Alternative Activation-Associated CC--1, a Novel Structural Homologue of Macrophage Inflammatory Protein-1α with a Th2-Associated Expression Pattern This information is current as of September 29, 2021. Vitam Kodelja, Carola Müller, Oliver Politz, Nahid Hakij, Constantin E. Orfanos and Sergij Goerdt J Immunol 1998; 160:1411-1418; ; http://www.jimmunol.org/content/160/3/1411 Downloaded from

References This article cites 52 articles, 17 of which you can access for free at: http://www.jimmunol.org/content/160/3/1411.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

by guest on September 29, 2021 *average

Subscription Information about subscribing to The Journal of 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 © 1998 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Alternative Macrophage Activation-Associated CC-Chemokine-1, a Novel Structural Homologue of Macrophage Inflammatory Protein-1␣ with a Th2-Associated Expression Pattern1

Vitam Kodelja, Carola Mu¨ller, Oliver Politz, Nahid Hakij, Constantin E. Orfanos, and Sergij Goerdt2

We have cloned a novel human CC-chemokine, alternative macrophage activation-associated CC-chemokine (AMAC)-1. The isolated cDNA clone (803 bp) shows a single open reading frame of 267-bp coding for 89 amino acid residues; mature AMAC-1 protein is predicted to consist of 69 amino acids with a m.w. of 7855. Sequence alignment and 3D-modeling show the typical Downloaded from structural characteristics of CC- with special features in the receptor-activating domain. AMAC-1 is most closely related to MIP-1␣ with a cDNA and protein sequence homology of 55% and 59%, respectively. However, the expression pattern of AMAC-1 is directly opposite to that of MIP-1␣. While MIP-1␣ is induced by classical macrophage mediators such as LPS and is inhibited by IL-4 and glucocorticoids, AMAC-1 is specifically induced in by alternative macrophage mediators such as IL-4, IL-13, and IL-10. Expression of AMAC-1 is inhibited by IFN-␥ while glucocorticoids exert a slightly positive synergistic effect in combination with IL-4. Peripheral blood monocytes do not express AMAC-1; time course experiments show http://www.jimmunol.org/ that monocyte-to-macrophage differentiation is a prerequisite for AMAC-1 expression. Expression of AMAC-1 by granulocyte-- macrophage CSF/IL-4-induced, monocyte-derived dendritic cells is complex; in mature adherent dendritic cells, however, only minor AMAC-1 mRNA expression was found. In vivo, AMAC-1 is expressed by alveolar macrophages from healthy persons, smokers, and asthmatic patients. In conclusion, AMAC-1 is a novel CC-chemokine whose expression is induced in alternatively activated macrophages by Th2-associated ; thus, AMAC-1 may be involved in the APC-dependent development in inflammatory and immune reactions. The Journal of Immunology, 1998, 160: 1411–1418.

he concept of Th2-associated alternative immunologic inhibited by IFN-␥. Furthermore, expression and synthesis of by guest on September 29, 2021 macrophage activation was introduced in 1992 by Stein et proinflammatory cytokines such as IL-1 (6, 7), IL-6 (8), and T al. (1). In contrast to classical macrophage activation by TNF-␣ (6, 7) are inhibited by IL-4. In contrast, IL-4 and IFN-␥ IFN-␥ and LPS, activation of macrophages by agents such as IL-4 sometimes exert synergistic effects such as accumulation of cyto- or glucocorticoids (GCs),3 was classified as alternative. Besides plasmic CD23 (9) or inhibition of CD14 expression (10). Vice IFN-␥ and IL-4, macrophage activation and differentiation is in- versa, GC effects on macrophages are sometimes antagonistic to fluenced by other cytokines. The effects of TNF-␣ and IL-12 par- IL-4; in contrast to IL-4, GCs, for example, inhibit macrophage tially overlap with those of IFN-␥. IL-10 and IL-13 resemble IL-4, expression of CD23 (9, 11). In general, IL-4-induced inflammatory ␤ ␥ while TGF- acts similarly to GC. Nevertheless, IFN- and IL-4 macrophages adopt an alternative phenotype characterized by a are the agents best known to exert a wide range of antagonistic high capacity for endocytic clearance and by reduced proinflam- effects on macrophages; for example, expression of the three spe- matory secretion (1). ␥ ␥ cies of Fc Rs (2, 3) is induced by IFN- , but is inhibited by IL-4. In previous reports, we have shown that MS-1 high m.w. On the other hand, expression of the macrophage mannose recep- protein (MS-1-HMWP) (9, 12–14) and RM 3/1 Ag (15, 16) tor (1, 4) and of 15-lipoxygenase (5) is induced by IL-4, but is characterize alternative macrophage phenotypes (17) in that their expression is induced by IL-4 and GC and inhibited by IFN-␥ (9, 18). In vivo, MS-1-HMWPϩ,RM3/1ϩ alternatively Klinik und Poliklinik fu¨r Dermatologie, Universita¨tsklinikum Benjamin Franklin, activated macrophages are found during the healing phase of Freie Universitat Berlin, Berlin, Germany acute inflammatory reactions (19), in chronic inflammatory dis- Received for publication September 2, 1997. Accepted for publication October 20, 1997. eases such as rheumatoid arthritis (20) and psoriasis (21), and in The costs of publication of this article were defrayed in part by the payment of page wound healing tissue (9). In addition, alternatively activated charges. This article must therefore be hereby marked advertisement in accordance macrophages are the cells of origin in cutaneous macrophage- with 18 U.S.C. Section 1734 solely to indicate this fact. derived tumors (14, 18). In contrast to IFN-␥-induced classi- 1 This work was supported by the Deutsche Forschungsgemeinschaft through grants cally-activated macrophages that occur during early phases of to S. G. (Go 470/2-1 and -2). inflammation (19) and in high turnover reactive granulomas 2 Address correspondence and reprint requests to Professor S. Goerdt, Klinik und Poliklinik fu¨r Dermatologie, Universita¨tsklinikum Benjamin Franklin, Freie Univer- (14, 18, 22), alternatively activated macrophages are associated sita¨t Berlin, Berlin, Germany. with a high degree of vascularization in vivo (9, 14, 18) and 3 Abbreviations used in this paper: GC, glucocorticoids; MS-1-HMWP, MS-1 high seem to be angiogenic in vitro (23). Furthermore, alternatively m.w. protein; AMAC-1, alternative macrophage activation-associated CC-chemo- activated macrophages do not costimulate, but actively inhibit kine; MIP, macrophage inflammatory protein; GM-CSF, granulocyte-macrophage ϩ CSFpfu, plaque-forming units. mitogen-induced proliferation of PBL and CD4 T cells by an

