EGF-TM7 family are expressed by cells of the immune system and by smooth muscle cells. In contrast to the *Laboratory for Experimental Internal Medicine, Departments of †Pathology and rather restricted expression profile of other family members (Ref. ‡Nuclear Medicine, and §Laboratory for Experimental Immunology, Academic Med- ical Center, University of Amsterdam, Amsterdam, The Netherlands 8 and references therein; Refs. 14, 17, 20), CD97 is found on a Received for publication August 8, 2003. Accepted for publication October 21, 2003. broad array of hematopoietic cells including activated lympho- The costs of publication of this article were defrayed in part by the payment of page cytes, granulocytes, monocytes, macrophages, and dendritic cells charges. This article must therefore be hereby marked advertisement in accordance (1, 2, 23, 24). High expression and elevated levels of soluble CD97 with 18 U.S.C. Section 1734 solely to indicate this fact. at sites of inflammation suggest a role for CD97 in the onset of 1 This work has been supported by a grant from The Netherlands Organization for protective as well as destructive immune responses (4, 25, 26). In Scientific Research (to J.C.L). J.H. is a fellow of the Royal Netherlands Academy of Arts and Sciences. addition, smooth muscle cells and malignant cells in various epi- 2 Address correspondence and reprint requests to Dr. Jo¨rg Hamann, Laboratory for thelial tumors express CD97 (23, 27–29). Experimental Immunology, G1-106, Academic Medical Center, University of Am- We previously identified decay accelerating factor (CD55) as a sterdam, P.O. Box 22700, 1100 DE Amsterdam, The Netherlands. E-mail address: cellular ligand of CD97 (30). CD55 is a GPI-linked molecule that [email protected] prevents complement deposition on self cells by inhibiting 3 Abbreviations used in this paper: EGF, epidermal growth factor; TM7, seven transmembrane; EMR, EGF module-containing mucin-like receptor; hCD97, human C3/C5 convertases (31, 32). The binding site for CD55 is formed by CD97; mCD97, mouse CD97; GAG, glycosaminoglycan; i.n., intranasal; DSS, dex- the EGF domain region (33). Lin et al. (34) showed that tran sulphate sodium; HMPAO, hexamethylpropylene amine oxime; MIP-2, mac- hCD97(EGF1,2,5) binds CD55 with low affinity (86 ␮M) and a rapid rophage-inflammatory protein-2; SPECT, single photon emission computed Ϫ tomography. off-rate (0.6 s 1). Affinity for CD55 differs between CD97 isoforms

mCD97(EGF1,2,3,4) cDNA in pcDNA3.1/Zeo(ϩ) (6). Cells were selected with zeocin (Invitrogen, Leek, The Netherlands) at 500 mg/ml in culture medium. Resulting stably transfected clones were tested for mCD97 expression by flow cytometry with the earlier generated mCD97 mAb 1A2 (6). One clone was selected and used for immunization. Three i.p. injec- tions with 10 ϫ 106 irradiated (50 Gy) cells in PBS were given at weekly intervals. Eight weeks after the third injection, the hamster was boosted i.p. with 10 ϫ 106 cells. Three days later, hamster spleen cells were fused with mouse myeloma SP2/0 cells by standard hybridoma technology. Binding of hybridoma supernatants to the ARHO12 clone stably expressing mCD97(EGF1,2,3,4) was tested by flow cytometry. Flow cytometry of COS cells expressing mCD97 isoforms or chimeras between mCD97 and hCD97 was used to identify hybridomas that recognize different regions of mCD97. Selected hybridomas were subcloned until they were monoclonal and stable. The hybridomas 1B2, 1C5, and 1D2 were grown at large amounts and Ig was purified using protein A-Sepharose CL-4B (Sigma- Aldrich, St. Louis, MO).

Generation of chimeric m-hCD97 constructs Generation of an expression construct encoding mCD97(EGF1h,2,4) by overlap extension-PCR has been described earlier (6). In a similar manner,

constructs in which EGF domain 1 or EGF domains 1 and 2 of mCD97 Downloaded from have been exchanged by homologues EGF domains of hCD97 were made. In a first step, two overlapping PCR fragments were generated in separate FIGURE 1. Schematic structure of CD97. Depicted are the three differ- PCRs. For hCD97(EGF1m,2,5), the N-terminal part of mCD97 including ent isoforms that have been identified in both human and mouse. For de- EGF domain 1 was amplified from mCD97 cDNA (6) using a standard T7 ϩ Ј tails, see text. primer (( ) strand) and the specific primer 5 -CGTTGATGTCTTCA CAGCTCTCTGCAGGGTTAG-3Ј (nt 214–237 of the (Ϫ) strand; the sequence overlapping with EGF domain 2 of hCD97 is underlined). EGF domains 2 and 5 and part of the stalk region of hCD97 were amplified from http://www.jimmunol.org/ and is significantly lower for the larger isoforms hCD97(EGF1,2,3,5), hCD97 cDNA (2) using the specific primers 5Ј-GCTGTGAAGACAT hCD97(EGF1,2,3,4,5), and mCD97(EGF1,2,X,3,4) (6, 33, 34). Effi- CAACGAGTGTGCAACAC-3Ј (nt 261–281 of the (ϩ) strand; the se- cient up-regulation of CD97 and CD55 under physiological and quence overlapping with EGF domain 1 of mCD97 is underlined) and Ј Ј Ϫ pathological conditions (1, 7, 24, 35, 36) implies that generation of 5 -GGAATGCAGGTTCAAGGAGGC-3 (nt 1031–1051 of the ( ) strand). For hCD97(EGF1m,2m,5), the N-terminal part of mCD97 includ- cellular interactions of high avidity in vivo is regulated at the expres- ing EGF domains 1 and 2 was amplified using a standard T7 primer ((ϩ) sion level of both molecules (37). More recently, chondroitin sulfate strand) and the specific primer 5Ј-CGTCCACATCTTGACATGTATTCT has been identified as a second ligand of CD97 (38). This cell surface CACTCTCG-3Ј (nt 368–390 of the (Ϫ) strand; the sequence overlapping glycosaminoglycan (GAG) side chain specifically interacts with the with EGF domain 5 of hCD97 is underlined). EGF domain 5 and part of the Ј

