USOO5766866A United States Patent (19) 11 Patent Number: 5,766,866 Center et al. 45) Date of Patent: Jun. 16, 1998

54 CHEMOATTRACTANT Center, et al. The Journal of Immunology. 128:2563-2568 FACTOR AND USES THEREOF (1982). 75 Inventors: David M. Center. Wellesley Hills; Cruikshank, et al. The Journal of Immunology, William W. Cruikshank. Westford; 138:3817-3823 (1987). Hardy Kornfeld, Brighton, all of Mass. Cruikshank, et al., The Journal of Immunology, 73) Assignee: Research Corporation Technologies, 146:2928-2934 (1991). Inc., Tucson, Ariz. Cruikshank, et al., EMBL Database. Accession No. M90301 (21) Appl. No.: 580,680 (1992). 22 Filed: Dec. 29, 1995 Cruikshank. et al. The Journal of Immunology, 128:2569-2574 (1982). Related U.S. Application Data Rand, et al., J. Exp. Med., 173:1521-1528 (1991). 60 Division of Ser. No. 480,156, Jun. 7, 1995, which is a continuation-in-part of Ser. No. 354,961, Dec. 13, 1994, Harlow (1988) Antibodies, a laboratory manual. Cold Spring which is a continuation of Ser. No. 68,949, May 21, 1993, Harbor Laboratory, 285, 287. abandoned. Waldman (1991) Science. vol. 252, 1657-1662. (51) Int. Cl...... G01N 33/53; CO7K 1700; CO7K 16/00: A61K 45/05 Hams et al. (1993) TIBTECH Feb. 1993 vol. 111, 42-44. 52 U.S. Cl...... 424/7.24; 435/7.1; 435/7.92: 435/975; 530/350:530/351; 530/387.1: Center, et al. (Feb. 1995) "The Lymphocyte Chemoattractant 530/388.23: 530/389.2: 424/85.1: 424/130.1 Factor”. J. Lab. Clin. Med. 125(2):167-172. 58) Field of Search ...... 435/7.24, 7.1. Cruikshank, et al. (May 24, 1994) "Molecular and Func 435/7.92.975; 530/350, 351,387.1388.23. tional Analysis of a Lymphocyte Chemoattractant Factor: 389.2: 424/130.1. 85.1 Association of Biologic Function with CD4 Expression", 56) References Cited Proc. Natl. Acad. Sci. USA 91(11):5109-5113. U.S. PATENT DOCUMENTS (May 1996) "Terminology Note: 16 (IL-16)". Eur: J. Inmunol. 26(5)1196. 4486,530 2A1984 David et al. . 4.745,051 5/1988 Smith et al. . FOREIGN PATENT DOCUMENTS Primary Examiner-Anthony C. Caputa Assistant Examiner-Khalid Masood WO 94/28134 12/1994 WIPO Attorney; Agent, or Firm-Scully. Scott, Murphy & Presser OTHER PUBLICATIONS 57 ABSTRACT Berman, et al., Cellular Immunology, 95:105-112 (1985). Berman, et al., AM Rev. Respir. Dis... 142:238-257 (1990). Disclosed is a substantially pure antibody which specifically Cambell, "Monoclonal Antibody Technology, The Produc binds a LCF polypeptide and methods of using such anti tion and Characterization of Rodent and Human Hybrido bodies. mas." Elsevier Science Publishers (Amsterdam), pp. 1-4 and 29 (1984). 5 Claims, 19 Drawing Sheets U.S. Patent Jun. 16, 1998 Sheet 1 of 19 5,766,866

28S

18S

FIG. U.S. Patent Jun. 16, 1998 Sheet 2 of 19 5,766,866

1. TTCCTCGAGAGCTGTCAACACAGGCTGAGGAATCTCAAGGCCCAGTGCTCAAGATGCCT 6O AGCCAGCGAGCACGGAGCTTCCCCCTGACCAGGTCCCAGTCCTGTGAGACGAAGCTACT 19 TGACGAAAAGACCAGCAAACTCTATCTATCACCAGCCAGTGTCATCGGCTGTCATGAA 178 ATCCTTGCTGGCCTTCCATCTTCTATCTCCTGTGCCCAGACTCCCTGCATCCCCAAGG 237 CAGGGGCATCTCCAACATCATCATCCAACGAAGACTCAGCTGCAAATGGTTCTGCTGAA 296 ACATCTGCCTTGGACACGGGGTTCTCGCTCAACC TTCAGAGCTGAGAGAATATACAGA 355 GGGTCTCACGGAAGCCAAGGAAGACGATGATGGGGACCACAGTTCCTTCAGTCTGGTCA 41.4 GTCCGT TATCTCCCTGCTGAGCTCAGAAGAATTAAAAAAACTCATCGAGGAGGTGAAGG 473 TTCTGGATGAAGCAACATAAAGCAATTAGACGGCATCCAGTCACCATCTTACACAAG 532 GAGGAAGGTCGTGGTCTGGGTTCAGCTTGGCAGGAGGAGCAGATCTAGAAAACAAGGT 59. GATTACGG CACAGAGTGTCCAAATGGGCTGGCCTCCCAGGAAGGGACTATTCAGA 6SO AGGGCAATGAGGTTCTTCCATCAACGGCAAGTCTCCAAGGGGACCACGCACCATGAT 709 GCCTTGGCCATCCTCCGCCAAGCTCGAGAGCCCAGGCAAGCTGTGATTGTCACAAGGAA 768 GCTGACTCCAGAGCC ATG CCC GAC CTC AAC TCC TCC ACT GAC CT GCA Met Pro Asp Leu Asn Ser Ser Thr Asp Ser Ala 816 GCC TCA GCC TCT GCA GCC AGT GAT GTT TCT GTA GAA TCT ACA GCA 12 Ala Ser Ala Ser Ala Ala Ser Asp val Ser wall Glu Ser Thr Ala 861 GAG GCC ACA GTC TGC ACG GTG ACA CTG GAG AAG ATG TCG GCA GGG 27 Gu Ala Thr Val Cys Thr Val Thr Leu Gu Lys Met Ser Ala Gly 906 CTG GGC TTC AGC CTG GAA GGA GGG AAG GGC. TCC CTA CAC GGA GAC 42 Leu Gly Phe Ser Leu Glu Gly Gly Lys Gly Ser Leu His Gly Asp 951 AAG CCT CTC ACC ATT. AAC AGG ATT TTC AAA GGA GCA GCC TCA GAA 57 lys Pro Leu Thir Ile Asn Arg Ile Phe Lys Gly Ala Ala Ser Glu 996 CAA AGT GAG ACA GTC CAG CCT GGA GAT GAA ATC TTG CAG CTG GGT 72 Gln Ser Gu Thr Val Gln Pro Gly Asp Glu Ile Leu Gln Leu Gly 1041 GGC ACT GCC ATG CAG GGC CTC ACA CGG TTG GAA GCC TGG AAC ATC 87 Gly Thr Ala Met Gln Gly Leu Thr Arg Phe Glu Ala Trp Asn Ile 1086 ATC AAG GCA CTG CCT GAT GGA CCT GTC ACG ATT GTC ATC AGG AGA 102 Ile lys Ala Leu Pro Asp Gly Pro Val Thr Ile vali Ile Arg Arg 1131 AAA AGC CTC CAG TCC AAG GAA ACC ACA GCT GCT GGA GAC TCC TAG 117 Lys Ser Leu Gln Ser Lys Glu Thr Thr Ala Ala Gly Asp Ser - 1176 GCAGGACATGCTGAAGCCAAAGCCAATAACACACAGCTAACACACAGCTCCCATAACC 1235 CGATTCTCAGGGTCTCTGCTGCCGCCCCACCCAGATGGGGGAAAGCACAGGTGGGCTT 1294 CCCAGTGGCTGCTGCCCAGGCCCAGACCTTCTAGGACGCCACCCAGCAAAAGGTTGTTC 1353 CTAAAATAAGGGCAGAGTCACACTGGGGCAGCTGATACAAATTGCAGACTGTGTAAAAA 1412 GAGAGCTTAATGATAATATTGTGGTGCCACAAATAAAATGGATTTATTAGAATTCCATA 1471. TGACA TCATGCCTGGCTTCGCAAAATGT CAAG ACTGAACTGTGTCATGATTCAC 1530 CCCCAAACAGTGACATTTATTTTTCTCATGAATCTGCAATGTGGGCAGAGATTGGAATG 1589 GGCAGCTCATCTCTGTCCCACTTGGCATCAGCTGGCGTCATGCAAAGTCATGCAAAGGC 1648 TGGGACCACCTGAGATCATTCACTCATACATCTGGCCGTTGATGTTGGCTGGGAACTCA 1707 CCTGGGGCTGCTGGCCTGAATGCTTATAGGTGGCCTCTCCTTGTTGCCTGGGCTCCTCA 1755 CAACATGGTGTCTGGATTCCCAGGATGAGCATCCCAGGATCGCAAGAGCCACGTAGAAG 1825 CTGCATCTTGT TATACCTTTGCCTTGGAAGTTGCATGGCATCACCTCCACCATACTCC 1884 ATCAGTTAGAGCTGACACAAACCTGCCTGGGTTTAAGGGGAGAGGAAATATTGCTGGGG 1943 TCATT TATGAAAAATACAGTTTGTCACATGAAACATTTGCAAAATTGTTTTTGGTTGGA 2002 TTGGAGAAGTAATCCTAGGGAAGGGTGGTGGAGCCAGTAAATAGAGGAGTACAGTGTAA 2061 GCACCAAGCTCAAAGCGTGGACAGGTGTGCCGACAGAAGGAACCAGCGTGTATATGAGG 2120 GTATCAAATAAAATTGCTACTACTTACCACC

FG.2 U.S. Patent Jun. 16, 1998 Sheet 3 of 19 5,766,866

FG3A U.S. Patent Jun. 16, 1998 Sheet 4 of 19 5,766,866

FIG.3B U.S. Patent Jun. 16, 1998 Sheet 5 of 19 5,766,866

FG4 U.S. Patent Jun. 16, 1998 Sheet 6 of 19 5,766,866

N

N N N

HNNN Nip

O O : N 3 Migration (%. Control) U.S. Patent Jun. 16, 1998 Sheet 7 of 19 5,766,866

-K

o

* -

o i :

ck

S. 8 8 s & 8 8 Cell Migration (/. Control) U.S. Patent Jun. 16, 1998 Sheet 8 of 19 5,766,866

Control

an r O O t Nu CD O Os 24 hr O CD h O 2 - O CD 48 hr

Green Fluorescence (IL-2R)

FG.7 U.S. Patent Jun. 16, 1998 Sheet 9. Of 19 5,766,866

5OOOOO 2OO

4OOOOO 8O s E O 3OOOOO 16O

O 5 2OOOOO l4O s

O IOOOOO 2O

OO 3O

Froction (minutes) FIG.8A U.S. Patent Jun. 16, 1998 Sheet 10 of 19 5,766,866

kDa A B 31

U.S. Patent Jun. 16, 1998 Sheet 11 of 19 5,766,866 sap??dad

O

O£ O6O8O/O9OGOt?O2O|

La––––------~------—] O 2"|-- U.S. Patent Jun. 16, 1998 Sheet 12 of 19 5,766,866

Ceil Migration (%. Control)

OO

8O

4O N N 2 N 3 N 4 N 5 N 6 N 7 N 8 N 9

BAL Somples From Normal Subjects

FIGO U.S. Patent Jun. 16, 1998 Sheet 13 of 19 5,766,866

Cell Migration (%. Control) 2OO

8O

6O

4 O

2O

OO

8O

L 2 3 4 5 L 6 L 7 BAL Somples Post Cho lenge

FG U.S. Patent Jun. 16, 1998 Sheet 14 of 19 5,766,866 Cell Migrotion (A Control ) 22O

2OO

8O

- b -2 b -3b L4b L5b L6 b L7 a BAL. Somples Post Antigen-Challenge FIG2O Cell Migration (%. Control ) 22O

2OO

8O

4 O

2O

OO L2 L3 b L 4b LS b L. 6 L 7 BAL Somples Post Antigen - Challenge FG.2b U.S. Patent Jun. 16, 1998 Sheet 15 of 19 5,766,866

Cel. Migration (%. Contro) 22O

2OO

8 O

6 O

4 O k

r 2 O wis A

ps s st a OO 3 E h L b 2 | 3 b - 4 b 5 b l S | L 7 O BAL. Somples Post Antigen - Cho lenge

FIG 3 U.S. Patent Jun. 16, 1998 Sheet 16 of 19 5,766,866

:

.

