CA2142860A1 - Human interleukin-13 - Google Patents

Abstract

Nucleic acids encoding human IL-13, and purified IL-13 proteins and fragments thereof. Antibodies, both polyclonal and monoclonal, are also provided. Methods of using the compositions for both diagnostic and therapeutic utilities are provided.

Description

,o 94/0~680 2 1 ~ 2 8 6 PCr/lJS93/07645 The present invention relates to compositions and methods for affecting the human immune system. In particular, it provides nucleic acids, proteins, and antibodies which regulate immune system response and development. Diagnostic and therapeutic uses of these materials are also disclosed.
BACKGROUND OF ~ INVI~TION
For some time, it has been known that the mammalian 0 immune response is based on a series of complex cellular interactions, called the "immune network." Recent research has provided new insights into the inner workings of this network.
While it remains clear that much of the response involves !:
network-like interactions of lymphocytes, macrophages, granulocytes, and other cells, immunologists now generally hold the opinion that soluble proteins, known as lymphokines, cytokines, or ! :
monokines, play a critical role in controlling these cellular interactions.
In view of their importance, there is a need to identify and isolate new lymphokines. --SUMMARY OF THE INV~1TION
The present invention fills this need by providing one such --new lymphokine. More particularly, this invention provides human inte~leukin-13 ~IL-13), and methods for its use.
2s This invention also provides nucleic acids coding for - -polypeptides themselves and methods for their production and use.
The nucleic acids of the invendon are characterized by ~h=eir--homology to cloned complementary DNA (cDNA) sequences--eac-losed herein, and/or by functional assays for IL-l 3 activity applied to the ;
3 o polypeptides, which are typically encoded by these nucleic acids.

2 i ~ ~ 8 ~ 0 2 PCI~US93tO7645 ` .
' .
. ''`' .: .
Methods for modulating or intervening in the control of an immune response are provided.
The invention is based, in part, on the discovery and cloning of human cDNAs which are capable of expressing proteins having IL-13 activity. cDNA clones include human cDNA inserts of plasmid ~-vectors pB21.2Bf and pA10.66, which contain partial and full length -sequences, respectively. Equivalent vectors may be constructed by using polymerase chain reaction (PCR) techniques and the sequences of the inserts.
0 The present invention provides an isolated nucleic acid -~
comprising a segment homologous to a sequence of human IL-13 disclosed in Tab!e 1. Typically, the segment is at least about 50 nucleotides, and will often encode a protein exhibiting a biological activity characteristic of a human IL-13, e.g., an amino acid -sequence of Table 1. In other embodiments, the segment is at least 80% homologous to the coding sequence disclosed in Table 1. In `-~
other embodiments, the nucleic acid will further encode a second `
protein. The invention also encompasses a vector or a cell containing the nucleic acid.
; ~ Alternatively, the nucleic acid can be a recombinant nucleic acid ~with~ a segment homologous to a sequence of human IL-13 ;~
disclosed in Table 1. Usually, this will encode a human IL-13 or ;
may encode a fusion protein. The invention also embraces vectors, e.g., expression vectors? and cells containing the nucleic acid.
2s ~ In alternative embodiments, the invention provides an ~`~
isolated human IL-13 protein or peptide. In-~~`some embodiments, the protein has a full length sequence di~closod- in ~Table 1, or will be a mutein thereof, and may include an altered post-translational r~' modificàtion pattern, e.g., glycosylatiQn variants. Other ! ~'4, 30 embodiments include a fusion protein~ comprising a peptide of human IL-13, and cells containing such.
In anothcr embodimcnt, the in~on provides a method of refolding a guanidine denatured moùse-P600 or human IL-13 ~;
protein comprising solubilizing said protein in 6M guanidine at a 3s concentration of about 2.5 mg/ml; diluting the guanidine to about 60 mM over a period of hours in the presence of both reduced and - .