Copyright © 1998 by The American Association of Immunologists 0022-1767/98/$02.00 1412 AMAC-1-CC-CHEMOKINE FROM ALTERNATIVELY ACTIVATED MACROPHAGES unknown mechanism (24). Therefore, we reasoned that alter- Germany); human recombinant macrophage CSF (M-CSF), used at 100 natively activated macrophages may host a not yet fully unrav- U/ml, was from Cellular Products (Buffalo, NY); human rIL-4, used at 15 eled molecular repertoire important in modulating diverse in- ng/ml for induction of alternative macrophage differentiation and at 45 ng/ml for derivation of dendritic cells from monocytes, was from Serva flammatory and immune responses. (Heidelberg, Germany); human rIL-10, used at 50 ng/ml, was from Biomol Here, we report the identification, cloning, and expression anal- (Hamburg, Germany). Dexamethasone, used at 5 ϫ 10Ϫ7 M or as indi- ysis of a novel CC-chemokine, alternative macrophage activation- cated, and LPS from Escherichia coli serotype 055:B5, used at 25 ␮g/ml, associated CC-chemokine (AMAC)-1, induced by Th2-associated were from Sigma. cytokines such as IL-4, IL-13, and IL-10 in alternatively activated Isolation of cDNA clones macrophages in vivo and found in alveolar macrophages in vitro. Total RNA from alternatively activated macrophages was isolated using Stratagene isolation kits. The poly(A)ϩ fraction was separated with Oli- Materials and Methods gotex-dT (Qiagen, Hilden, Germany) and 2 ␮g were used for synthesis of Tissues, cells, and cell lines cDNA library using ␭ZAP Express EcoRI/XhoI vector cloning (Strat- agene, Heidelberg, Germany). The library consisted of 1.2 ϫ 105 original Skin and bowel specimens were taken at routine surgery and bronchoal- clones, 86% of which contained inserts. This first cDNA library was am- veolar lavage cells were harvested at routine BAL after informed consent plified up to a titer of 2.5 ϫ 108 pfu/ml. The resultant alternatively acti- was obtained. Cell lines used were monocytic leukemia cell line THP-1 and vated macrophage cDNA library and a commercially available human melanoma cell line MEWO cultured in RPMI 1640 supplemented with spleen cDNA library (␭gt10, Clontech, Heidelberg, Germany) were differ- 10% FCS (Biochrom, Berlin, Germany) and appropriate concentrations of entially screened as follows. penicillin/streptomycin, glutamine, and nonessential amino acids (all from A total of 2 ϫ 106 pfu of each cDNA library were plated on NZY plates Biochrom); human umbilical vein endothelial cells were isolated as de- and three replicas on Nylon filters (Millipore, Eschborn, Germany) or Downloaded from scribed (25) and cultured in Clonetics endothelial basal medium (Clonetics, GeneScreen (DuPont/NEN/Life Science, Cologne, Germany) were taken. Walkersville, MD) supplemented with 10 ␮g/ml human epidermal growth For differential screening, 32P-labeled first cDNA was prepared using 2 to factor, 1 ␮g/ml hydrocortisone, bovine brain extract 12 ␮g/ml, 2% FCS 5 ␮g poly(A)ϩ from alternatively and classically activated macrophages and appropriate concentrations of gentamicin and amphotericin B. Isolation and from control macrophages. Poly(A)ϩ was isolated with use of Oligo- and culture of human monocytes/macrophages and monocyte-derived den- tex-dT and was reverse transcribed overnight at 42°C in 1 mM dATP, dritic cells as well as PBL was performed largely as described (9, 18, 26). dGTP, dTTP and 0.05 mM dCTP 200 ␮g/ml oligo(dT)18-mer (Pharmacia), The cells were purified using EDTA-anticoagulated blood from single do- 10 U AMV reverse transcriptase (Invitrogen, Leek, The Netherlands), 60 U http://www.jimmunol.org/ nors or pooled buffy coats. A total of 35 ml of blood were layered on top RNasin (Ambion, Austin, TX), and 125 ␮Ci [␣Ϫ32P]dCTP with a specific of 15 ml Ficoll-Paque (Biochrom) in a 50-ml Leuco-sep centrifuge tube activity of 800 ␮Ci/mmol (Amersham Buchler, Braunschweig, Germany). (Greiner, Nu¨rtingen, Germany) and were centrifuged in a swing out rotor Unincorporated radioactivity was separated using Sephadex-25-filled for 40 min at 650 ϫ g at room temperature without using the brake. PBMC NAP-5 columns (Pharmacia). Filters were prehybridized for 2 to6hat were collected at the serum/Ficoll interface and washed three times in 42°C in 10% dextransulfate, 5ϫ Denhardt’s solution, 2ϫ SSC, 1% SDS, ϩ ϩ sterile Ca2 and Mg2 free PBS (Biochrom). 3 ml of a suspension con- 50% deionized formamide, and 100 ␮g/ml salmon sperm DNA. For hy- taining 5 to 8 ϫ 108 PBMC in PBS were layered on top of 30 ml of a bridization, denatured labeled first cDNA probes were added and hybrid- preformed Percoll gradient (Pharmacia, Freiburg, Germany) in a 50-ml ized for 2 days at 42°C. After hybridization, the filters were washed twice tube; the Percoll gradient (13.5 ml Percoll, 1.5 ml 10ϫ Earle’s MEM, 15 in 2ϫ SSC, 1% SDS at room temperature, twice in 2ϫ SSC, 1% SDS at ml Spinner’s medium supplemented with appropriate concentrations of 65°C for 30 min, and one final wash was in 0.1 ϫ SSC at room temper- penicillin/streptomycin, glutamine, and nonessential amino acids (all from ature. Filters were dried, put into saran wrap, and exposed to X-AR Film by guest on September 29, 2021 Biochrom)) had been preformed in an SS-34 rotor of an ultracentrifuge (Kodak, New Haven, CT). After the exposure, differentially expressed (Sorvall, Frankfurt, Germany) at 11950 ϫ g for 12 min at 20°C without clones were picked. After amplification of these clones, ␭DNA was iso- using the brake. Percoll gradients with the cells on top were centrifuged at lated using a commercially available kit (Qiagen) and was PCR-amplified 650 ϫ g for 40 min at 20°C without using the brake. The upper cell layer with T3/T7 primer for ␭ZAP clones and ␭gt10 forward/reverse primers for contained approximately 80 to 90% monocytes while the lower cell layer spleen cDNA clones using GeneAmp XL PCR kit (Perkin Elmer, Branch- contained approximately 90 to 95% PBL. Both layers were collected sep- burg, NJ). In order to identify the extreme 5Ј end of the AMAC-1 mRNA, arately, and the cells were washed three times in PBS and further used or 5ЈRACE system version 2.0 (Life Technologies, Eggenstein, Germany) frozen in liquid nitrogen. was used with the following primers: REV (5Ј-TCA CAG TGA GAA TGC For culture, monocytes were resuspended in McCoy’s medium (Bio- TGG TTT ACC TTT TAT; 751–780); REV1 (5Ј-GAG TTG AAG GGA chrom) supplemented with 10% FCS and appropriate concentrations of AAG GGG AAA GGA TGA TAA; 584–613); REV3 (5Ј-CTC CAG GGT penicillin/streptomycin, glutamine, and nonessential amino acids, and were GGC AGG GCC ATT GCC CT; 406–431) (see Fig. 1). then transferred into Teflon-coated, UV-irradiated (1 min; transiluminator Nucleotide sequences were determined by automated sequencing of 2011 macrovue (LKB, Germany) plastic bags (Biofolie, Heraeus, Hanau, both DNA strands using an Applied Biosystems 373 sequencer (Foster Germany), which were sealed using Polystar 410 HM (Rische und Her- City, CA) using dye terminator cycle sequencing ready reaction kit (Perkin furth, Hamburg, Germany). Cell numbers varied between 0.2 ϫ 106 and Elmer) for dideoxy