Total cells ( millions) U.S. Patent Jun. 16, 1998 Sheet 17 of 19 5,766,866

s i CX s s |

i

Tolo cells (millions) U.S. Patent Jun. 16, 1998 Sheet 18 of 19 5,766,866

4uno0ºC]O OO9 O62 Og O9 O9

Toto number cells (milions) U.S. Patent Jun. 16, 1998 Sheet 19 of 19 5,766,866

7O O.

6OO T

SOO

G.E 4oo T 9. S.N 3OO I

2OO T

OO ahme 1.

O O 2 3 4 5 6 7 8 9 O

Week

FIG 7 5,766.866 1 2 LYMPHOCYTE CHEMOATTRACTANT capable of promoting or stimulating the migration of unac FACTOR AND USES THEREOF tivated or activated CD4“, , , and the like. By "polypeptide" is meant any This is a divisional of application Ser. No. 08/480.156. chain of amino acids, regardless of length or post filed on Jun. 7, 1995; which is a continuation-in-part of U.S. 5 translational modification (e.g. glycosylation). By a "sub Ser. No. 08/354,961, filed Dec. 13, 1994; which is a con stantially identical" amino acid sequence is meant an amino tinuation of U.S. Ser. No. 08/068.949, filed May 21, 1993 acid sequence which differs only by conservative amino acid now abandoned. substitutions. for example, substitution of one amino acid for This work was supported in part by a grant from the another of the same class (e.g., valine for glycine, arginine O for lysine and the like) or by one or more non-conservative Federal Government and the Government therefore has amino acid substitutions. deletions, or insertions located at certain rights in the invention. positions of the amino acid sequence which do not destroy BACKGROUND OF THE INVENTION the biological activity of the polypeptide. Such equivalent polypeptides can be isolated by extraction from tissues or This invention relates to recombinant LCF, DNA, and cells of any animal which naturally produce such a polypep uses thereof. 15 tide or which can be induced to do so. using the methods CD4, a cell-cell adhesion , is expressed on a subset described below, or their equivalent; or can be isolated by of T lymphocytes (Krensky et al., Proc. Natl. Acad. Sci. chemical synthesis; or can be isolated by standard tech U.S.A. 79:2365-2369, 1982: Biddison et al., J. Exp. Med. niques of recombinant DNA technology, e.g. by isolation of 156:1065-1076, 1982; and Wilde et al., J. Immunol. 2 cDNA or genomic DNA encoding such a polypeptide. 131:152-157, 1983), mononuclear cells (Stewart et al., J. In another aspect, the invention features a fragment or Immunol. 136:3773-3778, 1986), and eosinophils (Rand et analog of LCF which exhibits LCF agonist or antagonist al., J. Ep. Med. 173:1521-1528, 1991). In lymphocytes, activity. The invention thus includes any biologically active CD4 contributes to antigen signaling (Collins et al., fragment or analog of LCF polypeptide. By “biologically J. Immunol. 148:2159-2162, 1992; Anderson et al., J. Immu 25 active" is meant possessing any activity which is character nol. 139-678-682, 1987: Eichmann et al., J. Inmunol. istic of the 130-amino acid LCF polypeptide shown in FIG. 17:643-650, 1987; Walker et al., Eur. J. Immunol. 2 (SEQ ID NO: 1). Because LCF polypeptide exhibits a 17:873-880 1987; and Sleckman et al., Nature range of physiological properties and because such proper 328:351–353, 1987) by direct interaction with MHC Class II ties may be attributable to different portions of the LCF molecules (Doyle et al., Nature 330:256–259, 1987). In 30 polypeptide molecule, a useful LCF polypeptide fragment or addition, a natural soluble , lymphocyte LCF polypeptide analog is one which exhibits a biological chemoattractant factor (LCF), requires cell surface expres activity in any biological assay for LCF polypeptide activity, sion of CD4 to induce chemotactic activity in monocytes for example, those assays described herein. Most preferably (Cruikshank et al., J. Ep. Med. 173:1521-1528, 1991) and it possesses 10%, preferably 40%, or at least 90% of the T lymphocytes (Cruikshank et al., J. Immunol. 35 activity of LCF polypeptide (shown in FIG. 2; SEQD NO: 138:3817-3823, 1987: Cruikshank et al., J. Immunol. 1). in any LCF polypeptide assay. 146:2928-2934, 1991). In concert with its chemoattractant Preferred analogs include LCF polypeptide (or biologi activity LCF acts as a competence for human cally active fragments thereof) whose sequences differ from T lymphocytes (Cruikshank et al., J. Immunol. the wild-type sequence only by conservative amino acid 138:3817-3823, 1987). substitutions, for example, substitution of one amino acid for LCF is a cationic, 56-kD glycoprotein representing the another with similar characteristics (e.g., valine for glycine. tetrameric form of four 14-kD monomeric chains. LCF is arginine for lysine, and the like) or by one or more non produced by T lymphocytes and is specifically chemoattrac conservative amino acid substitutions, deletions, or inser tant for CD4+ T-cells, monocytes and eosinophills (see, e.g. tions which do not abolish the polypeptide's biological Berman et al. Cell Immunol. 95:105-112, 1985; Rand et al., 45 activity, Other useful modifications include those which JEM 173:1521-1528, 1991). Secretion of LCF by CD8+ T increase peptide stability; such analogs may contain, for cells occurs (Cruikshank et al., J. Inmunol. 138:3817, example, one or more non-peptide bonds (which replace the 1987) after stimulation by mitogen, antigen, histamine or peptide bonds) or D-amino acids in the peptide sequence. serotonin. The latter two are of particular interest because Analogs can differ from naturally occurring LCF polypep degranulated mast cells and basophils are present in tissue 50 tide in amino acid sequence or can be modified in ways that sites of delayed-type hypersensitivity reactions (see, e.g., do not involve sequence, or both. Analogs of the invention Askenase Prog. Allergy 23:199-320, 1977). Induction of will generally exhibit at least 70%. more preferably 80%, LCF by a mast cell or a basophil product provides a link more preferably 90%, and most preferably 95% or even between the early mediator phase of the immune response 99%, homology with a segment of 20 amino acid residues, and the development of the later T-lymphocyte-predominant 55 preferably more than 40 amino acid residues, or more inflammatory reaction. preferably the entire sequence of a naturally occurring LCF polypeptide sequence. SUMMARY OF THE INVENTION Alterations in primary sequence include genetic variants, In general, the invention features recombinant lympho both natural and induced. Also included are analogs that cyte chemoattractant factor (LCF) polypeptide, e.g., LCF include residues other than naturally occurring L-amino produced in a prokaryotic or baculovirus expression system. acids, e.g., D-amino acids or non-naturally occurring or Preferably, the polypeptide includes an amino acid sequence synthetic amino acids, e.g., Bor Yamino acids. Alternatively, substantially identical to the amino acid sequence shown in increased stability may be conferred by cyclizing the peptide FIG. 2 (SEQ ID NO: 1). By “lymphocyte chemoattractant molecule. Modifications include in vivo or in vitro chemical factor polypeptide" is meant all or part of a protein which 65 derivatization of polypeptides, e.g., acetylation. specifically binds CD4 and signals the appropriate LCF methylation, phosphorylation, phremy lation. mediated cascade of biological events, e.g., a polypeptide isupremylation. myristilation. carboxylation, or glycosyla 5,766,866 3 4 tion; glycosylation can be modified, e.g., by modifying the translation of the sequence (i.e., facilitates the production of. glycosylation patterns of a polypeptide during its synthesis e.g. LCF, or fragment or analog thereof). and processing or in further processing steps, e.g. by In still another aspect, the invention features a substan exposing the polypeptide to glycosylation affecting enzymes tially pure antibody which binds preferentially to a LCF (or derived from cells that normally provide such processing. a fragment or analog thereof). By "substantially pure anti e.g., mammalian glycosylation enzymes; phosphorylation body" is meant antibody which is at least 60%, by weight, can be modified by exposing the polypeptide to free from the and naturally-occurring organic mol phosphorylation-altering enzymes, e.g. kinases or ecules with which it is naturally associated. Preferably, the phosphatases, etc. By "substantially pure" is meant that the preparation is at least 75%, more preferably at least 90%. LCF polypeptide provided by the invention is at least 60%. 10 and more preferably at least 99%, by weight, antibody, e.g., by weight, free from the proteins and naturally-occurring LCF antibody. A substantially pure LCF antibody may be organic molecules with which it is naturally associated. obtained, for example, by affinity chromatography using Preferably, the preparation is at least 75%, more preferably recombinantly-produced LCF polypeptide and standard at least 90%, and most preferably at least 99%, by weight, techniques. Furthermore, the purified antibody is sufficiently LCF polypeptide. A substantially pure LCF polypeptide may 15 free of other proteins, carbohydrates, and lipids with which be obtained, for example, by extraction from a natural it is naturally associated to permit therapeutic administra source (e.g. a human peripheral blood mononuclear cell) tion. Such an antibody "preferentially binds" to an LCF using the methods outlined below; or can be isolated by polypeptide (or a fragment or analog thereof), i.e. does not expression of a recombinant nucleic acid encoding a LCF substantially recognize and bind to other antigenically polypeptide using the standard techniques of recombinant DNA technology, e.g. by isolation of cDNA or genomic unrelated molecules. DNA encoding such an LCF polypeptide, or by chemically Preferably, the antibody neutralizes the biological activity synthesizing the protein, fragment or analog thereof. Purity of the protein to which it binds. By "neutralize" is meant to can be measured by any appropriate method, e.g., column partially or completely block (e.g., the biological activity of chromatography, polyacrylamide gel electrophoresis, or a LCF polypeptide). high-performance liquid chromatography (HPLC) analysis. 25 In other aspects, the polypeptides or antibodies described above are used as the active ingredient of therapeutic In another aspect, the invention features substantially pure compositions. In such therapeutic compositions, the active DNA encoding a LCF polypeptide (or polypeptide fragment ingredient is formulated with a physiologically-acceptable or analog thereof) as described above. Preferably, the DNA carrier. These therapeutic compositions are used in a method comprises a nucleotide sequence substantially identical to 30 the nucleotide sequence shown in FIG. 2 (SEQ ID NO: 2). of suppressing or mimicking LCF-CD4 interaction mediated Moreover, such a DNA is cDNA and encodes a mammalian physiological response. In particular, these methods are used LCF polypeptide, e.g. a human. The invention also features to reduce an immune response, or inflammation, or growth a vector which includes such substantially pure DNA and of an unwanted cell. Compounds useful in practicing the which is capable of directing expression of the protein 35 method include, without limitation, an LCF antibody, or an encoded by the DNA in a vector-containing cell. The inven LCF fragment or analog, or a drug, e.g. an organic com tion features a cell which contains the substantially pure pound. DNA. The cell may be either prokaryotic, e.g., E. coli or In another aspect, the invention features an LCF immu eukaryotic, e.g., a mammalian cell or the cell of an noassay kit including an antibody of the invention. arthropod, e.g., a grasshopper. Preferably, such a kit includes a means for detecting the By "substantially pure DNA" is meant DNA that is free of binding of the antibody to the LCF polypeptide. the which, in the naturally-occurring genome of the In another aspect, the invention features a method of organism from which the DNA of the invention is derived. detecting LCF in a biological sample, the method involving flank the . The term therefore includes, for example, a (a) contacting the biological sample with an antibody of the recombinant DNA which is incorporated into a vector; into 45 invention; and (b) detecting immune complex formation an autonomously replicating plasmid or virus; or into the between the antibody and a sample constituent as indicative genomic DNA of a prokaryote or eukaryote; or which exists of the presence of LCF in the sample. Preferably, the method as a separate molecule (e.g. a cDNA or a genomic or cDNA involves an immune complex formation which is detected fragment produced by polymerase chain reaction-(PCR) by an ELISA or a Western blot analysis. methodologies or restriction endonuclease digestion) inde 50 In yet another aspect, the invention features a method of pendent of other sequences. It also includes a recombinant screening candidate compounds for their ability to inhibit DNA which is part of a hybrid gene encoding additional interaction between LCF and CD4. The method involves: a) polypeptide sequence. mixing a candidate antagonist compound with LCF; b) In another aspect, the invention features a method of measuring LCF-CD4 binding; and c) identifying antagonis producing a recombinant LCF polypeptide (or a fragment or 55 tic compounds as those that interfere with the binding. analog thereof). The method involves (a) providing a cell In still another aspect, the invention features a method of (e.g., E. coli or S. frugidera transformed with DNA encoding screening candidate compounds for the ability to mimick a LCF polypeptide or a fragment or analog thereof posi LCF activity, the method involving: a) mixing a candidate tioned for expression in the cell; (b) culturing the trans agonist compound with CD4 receptor; b) measuring binding formed