stalk region of hCD97 were amplified using the specific primers 5 -CAT by guest on September 23, 2021 larger isoforms of CD97 (for details see Discussion). GTCAAGATGTGGACGAGTGCAGCTCCGG-3Ј (nt 417–438 of the (ϩ) Based on the unusual structure of the ␣ subunit, the ability of strand; the sequence overlapping with EGF domain 2 of mCD97 is under- this region to bind cellular ligands, and the rapid up-regulation of lined) and 5Ј-GGAATGCAGGTTCAAGGAGGC-3Ј (nt 1031–1051 of the CD97 on activated lymphocytes, a role of CD97 in cell adhesion (Ϫ) strand). The PCR fragments were purified and mixed together to serve and migration has been suggested (2, 4, 7, 30, 39). Supportive as template for the second step of the overlap extension-PCR performed for both constructs with a standard T7 primer and the specific primer 5Ј- evidence for this assumption was recently received from studies in GGAATGCAGGTTCAAGGAGGC-3Ј. Chimeric constructs were ob- intestinal carcinomas. Aust and coworkers (28, 29) showed, first, tained by replacing the N-terminal part of hCD97(EGF1,2,5) cDNA in that expression levels of CD97 correlate with the in vitro migration pcDNA3.1/Zeo(ϩ) with the PCR amplicons, thereby using a HindIII site in and invasion capacity of colorectal tumor cell lines, second, that the multiple-cloning site of the vector and a EcoRV site in the hCD97 sequence. migration and invasiveness of the fibrosarcoma cell line HT-1080 To generate a mCD97(EGF1h,2h,4) construct, the N-terminal part of can be increased 2- to 3-fold by inducing (tetracycline-off control) mCD97 including EGF domain 1 and part of EGF domain 2 was amplified expression of CD97, and, third, that scattered tumor cells at the using a standard T7 primer ((ϩ) strand) and the specific primer 5Ј-CTT- invasion front of colorectal and gastric carcinomas are stronger GAATGTTTTTGCCCCAGAAAC-3Ј (nt 368–391 of the (Ϫ) strand). A CD97 positive than tumor cells in solid formations of the same chimeric construct was obtained by replacing the N-terminal part of mCD97(EGF1,2,4) cDNA in pcDNA3.1/Zeo(ϩ) with the PCR amplicon, tumor. thereby using a NheI site in the multiple-cloning site of the vector and a In this study, we report for the first time in vivo data on CD97. HpaI site in the mCD97 sequence (blunt-end ligation of the 3Ј end of the In two murine models we investigated the effect of newly gener- amplicon). ated mCD97 mAb on the migration of neutrophilic granulocytes. COS cell expression and mapping of mAb binding sites First, in experimental colitis, mCD97 mAb caused a delay in the homing of adoptively transferred neutrophils to the colon. Second, COS cells were transfected with mCD97, hCD97, or chimeric m-hCD97 in mice intranasally (i.n.) inoculated with Streptococcus pneu- constructs using Lipofectamine Plus reagent (Life Technologies, Gaithers- burg, MD). Three days after transfection, reactivity of mCD97 mAb 1A2, moniae, application of mCD97 mAb impaired the recruitment of 1B2, 1C5, and 1D2 was analyzed by standard flow cytometry on a FAC- neutrophils to the lungs thereby reducing resistance to pneumonia. Scan (BD Biosciences, Mountain View, CA). Control stainings were per- Together, these data are in favor of an essential role of CD97 in formed with hCD97 mAb CLB-CD97/1 and CLB-CD97/3 (40). PE-con- leukocyte migration. jugated goat anti-hamster Ig (Southern Biotechnology Associates, Birmingham, AL) or PE-conjugated goat anti-mouse Ig (Immunotech, Marseille, France) were used as second step reagent. Materials and Methods Generation of mAb Erythrocyte adhesion studies Novel mCD97 mAb were generated in Armenian hamsters (Cricetulus mi- Adhesion studies were performed as described previously (6, 30, 33) by gratorius; Cytogen, West Roxbury, MA) as described recently (6). In short, overlaying COS cells in 12-well cell culture plates three days after trans- the Armenian hamster fibroblast line ARHO12 was transfected with fection with 50 ϫ 106 mouse erythrocytes for 30 min at room temperature. The Journal of Immunology 1127

Nonadhering cells were removed by gentle washing with PBS before ex- in sterile isotonic saline at a concentration of ϳ1 ϫ 104 CFU/50 ␮l, as amination by microscopy. To test for blocking capacity, mCD97 mAb were determined by plating serial 10-fold dilutions on sheep-blood agar plates. added to the erythrocyte suspension at a final concentration 5 ␮g/ml. Mice were lightly anesthesized by inhalation of isoflurane (Upjohn, Ede, The Netherlands) and 50 ␮l were inoculated i.n. Induction of dextran sulfate sodium (DSS) colitis Abs 1B2, 1C5, and cIg were given i.p. at doses of 0.5 mg 24 h before, 24 and 72 h after induction of pneumonia. The size of the groups was n ϭ Pathogen-free 8-wk-old female BALB/c mice were obtained from Harlan 10 for the survival study and n ϭ 8 for the remaining assays. (Zeist, The Netherlands). The Animal Care and Use Committee of the University of Amsterdam (Amsterdam, The Netherlands) approved all ex- Preparation of lung homogenates and determination of bacterial periments described in this manuscript. To induce DSS colitis, mice were fed 4% (w/v) DSS (TdB Consultancy, Uppsala, Sweden) in their drinking outgrowth water for at least 7 days (41). At 20 and 44 h after inoculation, mice were anesthetized with hypnorm Adoptive transfer of technetium-99m (99mTc)-labeled neutrophils (Janssen Pharmaceutica) and midazolam (Roche) and sacrificed by bleed- ing out the vena cava inferior. Blood was collected in EDTA-containing and pinhole single photon emission computed tomography tubes. Whole lungs were harvested and homogenized at 4°C in four vol- (SPECT) umes of sterile saline using a tissue homogenizer (BioSpec Products, Bartlesville, OK). CFU were determined from serial dilutions of lung ho- Peritoneal granulocytes were harvested by rinsing the peritoneal cavity mogenates and blood, plated on blood agar plates and incubated at 37°Cat with 5 ml of sterile PBS 5 h after i.p. injection of 1 ml of 10% proteose 5% CO for 16 h before colonies were counted. peptone (Difco, Detroit, MI) in PBS. A total of 5–10 ϫ 106 cells/mice were 2 99m harvested and labeled with 75 MBq of freshly prepared Tc-hexameth- Histologic examination of lungs ylpropylene amine oxime (HMPAO) according to manufacturers’ guide-