PCR sequencing of DNA fragments isolated from aga- 2 ϫ 106 (usually 1.3 ϫ 106) monocytes/ml according to the length of the rose gels. From 20 to 100 ng of DNA template were used and amplified (30 culture period (3 and 6 days). Mediators were added directly to the medium s at 96°C, 15 s at 55 to 60°C, 4 min at 60°C) for 30 cycles without a final in the combinations indicated at the beginning of the culture period or at extension at 72°C using T3 and T7 primers and the following AMAC-1- the time points indicated. In some experiments, monocytes were also cul- specific primers: REV; REV1; REV3; REV4 (5Ј-ACC AAC TTG TGC tured in T75 tissue culture plastic flasks (Falcon). Incubation was at 7.5% ACA GGA GCA GAG GGC CAT G; 112–142); FORW (5Ј-TTA TCA CO2. Before harvest, the plastic bags were put on ice for at least 30 min and TCC TTT CCC CTT TCC CTT CAA CTC; 584–613); FORW3 (5Ј-AGG were lightly hit with a stick for a while to get the lightly adherent cells back GCA ATG GCC CTG CCA CCC TGG AG; 406–431); FORW4 (5Ј-CAT into suspension. The bags were cut open, the supernatant was collected and GGC CCT CTG CTC CTG TGC ACA AGT TGG T; 112–142); FORW5 the cells were washed in PBS and either used directly for coculture, frozen (5Ј-AGA GGA GTT GTG AGT TTC CAA GCC CCA GC; 9–37) (see Fig. as pellets for RNA isolation, or used for flow cytometric analysis or to 1). The sequencing reaction was separated on 6% polyacrylamide/urea prepare cytospin preparations in a cytocentrifuge at 700 rpm for 3 min sequencing gels. using 104 cells/cytospin. Monocytes cultured in plastic flasks were har- vested using trypsin-EDTA and were otherwise treated similarly. Northern hybridization Mediators Multiple tissue Northern blots of normal organs including lymphoid tissues were purchased from Clontech. For macrophages and some other cell lines Human IFN-␥, used at 1000 U/ml, was from Sigma (Deisenhofen, Ger- and tissues, total RNA was isolated using RNA Midi-Isolation kit (Qia- many); human rTNF-␣, used at 20 ng/ml, and human recombinant GM- gen). Ten micrograms of total RNA were electrophoresed in formaldehyde/ CSF, used at 200 U/ml, were from Tebu Peprotech (Frankfurt/M., Germa- agarose gels and blotted onto nylon membrane Gen screen plus (DuPont). ny); human rIL-2, used at 100 U/ml, was from Chiron (Ratingen, After2hat80°C, filters were hybridized with an AMAC-1 specific 200-bp Germany); human rIL-3, used at 10 ng/ml, human rIL-6, used at 200 U/ml, PCR-generated DNA probe (primers used were FORW and REV; see Fig. human rIL-12, used at 100 U/ml, human rIL-13, used at 50 ng/ml, and 1). The probe was labeled using a ready-to-go random priming kit (Phar- human rIL-1␣, used at 100 U/ml, were from R&D Systems (Wiesbaden, macia) with 50 ␮Ci [␣Ϫ32P]dCTP with a specific activity of 6000 ␮Ci/ The Journal of Immunology 1413 mmol (Amersham Buchler), and nonincorporated radioactivity was sepa- ing all this sequence information together, a contig for a sequence rated using G50 Sephadex-filled Nick columns (Pharmacia). The filters of 803 bp was constructed. Finally, a full length clone (␭ZAP were prehybridized for at least2hin1MNaCl, 1% SDS, and 10% dextran Express clone 202) was identified and fully sequenced from both sulfate at 60°C. Denatured probe and 100 ␮g/ml salmon sperm DNA were then added and hybridized at 60°C overnight. After hybridization, the fil- ends and using several internal primers. The resulting sequence ters were washed twice in 2ϫ SSC, 1% SDS at room temperature, twice in fully confirmed the previously established contig sequence (Fig. 1; 2ϫ SSC, 1% SDS at 60°C for 30 min, and the final wash was in 0.1 SSC accession no. Y13710). at room temperature. Filters were dried, put into SaranWrap, and exposed The isolated clone showed an open reading frame of 267 bp to X-AR film. coding for 89 amino acid residues. An analysis for the presence of RT-PCR a signal peptide sequence gave three possible cleavage points after Total RNA was isolated using Trizol (Life Technologies), and first-strand residues 18, 20, or 21 with the highest probability for cleavage cDNA was synthesized using commercially available kits (Pharmacia; or between residues 20 and 21 resulting in an N terminus for the Invitrogen, Leek) according to the manufacturers’ protocols. Several mature protein with the sequenceMkglaaallvlvctmal primer pairs were tested. Of these, putative intron-spanning primer pair csc͉ A Q V (Fig. 1). Further sequence analysis pointed toward a FORW5 and REV (see Fig. 1) turned out to give the most specific and possible CC-chemokine of 69 amino acid residues and a predicted reliable results. Primers were tested for exclusion of amplification of genomic sequences using 200 ng and 1 ␮g genomic DNA from PBMC (not m.w. of 7855. shown). PCR was performed in a 50-␮l total reaction volume containing 25 By scanning the sequence against a nonredundant database set to 250 ng cDNA templates, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 on the internet, the highest homology to our clone was found with mM MgCl2, 0.2 mM dNTP, 1.25 U of Amplitaq DNA polymerase (Perkin- the human and mouse DNA sequences of the CC-chemokine ¨ ␮ Downloaded from Elmer Cetus, Uberlingen, Germany), and oligonucleotide primers (0.2 M ␣ each). The amplification profile for the primer pair was: 95°C, 45 s; 60°C, MIP-1 (55% and 51%, respectively). Several expressed sequence 45 s; 72°C, 2 min for 40 cycles. Amplification was terminated by 10 min tags (ESTs) and other cDNA clones (accession nos. R83915, of final extension at 72°C. Twenty-microliter aliquots of the resulting PCR T89961, AA031820, AA031821, N44551, N33793, and G907797) reaction mixture were separated on ethidium bromide-stained 1.5% agarose with very high homology to our sequence were also found; how- gels (Life Technologies) by electrophoresis. Gels were photographed under ever, our sequence is the longest reported to date. When using only UV light (Polaroid 665 film, Sigma). Specific signals were adjusted ac- cording to expression of housekeeping genes ␤-actin or glyceraldehyde-3- the coding region of our cDNA sequence from nucleotide 71 to phosphate dehydrogenase. 337, a perfect match of the protein sequence to protein sequence 3 http://www.jimmunol.org/ from U.S. patent No. 5504003 (accession no. I19356) was found. Bioinformatics This protein sequence was tentatively named macrophage inflam- The DNA sequence and the deduced protein sequence of the clone were matory protein (MIP)-4 (31). However, we prefer to use the term compared with public databases using the internet programs BLASTN and alternative macrophage activation-associated CC-chemokine BLASTP at http://www.ncbi.nlm.nih.gov to find similar sequences. The database set was a nonredundant combination of GenBank, EMBL, DDBJ, (AMAC)-1 for our clone since we feel that it better reflects the PDB, dBEST, and SWISSPROT sequences (27). expression pattern of our gene that is opposite to other macrophage Sequences of known cytokines were obtained from the ENTREZ inflammatory proteins, especially, MIP-1␣. server at National Centre for Biotechnology Information at http://www.