cell under conditions for expressing the DNA; and 60 of the compound to CD4 receptor; and c) identifying agonist (c) isolating the recombinant LCF polypeptide. By "trans compounds as those that bind CD4 receptor and mediate cell formed cell" is meant a cell into which (or into an ancestor migration. of which) has been introduced, by means of recombinant In another aspect, the invention features a composition for techniques, a DNA molecule encoding (as used herein) an stimulating proliferation of CD4+ T-cells in a mammal, the LCF polypeptide. Such a DNA molecule is "positioned for 65 composition including LCF and a growth factor. In preferred expression" meaning that the DNA molecule is positioned embodiments, the composition includes LCF and a growth adjacent to a DNA sequence which directs transcription and factor in a ratio which causes synergy, e.g., ranging from 5,766,866 5 6 1:100 to 1:1 (LCF to growth factor). Preferably, the growth methods of the present invention include, without limitation, factor is a e.g., IL-2. IL-4, IL-6, IL-7. IL-8, insulin, any granulomatous immune reaction, e.g., as effected by and insulin-like growth factor I. tissue-invasive helminth parasites, cutaneous and respira The invention also features a method for stimulating tory late-phase reactions to allergens. asthma. Sarcoidosis, proliferation of CD4+ T cells in a mammal. the method hypersensitivity pneumonitis, interstitial pulmonary fibrosis, includes contacting cells with LCF and IL-2 together or tuberculosis, rheumatoid arthritis, and lupus erythematous, close enough in time to cause synergy. In preferred allogenic organ transplant rejection, contact (cell-mediated) embodiments, the method includes administering to a mam dermatitis,and immunologically mediated skin diseases (e.g. mal (e.g., a human patient) an effective amount of LCF and pemphigoid and bullous pemphigoid). A comprehensive text a growth factor, wherein the proliferative activity of LCF in 10 on the aforementioned disorders may be found in Principles combination with the growth factor is greater than the of Internal Medicine 12th ed. (Wilson et al., McGraw Hill, proliferative activity of the LCF in the absence of the growth Inc., New York). Preferred therapeutics include antagonists, factor and the proliferative activity of the growth factor e.g., peptide fragments, or antibodies, or drugs, which block in-the absence of LCF. In preferred embodiments the growth LCF or LCF:CD4 receptor function by interfering with the factor is a cytokine and, if desired, the administration of the 15 LCF:CD4 receptor interaction and any concomitant biologi composition occurs more than once. cal activity directed by LCF. Similarly, the antibodies of the In other preferred embodiments, the method for stimulat invention are useful for detecting the presence and clinical ing proliferation of CD4+ T cells involves (a) contacting course of any disease associated with LCF, e.g., those cells with LCF and IL-2 in vitro and returning the prolifer disseases described above. ated cells into the mammal. Preferably, the stimulated CD4+ Recombinant LCF can also be used as an immunosup is a PBMC or a HIV+PBMC. In other preferred pressive agent or as part of immunosuppressive therapy. In embodiments, the method further involves contacting the particular, recombinant LCF may serve to attenuate. cells with an anti-retroviral agent (e.g., AZT or dd). interrupt, or prevent the cascade of events that eventually In another aspect, the invention features a method for result in immunological rejection of tissue or organ trans stimulating proliferation of CD4+ T cells in a human 25 plants. For example, recombinant LCF may be used to infected with HTV. involving administering an effective attenuate, interrupt, or prevent a patient from rejecting a therapeutic amount of a composition including LCF and a kidney, lung, or combined heart-lung, or liver transplants. growth factor. In preferred embodiments, the infected Further, recombinant LCF by virtue of its ability to interact human is an asymptomatic human infected with HTV. In still and bind with CD4 receptors may be useful in the design of other preferred embodiments, the human infected with HIV 30 immunotoxins that selectively destroy CD4+ receptor bear has a CD4+ count greater than 50. ing cells. Finally, recombinant LCF may be used, alone or in In another aspect, the invention features a method for combination with other compounds (e.g., growth factors), to stimulating proliferation of CD4+ in a human having an activate and replenish a CD4 lymphocyte population in any immune disorder, the method involving administering an patient with a depleted population. effective therapeutic amount of a composition including 35 Because LCF may now be produced by recombinant LCF and a growth factor. techniques, and because candidate antagonists or agonists In another aspect, the invention features a method for may be screened according to the assays described herein, inducing the proliferation of CD4+ T cells in a human, the the instant invention provides a simple and rapid approach method involving administering an effective therapeutic to the identification of useful therapeutics. Such an approach was previously difficult because LCF was unavailable in amount of a composition including LCF. sufficient quantities to identify its role in disease in animal In still another aspect, the invention features a method of models, and antibodies and DNA and RNA probes were inhibiting a CD4+ bearing malignant cell in a mammal. previously unavailable for detection of LCF protein or gene involving administering to the mammal (e.g. a human expression in diseased tissues. patient), a therapeutically effective amount of an LCF anat S gonist (as described herein). In preferred embodiments, the Thus, once identified, a peptide- or antibody-based thera CD4+ T cell is a lymphoma or is a leukemia. Preferably, the peutic may be produced, in large quantity and inexpensively, antagonist or inhibitor is a LCF fragment or analog thereof using recombinant and molecular biological techniques, and the methods of the present invention. Finally, any chemical or is an anti-LCF antibody. In other preferred embodiments, compound, e.g., an organic compound, may be easily the method further involves administering to the mammal a SO chemotherapeutic agent in an effective dose which is lower screened according to the methods outlined herein in order than the standard dose when the chemotherapeutic agent is to evaluate their effect on LCF:CD4 interaction. used alone. Other features and advantages of the invention will be apparent from the following description of the preferred In another aspect, the invention features a method of embodiments thereof, and from the claims. protecting a mammal from developing a neoplasm, involv 55 ing administering to the mammal (e.g., a human patient) a DETALED DESCRIPTION OF THE therapeutically effective amount of an LCF antagonist. PREFERRED EMBODIMENTS DRAWINGS The proteins of the invention are involved in events leading to inducing the migration of specialized immune The drawings will first be described. cells, e.g., eosinophils, monocytes, and T lymphocytes, FIG. 1 shows a northern analysis of LCF from total which are important constituents and mediators of both the cellular RNA prepared from human T lymphocytes. Posi immune response and inflammation. Such proteins are there tions of 18S and 28S RNA visualized by ethidium bromide fore useful to treat or, alternatively, to develop therapeutics staining are shown at their respective arrows. to treat hyperresponsive immune reactions and inflammation FIG. 2 shows the nucleotide sequence of the LCF-A that pertain to the activation and subsequent infiltration of T 65 cDNA (SEQID NO: 2) and predicted amino acid sequence lymphocytes, monocytes and eosinophils. Particular disor of the encoded protein (SEQ ID NO: 1). Nucleotides are ders which may be treated using the proteins and/or the numbered on the left side beginning with the first nucleotide 5,766.866 7 8 of the cDNA. The poly A tail begins immediately after the level of 3% (top panel) to 17% (bottom panel) by 48 h. The last indicated nucleotide (252) and is omitted. Translation 24h time point demonstrated an increase in 9% of the cells. of the putative LCF coding sequence is indicated below the At no time did CD4 - cells show an increase in IL-2R corresponding nucleotide sequence starting with Met. Each expression. This is a representative FACs analysis of three amino acid is consecutively numbered. An Asn residue different experiments. Other experiments demonstrated (amino acid residue 5) represents a potential glycosylation increases in IL-2R-cells at the 48 h time point in 15% and site (marked with a dot). Two candidate polyadenylation 19% of the cells. signal sequences are underlined. FIG. 8A and FIG. 8B show the aggregation of recombi FG. 3A and FIG. 3B show a SDS-PAGE of recombinant nant LCF under physiological conditions. FIG. 8A shows a LCF expressed in E. coli and a rabbit reticulocyte in vitro O molecular sieve HPLC of S-labelled recombinant LCF translation of RNA synthesized from LCF cDNA. FIG. 3A (run in phosphate buffered saline, pH 8.0). Fractions were shows recombinant LCF protein run on a 15% SDS-PAGE collected and analyzed by scintillation counting (open followed by coomassie blue staining. In FIG. 3A. lane A squares). Parallel samples were collected and assayed for the shows crude supernatant from E. coli induced to express induction of lymphocyte (solid squares). FIG. LCF protein. lane B shows LCF protein generated as a 15 8B, lane A and lane B show an autoradiogram of the peak fusion protein conjugated to a polyhistidine linker purified fraction for both radioactivity and cell migration (fraction 13 by nickel affinity chromatography, and lane C shows LCF shown in FIG. 8A) and the second peak of radioactivity after Factor Xa cleavage. The band at 17.5 kDa was blotted, which had no corresponding chemoattractant activity excised and subjected to N-terminal amino acid sequencing. (fraction 17 shown in FIG. 8A) after separation by SDS FIG. 3B shows a rabbit reticulocyte in vitro translation of PAGE, respectively. LCF cDNA: the S-labeled protein product of LCF cDNA FIG. 9 shows a hydrophilicity plot of recombinant LCF translated by rabbit reticulocytes was run on a 15% SDS predicted by the method of Kyte and Doolittle (Kyte et al. PAGE. In FIG, 3B, lane A shows LCF protein translated J. Molec. Bio. 157:105-132 (1982)). Peptides were synthe under non-glycosylating conditions, and lane B shows LCF sized and rabbit anti-peptide specific anti-sera were gener translated under glycosylating conditions. 25 ated to four major hydrophilic regions designated by A.B. FIG. 4 shows the immunoprecipitation of recombinant C.D. LCF by rsCD4. In FIG. 4, lane 1 shows 10 g of recombi FIG. 10 shows induced chemotaxis of human T lympho nant LCF; lane 2 shows recombinant LCF incubated with 50 cytes by concentrated BAL fluid from normal individuals. grsCD4 immunoprecipitated with 10 ug rabbit polyclonal Fifty milliliters of BAL fluid was concentrated fold and then anti-CD4 antibody; lane 3 shows recombinant LCF incu 30 assayed diluted 1:1 in phosphate buffered saline in a micro bated with 10 ug rsCD4 immunoprecipitated with poly chemotaxis chamber. The data is expressed as a percent of clonal anti-CD4 antibody; lane 4 shows recombinant LCF random cell migration in the presence of PBS alone incubated with rsCD4 (10 ug) immunoprecipitated with (normalized to 100% in all experiments, for these experi rabbit polyclonal anti-IgG (10 g); lane 5, shows recombi ments control migration averaged 14.3 cells/hpf). Each BAL nant LCF incubated with rsCD4 and immunoprecipitated 35 fluid was assayed three times, with the asterisks denoting with monoclonal anti-CD4 (10 g); lane 6, shows recombi migration statistically different from control cell migration nant LCF incubated with rsCD4 and immunoprecipitated (p<0.05). with monoclonal anti-CD8 antibody (10 ug); and lane 7. FIG. 11 shows the induced migration of peripheral T cells shows rsCD4 (10 ug) incubated with monoclonal anti-CD8 40 by concentrated BAL fluids from asthmatics following antibody. either saline (solid bars) or specific antigen (hatched bars) FIG. 5 shows a dose response curve for recombinant LCF challenge. The BAL fluids were obtained 6 hrs after chal induced chemotaxis of human peripheral blood T lympho lenge and concentrated 100 fold prior to assaying. Each BAL cytes. In FIG. 5, an asterisk (*) represents statistical signifi fluid was assayed three times with the asterisks denoting cell cance at p<0.05 (using a Student's T test from control cell migration which was statistically different from control cell migration). 