lines (Ceretec; Amersham Health, Eindhoven, The Netherlands). After 15 Lungs for histologic examination were harvested at 20 h and 44 h after Downloaded from min, cells were washed and resuspended in saline, containing 5 ␮g/ml Ab. inoculation, fixed in 10% formalin and embedded in paraffin. Four-mi- Next to mCD97 mAb 1B2, hamster Ig (Rockland, Gilbertsville, PA) was crometer sections were stained with H&E and analyzed by a pathologist used as control (cIg). who was blinded for groups. For granulocyte staining (46), slides were Within 30 min after labeling, ϳ5 ϫ 106 cells, labeled with 40 MBq deparaffinized and endogenous peroxidase activity was quenched by a so- 99m Tc-HMPAO, were injected i.v. in a tail vein of mice with DSS colitis lution of methanol/0.03% H2O2. After digestion with a solution of pepsine (n ϭ 7–8). One hour later, scintigraphy was performed on rotating mice for 0.25% in 0.01 M HCl, the sections were incubated in 10% normal goat three-dimensional imaging using a recently described and validated high- serum and then exposed to FITC-labeled mouse Ly-6G mAb (BD PharM- resolution pinhole SPECT technique (42). During the whole procedure, ingen, San Diego, CA). Slides were incubated with a rabbit anti-FITC Ab http://www.jimmunol.org/ animals were sedated with fentanyl and fluanisone (Janssen Pharmaceutica, (DAKO, Glostrup, Denmark) followed by a further incubation with a bi- Beerse, Belgium) and diazepam (Roche, Mijdrecht, The Netherlands). otinylated swine anti-rabbit Ab (DAKO), rinsed again, incubated in a

SPECT reconstruction was done using a HERMES (Nuclear Diagnostics, streptavidin-ABC solution (DAKO) and developed using 1% H2O2 and Stockholm, Sweden) application program, using filtered back projection 3.3Ј-diaminobenzidin-tetra-hydrochloride in Tris-HCl. The sections were adapted to pinhole SPECT. mounted in glycerin gelatin with a light methylgreen counter staining and To determine the radioactivity uptake semiquantitatively, and to correct analyzed. for differences in total amount of radioactivity administered and variable numbers labeled granulocytes, the colon uptake ratio was determined as Cytokine and chemokine determinations follows. Five consecutive transverse slices with the highest colon uptake were selected and added. Regions of interest were set for the colon and For cytokine and chemokine measurements, lung homogenates were di- by guest on September 23, 2021 abdominal background (not containing the colon) and the number of counts luted 1 to 2 in lysis buffer containing 300 mM NaCl, 30 mM Tris, 2 mM in each region of interest was measured. The colon uptake ratio was cal- MgCl2, 2 mM CaCl2, 1% Triton X-100, and pepstatin A, leupeptin, and culated by subtracting background activity from the colon activity and aprotinin (all 20 ng/ml; pH 7.4) and incubated at 4°C for 30 min. Homog- subsequently dividing the corrected colon uptake by the background to enates were centrifuged at 1500 ϫ g at 4°C for 15 min, and supernatants yield a specific uptake ratio: (counts colon-counts background)/(counts were stored at Ϫ20°C until assays were performed. Cytokines (TNF-␣, background). IL-1␤) and chemokines (KC, macrophage-inflammatory protein-2 (MIP- 2)) were measured using specific ELISA (R&D Systems, Minneapolis, Induction of pneumonia MN) according to the manufacturer’s instructions. Pathogen-free 8-wk-old female BALB/c mice were obtained from Harlan Blood cell differentiation Sprague Dawley. Pneumonia was induced as described previously (43– 45). Briefly, S. pneumoniae serotype 3 was obtained from American Type EDTA blood was collected from mice 24 h after i.p. administration of 0.5 Culture Collection (ATCC 6303; Rockville, MD). Pneumococci were mg 1B2 or cIg per mouse (n ϭ 4). Whole blood counts were determined grown for6htomid-logarithmic phase at 37°Cin5%CO2 using Todd- and the number of neutrophils was calculated from these totals, using cy- Hewitt broth (Difco), harvested by centrifugation at 1500 ϫ g for 15 min, tospin preparations stained with modified Giemsa stain (Diff-Quick; Baxter and washed twice in sterile isotonic saline. Bacteria were then resuspended Diagnostics, McGraw Perk, IL).

Table 1. Characteristics of mCD97 mAba

mAb 1B2 1A2 1D2/1C5

Binding to COS cells expressing: mCD97(EGF1,2,4) ϩϩϪ mCD97(EGF1,2,3,4) ϩϩϩ mCD97(EGF1,2,X,3,4) ϩϩϩ mCD97(EGF1h,2,4) ϪϩϪ mCD97(EGF1h,2h,4) ϪϪϪ hCD97(EGF1m,2m,5) ϩϩϪ hCD97(EGF1m,2,5) ϩϪϪ hCD97(EGF1,2,5) ϪϪϪ Epitope speciﬁcity EGF domain 1 EGF domain 2 EGF domain 3 Effect on adhesion of mouse erythrocytes Blocking Blocking Nonblocking to COS cells expressing mCD97

a Binding of the mAb to COS cells expressing different mCD97 isoforms and m-hCD97 chimeras was determined by ﬂow cytometry. Adhesion of CD55ϩ mouse erythrocytes to COS cells expressing mCD97 was used to test the effect of the mAb on the CD97–CD55 interaction. 1128 REQUIREMENT OF CD97 FOR NEUTROPHIL MIGRATION AND HOST DEFENSE