In order to identify the relationship of AMAC-1 to other known by guest on September 29, 2021 ncbi.nlm.nih.gov/Web/Search/index.html. The sequences were then aligned using the ClustalW program (28) with standard parameters, and chemokines, an alignment of related sequences obtained from a the resulting data set was imported into the GeneDoc4 program. The BLAST search was made with ClustalW and processed in Tree- phylogenetic trees were drawn with Treeview5 using the guide tree file view and Genedoc. The resultant dendrogram of protein sequences of the ClustalW program. Modeling of the three-dimensional structure of known CC-chemokines (Fig. 2A) shows that the strongest rela- of AMAC-1 was possible using the Swissmodel server at ExPasy tionship of AMAC-1 is to human and mouse MIP-1␣ (59% and (http://www.expasy.ch/swissmod/SWISS-MODEL.html.) (29, 30) based on the known structures of the following sequence entries from 51% homology, respectively). The next closest similarity of the Brookhaven database: 11HUM.pdb (HS_MIP-1␤), 11DOK.pdb, AMAC-1 is to human and mouse MIP-1␤ (39% and 45% homol- 11MCA.pdb, 11DOM.pdb, 11DOL.pdb, 13IL8.pdb (HS_IL8), ogy, respectively). Together with MIP-1␣, MIP-1␤, RANTES, and 11IKL.pdb, and 11HRJ.pdb. The obtained coordinates for AMAC-1 macrophage-derived chemokine (32), AMAC-1 constitutes one were visualized with the Rasmol program (Roger Sayle, Glaxo Well- come Research and Development, Stevenage, Hertfordshire, U.K.). group of CC-chemokines that is clearly separated from two other groups in the dendrogram; one of these two latter groups com- Results and Discussion prises human HCC-1 (33), MIP-3/MPIF-1 (34), MIP-3␣, MIP-3␤ (35), MIP-5/HCC-2 (accession no. R91733) and murine C10 (36), By comparing the molecular repertoire of IFN-␥-induced, classi- and MIP-1␥ (37) while the other comprises human eotaxin and cally activated macrophages and of IL-4- and GC-induced, alter- MCP-1, -2, -3, and -4 (38). Between these groups, overall homol- natively activated macrophages using differential hybridization, ogy is in the 25% range. An optimized alignment of these chemo- several clones were found to be specifically expressed in alterna- kines together with the human CXC-chemokine IL-8 (Fig. 2B) tively activated macrophages. The greater part of these clones reveals six absolutely conserved positions in both CC- and CXC- showed positive cross-hybridization reactions using Southern blot- chemokines comprising the four cysteines, Val-58, and Leu-65 ting, and a high frequency of cross-hybridizing clones was de- (numbered according to mature AMAC-1 protein and marked in tected in our cDNA library made from alternatively activated mac- blue). Within the mature proteins, there are 14 residues more or rophage RNA, but not in a spleen cDNA library. Several of these less well conserved among the CC chemokines (green). These in- cross-hybridizing clones were sequenced from their 5Ј and 3Ј ends clude regions of putative importance for dimerization and structure and from various internal primers and showed partially overlap- determination (i.e., WV motive at position 57, 58) (39, 40). ping sequences. In order to identify the extreme 5Ј end of the Due to the high homology of AMAC-1 to human MIP-1␤,itwas respective mRNA, 5Ј RACE products were also sequenced. Tak- possible to generate a model of the putative three-dimensional structure of AMAC-1 using the Swissmodel server at ExPasy. The 4 Nicholas, K. B., and H. B. Nicholas, Jr. 1997. GeneDoc: a tool for editing and resulting structure file was visualized with Rasmol (Fig. 2C). The anotating multiple sequence alignments. Distributed by the author. structure-determining residues as well as the four cysteines are 5 Page, R. D. M. 1996. TreeView: tree drawing software for Apple Macintosh and Microsoft Windows. Distributed by the author at: http://taxonomy.zoology.gla.ac.uk/ well conserved in their position in the three-dimensional structure rod/rod.htm. between AMAC-1 and MIP-1␤ resulting in conservation of all 1414 AMAC-1-CC-CHEMOKINE FROM ALTERNATIVELY ACTIVATED MACROPHAGES Downloaded from http://www.jimmunol.org/