45 migration (p<0.05). For these experiments control migration FIG. 6 shows recombinant LCF-induced chemotaxis in averaged 12.5 cells/hpf. murine T cell hybridoma cells. Murine cell lines expressing FIG. 12 shows the blocking effect of a panel of anti either wild-type CD4 (13.13), truncated CD4 (delta-13), or cytokine antibodies on the induction of peripheral T cell mock infected cells lacking CD4 expression (155.16) were 50 migration by BAL fluids. Positive BAL samples, as deter stimulated by recombinant LCF (10M) (open bars) or mined in FIG. 11, were reassessed for the induction of T cell 2C11 antibody (10 pg/ml) (striped horizontal bars) and the migration (as shown in panel a) either alone (solid bars), in migratory response quantitated. Cells stimulated by recom the presence of anti-LCF polyclonal antibody (shaded bars), binant LCF in the presence of a fold excess of anti-CD4 Fab or with anti-MIPlot polyclonal antibody (left hatched bars), fragments (10 ug/ml) are also shown (solid bars). Cell 55 Panel (b) shows BALs alone (solid bars), with anti-DL-8 migration is expressed as mean of ten high power fields polyclonal antibody (stippled bars), or anti-RANTES mono +/-S.D. Migration which was significantly different (p<0.05 clonal antibody (horizontal bars). All antibodies were used by Student's T test) from control cell migration (designated at a concentration sufficient to neutralize bioactivity from 50 as 100%) is indicated by asterisks. ng/ml of protein. The experiment was conducted three FIG. 7 shows the specificity of recombinant LCF for different times and the asterisks denotes cell migration CD4+ human T cells using FACs analysis. Twox10' human statistically different from cell migration induced by the Tlymphocytes were cultured for 24 and 48 h in the presence same BAL sample assayed alone (p<0.05). Control migra of 10M recombinant LCF, Cells double-labelled with tion in these experiments averaged 15 cells/hpf. phycoerythrin-conjugated anti-CD4 antibody and fluoresce FIG. 13 shows the blocking of BAL fluid-induced T cell inconjugated anti-IL-2R antibody were analyzed on a Bec 65 migration by anti-LCF. anti-MIP1c or a combination of the ton Dickinson FACscan flow cytometer. Recombinant LCF two antibodies. Induction of cell migation was assessed for induced an increase in CD4+/IL-2R+ cells from a control BAL samples incubated either alone (solid bars), in the 5,766,866 9 10 presence of anti-MIP1o antibody (left hatched bars), in the assays described herein); such analogs are also considered to presence of anti-LCF antibody (shaded bars), or in a com be useful in the invention. bination of the two antibodies (horizontal bars). Antibodies There now follows a description of the cloning and were used at concentrations sufficient to neutralize bioac characterization of a human LCF cDNA useful in the instant tivity from 50 ng/ml of specific protein. The data is invention. This example is provided for the purpose of expressed as percent of control cell migration, with asterisks illustrating the invention, and should not be construed as denoting inhibition of migration which was statistically limiting. different from BAL-induced cell migration in the absence of Isolation of Human LCF cDNA blocking antibodies (p<0.05). Control migration in these The human LCF gene was isolated as follows. experiments averaged 13.8 cells/hpf. A cDNA library from mitogen-stimulated human periph FIG. 14 shows the effect of media, r-2 r CF, and rCF eral blood mononuclear cells (PBMC) was ligated into the and IL-2 on human nylon wool non-adherent T cells COS cell expression vector pXM (Wong et al., Science (NWNAT) proliferation. 228:801-815, 1985). Supernatants from cells transfected with pooled plasmids were screened for lymphocyte FIG. 15 shows the effect of media r-2, LCF, and r CF chemoattractant activity using a modified Boyden chamber and IL-2 on human HIV-PMBCs. 15 assay (Cruikshank et al. J. Immunol. 128:2569-2574. FIG. 16 shows the cell counts of long term cultures of 1982). Supernatants collected 24h after transfection were rLCF and r-2 treatments with CD4 counts obtained from placed in bottom wells of microchambers. The migration of patients infected with HIV. Data for 5 and 6 represent the human T cells through 8 um nitrocellulose filters in response same individual a month apart. to the presence of these supernatants was determined, com FIG. 17 shows p24 measurement by ELISA. pared to supernatant of mock (vector only) transfected COS LCF Polypeptides cells. Supernatants with chemoattractant activity were fur LCF polypeptides according to the invention include the ther screened for the capacity to induce IL-2R expression on full-length LCF polypeptide (as described in FIG. 2, SEQID resting T-cells by FACs analysis of cells incubated with NO: 1). Such polypeptides may be derived from any source. fluorescein-conjugated anti-Tac antibody, and for the ability These polypeptides are used, e.g., to screen for antagonists 25 of Fab fragments of monoclonal OKT4 antibody to block which disrupt a LCF:CD4 receptor interaction or an LCF this induction (Cruikshank et al... J. Immunol. :mediated physiological response (see below). LCF frag 138:3817-3723, 1987). Seven different subclonings were ments or analogs may also be useful candidate antagonists of screened, approximately 200 clones per supernatant in origi LCF:CD4 receptor activity. The efficacy of a LCF fragment nal supernatants that were subcloned were found to be or analog antagonist is dependent upon its ability to interact positive. Next, the supernatants were sequentially subcloned with CD4; such an interaction may be readily assayed using and diluted until one clone per supernatant was obtained. any number of standard binding methods and LCF-mediated The criteria established for the presence of LCF-containing CD4 receptor functional assays (e.g., those described supernatant included a positive response for both assays and, below). Polypeptides of the invention also include any in addition, that the activity could be blocked by coincuba fragment or analog capable of interacting with the CD4 35 tion with Fab fragments generated from OKT4 antibodies receptor and mediating the LCF biological cascade, i.e. LCF (Ortho Pharmac, Raritan, N.J.). A single clone (LCF-7) with agonists. these characteristics was isolated and both strands were Specific LCF polypeptide fragments of interest include sequenced by the dideoxynucleotide chain termination any portion of the LCF polypeptide which are capable of method (Sanger et al., Proc. Natl. Acad. Sci. U.S.A. interaction with CD4 receptor. e.g. all or part of the 74:5463-5467, 1977). Sequence analysis and northern blot N-terminus or e.g., a hydrophilic domain. Based on the ting (FIG. 1) indicated that the LCF-7 cDNA was not hydrophilicity analysis (see FIG. 9) and biologic inhibition full-length (corresponding to nucleotide 441 to 1450 of the data, other likely candidates include without limitation, the indicated sequence). Then, the LCF-7 cDNA was used to four hydrophilic regions, A, B, C and D (see FIG. 5) and the probe a second mitogen-stimulated human PBMC cDNA FEAW (Phe, Glu, Ala, Trp) sequence from amino acids 45 library ligated into bacteriophage lambda ZAP 125,000 96-99 of LCF (FIG. 2 and SEQID NO: 1). Such fragments plaques were screened with full length LCF-7. Upon may be useful as agonists or antagonists (as described screening, three clones were isolated ranging in size from above), and are also useful as immunogens for producing 0.6- to 2.2-kb. The largest clone was sequenced on both antibodies which neutralize the activity of LCF; see infra). strands (see FIG. 2; SEQID NO: 2). Partial sequencing of Alternatively, from the primary amino acid sequence the 50 two shorter clones revealed that they were identical to secondary protein structure and, therefore, the domains of LCF-A, but incompletely extended in the 5' direction. LCF may be deduced semi-empirically using any standard As described above, LCF cDNA was isolated by screen hydrophobicity/hydrophilicity calculation, e.g., the Chou ing a COS cell expression library of mitogen-stimulated Fasman method (see,e.g., Chou and Fasman, Ann. Rev. human peripheral blood mononuclear cells (PMBC). Super Biochem, 47:251, 1978). Hydrophilic domains present 55 natants were assessed for the presence of LCF by the themselves as strong candidates for antigenicity and hydro induction of human CD4+ T cell chemotaxis and cell cycle phobic regions for binding domains, and therefore, useful changes as determined by upregulations of IL-2 receptors antagonists or agonists. (IL-2R) (Cruikshank et al., J. Immunol. 138:3817-3823. Candidate fragments (e.g., all or part of Domains A or D; 1987). Following four rounds of screening, a positive super see. FIG.9) are then tested for interaction with CD4 receptor natant from a single clone of 1-kb was identified. The LCF and their ability to induce an LCF-mediated physiological cDNA was used to probe a northern blot of total RNA response, i.e., serve as LCF agonists, by assays described isolated from human T cells (FIG. 1). A single band of herein. Fragments are also tested for their ability to antago 2.2-kb was detected. In order to isolate a full length clone the nize the interaction between LCF and CD4 using the assays 1-kib LCF cDNA was used to probe a second mitogen described herein. Analogs of useful LCF fragments (as 65 stimulated human PBMC cDNA library. Three clones were described above) may also be produced and tested for isolated, and the sequence of the largest clone is shown in efficacy as screening components or antagonists (using the FIG. 2 and SEQED NO: 2. 5,766.866 11 12 Within the LCF cDNA there is an open reading frame of (SF9) cells). Such cells are available from a wide range of 393 base pairs extending from nucleotide 783 to 1176 that sources (e.g., the American Type Culture Collection, codes for a 130 residue protein with a predicted molecular Rockland. M.d.; also see, e.g., Ausubel et al. supra). The mass of 13,385 daltons. The methionine at nucleotide 783 is method of transfection and the choice of expression vehicle in good context for initiation by Fickett analysis (Fickett, will depend on the host system selected. Transformation and Nucleic Acids Res. 10:5303-5318, 1982). The only other transfection methods are described, e.g. in Ausubel et al., possible initiation site lies downstream and is in-frame. supra; expression vehicles may be chosen from those representing residue 38 of the predicted amino acid provided, e.g. in Cloning Vectors: A Laboratory Manual (P. sequence. There is one potential N-linked glycosylation site H. Pouwels et al., 1985. Supp. 1987). on the serine located five residues after the start methionine. O One preferred LCF expression system is a prokaryotic While previous work suggests that native LCF is a secreted expression system as described by Ausubel et al. (supra). cytokine (Cruikshank et al. J. Immunol. 128:2569-2574. Thus, a DNA fragment containing the LCF cDNA open 1982), in the predicted amino acid sequence there is no reading frame with flanking BamH1 and Ndel restriction consensus hydrophobic signal sequence; however, nor is sites was generated by PCR according to standard methods there a potential transmembrane domain. While most 15 and ligated into the E. coli expression vector pt-16b secreted contain a signal sequence, the absence of (Novagen). This plasmid, pFT-166-ICF, was then used to a signal sequence has been reported for both secreted transform E. coli JM109. In order to stimulate the produc IL-1oland IL-1d. Similarly searches of the Genbank nucleic tion of recombinant LCF the transformed bacterial were acid and protein data bases failed to find any related stimulated with IPTG, grown in culture media and subse sequences. DNA and protein homology searches were con quently lysed. Recombinant protein was isolated by metal ducted using the programs FASTA, SEARCH, and BLAST chelation chromatography according the well known meth in the Genbank and PIR databases. ods (see, e.g., Studier Meth. Enzymol. 185:60-89, 1990). RNA Isolation and Northern Analysis Recombinant LCF was then subjected to SDS-PAGE (FIG. Human peripheral blood mononuclear cells (PBMC) were 3A) and blotted to Problott transfer filters (Applied prepared by Ficoll-Paque density centrifugation as previ 25 Biosystems). A prominent band found at an apparent ously described (Cruikshank et al., J. Immunol. molecular weight of 17.5 kDa was excised and subjected to 138:3817-3823, 1987: Cruikshank et al., J. Inn unol. N-terminal amino acid sequencing according to standard 146:2928-2934, 1991). The T lymphocyte population was techniques. Twenty-five amino acid residues at the purified by plastic adherence followed by nylon wool adher N-terminus of the recombinant LCF were sequenced and ence and finally by sheep erythrocyte rosetting and centrifu 30 were found to be identical to the predicted amino acid gation. Cells recovered from the pellet were >99% T lym sequence shown in FIG. 2 (SEQ ID NO: 1). While the phocytes as determined by fluorescent analysis. Monocytes SDS-PAGE mass of 17.5 kDa is larger than the expected were purified from PBMC using sheep erythrocyte rosetting 13.4 kDa based on nucleotide sequence, it is identical to the to deplete T lymphocytes, followed by plastic adherence of migration pattern of S-labeled in vitro translated protein the cells remaining in the supernatant after the rosetting step. 35 (FIG. 3B). The discrepancy in mass determined by SDS Adherent cells recovered from the plastic were >92% mono PAGE-from the predicted sequence may be due to aberrant cytes by fluorescence analysis. All cells were lysed with cold migration of recombinant LCF in the SDS acrylamide gel 4 M guanidinium isothiocyanate and RNA was isolated by system. CsCl centrifugation (Ausubel et al., Current Protocols in Another preferred LCF expression system is abaculovirus Molecular Biology. John Wiley & Sons, New York, 1989). expression system as described by Ausubel et al. (supra). Ten pig of RNA from each sample was loaded on a 1% DNA encoding an LCF polypeptide is inserted into an agarose-formaldehyde gel for electrophoresis, and blotted appropriate transfer vector, e.g., pVL1392 (Invitrogen onto nylon membrane. A cDNA probe from a 704 bp Pst I Corp., San Diego, Calif.). Next, the vector is co-transfected fragment of recombinant LCF-7 was (PdCTP-labeled by with wild type baculovirus genomic DNA into Spodoptera the random primer method (Feinberg et al. Anal. Biochem. 