FIGURE 2. Effect of preincubation with mCD97 mAb on the homing of radioactively labeled neutrophils to the colon in experimental colitis. Pinhole SPECT was performed of mice with DSS colitis 1 h after adoptive transfer of 5 ϫ 106 99mTc-labeled neutrophils. A, Pinhole SPECT images of the pelvic region of mice 1 h after injection of neutrophils treated with cIg (upper panels) or 1B2 (lower panels). Corresponding transverse, sagittal, and coronal slices

are shown. Bladder (B), colon (C), limbs (L), pelvis (P), and spleen (S) are indicated. Whereas radioactivity uptake in colon is clearly visible with cIg-treated Downloaded from neutrophils, 1B2 efﬁciently blocked neutrophil homing. Images are representative for both groups. B, Radioactivity uptake in the colon. Speciﬁc colon radioactivity uptake was calculated of mice injected with neutrophils treated with cIg (E) or 1B2 (ࡗ). Medians are indicated with horizontal lines. Data .p Ͻ 0.05 ,ءء .are pooled from two separate experiments. n ϭ 3–4/group/experiment

Statistical analysis adoptive transfer into the blood circulation of acceptor mice sen- http://www.jimmunol.org/ Differences between groups were calculated by Mann-Whitney U test. For sitized with DSS. As depicted in Fig. 2, homing of neutrophils survival analyses, Kaplan-Meier analysis followed by log rank test was treated with 1B2 to the colon but also to other sites was signifi- performed. Values are expressed as mean Ϯ SEM. A two-tailed p value cantly diminished when compared with preincubation with cIg. of Ͻ 0.05 was considered to represent a significant difference. Application of mCD97 mAb results in increased severity of Results S. pneumoniae-induced pneumonia Generation and characterization of novel mCD97 mAb Based on the delaying effect on neutrophil migration, we expected To generate mAb with different binding sites on the mCD97R, an that application of mCD97 mAb would be detrimental for anti- Armenian hamster was immunized with ARHO12 cells stably ex- by guest on September 23, 2021 pressing mCD97(EGF1,2,3,4). Screening of hybridoma supernatants for mCD97 specificity identified Ͼ 100 positive clones. mAb with disparate binding sites were detected when analyzing binding to COS cells expressing different mCD97 isoforms and m-hCD97 chimeras. The mAb 1B2, 1C5, and 1D2 were selected and analyzed in detail, together with the earlier generated mAb 1A2 (6), for binding specificity using an extended panel of m-hCD97 chimeras. As shown in Table I, 1B2 binds EGF domain 1, 1A2 binds EGF domain 2, and 1C5 and 1D2 bind EGF domain 3. Accord- ingly, 1C5 and 1D2 recognize mCD97(EGF1,2,3,4) and mCD97 (EGF1,2,X,3,4) but not the smallest isoform mCD97(EGF1,2,4). We previously reported that the mAb 1A2 blocks adhesion of mouse erythrocytes to COS cells expressing mCD97(EGF1,2,4) or mCD97(EGF1,2,3,4) (6). Using the same assay, the ability of the novel mAb to interfere with the interaction between mCD97 and mouse CD55 was tested (Table I). Whereas adhesion was blocked by 1B2, no blocking effect was seen with 1C5 and 1D2, indicating that EGF domain 1 and 2 form the binding site for mCD55. This finding is in good agreement with a recent mutational analysis study demonstrating the importance of hCD97 EGF domains 1 and 2 for human CD55 binding (34).

Binding of CD97 mAb inhibits neutrophil homing in FIGURE 3. Effect of application of mAb to mCD97 on the course of experimental colitis pneumococcal pneumonia. Mice were inoculated i.n. with 1 ϫ 104 CFU S. pneumoniae at day 0. A, Survival study. Mortality of mice treated with cIg Using the newly generated mCD97 mAb, we evaluated the role of (E), 1B2 (ࡗ), or 1C5 (f) was assessed 1–2 times daily for 20 days. The CD97 in neutrophil migration. First, we investigated the homing of p value indicates the difference between the mCD97 mAb-treated groups radioactively labeled neutrophils that had been incubated with and the control group, n ϭ 10/group. B, Bacterial outgrowth in lungs. CFU mAb, to the colon in mice with experimental colitis. In this model, S. pneumoniae in lungs of mice treated with cIg, 1B2, or 1D2 20 and 44 h neutrophils, recruited from the peritoneum of donor mice and la- after inoculation. Medians are indicated with horizontal lines, n ϭ 8/group/ .p Ͻ 0.0005 ,ءءء ;p Ͻ 0.05 ,ء .beled with 99mTc-HMPAO, were incubated with cIg or 1B2 before time point The Journal of Immunology 1129 Downloaded from