~ 94~04680 2 1 4 2 8 6 1) P~/US93/0764~

oxidized glutathione; and incubating the~diluted guanidine solution for at least about 12 hrs.
The invention also provides an antibody which specifically binds to human IL-13, e.g., a mouse, a monoclonal, or a chimeric s antibody. It also provides a method of supporting monocyte or B
cell proliferation in a sample, or sustaining viability of said cell, by contacting the sample with an effective amount of human IL-13, alone or in combination with another cytokine, e.g., IL-4 or IL-10.
In some embodiments, methods are provided for detecting human 0 IL-13 in a sample by contacting the sample with a binding composition which recognizes human IL-13 or a nucleic acid which hybridizes to a nucleic acid encoding a human IL-13. The binding composition can be a monoclonal antibody, and the sample can be a blood sample.
In other embodiments, the invention provides methods of moduladng the growth of a hemopoietic B cell or T cell by contacting the cell with an effective amount of an IL-13 and ~-4 combination or antagonists thereto, including an IL-4 antagonist. The `
hemopoietic cell growth can be accompanied by cell differentiation to antibody producing cells.
The invention further provides methods of modulating proliferadon of a mycloid precursor cell by contacting the eell with an~ effective amount of a human IL-13, mouse P600, or agonists or antagonists thereof. ~ Often the modulating proliferation is 2s accompanied by cell differentiation.
Methods of modulating the immune response to an infection or~
allergen are provided, e.g., by administering an effective amount of ~
a human IL-13, mouse P600, or agonists or antagonists thereof, including an IL-4 antagonist. And the ihvention provides methods of sustaining cell viability of a myeloid precursor cell by- contacting the cell with an effective amount of a human IL-13, a mouse P600, or an agonist or antagonist thereof, including an IL-4 antagonist-,~ and combinations with additional cytokines, including IL-4 a-d IL-10.-wo 94/04680 2 1 4 2 8 6 ~) 4 Pcr/uss3/076~s DES~ TIQN OF THE INVEN~ON
A11 references cited herein are hereby incorporated in their entirety by reference.
I. Gen~al The present invention provides the amino acid se~uence and DNA sequence of human interleukin molecules having particular defined properties, both structural and biological, designated herein as human interleukin-13 (IL-13). This molecule was obtained using ~"~
a mouse gene encoding a related mouse protein designated P600. -~
0 Some of the standard methods are described or referenced, e.g., in Maniatis et al. (1982) Molecular Cloning, A Laboratory Manual, Cold Spring ~arbor Laboratory, Cold Spring EIarbor :Press;
~ambrook et al. (1989) Molecular Cloning: A Laboratory Manual, (2d -~
ed.), vols 1-3, CSH Press, NY; Ausubel e~ al., Biology, Greene , Publishing Associates, Brooklyn, NY; or Ausubel et al. (1987 and periodic supplements) Current Protocols in Molecular Biology, Greene/Wiley, New York.
Isolation of the human gene presented obstacles which prevented others from succeeding. Earlier attempts to isolate the `
2 o human homolog of the mouse P600 protein using oligonucleotide probes and primers ended in failure. Difficulties in using such methods often arise from the inability-- to selec~ probes of sufficient length which provide a sufficient signal to noise ratio to allow isolation of correct clones. Moreover,- mar~y mouse cell lines fail to produce detectable levels of mRNA-for the mouse gene, even using highly sensitive polymerase chain reaction (PCR) techniques.
Because the homology of the mouse and human genes is relatively low, about 60%, relatively-long probes ale needed to I ~
provide sufficiently high homology to assure a discernable positive ¦ `
signal by hybridization. But before- isolation of the human gene, it was impossible to know the degree of homology or to predict which regions of the target gene exhibit high homology from which the probe should be selected. In fact, multiple attempts using various ~ 94/04680 2 1 ~ 2 8 6 o PCI`/US93/07645 pro~es, alone or in combina~ion, to isolate the gene from various libraries ended in failure.
The library from which an intermediate clone of less than full length was isolated failed to provide the correct clone when s screened with oligonucleotide or genomic DNA probes. In fact, clones isolated using the genomic mouse sequence as a probe turned out to be false positives, i.e., they did not encode the human equivalent as evaluated by sequencing. At least one other research group also failed to isolate the gene using a similar approach.
0 A different approach was devised which successfully led to isolating a human homolog to the mouse P600 gene. Instead of using oligonucleotide probes of relatively short length, a probe -corresponding to nearly the full length coding region of the nouse gene was used. Moreover, the cell type used to make the cDNA
library was quite important. As indicated above, expression of the mouse gene varied dramatically in different cell types. The human B21 cdls used to produce the cDNA library which provided the positive clone described herein turn out to be a cell type which expresses the human gene at a relatively high level.
2 o However, this fact was not apparent when the earlier screening was performed. In addition, the positive signal arising ;~ from the hybridization was difficult to distinguish from background.
The hybridization and wash conditions used in the screening were quite important, and slightly more harsh wash conditions could hàve easily eliminated any positive signal. See, e.g., Wetmur et al., J.
Molecular Biology 31 :34g (1968).
The initially isolated clone, designated pB21.Bf2, was less than ~
full length. lsolating a full length clone required use of yet another cDNA library. Thus, the isolation of the full length human clone, designated pAl0.66 rcquired investment of significant time and ~` l;
resources. After knowing the regions of high homology between- the mouse and human genes, isolation using oligonucleotide probes~--~f--reladvely short length would now be relatively straightforwar~~~
The procedure used to isolate the human IL-13 is broadly set 3s forth below. A cDNA library, constructed in a pCD vector,~was preparcd from RNA isolated from human B21 cells. These cells are wo 94/04680 Pcr/uss3/0764~
21~286U 6 .