FIGURE 1. Nucleotide and deduced amino acid sequence of AMAC-1. The 5Ј nontranslated region consists of 70 bp, the coding region of 267 bp, and the 3Ј untranslated region of 466 bp. The positions and orientations of the PCR primers used are indicated by arrows. The predicted site for a signal peptidase by guest on September 29, 2021 cleavage is marked by an asterisk. The sequence has been deposited in Genbank/EMBL/DDBJ under the accession no. Y13710.

decisive structural features between these two chemokines. These or IFN-␥, and, interestingly, MIP-1␣ with its high homology to ␤ ␤ features are the 310-helical turn, the three antiparallel -sheets 1, AMAC-1 is frankly inhibited by alternative macrophage activators ␤2, and ␤3 as well as the C-terminal ␣-helix. such as IL-4, IL-13, and GC (42–44). Classically activated mac- With respect to receptor selectivity of chemokines, a two-site rophages, , endothelial cells, melanoma cells, and nor- paradigm has been postulated (41). Site I serves initial complex mal human organs do not express AMAC-1 (Fig. 3A). However, in formation (address) while site II is important for agonist activity of some specimens, especially endothelial cells and melanoma cells, the ligand (message). The region of AMAC-1 known from MIP-1␤ weak cross-hybridization was seen with an as yet unidentified to be directly or indirectly involved in interaction at site I is cen- mRNA species of 1350 bp. RT-PCR with carefully selected puta- tered around Tyr27 (including also Asp26, Glu29, and Ser32) and tive intron-spanning AMAC-1-specific primers revealed only the is highly conserved in most CC-chemokines including AMAC-1 expected single band in alternatively activated macrophages (Fig. (40). In CXC-chemokines such as IL-8, Tyr27 is uniformly re- 3B). Due to the high sensitivity of RT-PCR, weak AMAC-1 ex- placed by leucine. This receptor recognition site is located within pression was also detected in some unstimulated control macro- ␤ ␤ -sheet 1 opposite to the dimerization region preceding the 310- phage specimens and very weakly in spleen. helical turn. Alteration of this sequence motif by site-directed mu- Expression of AMAC-1 is induced in alternatively activated tagenesis influences receptor specificity in case of IL-8 (41). In macrophages by IL-4, while GC are not able to induce AMAC-1 contrast, the flexible N terminus of AMAC-1 involved in agonist on their own, but may enhance the effects of IL-4 (Fig. 3A). In activity at site II is considerably different from all other CC-che- contrast to macrophage migration inhibitory factor (45), GC do not mokines and rather resembles the N terminus of CXC-chemokines induce AMAC-1 even if they are used at lower concentrations than such as IL-8 including an incomplete ELR motif and a leucine/ usual (10Ϫ8 to 10Ϫ16 M) (Fig. 3C). In contrast to other chemokines isoleucine at position 12 vs a tyrosine/phenylalanine in other CC- such as MIP-1␣ (44) and to other proinflammatory cytokines such chemokines (AMAC-1: TNKEL-C-CL; IL-8: SAKELRCQCI; as IL-1 or TNF-␣, AMAC-1 expression is not inhibited by GC. MIP-1␣: DTPTA-C-CF). In time course experiments, AMAC-1 expression by alterna- Expression of AMAC-1 was studied by Northern analysis in a tively activated macrophages was shown to start only at day 2 and variety of cultured cells and tissues and was found highly specific to be fully developed at day 3 when IL-4 was present throughout for alternatively activated macrophages (Fig. 3A). In contrast, most the culture period (Fig. 3D). When IL-4 was added only at day 2 chemokines including MIP-1␣ expressed by macrophages are in- of culture, expression of AMAC-1 was nevertheless fully devel- duced by classical macrophage activators such as LPS (not shown) oped at day 3 (Fig. 3E) indicating that monocyte-to-macrophage The Journal of Immunology 1415 Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 2. A, Phylogram of CC-chemokines. AMAC-1 constitutes a separate arm in a group comprising MIP-1␣, MIP-1␤, and RANTES. B, Alignment of deduced amino acid sequence of AMAC-1 with other CC-chemokines and the CXC-chemokine IL-8. Amino acids conserved among all CC- and CXC-chemokines are indicated in blue, among part of CC- and CXC-chemokines in green and yellow. C, Schematic representation of modeled 3D structure of AMAC-1 with emphasis on putative secondary structure elements. Amino acids unique to AMAC-1 among CC-chemokines and involved in putative receptor-ligand interactions are depicted in red. 1416 AMAC-1-CC-CHEMOKINE FROM ALTERNATIVELY ACTIVATED MACROPHAGES Downloaded from http://www.jimmunol.org/

FIGURE 4. Northern blot analysis of AMAC-1 expression. A, Induc- FIGURE 3. Northern blot (A, C, D) and RT-PCR (B) analysis of ibility of AMAC-1 expression by cytokines. Monocytes/macrophages cul- ␥ AMAC-1 expression. A, Monocytes/macrophages cultured for 3 (1–5) and tured for 3 days without any mediators (1), or in the presence of IFN- (2), ␣ 6 (6–10) days without any mediators (1, 6), or in the presence of IFN-␥ (2, dexamethasone and IL-4 (3), IL-4 (4), IL-1 (5), IL-2 (6), IL-3 (7), IL-6 ␣ 7), dexamethasone and IL-4 (3, 8), dexamethasone (4, 9), or IL-4 (5, 10); (8), IL-10 (9), IL-12 (10), IL-13 (11), TNF- (12), macrophage-CSF (13), spleen (11), granulocytes (12), melanoma cell line MEWO (13), human GM-CSF (14), GM-CSF and IL-4 (15). B, Inhibition of AMAC-1 expres- ␥ umbilical vein endothelial cells (14), PBL (15) and PBMC (16). B, Mono- sion by IFN- . Monocytes/macrophages cultured for 3 days without me- ␥ cytes/macrophages cultured for 3 (1–3) and 6 (4–6) days without any diators (1, 5–7), or in the presence of IFN- (2), dexamethasone and IL-4 by guest on September 29, 2021 mediators (1, 4), or in the presence of IFN-␥ (2, 5), or dexamethasone and (3, 8–10), and IL-4 (4, 11–13); some monocyte/macrophage cultures were ␥ IL-4 (3, 6); phorbol ester-treated, IL-4-stimulated THP-1 cells (7), spleen additionally treated with IFN- at 1000 U/ml (5, 8, 11), 100 U/ml (6, 9, (8), granulocytes (9), human umbilical vein endothelial cells (10), PBMC 12), and 10 U/ml (7, 10, 13) during the whole culture period. C, Expression (11), monocytes (12), PBL (13), melanoma cell line MEWO (14), and of AMAC-1 in dendritic cells. Nonadherent (1–4) and adherent (8–11) fibroblasts (15). C, Monocytes/macrophages cultured for 3 days without monocytes/macrophages cultured for 6 days without any mediators (1, 8), ␥ any mediators (1), or in the presence of IFN-␥ (2), dexamethasone and IL-4 or in the presence of IFN- (2, 9), dexamethasone and IL-4 (3, 10), or (3), dexamethasone (4), IL-4 (5), or of reduced concentrations of dexa- dexamethasone and IL-4 (45 ng/ml) (4, 11). Nonadherent (5–7) and ad- methasone (10Ϫ8 M (6), 10Ϫ10 M (7), 10Ϫ12 M (8), 10Ϫ14 M (9), 10Ϫ16 M herent (12–14) monocyte-derived dendritic cells cultured for 6 (5, 12) and (10)). D, Monocytes/macrophages cultured for 16 h (1), 24 h (2), 48 h (3), 8 (6, 13) days with GM-CSF and IL-4 [45 ng/ml]; some monocyte-derived ␣ and 72 h (4) in the presence of dexamethasone and IL-4. E, Monocytes/ dendritic cell cultures were additionally treated with TNF- from day 6 to ␣ macrophages cultured for 3 days without any mediators (1), or in the pres- day 8 (7, 14). D, Effect of TNF- on AMAC-1 expression. Monocytes/ ence of IFN-␥ (2), dexamethasone and IL-4 (3), dexamethasone (4), IL-4 macrophages cultured for 3 days without any mediators (1, 5), or in the ␥ (5); monocytes/macrophages cultured for 2 days without any mediators, presence of IFN- (2, 6), dexamethasone and IL-4 (3, 7), and IL-4 (4, 8); ␣ and for an additional 24 h with dexamethasone (6), IL-4 (7), or LPS (8). some monocyte/macrophage cultures were additionally treated with TNF- during the whole culture period (5–8). E, Expression of AMAC-1 in al- veolar macrophages. Asthmatics (1, 2), nonsmokers (3–5), and a smoker (6). differentiation is a prerequisite of AMAC-1 expression. This is also reflected by the finding that expression of AMAC-1 by the monocytic leukemia cell line THP-1 requires preceding induction stimulated macrophages (Fig. 4D). This latter result is unexpected, of macrophage differentiation by phorbol esters (Fig. 3B). Simi- since expression of alternatively activated macrophage Ags MS- larly, even monocytes of atopic patients with high serum IgE levels 1-HMWP and RM 3/1 is inhibited by both IFN-␥ and TNF-␣ (9, do not express AMAC-1 (not shown). Besides IL-4, other medi- 18). ators of alternative immunologic macrophage activation such as Since recent research has revealed a closer relationship of IL-13 and IL-10 are able to induce expression of AMAC-1 while monocytes/macrophages and dendritic cells than hitherto recog- all other cytokines tested such as IFN-␥, IL-1, IL-2, IL-3, IL-6, nized (17, 46) and since dendritic cells can be derived from mono- IL-12, macrophage-CSF, GM-CSF, and TNF-␣ are not (Fig. 4A). cytes by cytokine stimulation including IL-4 (26), we analyzed Besides lacking correspondence in the homology-based CC-che- AMAC-1 expression in monocyte-derived dendritic cells in vitro. mokine dendrogram, these findings confirm that C10 is not the GM-CSF- and IL-4-induced monocyte-derived dendritic cells murine homologue of human AMAC-1, since C10 is inducible by showed expression of AMAC-1 at day 3 comparable in strength to IL-3 and GM-CSF (36) in addition to IL-4. On the other hand, alternatively activated macrophages (Fig. 4A), while expression of IFN-␥ strongly inhibits expression of AMAC-1 (Fig. 4B), while AMAC-1 at days 6 and 8 was less intense and was further down- TNF-␣ does rather enhance AMAC-1 expression in IL-4- and GC- regulated by addition of TNF-␣ (Fig. 4C). The Journal of Immunology 1417