45 frugiperda (SF9) cells (ATTCAccession No: CRL 1711) and 132:613, 1983) and the blot was hybridized with 1x10 recombinant viruses are isolated by standard techniques, cpm/ml for 24 hr. After hybridization the blot was washed e.g., see Ausubel et al. (supra). Recombinant LCF produced with 0.2XSSC (30 mM NaCl, 3 mM sodium citrate, 0.05% in a baculovirus system was found to synthesize a protein sodium pyrophosphate, 0.1% sodium lauryl sarcosine) at with an apparent molecular weight of 17.5 kDa which is 56° C. and hybridization was visualized by autoradiogra 50 similar to the protein synthesized using the E. cole expres phy. As shown in FIG. 1. the probe hybridized specifically sion system shown in FIG. 3A and FIG. 3B. Sequencing of to a lymphocyte RNA of approximately 2.2 kilobases. This the first five N-terminal amino acid residues of the bacu confirmed that LCF was expressed in T lymphocytes and lovirus recombinant LCF was performed. The sequences indicated that the clone was full-length. were found to be identical to the predicted amino acid LCF Polypeptide Expression and Synthesis 55 sequence shown in FIG. 2 (SEQ. ID No.: 1) with a methion Polypeptides according to the invention may be produced ine at position 783 as the initiation site. by transformation of a suitable host cell with all or part of Alternatively, an LCF polypeptide may be produced by a an LCF-encoding cDNA fragment (e.g., the cDNA described stably-transfected mammalian cell line. A number of vectors above) in a suitable expression vehicle. suitable for stable transfection of mammalian cells are Those skilled in the field of molecular biology will available to the public, e.g., see Pouwels et al. (supra); understand that any of a wide variety of expression systems methods for constructing such cell lines are also publicly may be used to provide the recombinant LCF protein. The available, e.g. see Ausubel et al. (supra). In one example, precise host cell used is not critical to the invention. The cDNA encoding the LCF polypeptide is cloned into an LCF polypeptide may be produced in a prokaryotic host expression vector which includes the dihydrofolate reduc (e.g., E. coli), or in a eukaryotic host (e.g., S. cerevisiae or 65 tase (DHFR) gene. Integration of the plasmid and, therefore, mammalian cells, e.g. COS1, NTH3T3, and JEG3 cells, or the LCF-encoding gene into the host cell is in the cells of an arthropod, e.g. Spodoptera frugiperda selected for by inclusion of 0.01-300 uM methotrexate in 5,766.866 13 14 the cell culture medium (as described in Ausubel et al., In vitro assays to determine the extent of LCF (fragment supra). This dominant selection can be accomplished in most or analog thereof) binding to rsCD4 or CD4 receptor cell types. Recombinant protein expression can be increased bearing cells is then performed. For example, a whole cell by DHFR-mediated amplification of the transfected gene. assay is preferably performed by fixing the cell expressing Methods for selecting cell lines bearing gene amplifications the CD4 receptor, e.g., eosinophils, to a solid substrate (e.g., are described in Ausubel et al. (supra); such methods gen a test tube, or a microtiter well) by means well known to erally involve extended culture in medium containing gradu ally increasing levels of methotrexate. DHFR-containing those in the art (see, e.g., Ausubel et al. supra) and presenting expression vectors commonly used for this purpose include labelled LCF polypeptide (e.g. 'I-labelled LCF). Label pCVSEII-DHRF and padD26SV(A) as described in ling of LCF, e.g., with 'I, is performed according to Ausubel et al. (supra), Any of the host cells described above standard techniques known in the art. Binding is assayed by or, preferably, a DHFR-deficient CHO cell line (e.g. CHO the detection label in association with the receptor compo DHFR-cells, ATCC Accession No. CRL 9096) are among nent (and, therefore, in association with the solid substrate the host cells preferred for DHFR selection of a stably and CD4 receptor) by techniques well known in the art. transfected cell line or DHFR-mediated gene amplification. The assay format may be any of a number of suitable Once the recombinant LCF polypeptide is expressed, it is 15 formats for detecting suitable binding, such as a radioim isolated, e.g., by using affinity chromatography. In a working munoassay format (see, e.g., Ausubel et al., Supra). example, a CD4 affinity column was prepared by coupling Preferably, cells bearing CD4 receptor are immobilized on a recombinant soluble CD4 (rsCD4) to CNBr Sepharose 4B solid substrate (e.g. the well of a microtiter plate) and according to previously described methods (see, e.g., Cruik reacted with LCF polypeptide which is detectably labelled, shank et al., Journal of Immunology 1991). Thus, 100g e.g., with a radiolabel such as 'I or an enzyme which can rsCD4 was covalently conjugated to a CNBr activated be assayed, e.g., alkaline phosphatase or horseradish per Sepharose 4B (Pharmacia, Piscataway, N. J.). Next, an in oxidase. Thus, 'I-labelled LCF is bound to the cells and vitro RNA transcript of LCF was generated and used for in assayed for specific activity; specific binding is determined vitro translation with rabbit reticulocyte lysate in the pres by comparison with binding assays performed in the pres ence of SI methionine according to standard methods. 25 ence of excess unlabelled LCF polypeptide. S-labeled in vitro LCF was applied to the column for 3hr Alternatively, LCF polypeptide (fragment or analog at 37° C. at which time the column was extensively washed thereof) may be adhered to the solid substrate (e.g., to a with wash buffer (0.01 M Tris-Cl, pH 8.0, 0.14 M NaCl, microtiter plate using methods similar to those for adhering 0.025% NaN 0.5% Triton X-100. 0.5% sodium cells for an ELISA assay; Ausubel et al. supra) and the deoxycholate). LCF was eluted with a triethanolamine solu 30 ability of labelled rsCD4 receptor to bind LCF can be used tion (50 mM triethanolamine, pH 11, 0.1% Triton X-100. to detect specific rsCD4 receptor binding to the immobilized 0.15 MNaCl) into tubes containing 1M Tris-Cl, pH 6.7 and LCF. analyzed. There now follows an example demonstrating still another Once isolated, the recombinant protein can, if desired, be method useful for analyzing the LCF:CD4 interaction. In further purified, e.g., by high performance liquid chroma 35 this method recombinant LCF-containing E. coli crude tography. These general techniques of polypeptide expres supernatant was incubated with 10g of rsCD4 for 1 h at 40 sion and purification can also be used to produce and isolate C. Next, the recombinant LCF-CD4 complex was added to useful LCF fragments or analogs (as described below). protein A Sepharose beads which had been incubated with Furthermore, the eluate may then, if desired, be run on a 1 grabbit anti-CD4 polyclonal antibody and washed with a SDS-PAGE and visualized by autoradiography (see, e.g., the suitable buffer. The mixture was then incubated for 2h at 40 results from the above experiment presented in FIG. 3B). C., washed four times with TSB (0.01M Tris, (pH 8.0), Finally, LCF polypeptides, particularly short LCF 0.14M NaCl. 0.025% NaN) prior to running on a 15% fragments, can be produced by chemical synthesis (e.g., by SDS-polyacrylamide gel system. Protein separated on the the method described in Solid Phase Peptide Synthesis, SDS-gel was then transferred to Problott transfer filters and 1984. 2nd ed. . Stewart and Young, eds. Pierce Chemical 45 probed using rabbit anti-peptide Dantibody (1:200 dilution) Co., Rockford, Ill.). (also see section infra anti-LCF Antibodies) followed by Assays for LCF Binding and Function goat anti-rabbit 'I-IgG antibody. The results of this experi Useful LCF polypeptide fragments or analogs in the ment are presented in FIG. 4. As shown in FIG. 4. there is invention are those which interact with CD4 receptor, e.g., a detectable specific physical interaction between recombi LCF agonists or antagonists. Such an interaction may be 50 nant LCF and rsCD4. detected by an in vitro binding assay (as described infra) LCF polypeptide (or fragment or analog thereof) may also followed by functional analysis. Thus, the fragments or be assayed functionally for its ability to mediate migration analogs thereof may also be assayed functionally, i.e., for its of CD4+ lymphocytes, monocytes, eosinophils and the like. ability to bind a CD4 receptor and to induce the migration Migration assays may be employed using any suitable cell. of T4+lymphocytes, monocytes, eosinophils and the like (as 55 e.g., T lymphocytes. monocytes or eosinophils as described described infra). These assays include, as components, LCF in (Cruikshank et al., 1987, J. Immunol. 128: 2569-2571; (or a suitable LCF fragment or analog thereof) and recom Rand et al., 1992. J. Ep. Med. 173:1521-1528) follows. For binant soluble CD4 receptor (rsCD4) or CD4 receptor example, recombinant LCF synthesized in an expression bearing cell, e.g., an , configured to permit detec system, e.g., E. coli or baculovirus expression systems (as tion of binding. Thus, the invention includes methods for described supra), can be assayed for the ability to induce cell screening compounds useful as LCF agonists. migration. In one working example, murine cell chemotaxis One such assay method is as follows. Full-length LCF was performed using a modified Boyden chemotaxis cham polypeptide (fragment or analog thereof) is produced as ber (Cruikshank et al. J. Inmunol. 128: 2569-2571). The described supra. CD4 receptor component is produced either cells were suspended in RPMI 1640 containing 10% FBS at as a recombinant soluble component or is produced as a 65 a concentration of 5x10 cells/ml. A 12 um nitrocellulose membrane component by a cell, e.g., a T lymphocyte, membrane was used and the cells were incubated for 4 h. or eosinophil. Next, the membranes were stained with hematoxylin and 5,766.866 16 dehydrated using sequential washing with ethanol, propanol. multimer, but some LCF may exist as monomers. The and finally xylene to clarify the filters and allow for cell multimeric form, however, is believed to possess chemoat counting by light microscopy. Cell migration was quanti tractant activity. tated by counting the number of cells which had migrated Screening For Compounds that Inhibit LCF:CD4 Interaction beyond 50 m. All counts were compared with control cell As discussed above, one aspect of the invention features (unstimulated) migration which was always normalized to screening for compounds that antagonize the interaction 100%. In addition, all samples were performed in duplicate between LCF and CD4 receptor, thereby preventing or and five high-powered fields were counted for each dupli reducing the cascade of events that are mediated by that cate. FIG. 5 shows a representative dose response curve for interaction. Chemical antagonists to LCF which bind to LCF protein generated from the E. coli expression system (supra). O or LCF/CD4 receptor or CD4 receptor without triggering a As indicated from the dose response curve, maximal migra response are used to reduce, attenuate or interfere with the tion was induced with a concentration of recombinant LCF effects of LCF or cross-linked LCF agonists or biologically at 10M, and ED of 10 'M. Statistics were performed active LCF polypeptide fragments or analogs thereof which using Student's TTest (or analysis of variance modifications act to stimulate or activate LCF-mediated events of the when data from multiple experiments were pooled) and 5 immune response and inflammation. Thus, the invention counts statistically different from control cell migration provides for methods to screen for such useful compounds. (p<0.05) are designated by an asterisk. Similar results were These antagonists include, without limitation, e.g., cross obtained when baculovirus-produced LCF was substituted linked LCF synthetic LCF, anti-LCF antibodies, or other for E. coli-produced LCF. drugs, e.g. organic compounds. In order to demonstrate that this physical association 20 Thus, LCF polypeptide can be used to prepare compounds between recombinant proteins in solution corresponds to a that tend to neutralize or impede its activity. For example, specific functional association between recombinant LCF one approach pertains to identification of the active sites of and cell surface CD4 the effects of recombinant LCF on LCF, followed by the alteration of those sites of the LCF murine T cell hybridoma cell lines expressing either full amino acid sequence by substitution of amino acids within length or truncated human CD4 was examined (Sleckman et 25 the active site by other amino acids, so that the peptide does al.. 1987, 1988). Three cell lines were employed: a mock not lose its binding affinity for the CD4 receptor, but upon infected cell line which lacked expression of CD4; a cell line binding is unable to promote activity, and thereby blocks the expressing intact (wild type) CD4; and a cell line expressing effect of LCF. LCF activity may also be blocked, attenuated, truncated CD4 (delta 13) in which the 31 most distal or interfered with by using antibodies, e.g., monoclonal, or residues of the cytoplasmic tail of CD4 have been deleted. 