human T cells which, it is now understood, exhibit many of the same markers as the cells providing the mouse clone. Several modifications and unusual techniques were utilized to overcome problems associated with isolating a cDNA clone when probing the .
library with oligonucleotides.
In particular, instead of using oligonucleotide probes of relatively short length, a near full length double stranded probe of :
about 400 nucleotides was selected. Although previous attempts using a B21 derived cDNA library had failed, the near full length o double stranded probe provided faint positive signals. Although ;~
several experienced molecular biologists were highly skeptical that the faint signals were real, continued pursuit of those signals led to ultimate success. - ;
The initial human isolate showed homology to the mouse gene, but lacked part of the amino terminal coding portion. Thus, this intermediate isolate was less than a full length clone. Attempts to isolate a full length clone from the B21 derived library failed. 1 `
However, upon selection of another cDNA library, the near full , `
length human probe allowed isolation of the full length human clone.
A complete nucleotide and deduced amino acid sequence of ¦ `
the pA10~66 clone is shown in Table 1. This nucleotide sequence corresponds to the sequence ~efined by SEQ ID NO: 1. Table 2 compares the gene sequence of Table 1 to a published gene sequence of the mousè P600 protein. -Table 3 compares the deduced ~
2 5 amino acid sequence of the human IL- 13 and the published mouse ~;
P600 amino acid sequence.

;
l`.
.
. ..

~ 94/n4680 7 2 1 ~ 2 8 6 ~

Table 1: Nucleotide and Amino Acid Sequence~of.huIL-13.

TTCGGCAICC GCTCCTCAAT CCTCTCCTGT lGGcacTGGG CCTC ATG GCG CTT TTG 56 Met Ala Leu Leu ~ '.

TTG ACC ACG GTC ATT GCT CTC ACT TGC CTT GGC GGC m GCC TCC CCA 104.
Leu Thr ~hr Val Ile Ala Leu Thr Cys Leu Gly Gly Phe Ala Ser Pro GGC CCT GTG CCT CCC TCT ACA GCC CTC AGG GAG GTC AIT GAG GA~ CTG 152 ~;
Gly Pro Val Pro Pro Ser Thr Ala Leu Arg Glu L2u Ile Glu Glu Leu GTC AAC ATC ACC QG AAC CAG A~G GCT CCG CTC TGC A~r GGC AGC ATG 200 Val Asn Ile Thr Gln Asn Gln Lys Ala Pro Leu Cys Asn Gly Ser Me.t GTA TGG AGC ATC A~C CTG ACA GCT GGC ATG TAC TGT GCA GCC CTG G~A 248 Val Trp Ser Ile Asn Leu mr Ala Gly Met Tyr Cys Ala Ala Leu ~lu TCC CTG ATC AAC GTG TCA GGC TGC AGT GCC ATC G~G AAG ACC C~G AGG 296 25 Ser Leu Ile Asn Val Ser Gly Cys Ser Ala Ile Glu Lys T~ Gln Arg .