In addition to cultured human cells, a variety of inflamed human References tissues was examined for expression of AMAC-1 including acute 1. Stein, M., S. Keshav, N. Harris, and S. Gordon. 1992. -4 potently appendicitis, Crohn’s disease, ulcerative colitis, lesional skin from enhances murine macrophage mannose receptor activity: a marker of alternative psoriasis, and bronchoalveolar lavage cells from various donors. immunologic macrophage activation. J. Exp. Med. 176:287. Only alveolar macrophages from asthmatics, smokers and healthy 2. Becker, S., and E. G. Daniel. 1990. Antagonistic effects of IL-4 and - gamma on human monocytes and macrophages: effects on Fc receptors, HLA-D persons, but not inflammatory skin and bowel disease specimens antigens, and superoxide production. Cell. Immunol. 129:351. strongly expressed AMAC-1 in vivo (Fig. 4E). 3. Te Velde, A. A., R. J. F. Huijbens, J. E. de Vries, and C. G. Figdor. 1990. IL-4 During preparation of this manuscript, Adema et al. (47) re- decreases Fc␥R membrane expression and Fc␥R-mediated cytotoxic activity of human monocytes. J. Immunol. 144:3046. ported on the identification and analysis of a novel CC-chemokine, 4. Cowan, H. B., S. Vick, J. T. Conary, and V. L. Shepherd. 1992. Dexamethasone DC-CK1, which was cloned from a dendritic cell-derived cDNA upregulates mannose receptor activity by increasing mRNA levels. Arch. Bio- library. Based on a limited expression analysis, Adema et al. claim chem. Biophys. 296:314. 5. Conrad, D. J., H. Kuhn, M. Mulkins, E. Highland, and E. Sigal. 1992. Specific that DC-CK1 is a dendritic cell-selective chemokine in vitro and in inflammatory cytokines regulate the expression of human monocyte 15-lipoxy- vivo and they show that it functions in selectively attracting and genase. Proc. Natl. Acad. Sci. USA 89:217. activating naive T cells. Summarizing, Adema et al. hold that DC- 6. Hart, P. H., G. F. Vitti, D. R. Burgess, G. A. Whitty, D. S. Piccoli, and CK1 is part of the potent immunostimulatory armament of den- J. A. Hamilton. 1989. Potential antiinflammatory effects of interleukin-4: sup- pression of human monocyte -a, interleukin-1, and prosta- dritic cells during the development of naive T cells into Th1 ef- glandin E2. Proc. Natl. Acad. Sci. USA 86:3803. fector cells. 7. Hart, P. H., G. A. Whitty, D. R. Burgess, M. Croatto, and J. A. Hamilton. 1990. Interestingly, DC-CK1 and AMAC-1 show sequence identity. Augmentation of glucocorticoid action on human monocytes by interleukin-4. Lymphokine Res. 9:147. Downloaded from Insofar, the data presented in this paper help to broaden and correct 8. Cheung, D. L., P. H. Hart, G. F. Vitti, G. A. Whitty, and J. A. Hamilton. 1990. the oversimplified view of DC-CK1/AMAC-1 put forward by Contrasting effects of interferon-␥ and interleukin-4 on the interleukin-6 activity Adema et al. (47). In contrast to Adema et al., we have shown here of stimulated human monocytes. Immunology 71:70. 9. Goerdt, S., R. Bhardwaj, and C. Sorg. 1993. Inducible expression of MS-1 high that AMAC-1 is preferentially expressed by IL-4-induced alterna- molecular weight protein by endothelial cells of continuous origin and by den- tively activated macrophages; admittedly, AMAC-1 is also ex- dritic cells/macrophages in vivo and in vitro. Am. J. Pathol. 142:1409. pressed by GM-CSF/IL-4-induced monocyte-derived dendritic 10. Landmann, R., A. E. Fisscher, and J.-P. Obrecht. 1992. Interferon-gamma and