FIGURE 4. Histopathology of lungs 20 h after induction of pneumococcal pneumonia. Mice treated with cIg (A) show interstitial and intraalveolar inflam- http://www.jimmunol.org/ mation with endothelialitis. In contrast, mice treated with 1B2 (B) and 1C5 (C) displayed only slight interstitial inflammation. This was confirmed by immuno- histochemistry for granulocytes showing massive influx of granulocytes in mice treated with cIg (D) and little granulocyte infiltration in mice treated with 1B2 (E) and 1C5 (F). A–C, H&E staining, D–F, Ly-6G immunostaining. Representative sections from all groups are shown. Magnification ϫ20. bacterial immunity. This assumption was tested in the well estab- areas in mCD97 mAb-treated mice (Fig. 4, E and F). At 44 h after lished S. pneumoniae pneumonia model (43–45). We first assessed inoculation, all mice showed moderate inflammatory infiltrates es- the effect of the mAb 1B2 and 1C5 on survival of mice after i.n. sentially composed of monocytes and macrophages. There was no ϫ 4 inoculation with 1 10 CFU S. pneumoniae. As shown in Fig. difference between control and treated mice (data not shown). The by guest on September 23, 2021 3A, a dramatic difference was observed between mice treated with observation of similar degrees of infiltration in all three groups at mCD97 mAb or cIg. Whereas all mice of the control group sur- 44 h may reflect a relative deficiency in neutrophil influx as the vived to day 20, 70% of animals from the 1B2- and the 1C5-treated bacterial load was much higher in mCD97 mAb-treated mice at group died within 2 wk. that time point. The rapid disease progression in mCD97 mAb-treated mice was We next determined expression levels of soluble inflammatory reflected by pneumococci titers cultured from the lungs as com- mediators implicated in pulmonary host defense (47). As depicted pared with the control group (Fig. 3B). At 20 h after inoculation, in Table II, levels of cytokines (TNF-␣, IL-1␤) and neutrophil- the lung S. pneumoniae burden was similar in all three groups. In attracting chemokines (KC, MIP-2) in the lungs did not signifi- contrast, after 44 h, mCD97 mAb-treated mice had four-log greater cantly differ between the three groups. These data imply that al- lung S. pneumoniae titers than cIg-treated mice. Lung pneumo- terations in the expression of humoral factors are unlikely to cocci titers below the detection limit, indicating bacterial clear- contribute to the diminished response in mCD97 mAb-treated ance, were observed in 7 of 8 mice from the control group, but mice during pneumococcal pneumonia. only in 0 of 8 mice of the 1B2-treated and in 2 of 8 mice of the In a final experiment, we confirmed that i.p. administration of 1C5-treated group. Blood cultures obtained at 20 h, were negative mCD97 mAb did not result in neutropenia. One day after appli- for S. pneumoniae in all three groups. After 44 h, pneumococci cation of 0.5 mg Ig, peripheral blood neutrophil counts (ϫ 106 Ϯ were detectable in the blood of 0 of 8 mice, 4 of 8 mice, and 2 of SEM) were 1.0 Ϯ 0.3 in cIg-treated and 0.9 Ϯ 0.3 in 1B2-treated 8 mice of the groups treated with cIg, 1B2, and 1C5, respectively. mice ( p ϭ 0.4, n ϭ 4). mCD97 mAb impair the recruitment of neutrophils to the lungs during pneumococcal pneumonia Discussion Histological analysis revealed that at 20 h after infection with S. The molecular structure, ligand specificity, and expression led to pneumoniae, lungs of all control mice displayed interstitial and the idea that the EGF-TM7 receptor CD97 has a role of in cell peribronchial inflammation together with some degree of endothe- trafficking (2, 4, 7, 30, 39). Further evidence was provided when lialitis. Moreover, in 50% of these mice multiple foci of pneumo- migration and invasiveness of tumor cells was found to correlate nia were present in the lungs (Fig. 4A). In contrast, mice treated with CD97 expression (28, 29). In the current investigation, we with mCD97 mAb presented less inflammation in the lungs and used a panel of newly generated mCD97 mAb to evaluate the none of the animals had developed foci of pneumonia (Fig. 4, B function of CD97 in leukocyte migration in vivo. Because of their and C). These findings were confirmed by immunostaining for fast migration and their important role in the eradication of patho- granulocytes showing a prominent granulocytic infiltration in cIg- gens we choose neutrophils as a model system. Neutrophils are treated mice (Fig. 4D) but only few granulocytes in interstitial recruited from the bloodstream in response to molecular changes at 1130 REQUIREMENT OF CD97 FOR NEUTROPHIL MIGRATION AND HOST DEFENSE

Table 2. Cytokine and chemokine levels in lung homogenatesa

20 h 44 h

cIg 1B2 1C5 cIg 1B2 1C5

TNF-␣ 0.6 Ϯ 0.07 0.5 Ϯ 0.03 0.7 Ϯ 0.05 1.3 Ϯ 0.1 0.9 Ϯ 0.3 1.5 Ϯ 0.3 IL-1␤ 6.3 Ϯ 1.3 4.0 Ϯ 0.4 5.1 Ϯ 0.5 3.6 Ϯ 0.4 4.1 Ϯ 1.2 5.9 Ϯ 0.9 KC 3.1 Ϯ 0.2 2.7 Ϯ 0.1 3.0 Ϯ 0.1 3.0 Ϯ 0.1 3.0 Ϯ 0.7 3.5 Ϯ 0.3 MIP-2 5.5 Ϯ 0.5 4.0 Ϯ 0.2 5.1 Ϯ 0.1 6.4 Ϯ 0.4 4.7 Ϯ 0.9 6.1 Ϯ 0.7

a Data (measured in nanograms per milliliter) are mean Ϯ SEM of eight mice per group, measured 20 and 44 h after i.n. inoculation with 1 ϫ 104 CFU S. pneumoniae. the surface of the vascular endothelium that signal injury or infec- sical view, chemokines are presented to leukocytes in two forms, tion. The steps involved in the trafficking process sequentially in- either immobilized through binding to GAG at the endothelial ex- clude selectin-carbohydrate, chemokine-chemokine receptor, and tracellular matrix or as soluble molecules (51). This view has been integrin-Ig family interactions and are basically explored by other challenged by studies showing that the formation of complexes types of leukocytes as well (48, 49). with the GAG side chains of proteoglycans is generally needed for In initial experiments, we adoptively transferred radioactively the in vivo activity of certain chemokines. Previously, it was found labeled neutrophils into mice with experimental colitis. Ex vivo that IL-8 must bind GAG to elicit directed neutrophil migration Downloaded from incubation with mCD97 mAb significantly diminished homing, (52). More recently, Li et al. (53) could demonstrate that the re- demonstrating that binding of mAb to CD97 on neutrophils causes lease of complexes between the cell surface proteoglycan synde- a defect in the migration of these cells. Based on this observation, can-1 and the chemokine KC from the epithelium in lung injury is we expected that application of mCD97 mAb would diminish the regulated by the matrix metalloproteinase matrilysin (MMP-7). In resistance to bacterial infections. To test this idea, we used a mu- syndecan-1- or matrilysin-deficient mice, KC was hardly detect-