Met Leu Ser Gly Phe Cys Pro His Lys Val Ser Ala Gly Gln Phe Ser AGC TTG CAT GTC CGA GAC ACC AAA ATC G~G GTG GCC CAG TTT GTA AAG 392 Ser Leu Hls Val Arg Asp Thr Lys Ile Glu Val Ala Gln Phe Val Lys lQ5 ~ 110 115 GAC CTG CTC TTA C~T TTA AAG AAA CTT m CGC GAG GGA CGG TTC AAC 440 Asp Leu Leu Leu His Leu Lys Lys Leu Phe Arg Glu Gly Arg Phe Asn 5~AAa~T~CG AAAGCATC~T T~IIIGCAGA GACAGGACCT GACTATTGAA GTTGCaGA~T 500 -CAITITTC~T TCTGalGT~A AAAAIGTCTT GGGTAGGCGG GAAGGAGGGT TA~GG~GGGG 560 TAAAATTCCT TA~CTTA~AC CTCAGCCTGT GCTGCCCGTC TTCAGCCTAG CCG~CCTCAG 620 -- ~ ~`
CCTTCCCCTT GCCCAGGGCT C~GCCTGGTG GGCCTCCTCT GTCCAGGGCC CTGAECTCGG 680 TGGACCCAGG GalGACAIGT CCCTA~A~CC CTCCCCTGCC CTAGAGCACA CTGTA~C~T 740 --ACaGlGGGTG CCCCCCTTGC CaGPCAIGTG GTGGGACAGG G~CCACTTC ALACACAGGC 800 Aa~TGAGGCA GACAGCAGCT CPGGCACACT TCTTCIIGGT CTTAITTAIT ATIGTGIGTT 860 W O 94/04680 PCT/US93tO764~
21~2860 `~ ~

A m AAAlGA GTGT~ l~T CACCG~TGGG GhIlG3GGL~ GPCIGTGGCT GCTGGCPCTT 920 .~`
GGAGCCAAGG GITCAG~G~C TCAGGGCCCC ACCACTAAAG CAGTGG~CCC CAGGA~TCCC 980 ;`
5 TGGT~AI~AG TA~IGIGTAC AGa~SClGC TACCTCACTG GGGTCCTGGG GCCTCGGAGC 1040 .".
CTCA~CCG~G GC~EGGTCAG GPGAGGGGCA GA~CAGCCGC TCCTGTCTGC CAGCCAGCAG 1100 CCaaCTCTC~ GCC~ACG~GT AAITTAIIGT TTTTCCTCGT AITTA~ATAT TA~ATAIGIT 1160 ''`
AsCAaAGaGT TAAI~T~TAG AA~GGTACCT TGAACACTGG GGG~GGGGAC AIIG~ACAAG 1220 TrGrTTCaTT GACTAICAAA CTG~AGCCA~ A~ATAAA~TT GGTGa~AEAT AL~UL~AaAA 1280 15 AL~a~ 12 90 `

.

~ 94/04680 2 1 'I ~ 8 6 ~ PCT/US93/07~5 Table 2: Comparison of Human IL-13 and Mouse P600 Nucleic Acid Seqllences (human abo~e; mouse below).

GACAA&CCAG CAGCCTAGGC CAGCCCACAG TTCTACAGCT CCCTGGTTCT 50 IIII IIIIII III III~111 1 1 1111 1 111 1 1111 TCACTTGCCT TGGCGGCTTT GCCTCCCCAG 5CC~ ---- CTGTG 113 III I II IIIIII IIIIIIIII

CCAGAACCAG AAGGCTCCGC TCTGCAATGG CAGCATGGTA TGGAGCATCA ~13 Il 11111 1 1111 1 1 11111 11 1111111111 Illli I

IIIII I II IIII II I II11 11111 1111 11111 11 11'11 1 II IIII IIIIIII I I IIIIIII.11111 11 11 1111 I I I I I I I I I I I I I I I I I I I I I

CCAAAATCGA GGTGGCCCAG TTTGTAAAGG ACCTGCTCTT ACATTTA~AG 413 AAACTTTTTC GCGAGGGACG GTTCAACTGA AACTTCGA~A GCATCATTAT 463 CAACTGTTTC GCCACGGCCC CTTCTAATGA ~ --------- 468 W094/~680 PCT/~S93/07~5 10 , ' ` ' ~14286Q ;

IIIIII IIII -:.