interleukin-4 down-regulate soluble CD14 release in human monocytes and mac- http://www.jimmunol.org/ cells, but the level of expression is much lower and decreases with rophages. J. Leukocyte Biol. 52:323. dendritic cell maturation. The common denominator for expression 11. Delespesse, G., and M. Sarfati. 1991. An update on human CD23 (Fc⑀RII). of AMAC-1 in alternatively activated macrophages and monocyte- Fc⑀RII and IgE-BFs (soluble CD23) play an essential role in the regulation of derived dendritic cells obviously is its inducibility by IL-4. This human IgE synthesis. Clin. Exp. Allergy 21(Suppl. 1):153. 12. Goerdt, S., L. J. Walsh, G. F. Murphy, and J. S. Pober. 1991. Identification of a notion is substantiated by our finding that AMAC-1 is also induc- novel high molecular weight protein antigen preferentially expressed by sinusoi- ible by other Th2 cytokines such as IL-10 and IL-13 and that its dal endothelial cells in normal human tissues. J. Cell Biol. 113:1425. 13. Walsh, L. J., S. Goerdt, J. S. Pober, H. Sueki, and G. F. Murphy. 1991. MS-1 expression is inhibited by the Th1 cytokine IFN-␥. Whether den- ϩ ϩ sinusoidal endothelial antigen is expressed by factor XIIIa , HLA-DR dermal dritic cells such as Langerhans cells that have not seen IL-4 and do perivascular dendritic cells. Lab. Invest. 65:732. 14. Goerdt, S., G. Kolde, G. Bonsmann, K. Hamann, B. Czarnetzki, R. Andreesen, not express the IL-4-inducible macrophage mannose receptor (48) by guest on September 29, 2021 will express AMAC-1 remains to be determined. In further con- T. Luger, and C. Sorg. 1993. Immunohistochemical comparison of cutaneous histiocytoses and related skin disorders: diagnostic and histogenetic relevance of trast to Adema et al. (47), we have found that AMAC-1 expression MS-1 high molecular weight protein expression. J. Pathol. 170:421. in vivo is strong in peripheral organs such as lung (alveolar mac- 15. Zwadlo, G., R. Voegeli, K. Schulze-Osthoff, and C. Sorg. 1987. A monoclonal rophages) and placenta (not shown) and is therefore not restricted antibody to a novel differentiation antigen on human macrophages associated with the down-regulatory phase of the inflammatory process. Exp. Cell Biol. to some sparsely distributed putative dendritic cells in lymphatic 55:295. organs. 16. Wenzel, I., J. Roth, and C. Sorg. 1996. Identification of a novel surface molecule, In summary, AMAC-1 is a novel CC-chemokine with several RM 3/1, that contributes to the adhesion of glucocorticoid-induced human mono- cytes to endothelial cells. Eur. J. Immunol. 26:2758. unconventional features. In contrast to most other known CC-che- 17. Goerdt, S., V. Kodelja, M. Schmuth, C. E. Orfanos, and C. Sorg. 1996. The mokines, AMAC-1 expression is specifically induced in alterna- mononuclear phagocyte-dendritic cell dichotomy: myths, facts, and a revised tively activated macrophages by Th2-associated cytokines in vitro concept. Clin. Exp. Immunol. 105:1. 18. Kodelja, V., and S. Goerdt. 1994. Dissection of macrophage differentiation path- and is naturally present in alveolar macrophages in vivo. Interest- ways in cutaneous macrophage disorders and in vitro. Exp. Dermatol. 3:257. ingly, both alternatively activated and alveolar macrophages are 19. Topoll, H., G. Zwadlo, D. E. Lange, and C. Sorg. 1989. Phenotypic dynamics of known to be involved in suppression of Th1-associated immune macrophage subpopulations during human experimental gingivitis. reactions (24, 49–52). To a lesser extent, AMAC-1 is also ex- J. Periodontal Res. 24:106. 20. Szekanecz, Z., G. K. Haines, T. R. Lin, L. A. Harlow, S. Goerdt, G. Rayan, and pressed in monocyte-derived dendritic cells. As Adema et al. have A. E. Koch. 1994. Differential distribution of intercellular adhesion molecules shown (47), AMAC-1 may be functional in preferentially attract- (ICAM-1, ICAM-2, and ICAM-3) and the MS-1 antigen in normal and diseased human synovia: their possible pathogenetic and clinical significance in rheuma- ing and activating naive T cells. Unexpectedly, double-negative toid arthritis. Arthritis Rheum. 37:221. knockout mice lacking the CC 1, which is the 21. Djemadji-Oudjiel, N., S. Goerdt, V. Kodelja, M. Schmuth, and C. E. Orfanos. major receptor for MIP-1␣, experience a shift of the Th1/Th2 bal- 1996. Immunohistochemical identification of type II alternatively activated den- ϩϩϩ ϩ/Ϫ Ϫ ance toward Th1 predominance (53) indicating that MIP-1␣ sup- dritic macrophages (RM 3/1 , MS-1 , 25F9 ) in psoriatic dermis. Arch. Dermatol. Res. 288:757. ports Th2 reactions. Due to the fact, however, that both Th2-as- 22. Cooper, A. M., D. K. Dalton, T. A. Stewart, J. P. Griffin, D. G. Russel, and sociated, preferentially anti-inflammatory alternatively activated I. M. Orme. 1993. Disseminated tuberculosis in interferon-␥ gene-disrupted mice. macrophages and Th1-associated, potent immunostimulatory den- J. Exp. Med. 178:2243. 23. Kodelja, V., C. Mu¨ller, S. Tenorio, C. Schebesch, C. E. Orfanos, and S. Goerdt. dritic cells express AMAC-1, AMAC-1 may not be directly in- 1997. Differences in angiogenic potential of classically vs alternatively activated volved in determining Th1 or Th2 skewing during T cell devel- macrophages. Immunobiology 197:478. opment. Full clarification of this open question must await 24. Schebesch, C., V. Kodelja, C. Mu¨ller, N. Hakij, C. E. Orfanos, and S. Goerdt. 1997. Alternatively activated macrophages do not co-stimulate, but actively in- identification of the AMAC-1 receptor. Special structural features hibit proliferation of peripheral blood lymphocytes and CD4ϩ T-cells in vitro. of AMAC-1 at sites influencing receptor specificity suggest un- Immunology. In press. usual CC-chemokine receptor-ligand interactions which might 25. Goerdt, S., F. Steckel, K. Schulze-Osthoff, H.-H. Hagemeier, E. Macher, and C. Sorg. 1989. Characterization and differential expression of an endothelial cell- comprise unknown or orphan chemokine receptors (54, 55) or even specific surface antigen in continuous and sinusoidal endothelia, in skin vascular chemokine receptor antagonism. lesions and in vitro. Exp. Cell Biol. 57:185. 1418 AMAC-1-CC-CHEMOKINE FROM ALTERNATIVELY ACTIVATED MACROPHAGES