30 chemical antagonists to LCF. These chemical antagonists The cell lines expressing either intact CD4 or delta 13 CD4 include any organic compounds, or any of the other afore were chosen for their comparable levels of CD4. As shown mentioned compounds, which can be assayed or screened in FIG. 6 cells which expressed intact CD4 migrated in for their ability to interfere with LCF:CD4 mediate events response to recombinant LCF stimulation. Cells either lack by the methods that follow. ing CD4 or expressing delta 13 CD4 were unresponsive to 35 The elements of the screen are LCF polypeptide (or a recombinant LCF. These cells were responsive to murine T suitable fragment or analog thereof) and rsCD4 or, a CD4 cell receptor-stimulated migration as the antibody 2C11 receptor expressing cell, e.g. CD4" lymphocyte, monocyte, induced migratory responses of 198%+4% and 192%+3% eosinophil and the like, configured to permit detection of for the mock transfected and delta 13 CD4 cell lines respec binding. A full-length LCF polypeptide (fragment or analog tively (FIG. 6). These studies demonstrate that CD4 must be 40 thereof) and rsCD4 may be produced as described above. expressed for LCF-induced cell motile responsiveness and Binding of LCF to its receptor may be assayed by any that the cytoplasmic tail is required. suitable method (as described above). For example, cells CD4 specificity for LCF stimulation in human T cells was expressing CD4 receptor, e.g., eosinophils, are immobilized demonstrated using the expression of IL-2R to identify LCF on a solid substrate (e.g., the well of a microtiter plate) and responsive cells. Mixed T cells were cultured in the presence 45 reacted with detectably-labelled LCF polypeptide (fragment of recombinant LCF (10M) for 24 and 48 hrs at which or analog thereof) as described above. Binding is assayed by time the cells were labeled for their expression of both CD4 the detection label in association with the receptor compo and IL-2R. As shown in FIG. 7, only cells which were CD4" nent (and, therefore, in association with the solid substrate). demonstrated an increase in surface expressed IL-2R. In this Binding of labelled full-length recombinant LCF polypep particular experiment an increase in IL-2R was observed for 50 tide to CD4 receptor bearing cells is used as a "control" 17% of the CD4 cells. This indicates not only LCF speci against which antagonist assays are measured. The antago ficity for CD4 cells, but also suggests that recombinant nist assays involve incubation of the CD4 receptor bearing LCF acts only on a subset of CD4" cells. cells with an appropriate amount of candidate antagonist, Finally, molecular weight sieve chromatography of e.g., an antibody or an organic compound. To this mix. an recombinant LCF shows that most chemoattractant activity 55 equivalent amount of labelled LCF is added. An antagonist elutes in the 50-60 kDa region. This peak of chemoattractant useful in the invention interferes with labelled-LCF binding activity corresponds to the elution profile of S-labeled to the immobilized receptor-bearing cells. Alternatively, an recombinant LCF subjected to identical chromatography as antagonist may bind but not activate a biological response. shown in FIG. 8A and FIG.8B. A small peak of radioactivity Subsequently, an antagonist, if desired, may be tested for was present with no corresponding chemoactivity in the its ability to interfere with LCF function, i.e., to specifically 14-18 kDa region. The peak fraction for both chemotaxis interfere with labelled LCF binding without resulting in and radioactivity (fraction 13) and the fraction containing signal transduction normally mediated by a full-length LCF only radioactivity (fraction 17) were applied to SDS-PAGE polypeptide. and subjected to autoradiography. The LCF proteins from Appropriate candidate antagonists include e.g. the each fraction appeared as single bands at 17.5 kDa (FIG. 65 polypeptides FEAW (Phe-Glu-Ala-Trp at amino acids 8B). These data suggest that under physiologic conditions 96-99) and RKSLQSKETTAAGDS (Arg-Lys-Ser-Leu-Gln LCF exists predominantly as a non-covalently linked Ser-Lys-Glu-Thr-Thr-Ala-Ala-Gly-Asp-Ser at amino acids 5,766.866 17 18 116-130) see e.g. SEQID No.:1 analogs of LCF, and other Identification of LCF in Bronchial Alveolar Lavage Fluid peptides as well as nonpeptide compounds, and anti-LCF (BAL) of Antigen Challenged Asthmatics polypeptide antibodies designed or derived from analysis of Below we describe the identification of two lymphocyte LCF/CD4 receptor interaction or the primary structure of chemoattractants present in the BAL of asthmatics by 6 hrs LCF. following antigen challenge. One chemoattractant (LCF), a CD8 cell product, acts exclusively on CD4" cells, while the Anti-LCF Polypeptide Antibodies second chemoattractant (MP1c. ), a monocyte product, Human LCF (or fragments or analogs) may be used to appears to act on both CD4 and CD8 cells. An important raise antibodies useful in the invention; such polypeptides finding of these studies is that the chemical stimuli which may be produced by recombinant or peptide synthetic tech 1 result in T cell accumulation in the lung in asthma are niques (see, e.g. Solid Phase Peptide Synthesis, supra; products of an inflammatory cascade which begins very Ausubel et al., supra). The peptides may be coupled to a early following antigen stimulation. Furthermore, we dem carrier protein such as KLH as described in Ausubel et al. onstrate that the majority of the lymphocyte chemoattractant supra. The KLH-peptide is mixed with Freund's adjuvant activity found in BAL fluid following antigen challenge is and injected into guinea pigs, rats, donkeys and like or 15 attributable to LCF. This example is intended to illustrate. preferably rabbits. Antibodies may be purified by peptide not limit, the invention. antigen affinity chromatography. Materials and Methods For example, Kyte-Doolittle analysis (Kyte, J. Molec. Subjects. Bio. 157:105-132, 1982) of the predicted amino acid Nine normal subjects (Table 1) and seven mild asthmatics sequence revealed four major hydrophilic regions (FIG. 9). (Table 2) were recruited for the study in Southampton Based on the LCF hydrophilicity plot, rabbit antibodies to General Hospital. At the time of enrollment all the asthmatic synthetic polypeptides of the four major hydrophilic regions subjects had stable pulmonary function with a forced expi from residues 3-11.47-58.68-81 and 115-130 (designated in ratory volume in one second (FEV) greater than 70% of that FIG. 9 as A, B, C, D, respectively) were generated. Peptide predicted for their age and height (Table 2). None of the specific polyclonal antisera were identified by ELISA for 25 asthmatics were treated with inhaled or oral corticosteroids, each peptide and then purified by protein A sepharose sodium cromoglycate or theophylline for at least 6 weeks chromatography. In one example demonstrating the utility of prior to their participation in this study, They had hyperre such antibodies, it was determined that antibodies generated active airways to inhaled methacholine with a geometric to region D blocked recombinant LCF (10M)-induced mean provocative concentration of agonist required to migation from 194%+7% (mean+S.D., N=4) to 112%+5% 30 reduce FEV from baseline by 20% (PC) of 1.20 mg/ml in the chemotaxis indicator assay system (described supra). (range 0.02-3.25 mg/ml). The asthmatic subjects were all Furthermore, the anti-peptide D antibody was found to be atopic as defined by a >3 mm skin wheal response to one or suitable for western blotting and identified the same 17.5 more of 5 common allergens (Dermatophagoides kDa band as was observed following protein staining in FIG. pteronyssinus, mixed grass pollen, dog, feathers and cat 3A and FIG. 3B. 35 dander (Hollister Stier). None of the asthmatics had expe Alternatively, monoclonal antibodies may be prepared rienced an upper respiratory tract infection within six weeks using LCF polypeptides described above and standard hybri of investigations. doma technology (see, e.g. Kohler et al., Nature, 256:495. Screening. 1975; Kohler et al., Eur, J. Inmunol. 6:511, 1976: Kohler et Subjects attended the laboratory 4 days before the first al. Eur, J. Immunol. 6:292, 1976; Hammerling et al. In bronchoscopy when allergen skin prick testing, baseline Monoclonal Antibodies and T Cell Hybridonnas. Elsevier, spirometry and methacholine reactivity testing were per N.Y., 1981; Ausubel et al. supra). Thus, in one example, formed. The technique used for bronchial challenge was monoclonal antibodies to LCF (fragments or analogs adapted from the 5 breath procedure of Chai et al. (J. Allergy thereof) can be raised in Balb/C or other similar strains of Clin. Immunol. 56:323-327, 1975) using an Inspiron nebu mice by immunization with purified or partially purified 45 lizer (CR Bard, Sunderland, U.K.). After recording baseline preparations of LCF (fragments or analogs thereof). The FEV subjects inhaled 5 breaths of 0.9% sodium chloride spleens of these mice can be removed, and their lympho (saline) from functional capacity to total lung capacity from cytes fused to a mouse myeloma cell line. After screening of the nebuliser via a mouthpiece. Measurements of FEV1 were hybrids by known techniques, a stable hybrid will be iso made at 1 and 3 mins and, provided this value did not fall lated that produces antibodies against LCF (fragments or 50 by >10% of baseline, the methacholine provocation was analogs thereof). Such activity can be demonstrated by the undertaken. Subjects inhaled sequential (doubled) concen ability of the antibody to prevent the binding of radio trations (0.02-32 mg/ml saline) of methacholine (Sigma labelled LCF (e.g. 'I-LCF) to the CD4 receptor. The Chemical Co.), with FEW measurements made 1-3 mins monoclonal antibody can then be examined for its ability to after each inhalation. The stepwise methacholine inhalations prevent the biological activity of LCF, e.g., cell migration 55 continued until the FEV had fallen by at least 20% of the (as discussed above). Once produced, polyclonal or mono post-saline value. The concentration of methacholine was clonal antibodies are tested for specific LCF polypeptide plotted against the percentage fall in FEV from post-saline recognition by Western blot or immunoprecipitation analysis baseline, and that concentration causing a 20% fall in FEV1 (by methods described in Ausubel et al., supra). Antibodies (PC) was derived by linear interpolation of the last two which specifically recognize an LCF polypeptide (fragment data points. or analog thereof) are considered to be likely candidates for The allergen used for local bronchial challenge (mixed useful antagonists; or such antibodies may be used, e.g., in grass pollen or D. pteronyssinus) was that which produced an immunoassay to monitor the level of LCF polypeptide the largest wheal response on skin prick testing. In each produced by a mammal. e.g. a human. Antibodies which subject a skin wheal dose response series was then under antagonize LCF/CD4 receptor binding or LCF mediated 65 taken using 10-fold dilutions of allergen and the concentra CD4 receptor function are considered to be useful antago tion chosen for the segmental bronchial challenge was one nists in the invention. tenth of the dilution producing a 3 mm wheal response. 5,766,866 19 20 Bronchoscopy and Local Challenge. Antibodies. Study Design. A rabbit polyclonal anti-rLCF antibody generated against Volunteers taking part in this study were divided into two groups. A) Normal controls had a single bronchoscopy and arl CF-KLH conjugate purified by protein Asepharose and BAL and, B) asthmatics had two bronchoscopies 6hrs apart. rLCF affinity chromatography (described herein) was used Fiberoptic bronchoscopy was undertaken on subjects with for ELISAS. western analysis and lymphocyte migration FEV greater than 70% of predicted, and the platelet and inhibition studies according to standard methods known in clotting studies were within normal limits. Fiberoptic bron the art. Neutralizing anti-MIP1o. RANTES (R&D. choscopy was undertaken using a standardized protocol. Minneapolis. Minn.) and IL-8 (Endogen, Boston, Mass.) Subjects received intravenous atropine (0.6 mg and mida Zolam 3-8 mg) prior to bronchoscopy. Oxygen (100%) was O antibodies were used according to manufactures specifica administered via nasal prongs throughout the procedure and tions. These antibodies were used at concentrations (anti oxygen saturation was monitored with a digital oximeter MIP10 at 20g/ml, anti-RANTES at 100g/ml, and anti (Minolta. Middlesex, U.K.). Fiberoptic bronchoscopy was IL-8 at 10pg/ml) sufficient to neutralize lymphocyte performed with Olympus IT-20 bronchoscope (Olympus migration induced by 50 ng/ml of the respective cytokines. Optical Co. Tokyo, Japan). Care was taken to ensure that the 15 There was no detectable cross-neutralization between any of larynx and upper airways were adequately anesthetised using lignocaine spray (4%). The bronchoscope was passed these antibodies for any other cytokine tested. through the nares and up to 12mls of lignocaine (1%) was ELISAS. introduced through the bronchoscope into the larynx and ELISAS for LCF were performed using the antibody lower airways. Immediately after this procedure the bron described above as follows. Recombinant LCF and BAL choscope was wedged into the anterior division of the right samples were dissolved in PBS to the appropriate concen upper lobe (RUL) to undertake sham challenge with 20ml of trations. Serial dilutions of r CF were used for the standard sterile saline solution prewarmed to 37° C. The instrument curve to which the unknowns were compared. 100 pil of was then passed into the medial subdivision of the right concentrated samples were incubated in duplicate in a 96 middle lobe (RML) and 20ml of prewarmed allergen solu 25 well microtitre plate (Nunc) at 37° C. for hr. All subsequent tion was instilled. Five minutes after the introduction of the steps were conducted at room temperature. The antigen was two solutions, the appearance of the airways was observed removed by washing four times with a PBS-Tween 20 and photographed to record airway narrowing. Six hours solution. Nonspecific binding was reduced by blocking with later a second bronchoscopy was performed with the same 100 ul of 1% BSA for 1hr. Following washing, 100 ul of a premedication and oxygenation. BAL was preformed with 30 6x20 ml aliquots of prewarmed 0.9% saline solution in both rabbit anti-LCF polyclonal antibody (10 g/ml) diluted in the allergen and saline challenged bronchial segments. PBS +0.05% Tween 20 was added to each well. The pres Returned fluid was aspirated through the suction channel. ence of a LCF-anti-LCF complex was detected by the Pulmonary function tests (FEV) were performed 3 hrs after addition of biotinylated goat anti-rabbit IgG (Sigma) diluted the first bronchoscopy and 3 and 24 hrs after the second. 