rine model of S. pneumoniae-induced pneumonia. Application of able in alveolar fluid and neutrophil influx was confined. Synde- http://www.jimmunol.org/ mCD97 mAb caused an early and persistent defect in neutrophil can-1 is a component of the extracellular matrix. These findings migration to the lungs that was associated with the inability to indicate that extracellular matrix fragments cooperate with chemo- control the growth of S. pneumoniae. kines to create gradients that direct cell movement. Supportive Together, these data reveal an important role of CD97 in the evidence for this concept comes from an investigation showing migration of neutrophils to sites of infection and injury. Neutro- that important monocyte attractants are devoid of their in vivo phils constitutively express substantial levels of CD97 (1, 6). After activity when the GAG-binding site is mutated (54). The question cellular stimulation, expression of CD97 on human neutrophils arises, whether leukocytes bind chemokines complexed with pro- was found to increase 1.5- to 2-fold (20). Expression of CD97 on teoglycans exclusively through chemokine receptors or whether most types of myeloid cells and on activated lymphocytes (1, 23) additional molecules are involved. Based on its ability to bind by guest on September 23, 2021 indicates that the role of CD97 in leukocyte migration might not be chondroitin sulfate, it seems possible that CD97 has a role in the restricted to neutrophils. fixation of soluble proteoglycan-chemokine complexes on the leu- On current evidence, the observed effect of mCD97 mAb on kocyte surface. Such a mechanism would support trafficking along neutrophil migration cannot be causally linked to one specific mo- chemokine gradients under the shear conditions caused by blood lecular interaction of CD97. Two cellular ligands of CD97 have flow. In addition, it would promote the efficient parallel engage- been identified hitherto. EGF domains 1 and 2 mediate binding to ment of numerous chemokine receptors on individual cells. the complement regulator CD55 (Refs. 6, 33, 34 and this study). In conclusion, this study for the first time provides in vivo ev- The more recently identified interaction with the GAG side chain idence for a role for CD97 in cell migration. Whether the function chondroitin sulfate is mediated by EGF domain 4 in humans and of CD97 indeed relates to the binding of proteoglycan-chemokine its homologue in mice, EGF domain 3, respectively (Ref. 38 and complexes remains to be demonstrated in further investigations. In M. J. Kwakkenbos and J. Hamann, unpublished observation). For addition, it needs to be shown whether other EGF-TM7 receptors the in vivo experiments, mAb that recognize EGF domain 1 (1B2) have a comparable role in leukocyte migration. Suggestively, and EGF domain 3 (1C5) were used. Despite different effects on EMR2, a molecular twin of hCD97 (not existing in mice) with a the molecular interactions of CD97—1B2 blocks binding to CD55, nearly identical EGF domain region (12), also binds chondroitin whereas 1C5 is expected to interfere with chondroitin sulfate bind- sulfate (38). ing—application of both mAb had similar, impairing consequences for the course of pneumococcal pneumonia. This result Acknowledgments might be explained in different ways. Possibly, mAb to the EGF- We thank Joost Daalhuisen for excellent technical assistance with the domain region, that do not block an interaction in cell-based ad- pneumonia model; Els de Groot, Chris van Zeventer, Astrid Bijl, and hesion assays, sterically hinder the same interaction in situ. Alter- Walter Pouwels for experimental help with the generation and character- natively, both interactions could be engaged in the role of CD97 in ization of mAb; Nike Claessens for performing histological stainings; and neutrophil migration. Prof. Lucien Aarden, Mark J. Kwakkenbos, and Dr. Rene Lutter for helpful A possible interpretation for our data could be that CD97 facil- comments and suggestions. itates the binding of chemokines to neutrophils. Chemokines are small, secreted proteins that are critically involved in the recruit- References ment and activation of leukocytes (50). Divided into several fam- 1. Eichler, W., G. Aust, and D. Hamann. 1994. Characterization of an early acti- ilies, different chemokines act on different types of leukocytes. For vation-dependent antigen on lymphocytes defined by the monoclonal antibody example, the most important neutrophil attractants are the glutamic BL-Ac(F2). Scand. J. Immunol. 39:111. ϩ 2. Hamann, J., W. Eichler, D. Hamann, H. M. J. Kerstens, P. J. Poddighe, acid-leucine-arginine-positive (ELR ) CXC chemokines KC and J. M. N. Hoovers, E. Hartmann, M. Strauss, and R. A. W. van Lier. 1995. Ex- MIP-2 in the mouse and IL-8 in man (47). According to the clas- pression cloning and chromosomal mapping of the leucocyte activation antigen The Journal of Immunology 1131