26. Sallusto, F., and A. Lanzavecchia. 1994. Efficient presentation of soluble antigen 42. Standiford, T. J., S. L. Kunkel, J. M. Liebler, M. D. Burdick, A. R. Gilbert, and by cultured human dendritic cells is maintained by granulocyte/macrophage col- R. M. Strieter. 1993. Gene expression of macrophage inflammatory protein-1a ony-stimulating factor plus and downregulated by tumor necrosis from human blood monocytes and alveolar macrophages is inhibited by inter- factor-␣. J. Exp. Med. 179:1109. leukin-4. Am. J. Respir. Cell Mol. Biol. 9:192. 27. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic 43. Berkman, N., M. John, G. Roesems, P. Jose, P. J. Barnes, and K. F. Chung. 1996. local alignment search tools. J. Mol. Biol. 215:403. inhibits macrophage inflammatory protein-1a production from hu- 28. Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improv- man alveolar macrophages. Am. J. Respir. Cell Mol. Biol. 15:382. ing the sensitivity of progressive multiple sequence alignment through sequence 44. Berkman, N., P. J. Jose, T. J. Williams, T. J. Schall, P. J. Barnes, and weighting, position-specific gap penalties and weight matrix choice. Nucleic Ac- K. F. Chung. 1995. Corticosteroid inhibition of macrophage inflammatory pro- ids Res. 22:4673. tein-1␣ in human monocytes and alveolar macrophages. Am. J. Physiol. 29. Peitsch, M. C. 1995. Protein modelling by email. Bio/Technology 13:658. 269:L443. 30. Peitsch, M. C. 1996. Promod and Swiss-Model: internet-based tools for auto- 45. Calandra, T., J. Bernhagen, C. N. Metz, L. A. Spiegel, M. Bacher, T. Donnelly, mated comparative protein modelling. Biochem. Soc. Trans. 24:274. A. Cerami, and R. Bucala. 1995. MIF as a glucocorticoid-induced modulator of 31. Wells, T. N. C., and M. C. Peitsch. 1997. The chemokine information source: cytokine production. Nature 376:68. identification and characterization of novel chemokines using the WorldWide- 46. Peters, J. H., R. Gieseler, B. Thiele, and F. Steinbach. 1996. Dendritic cells: from Web and expressed sequence tag databases. J. Leukocyte Biol. 61:545. ontogenetic orphans to myelomonocytic descendents. Immunol. Today 17:273. 32. Godiska, R., D. Chantry, C. J. Raport, S. Sozzani, P. Allavena, D. Leviten, 47. Adema, G. J., F. Hartgers, R. Verstraten, E. de Vries, G. Marland, S. Menon, A. Mantovani, and P. W. Gray. 1997. Human macrophage-derived chemokine J. Foster, Y. Xu, P. Nooyen, T. McClanahan, K. B. Bacon, and C. G. Figdor. (MDC), a novel chemoattractant for monocytes, monocyte-derived dendritic 1997. A dendritic-cell-derived C-C chemokine that preferentially attracts naive T cells, and natural killer cells. J. Exp. Med. 185:1595. cells. Nature 387:713. 33. Schulz-Knappe, P., H.-J. Ma¨gert, B. Dewald, M. Meyer, Y. Cetin, M. Kubbies, 48. Mommaas, A. M., A. Tan, A. A. Mulder, R. Jordens, J. J. M. Onderwater, J. Tomeczkowski, K. Kirchhoff, M. Raida, K. Adermann, A. Kist, M. Reinecke, J. W. Drijfhout, D. Verwoerd, T. H. M. Ottenhof, P. H. Kluin, A. Tulp, R. Sillard, A. Pardigol, M. Uguccioni, M. Baggiolini, and W.-G. Forssmann. J. J. Neefjes, and F. Koning. 1997. In contrast to human blood derived dendritic 1996. HCC-1, a novel chemokine from human plasma. J. Exp. Med. 183:295. cells, epidermal Langerhans cells do not employ the mannose receptor endocy-

34. Patel, V. P., B. L. Kreider, Y. Li, H. Li, K. Leung, T. Salcedo, B. Nardelli, Downloaded from tosis pathway. J. Invest. Dermatol. 109:261. V. Pippalla, S. Gentz, R. Thotakura, D. Parmelee, R. Gentz, and G. Garotta. 1997. 49. Poulter, L. W., G. Janossy, C. Power, S. Sreenan, and C. Burke. 1994. Immu- Molecular and functional characterization of two novel human C-C chemokines nological/physiological relationships in asthma: potential regulation by lung mac- as inhibitors of two distinct classes of myeloid progenitors. J. Exp. Med. rophages. Immunol. Today 15:258. 185:1163. 35. Rossi, D. L., A. P. Vicari, K. Franz-Bacon, T. K. McClanahan, and A. Zlotnik. 50. Fireman, E. M., S. B. Efraim, J. Greif, S. Kivity, and M. R. Topilsky. 1988. 1997. Identification through bioinformatics of two new macrophage proinflam- Suppressor cell activity of human alveolar macrophages in interstitial lung dis- matory human chemokines, MIP-3a and MIP-3b. J. Immunol. 158:1033. eases. Clin. Exp. Immunol. 73:111. 36. Orlofsky, A., E. Y. Lin, and M. B. Prystowsky. 1994. Selective induction of the 51. Geertsma, M. F., W. L. Teeuw, P. H. Nibbering, and R. van Furth. 1994. Pul- b chemokine C10 by IL-4 in mouse macrophages. J. Immunol. 152:5084. monary surfactant inhibits activation of human monocytes by recombinant inter- http://www.jimmunol.org/ ␥ 37. Poltorak, A. N., F. Bazzoni, I. I. Smirnova, E. Alejos, P. Thompson, G. Luheshi, feron- . Immunology 82:450. and N. Rothwell. 1995. MIP-1gamma: molecular cloning, expression, and bio- 52. Holt, P. G., M. A. Schon-Hegard, and J. Oliver. 1988. MHC class II antigen- logical activities of a novel CC chemokine that is constitutively secreted in vivo. bearing dendritic cells in pulmonary tissues of the rat: regulation of antigen pre- J. Inflamm. 45:207. sentation activity by endogenous macrophage populations. J. Exp. Med. 167:262. 38. Uguccioni, M., P. Loetscher, U. Forssmann, B. Dewald, H. Li, S. Hensche Lima, 53. Gao, J.-L., T. A. Wynn, Y. Chang, E. J. Lee, H. E. Broxmeyer, S. Cooper, Y. Li, B. Kreider, G. Garotta, M. Thelen, and M. Baggiolini. 1996. Monocyte H. L. Tiffany, H. Westphal, J. Kwon-Chung, and P. M. Murphy. 1997. Impaired chemotactic protein 4 (MCP-4), a novel structural and functional analogue of host defense, hematopoiesis, granulomatous inflammation and type 1-type 2 cy- MCP-3 and eotaxin. J. Exp. Med. 183:2379. tokine balance in mice lacking CC chemokine receptor 1. J. Exp. Med. 185:1959. 39. Chung, C.-w., R. M. Cooke, A. E. I. Proudfoot, and T. N. C. Wells. 1995. The 54. Liao, F., G. Alkhatib, K. W. C. Peden, G. Sharma, E. A. Berger, and J. M. Farber. three-dimensional structure of RANTES. Biochemistry 34:9307. 1997. STRL33, a novel chemokine receptor-like protein, functions as a fusion 40. Clore, G. M., and A. M. Gronenborn. 1997. Three-dimensional structures of a and cofactor for both macrophage-tropic and T cell line-tropic HIV-1. J. Exp. Med. by guest on September 29, 2021 b chemokines. FASEB J. 9:57. 185:2015. 41. Wells, T. N. C., C. A. Power, M. Lusti-Narasimhan, A. J. Hoogewerf, 55. Raport, C. J., V. L. Schweickart, D. Chantry, R. L. J. Eddy, T. Shows, R. M. Cooke, C.-w. Chung, M. C. Peitsch, and A. E. I. Proudfoot. 1996. Selec- R. Godiska, and P. W. Gray. 1996. New members of the chemokine receptor gene tivity and antagonism of chemokine receptors. J. Leukocyte Biol. 59:53. family. J. Leukocyte Biol. 59:18.