35 1:500 in PBS, incubated for hr. After washing with PBS, Lavage fluid processing. Extravidin peroxidase (Sigma) diluted 1:250 was incubated The recovered BAL fluid was centrifuged at 600 g for 15 in each well for 30 mins, the plate was washed and 10pil of min at 4°C., the cells separated and the supernatant stored freshly prepared substrate was added to the wells. The at -70° C. Lavage fluids were concentrated a hundred fold substrate consists of 0.2mg/ml 2,2'-azino-bis-(3-ethyl by lyophilization following extensive dialysis, against 40 benzthiazoline-6-sulphuric acid) in 0.05M citrate-phosphate ddH2O, using Spectapor membranes with a M.W. exclusion buffer, pH 5.3, and 0.015% hydrogen peroxide. The sub point of 3kDa. strate was incubated in the dark for up to 30 min. The results Lymphocyte chemotaxis. of the ELISA were read at 405 nm with a microplate reader. Cell migration was performed according to standard Using the Softmax program the standard curve of known methods (Cruikshank et al., J. Immunol. 138:3817-3825, 45 LCF was established and used to determine the concentra 1987). Migration was assessed using a modification of the tions of LCF in BAL samples. Boyden chamber assay using a microchemotaxis chamber (Neuroprobe, Cabin John, M. D.). Normal human T lym Quantitation for IL-3, IL-5, MIP10. RANTES and phocytes were isolated using hypaqueficol separation of GM-CSF levels were also assessed in the BAL samples peripheral blood mononuclear cells followed by nylon wool 50 using ELISAs. Commercial ELISA kits and reagents avail adherence, resulting in >97% CD3" cells by FACs analysis, able from Genzyme (Cambridge, MA) for IL-3 and IL-5, and then cultured overnight in RPMI 1640 containing 10% Biosource International (Camarilo, Calif.) for GM-CSF, and bovine fetal serum. The T cells (10x10/ml in RPMI 1640) R & D Systems (Minneapolis, MN) for MIP1o (Endogen, were loaded into the upper well of the chamber, with 30 ul Boston, Mass.) were utilized. All commercial reagents were of the BAL fluids placed in the bottom chamber. The two 55 used according to manufacturers specifications. For all wells were separated by a nitrocellulose filter paper with a cytokines, control species-specific-antibodies were used to pore size of 8 um. The chamber was incubated at 37° C. for establish background levels. To more accurately quantitate 3hr, after which the filter was stained and migration was cytokine concentrations, background levels were subtracted assessed by counting the number of cells that had migrated from sample measurements. beyond a certain depth into the filter (50m). For most Results experiments between 15-20 cells/hpf were counted in the control wells. In inhibition experiments the chemoattractant Lymphocyte Chemotactic Activity From BAL of Normal BAL was mixed with anti-cytokine antibodies (sufficient to Individuals. neutralize bioactivity of 50 ng/ml of specific protein) for We first determined whether lymphocyte chemoattractant 30min at 37° C. prior to loading the chemotaxis chamber. All 65 activity could be identified in the segmental BAL of normal migration is expressed as percentage values of cell migration individuals. The demographics of the normal subjects are in control buffer. listed in Table 1. 5,766,866 21 22 challenge of asthmatics (FIG. 11) the subsequent concen TABLE 1. trated saline BAL were less inhibitory and in some samples. chemoattractant activity was detected. Specifically, in indi Demographics of Normal Subjects viduals L1, L3, LA and L6 the BAL following saline challenge reduced migration to below control. appearing Subject Age Sex FEV (%)" PC20 (mg/mly like the BAL of normal individuals. The BAL fluid of the N1 29 M 1179 >32 saline challenged lobes of individuals, L2 and L5, and L7 N2 25 M 14.5 >32 had some baseline lymphocyte chemoattractant activity; N3 32 M 134.2 32 N4 28 M 106. 32 L7's BAL fluid contained significant T cell chemoattractant N5 60 F 13.5 32 O activity. However, we did detect an increase in lymphocyte N6 35 M 14.0 >32 chemoattractant activity in the antigen challenged. N7 22 F 18.0 >32 unconcentrated, subsegmental BAL from six of the seven N8 2O M 104.0 32 asthmatics (FIG. 11) compared to their saline challenged N9 27 F 118.5 32 lobes (p<0.05for each except L7). Four of the seven antigen "FEV% of predicted based on forced expiratory volume in the 1" second. 15 samples (L1, L2, LA. and L5) demonstrated significant Methacholine concentration (mg/ml) required to produce 20% fall in FEV. increases in migration compared to control migration buffer. BAL samples L3b and L6b did induce significant increased They all were non-atopic and had PC20's to methacholine of migratory responses compared to their saline control (p<.05) greater than 32 mg/ml. We assayed the chemoattractant but compared to chemotaxis buffer, the levels did not reach activity of unconcentrated and one hundred fold concen statistical significance (p>0.05). None of the saline or anti trated BAL to human peripheral T cells (FIG. 10). Samples gen challenge solutions had any intrinsic lymphocyte were concentrated by first dialyzing against double distilled chemoattractant activity. water followed by lyophilization to effect a 100 fold con As shown in Table 3, there were no detectable changes in centration. Chemotaxis analyses of each undiluted sample cell differentials following antigen challenge. On average did not demonstrate any increase in cell motility as com 25 the recovered cell population in the BAL fluid remained pared with migration in the presence of chemotaxis assay consistent pre- and post-antigen challenge with cell differ buffer alone. In fact, all of the samples were inhibitory, with entials of 65-70% macrophages, 13-20% lymphocytes, significant inhibition of migration (<80% of control 17-21% neutrophils, and 4-10% eosinophils. This finding migration. p-0.05) detected in 4 of 9 samples. The inhibitory suggests that the cytokines present in the BAL fluid at the 6 effect did not appear to mask the presence of T cell specific 30 hr time point were produced by either pre-existing or newly lymphocyte chemoattractant activities of IL-8, RANTES. recruited cells at the site of antigen challenge. This also LCF or MIPlot, in any of the concentrated BAL samples by indicates that chemoattractants detected at this early time ELISA (sensitivity >10pg/ml and 40pg/ml respectively). point play a role in recruitment of responding cells. Higher concentrations of normal BAL fluid were more inhibitory; while greater dilutions were less inhibitory, even 35 TABLE 3 tually reaching (buffer) control cell migration at a dilution of 1:1.000. For none of the dilutions did BAL fluid from Differential Cell Counts 6 hrs Post Saline or Allergen Challenge' normals induce enhanced migration. In addition there was Macro- Lympho- Neutro- Cels Total no detectable lymphocyte chemoattractant activity, for any Subjects phage cyte phil Eosinophil Counted Cells/ml dilution tested, in BAL six hours following saline challenge Lla 522 108 139 3 800 9.6 x 10' of 3 normals who volunteered for dual bronchoscopy stud L 539 26 93 45 803 9.6 x 10 ies. These dual BAL samples did demonstrate migration L2a 557 95 48 5 805 6.8 x 10 inhibitory activity in a similar fashion as seen with BAL L2b 271 85 403 46 805 8.2 x 10' L3a 515 237 4. 2 801 6.8 x 10 samples obtained from a single lavage from normals. L3b 453 162 195 4 8. 4.O x 10 Lymphocyte Chemoattractant Activity In BAL Fluid Of 45 Laa 457 284. 59 5 805 13.9 x 10 Asthmatics 6hr Following Antigen Challenge. Lab 384 289 118 13 804. 25.9 x 10' Table 2 describes the demographics of the asthmatic LSa 41 226 16 l 805 8.8 x 10 subjects. LSb 552 18 8.4 52 806 7.9 x 10 L6a 750 25 22 7 804 20 x 10 L6b 640 23 13T 10 810 3.3 x 10' TABLE 2 5 La 29 236 270 22 819 11.3 x 10' b 460 173 O5 65 803 8.6 x 10 Demographics and Characteristics of Asthmatics Subject Cell counts were performed on recovered BAL fluid from an instillation Subject Age Sex FEV PC20 Allergen (mg/ml) volume of 120 ml. Following centrifugation, cell pellets were cytospin gentrifuged and stained with a Wright's Giemsa's stain. Ll 27 M 84. 3.45 grasses 104 Cells obtained 6 hrs following saline challenge are designated by an "a". L2 42 F O5.5 3.48 Derm. Ptery 10-5 55 Cells obtained 6 hrs following antigen challenge are designated by a "b". L3 30 M 99.3 2.99 grasses 10-5 La 33 F 040 1.47 Derm. Ptery 10-5 Characterization Of The Lymphocyte Chemoattractant L5 25 M 76.2 2.28 Derm. Ptery 10-5 Activity. L6 37 F 889 2.96 grasses 10-5 L 28 M 9O.O 1.70 grasses 10 Initial characterization of the chemoattractant activity from the BAL of antigen challenged lobes was conducted FEW, 96 of predicted based on forced expiratory volume in the "second. using neutralizing antibodies to known lymphocyte Methacholine concentration (mg/ml) required to produce 20% fall in FEV. chemoattractants. Neutralizing antibodies to LCF, IL-8, MIP10, and RANTES were used in this study. FIG.2 depicts the lymphocyte chemoattractant data from the Co-incubation of the BAL samples with 1 g/ml anti-LCF BAL fluid of asthmatics challenged with antigen as 55 polyclonal antibody (sufficient to neutralize 50 ng/ml of described above. While the BAL fluid from normals dem LCF bioactivity) for 30min prior to the chemotaxis assay, onstrated predominantly inhibitory activity, following saline reduced the chemoattractant activity for each of the samples 5,766,866 23 24 (FIG. 12). However, anti-LCF antibody did not completely Detectable levels of RANTES were not observed at the inhibit the migratory response to any of the BAL samples, as 6hr time point following either saline or antigen challenge in compared with the saline control values, indicating that the BAL fluid. These data combined with the lack of other chemoattractants were also present. Co-incubation neutralizing effects by the anti-RANTES antibodies indi with neutralizing antibodies for IL-8 or RANTES bioactivity 5 cates that RANTES were not present at the 4-6hr time point. had no effect on the lymphocyte chemoattractant activity, The quantitation of MIP1o protein by ELISA is also while antibodies to MIP1o did have an inhibitory effect. shown in Table 3. The two BAL samples which demon (FIG. 12). Samples L2b and LAb demonstrated the greatest strated the largest percent blocking by anti-MIP10. percent blocking by anti-MIP1o alone. A combination of antibodies, L2b and Lab, exhibited MIP1oprotein in the anti-LCF and anti-MIP10 inhibited 90-95% of all induced range of 600pg/ml (corrected to original BAL volume) O (Table 3). Several other samples, L3b. L5b, and L7b, had migration of each of the BAL's following antigen challenge trace amounts of detectable MIPlot protein. For both (FIG. 13). The addition of the other antibodies to either chemoattractants the samples which displayed the greatest anti-LCF or anti-MIPlot did not further reduce the migratory chemoattractant activity also demonstrated the highest response. Interestingly, individual L7 had demonstrated sig detectable levels of protein. At this six hour time point nificant chemoattractant activity following saline challenge. following antigen challenge there were no measureable When this sample was co-incubated with anti-LCF or anti 15 levels of cytokines IL-3. IL-5 or GM-CSF by ELISA, with MIPlot antibodies, the migratory response was partially a lower sensitivity limit of 40pg/ml. in aliquots of the same reduced (FIG. 12). A combination of anti-LCF and anti BAL samples. Detectable levels of IL-8 protein were present MIP1o did not completely eliminate all induced migration in all antigen challenge and most saline challenged lobes. (FIG. 13) indicating that other chemoattractants are present. Stimulation of Cell Division Using LCF and A Growth Overall, the major lymphocyte chemoattractant activity was Factor LCF in the antigen challenged BAL. In total, approximately We have discovered that recombinant LCF induces the expression of cell receptors, e.g., IL-2R, which subsequently 90-95% of all the chemoattractant activity was attributed to render a cell-bearing the receptor, e.g., a T cell, competent a combination of LCF and MIPO. to respond to its cognate growth factor. e.g., IL-2. In one Detection Of LCF And MIP10. Protein By ELISA. working example, human T cells were stimulated with We next determined whether these same samples had 25 recombinant LCF (a concentration range of 10M to detectable levels of LCF and MIP1o protein. Table 4 shows 10'M was used with similar results, data for 10M is the results of these ELISAs of the BAL from antigen and shown) for 24h at which time r-2 (2U/ml) or anti-CD3 saline challenged lobes expressed as pg/ml of total BAL (OKT3.50 ng/ml) were added to the cell cultures. Four days volume.The LCF concentrations were determined from con after the addition of either rIL-2 or OKT3 antibody cell centrated BAL corrected to starting volume. 30 proliferation was assayed by H thymidine uptake. Averag ing the results of all three experiments shown in Table 5. TABLE 4 showing the effects of recombinant LCF on anti-CD3 and rL-2 induced thymidine incorporation, recombinant LCF ELISA-quantitated LCF and MIPlo. Concentration in BAL" in preincubation resulted in enhanced IL-2 responsiveness. asthmatic subjects 35 Human T cells do not increase the incorporation of H thymidine following incubation with recombinant LCF LCF (pg|ml) MIPO (pgml) alone at either 24 or 48h, but following preincubation with Subjects Saline Antigen Saline Antigen recombinant LCF, r -2 stimulated T cells increase their L1 <10 OO3 gaO a40 incorporation of H thymidine from 1,079 cpm to 13.818 L2 aO O42 C40 503 cpm. However, in the recombinant LCF treated Cell cultures L3