CD97, a new seven-span transmembrane molecule of the secretin receptor su- 27. Aust, G., W. Eichler, S. Laue, I. Lehmann, N. E. Heldin, O. Lotz, perfamily with an unusual extracellular domain. J. Immunol. 155:1942. W. A. Scherbaum, H. Dralle, and C. Hoang-Vu. 1997. CD97: a dedifferentiation 3. Hamann, J., E. Hartmann, and R. A. W. van Lier. 1996. Structure of the human marker in human thyroid carcinomas. Cancer Res. 57:1798. CD97 gene: exon shuffling has generated a new type of seven-span transmem- 28. Steinert, M., M. Wobus, C. Boltze, A. Schu¨tz, M. Wahlbuhl, J. Hamann, and brane molecule related to the secretin receptor superfamily. Genomics 32:144. G. Aust. 2002. Expression and regulation of CD97 on colorectal carcinoma cell 4. Gray, J. X., M. Haino, M. J. Roth, J. E. Maguire, P. N. Jensen, A. Yarme, lines and tumor tissues. Am. J. Pathol. 161:1657. M. A. Stetler-Stevenson, U. Siebenlist, and K. Kelly. 1996. CD97 is a processed, 29. Aust, G., M. Steinert, A. Schu¨tz, M. Wahlbuhl, J. Hamann, and M. Wobus. 2002. seven-transmembrane, heterodimeric receptor associated with inflammation. CD97, but not its closely related EGF-TM7 family member EMR2, is expressed J. Immunol. 157:5438. on gastric, pancreatic, and esophageal carcinomas. Am. J. Clin. Pathol. 118:699. 5. Qian, Y.-M., M. Haino, K. Kelly, and W.-C. Song. 1999. Structural character- 30. Hamann, J., B. Vogel, G. M. W. van Schijndel, and R. A. W. van Lier. 1996. The ization of mouse CD97 and study of its specific interaction with murine decay- seven-span transmembrane receptor CD97 has a cellular ligand (CD55, DAF). accelerating factor (DAF, CD55). Immunology 98:303. J. Exp. Med. 184:1185. 6. Hamann, J., C. van Zeventer, A. Bijl, C. Molenaar, K. Tesselaar, and 31. Lublin, D. M., and J. P. Atkinson. 1989. Decay-accelerating factor: biochemistry, R. A. W. van Lier. 2000. Molecular cloning and characterization of mouse CD97. molecular biology, and function. Annu. Rev. Immunol. 7:35. Int. Immunol. 12:439. 32. Liszewski, M. K., T. C. Farries, D. M. Lublin, I. A. Rooney, and J. P. Atkinson. 7. Hamann, J. 2002. CD97. In Encyclopedia of Molecular Medicine. 1996. Control of the complement system. Adv. Immunol. 61:201. T. E. Creighton, ed. John Wiley & Sons, Inc., New York. p. 692. 33. Hamann, J., C. Stortelers, E. Kiss-Toth, B. Vogel, W. Eichler, and R. A. W. van Lier. 1998. Characterization of the CD55 (DAF)-binding site on the 8. McKnight, A. J., and S. Gordon. 1998. The EGF-TM7 family: unusual structures seven-span transmembrane receptor CD97. Eur. J. Immunol. 28:1701. at the leukocyte surface. J. Leukocyte Biol. 63:271. 34. Lin, H.-H., M. Stacey, C. Saxby, V. Knott, Y. Chaudhry, D. Evans, S. Gordon, 9. Baud, V., S. L. Chissoe, E. Viegas-Pequignot, S. Diriong, V. C. N’Guyen, A. J. McKnight, P. Handford, and S. Lea. 2001. Molecular analysis of the epi- B. A. Roe, and M. Lipinski. 1995. EMR1, an unusual member in the family of dermal growth factor-like short consensus repeat domain-mediated protein-pro- hormone receptors with seven transmembrane segments. Genomics 26:334. tein interactions: dissection of the CD97-CD55 complex. J. Biol. Chem. 10. McKnight, A. J., A. J. Macfarlane, P. Dri, L. Turley, A. C. Willis, and S. Gordon. 276:24160. 1996. Molecular cloning of F4/80, a murine macrophage-restricted cell surface 35. Berger, M., and M. E. Medof. 1987. Increased expression of complement decay- glycoprotein with homology to the G-protein-linked transmembrane 7 hormone accelerating factor during activation of human neutrophils. J. Clin. Invest. Downloaded from receptor family. J. Biol. Chem. 271:486. 79:214. 11. Lin, H.-H., L. J. Stubbs, and M. L. Mucenski. 1997. Identification and charac- 36. Li, L., I. Spendlove, J. Morgan, and L. G. Durrant. 2002. CD55 is over-expressed terization of a seven transmembrane hormone receptor using differential display. in the tumour environment. Br. J. Cancer 84:80. Genomics 41:301. 37. Davis, S. J., S. Ikemizu, M. K. Wild, and P. A. van der Merwe. 1998. CD2 and 12. Lin, H.-H., M. Stacey, J. Hamann, S. Gordon, and A. J. McKnight. 2000. Human the nature of protein interactions mediating cell-cell recognition. Immunol. Rev. EMR2, a novel EGF-TM7 molecule on chromosome 19p13.1, is closely related 163:217. to CD97. Genomics 67:188. 38. Stacey M., G.-W. Chang, J. Q. Davies, M. J. Kwakkenbos, R. Sanderson,

13. Stacey, M., H.-H. Lin, K. L. Hilyard, S. Gordon, and A. J. McKnight. 2001. J. Hamann, S. Gordon, and H.-H. Lin. 2003. The epidermal growth factor-like http://www.jimmunol.org/ Human epidermal growth factor (EGF) module-containing mucin-like hormone domains of human EMR2 receptor mediate cell attachment through chondroitin receptor 3 is a new member of the EGF-TM7 family that recognizes a ligand on sulphate glycosaminoglycans. Blood 102:2916. human macrophages and activated neutrophils. J. Biol. Chem. 276:18863. 39. Van Lier, R. A., W. Eichler, and J. Hamann. 1996. Seven span transmembrane 14. Caminschi, I., K. M. Lucas, M. A. O’Keeffe, H. Hochrein, Y. Laabi, F. Ko¨ntgen, molecules: novel receptors involved in leukocyte adhesion. Immunol. Lett. A. M. Lew, K. Shortman, and M. D. Wright. 2001. Molecular cloning of F4/80- 54:185. like-receptor, a seven-span membrane protein expressed differentially by den- 40. Kwakkenbos, M. J., R. A. W. van Lier, and J. Hamann. 2001. Characterization of dritic cell and monocyte-macrophage subpopulations. J. Immunol. 167:3570. EGF-TM7 family members by novel monoclonal antibodies. In Leucocyte Typing 15. Stacey, M., G.-W. Chang, S. L. Sanos, L. R. Chittenden, L. Stubbs, S. Gordon, VII. D. Mason, ed. Oxford University Press, Oxford, p. 381. and H.-H. Lin. 2002. EMR4, a novel epidermal growth factor (EGF)-TM7 mol- 41. Ten Hoeve, T., P. Drillenburg, J. Wijnholds, A. A. te Velde, and ecule up-regulated in activated mouse macrophages, binds to a putative cellular S. J. van Deventer. 2002. Differential susceptibility of multidrug resistance pro- ligand on B lymphoma cell line A20. J. Biol. Chem. 277:29283. tein-1 deﬁcient mice to DSS and TNBS-induced colitis. Dig. Dis. Sci. 47:2056.