111 1 1. Il 111 1111 11 111111111 11111 AAAATTCCTT AGCTTAGACC TCAGCCTGTG CTGCCCGTCT TCAGCCTAGC 611 ~:~
II III IIIII I II IIIIIIII IIIII I II I I `:' CGACCTCAGC CTTCCC~TTG CCCAGGGCTC AGCCTGGTGG GCCTCCTCTG 661 `
II III IIIIIII ''':
20 --~ -GGTCTC AGCCTGG--- -~ 628 ~ .
. .
~ "
,;

II III I I I IIII I I II I IIII '`~' --CAACACCC CCACCCC--- --ACCC---- ----CCACCC CCGCCGCCCC 663 :
..

I I I I I I I I I I I 1 1 1 1' 1 1 1 1 1 'I I I I I I I I I '` .
ATCCCATCC~ TACAGAAAAC TGCAGCAAGA CCGTGAGTCC AGCC--~ 707 11111 11 1 11 llil 1111 1111111111 ' , -----TGTGG ---~ --- -CCTGGTCCA CACA-GGGCA ACTGAGGCAG 74Q

I 11 1 1 1 1 1 1 ~ I I I I I I I I I I I I I I I I I I I I I I I I I I I I
GCAGCAGCTT GAGCACATTT CTTCTTGATC TTATTTATTA TGGTTGTGTG ~90 I I I I I I I I I I I I I I I I I I I I I I I 1'1 1 1 1 1 1 -=- - -`

~94/~680 2 1 '~ 2 ~ 6 0 Pcr/us93/~)764g 111111111 1 1 11 11 111111 111 1 111111 1111 11 11 ;

TGCTACCTCA ~ CT GTAGCCTCCA GGTCTCACCC CAGGCAGGAG 953 AGCTCTCAGC CAACGAGTAA TTTATTGTTT T-TCCTCGTA TTTAAA-TAT llS0 AGCCCTCAGC CATGAAATAA CTTATTGTTT TGTTCTTATA TTTAAAGTAT`1051 TAAATATGTT AGCAAAGAGT ---TAATATA TAGAAGGGT- ACCTTGAACA 1`196 111111 11 1111111111 1111111 1 1111 1 111 111 . I
TAAATAGCTT AGCAAAGAGT TAATAATATA TGGAAGAATG GC~TGTTACA 1101 CT-------~ ----- ---GGGGGAG GG----~ --GACATTGA 1215 Il 1111111 11 1 11 111 -- I

1111 1 11 111111111 1111111 11 ~ 11 11111 1111`11111 ''.

40 ACAGATAAAA AA 1212 ` `~

' . 50 wo 94/04~80 PCr/US93/07645 21~286~) 12 Table 3: Companson of Human IL-13 and Mouse P600 Amino Acid :
Sequences (human above; mouse below). Another form of human IL- 13 has a GLN between amino acids numbered 97 and 98 (position indicated by 1), caused by alternative mRNA splicing. .

ME~ AI~ LEU LEU LE~ IR THR V~L ILE AI~ LEU THR CYS LEU GLY 15 * * * * * * * * * * ,.
10MET AI~ LEU TRP V~L THR AI~ V~L LEU AI~ LEU AI~ CYS LE~U GLY 15 ".
:,;
GLY PHE ALA SER PRO GLY PRO V~L PRO PRO SER -- -- -- 26 * * * * * * * * , .
15 OE.Y IEU AI~ AI~ PRO GIJY PRO V~L PRO ARG SER VAL SER LEU PRO 30 THR AI~ LEU ARG GLU LE~U ILE GLU GLU IEU ~L ASN ILE THR GL~I 41 * * * * * * ,~ * * * *
20 LE~U THR LEU LYS GLU LEU ILE GLU GLU LEU SER ASN ILE THR GLN 45 ASN ~LN LYS AIA PRO LEU CYS ASN GLY SEP~ MET V~L TRP SER ILE 56 * * * * * * * * * ~ * *
25 ASP GI~N THR -- PRO LEU CYS ASN GLY SER ~T VAL TRP SER V~L 59 ASN LEU THR AI~ GLY MET TYR CYS AI~ AI~ LEU GLU SER LEU ILE 71 , `
* * * * * * * * ~ :
30 ASP LEU AIA AI~ GLY GLY PHE CYS V~L AI~ LEU ASP SER IE:U THR 74 ASN VAL SER GLY CYS 51~R A~ ILE GLU LYS THR GLN ARG ME~T LEU -86 35 A5N TTF SER ASN CYS ASN AI~ IIE TYR ARG THR GLN ARG ILE LEU 89 S~R GLY PHE CYS PRO HIS LYS t~L S~R AI~ GLYIPHE SER SER LEU- 101 .:
HIS GLY LEU CYS ASN ARG LYS ALA PRO THR THR V~L SER SER LEU 104 `:
HIS VAL ARG ASP THR LYS ILE GLU V~L AL~ GIN P Æ V~L LYS ASP 116 * * * * * * * *
PRO - ASP THR LYS TT-F GLU U~L ALA HIS PHE ILE THR LYS 117 * * * * * * * *