SEQUENCE LISTING

1 GENERAL INFORMATION: ( i i i ) NUMBER OF SEQUENCES: 2

2 ) INFORMATION FOR SEQID No:1: ( i ) SEQUENCE CHARACTERISTICS: ( A LENGTH: 130 amino acids (B) TYPE:amino acid ( C ) STRANDEDNESS: Not Relevant (D TOPOLOGY: linear ( i i y MOLECULE TYPE: protein ( xi ) SEQUENCE DESCRIPTION: SEQ ID NO:1: Me t Pro A s p Le u As n Se r S e Th r A s p S e r A 1 a. A 1 a Ser A a Se r A 1 a 5 O 1 5 A 1 a Se r A s p w a 1 S e r V a 1 G1 u S e r Th T A a G 1 u A a Thr Wall Cy s Thr 2 25 3 O v a 1 Thir Le u G 1 u Lys Met S e r A a G 1 y Le u G y P h e Sier Le u G 1 u G 1 y 3 5 4 O 45 G y Lys G 1 y Ser Le u H is G 1 y A s p Lys Pro Le u Thir I 1 e A s in Arg e 5 O 55 5 P he y s G 1 y A a A a Ser G 1 u G in Se T G i u Thr V a Gl in P to G 1 y As p 65 7 O 75 8 0. G u I e Le u G 1 in Le u G 1 y Gil y Th r A 1 a Me t G in G 1 y L. eu Thr Arg Phe 85 9 O 95 G 1 u A 1 a T r p As n 1 e I le lys A 1 a Le u Pro A s p G 1 y Pro a 1 Th e 0 0 O 5 1 O Wa 1 e rig Arg Lys Ser Le u G 1 in Ser Lys G 1 u Thir Th r A a A a Gly 5 1 O 1 2 5

A s p Ser 3 0

( 2) INFORMATION FOR SEQID No.2: ( ; ) SEQUENCE CHARACTERISTICS: ( A LENGTH: 2150 base pairs (B) TYPE: nucleic acid ( C ), STRANDEDNESS: single (D) TOPOLOGY: Finear ( i i MOLECULETYPE: DNA (genomic) ( x i SEQUENCE DESCRIPTION: SEQID No:2: TT C C T C GAGA GC GT CAA CA CAG CT GA. G. G. AACT CAA. G. G. C C C A G T G C T C AAG AT G. C. CTA. 6

GCCA G C A G C A C G GA. G. C C C C C C T G A C C A G C C C CA C C C G T G A GAC G A A G CTA C G 2

A CGAAAA GA C C A G CAAA C C T A TT CA CA C CAGCCA GT G T CAT C G GCT G T CAT GAAATC 8 O

CGC T G T G C C T C CA CAT C C C T G T G C C C A GA C C C C T G CAT C C C CAAG G CA. G. G 2 4 3

G GCA C C CA. A. CAT CAT CAT C CAA, C GAA GA C C A G C G CA AA T G G C G C GAAA CATC 3 O.

G C C T G GA, C A C C GC GT C T C G CT CAA C C T CA, GA. G. C. G. AGA GAA. TATA CA. G. A. GG G CT 3 6 O

CA C G GAA. G. C. C A A G GAA GA C C A A G C GCA C CACA GT C C TT CA GT C T G G CA GT C C GT T 4 O

A T C T C C C T G C T G A G C. CAGA A GAA. T AAAA AAA CT CA C G A GG A G G T GAA. G. G. C. GG AT 4 8 0 5,766,866 33 34 -continued

GAA G CAA CAT TAAA G CAATT A GA C G G CATC CAT G CAC CA. T. CTTACA CAA G GA G GAA. G. GT 5 4 O

CGT G GT CTTG GGTT CAG CTT G G CAG GAG GA G CAGAT CT AG AAAACAAG G T G AT TAC GGTT 6 O O

CA CAA T GT T C CAAAG G CT GCC T C C CAC GAA Go GA CTA TT CA GAA C G G CAA T G A G 6 6 O

GTT CTTT C CA T CAA C G G CAA GT C C CAAG GG GACCAC GC ACC AT GAT C C C T C C CAT C 2

C T C C G C CAAG C. C. G A GA GCC CA G O CAA G C GGA T G T CA CAA AA G GA C T C CA. G. AG 8 O

C CAT G. C C C CA CCT CAA C C C T C C A CT CAC C T G CAG C C T C A C C T C T CA CAC GA G 8 4 O

T C C A GA A T C T A C A G CA A G CCA CAG T C GCAC GC GAC ACT G. G.A. G. A. A. A T G T C GG 9 OO

CAGGG C G C T C A C C CTG GAA G GAGGGA A G. G. G. CT C C C A CAC GAA. C. A. A G C C C T CA 9 6 O

C CATA ACA G GATT T CAAA GA G CAGCCT CA CAA. CAAA G T GAGA CAT C C A G C C G GA. G. 0

AT CAAAT CTT G CAG C T G G GT GGCA CT GCC A G CA. G. G. C. CT CACA CG G T G CAAGC CT GA O 3 O

A CAT CAT CAA G G CA CT G. C. CT GA T G GA. C. CT G T CA CGA T G T CAT CA. G A GA AAAA G C C C C 1 4 0

A GT C CAAG GA AA C CACA G CT G CT CAA C C C A G CA. G. G. A. CA C GAA GCC AAA G C CA 1 2 O O

ATA ACACA CA GCA ACA CA. C. A G C T C C CATA A C C GCT GATT CT CAG G GT C T C T G CTG CCGC 6 O

CCCA C C C A GA T G. GGG (GAAA G CA CA. G. G. G. G. G. C CCCA GT C C C C T G C C C A G C C C CAA C 3 O

C T C T AG GA GCCA C C C A GC AAAAGGTT ST C C T AA AATA AG G G CA, GA GT CACA CT G. G. C. G 3 8 O

CAGCT GAA C AAA TT CAGA C T G T G T AAAA A GA. G.A. G. CTTA AT GATAATA T G T G G GC CA 4 4 0

CAAATAAA AT G GATT TATA GAA TTC CATA GA CA CAT G. C. C C C C CAAAA T G T 5

T CAA CA CT G TAACGG C A GAT CA C C C C CAAA C A G CA CAT AT TT T. CT CA GA 1 5 6

A T C T G CAATG T G G GCA A GA T G GAA T G G G CA G CT CAT C T C T G T C C CA CT T G GCA CAGC 1 6 2 O

T G GC GT CAT G. CAAA GT CAT G. CAAAG G C T G CA C CAC CTGA CAT CATT CA, C T CATA CAT CT 1 68 0

GGCC GT TAT GT GGCT G. G. AACT CAC CT G G CT G. C. G. G. C C GAA CT TAAG G GC 1 4 O

C T C T C C T TGT T GCC TGG G CT (CCT CACAA CA TGG GT C GG AT T C C CAG GA GA G CA C C C 8

A GAT C G CAA GAGCCAC GTA GAA G C T G CAT CT T G T TATA CCTTT G CCTT G GAA GT GCA 8 6 O

T G. G. CAT CACC CCA C CATAC C CAT CA GT T A. GAC CT GACA CAAA C C T G C C T G G G T AA 1 9 2 3

GGGAGA AA A A T G C. GG G GT CATT TAT GAAAAAT ACA GTT GT CACA GAA. A. CAT 1 9 8 O

G CAA AA TT GT TTTT G GT T G G A T G GAGA AG AAT C C T A G GAA G G G G GAGCCA GTA 2 O 0.

AAT AGAG GA G TACA C G T AA GCA C CAA. G. CT CAAA C C C T G A CAG G T G C CGACA GAA GG 2 : 00

AA C CA C C G G TATA T GAG G G TAT CAAATAA A.A. G. C. A C AC TAC CACC 2 15 O

We claim: 50 which binds to lymphocyte chemoattractant factor hav 1. Alymphocyte chemoattractant factor (LCF) immunoas ing SEQID NO. 1; and say kit comprising an antibody which binds a LCF polypep (b) detecting immune complex formation, between said tide having SEQID NO. 1. 2. The lymphocyte chemoattractant factor immunoassay antibody and a sample constituent, as indicative of the kit of claim 1, further comprising means for detecting 55 presence of said lymphocyte chemoattractant factor in binding of said antibody to said lymphocyte chemoattractant said sample. factor polypeptide. 4. The method of claim3, wherein said immune complex 3. A method of detecting lymphocyte chemoattractant formation is detected by ELISA. factor (LCF) in a biological sample, said method compris 5. The method of claim3, wherein said immune complex ling: 60 formation is detected by Western blot analysis.

(a) contacting said biological sample with an antibody s: ck :: *k UNITED STATES PATENT AND TRADEMARK OFFICE CERTIFICATE OF CORRECTION

PATENT NO. : 5,766,866 Page 1 of 1 DATED : June 16, 1998 INVENTOR(S) : David M. Center, et al.

It is certified that error appears in the above-identified patent and that said Letters Patent is hereby corrected as shown below:

Column 7 Line 36: "(10g)” should read -- (10g) -- Line 54: 'a fold' should read -- a 100 fold --

Column 8 Line 29: “fold' should read -- 100 fold -- Column 11, Line 18: “a” should read -- 1 f-- Column 12 Line 50: “E. cole' should read -- E. coli --

Column 20 Lines 25: “hr. should read -- 1 hr. -- Line 35: "hr.' should read -- 1 hr -- Line37: “10pul” should read -- 100pal -- Column 22 Line 5: “below control' should read -- below chemotaxis control --

Column 30 Line 59: “full-length the should read -- full length polypeptides, the --

Signed and Sealed this Eighteenth Day of September, 2001 Attest 7c44, f2 abée

NCHOLAS P. GODIC Attesting Officer Acting Director of the United States Patent and Trademark Office