16. Hamann, J., M. J. Kwakkenbos, E. C. de Jong, H. Heus, A. S. Olsen, and 42. Habraken, J. B., K. de Bruin, M. Shehata, J. Booij, R. Bennink, by guest on September 23, 2021 R. A. W. van Lier. 2003. Inactivation of the EGF-TM7 receptor EMR4 after the B. L. van Eck-Smit, and E. Busemann-Sokole. 2001. Evaluation of high-resolu- Pan-Homo divergence. Eur. J. Immunol. 33:1365. tion pinhole SPECT using a small rotating animal. J. Nucl. Med. 42:1863. 17. Nechiporuk, T., L. D. Urness, and M. T. Keating. 2001. ETL, a novel seven- 43. Van der Poll, T., C. V. Keogh, X. Guirao, W. A. Buurman, M. Kopf, and transmembrane receptor that is developmentally regulated in the heart: ETL is a S. F. Lowry. 1997. Interleukin-6 gene-deficient mice show impaired defense member of the secretin family and belongs to the epidermal growth factor-seven- against pneumococcal pneumonia. J. Infect. Dis. 176:439. transmembrane subfamily. J. Biol. Chem. 276:4150. 44. Schultz, M. J., J. Wijnholds, M. P. Peppelenbosch, M. J. Vervoordeldonk, 18. Terskikh, A. V., M. C. Easterday, L. Li, L. Hood, H. I. Kornblum, P. Speelman, S. J. van Deventer, P. Borst, and T. van der Poll. 2001. Mice lacking D. H. Geschwind, and I. L. Weissman. 2001. From hematopoiesis to neuropoi- the multidrug resistance protein 1 are resistant to Streptococcus pneumoniae- esis: evidence of overlapping genetic programs. Proc. Natl. Acad. Sci. USA induced pneumonia. J. Immunol. 166:4059. 98:7934. 45. Rijneveld, A. W., S. Florquin, J. Branger, P. Speelman, S. J. van Deventer, and ␣ 19. Stacey, M., H.-H. Lin, S. Gordon, and A. J. McKnight. 2000. LNB-TM7, a group T. van der Poll. 2001. TNF- compensates for the impaired host defense of IL-1 of seven-transmembrane proteins related to family-B G-protein-coupled recep- type I receptor-deficient mice during pneumococcal pneumonia. J. Immunol. tors. Trends Biochem. Sci. 25:284. 167:5240. 46. Rijneveld, A. W., M. Levi, S. Florquin, P. Speelman, P. Carmeliet, and 20. Kwakkenbos, M. J., G.-W. Chang, H.-H. Lin, W. Pouwels, E. C. de Jong, T. van der Poll. 2002. Urokinase receptor is necessary for adequate host defense R. A. W. van Lier, S. Gordon, and J. Hamann. 2002. The human EGF-TM7 against pneumococcal pneumonia. J. Immunol. 168:3507. family member EMR2 is a heterodimeric receptor expressed on myeloid cells. 47. Strieter, R. M., S. L. Kunkel, M. P. Keane, and T. J. Standiford. 1999. Chemo- J. Leukocyte Biol. 71:854. kines in lung injury: Thomas A. Neff lecture. Chest 116(Suppl. 1):103S. 21. Chang, G.-W., M. Stacey, M. J. Kwakkenbos, J. Hamann, S. Gordon, and H.-H. 48. Butcher, E. C. 1991. Leukocyte-endothelial cell recognition: three (or more) steps Lin. 2003. Proteolytic cleavage of the EMR2 receptor requires both the extra- to specificity and diversity. Cell 67:1033. cellular stalk and the GPS motif. FEBS Lett. 547:145. 49. Springer, T. A. 1994. Traffic signals for lymphocyte recirculation and leucocyte 22. Ichtchenko, K., M. A. Bittner, V. Krasnoperov, A. R. Little, O. Chepurny, emigration: the multistep paradigm. Cell 76:301. ␣ R. W. Holz, and A. G. Petrenko. 1999. A novel ubiquitously expressed -latro- 50. Luster, A. D. 1998. Chemokines-chemotactic cytokines that mediate inflamma- toxin receptor is a member of the CIRL family of G-protein-coupled receptors. tion. N. Engl. J. Med. 338:436. J. Biol. Chem. 274:5491. 51. Tanaka, Y., D. H. Adams, and S. Shaw. 1993. Proteoglycans on endothelial cells 23. Jaspars, L. H., W. Vos, G. Aust, R. A. W. van Lier, and J. Hamann. 2001. Tissue present adhesion-inducing cytokines to leukocytes. Immunol. Today 14:111. distribution of the human CD97 EGF-TM7 receptor. Tissue Antigens 57:325. 52. Webb, L. M., M. U. Ehrengruber, I. Clark-Lewis, A. Baggiolini, and A. Rot. 24. Eichler, W., J. Hamann, and G. Aust. 1997. Expression characteristics of the 1993. Binding to heparan sulfate or heparin enhances neutrophil responses to human CD97 antigen. Tissue Antigens 50:429. interleukin 8. Proc. Natl. Acad. Sci. USA 90:7158. 25. Hamann, J., J. O. Wishaupt, R. A. W. van Lier, T. J. M. Smeets, F. C. Breedveld, 53. Li, Q., P. W. Park, C. L. Wilson, and W. C. Parks. 2003. Matrilysin shedding of and P. P. Tak. 1999. Expression of the actvation antigen CD97 and its ligand in syndecan-1 regulates chemokine mobilization and transepithelial efflux of neu- rheumatoid synovial tissue. Arthritis Rheum. 42:650. trophils in acute lung injury. Cell 111:635. 26. Visser, L., A. F. de Vos, J. Hamann, M.-J. Melief, M. van Meurs, 54. Proudfoot, A. E., T. M. Handel, Z. Johnson, E. K. Lau, P. Li Wang, R. A. W. van Lier, J. D. Laman, and R. Q. Hintzen. 2002. Expression of the I. Clark-Lewis, F. Borlat, T. N. Wells, and M. H. Kosco-Vilbois. 2003. Glycos- EGF-TM7 receptor CD97 and its ligand CD55 (DAF) in multiple sclerosis. aminoglycan binding and oligomerization are essential for the in vivo activity of J. Neuroimmunol. 132:156. certain chemokines. Proc. Natl. Acad. Sci. USA 100:1885.