.

O 94/0'~680 2 1 1 2 8 6 ~ PCI/U593/07645 The amino acid sequence of the other form of human IL-13 mentioned in the legend to Table 3 is defined in the Sequence Listing by SEQ ID NO~
As used herein, the term "IL-13" describes a protein 5 comprising a protein or peptide segment having the amino acid sequence shown in Table 1, or a ~ragment thereof. It also refers to a polypeptide which functionally affects cells or subcellular components in a manner similar to the IL-13 allele whose sequence is provided. It also encompasses allelic and other variants, e.g., 0 metabolic, of the protein described. Typically, it will bind to its corresponding biological receptor with high affinity, e.g., at least about 100 nM, usually better than about 30 nM, preferably better than about 10 nM, and more preferably at better than about 3 nM.
The term shall also be used herein to refer to related naturally -occurring forms, e.g., allelic and metabolic variants of the human protein.
This invention also encompasses proteins or peptides having substantial amino acid sequence homology with the amino acid sequence in Table 1, but excluding any protein or peptide which 2 o exhibits substantially the same or lesser amino acid sequence homology than does the corresponding P600 protein found in the mouse.
A polypeptide "fragment", or "segment", is a stretch of amino acid residues of at least about 8 amino acids, generally at least 10 amino acids, more generally at least 12 amino acids, often at least 14 amino acids, more often at least 16 amino acids, typically at least lB
amino acids, more typically at least 20 amino acids, usually at least 22 amino acids, more usually at least 24 amino acids, preferably at least 26 amino acids, more preferably at least 2~ amino acids, and, in particularly preferred embodiments, at least about 30 or more amino acids. Sequences of segments of different proteins can be compared to one another over appropriate length stretches.
Amino acid sequence homology, or sequence identity, is determined by optimizing residue matches, if necessary, by ~ ~-3s in~ôducing gaps as required. See, e.g., Needleham et al., J. Mol. Biol. ~-WO 94~4680 PCT/US93/07~i45 -.
21~861~ 14 .;-`

:
48:443 (1970); Sankoff et al.,-~chapter one in Time~ Warps, String ' Edi~s, and Macromolecules: The Theory and Practice of Sequence Comparsiora, 1983, Addison-Wesley, Reading, MA; and software packages from IntelliGenetics, Mountain View, CA; and the ~--5 University of Wisconsin Genetics Computer Group, Madison, WI. This changes when considering conservative substitutions as matches.
Conservative substitutions typically include substitutions within the following groups: glycine, alanine; valine, isoleucine, leucine; aspartic acid, glutamic acid; asparagine, glutamine; serine, 0 threonine; lysine, arginine; and phenylalanine, tyrosine. Homologous amino acid sequences are intended to include natural allelic variations in the provided sequence. Typical homologous proteins or peptides will have from 50-100% homology (if gaps can be introduced), to 60-100% homology (if conservative substitutions are 5 included) with an amino acid sequence segment of Table 1.
Homology will be at least about 50%, generally at least 58%, more generally at least 63%, often at least 69%, more often at least 75%, typically at least 81%, more typically at least 86%, usually at least 90%, more usually at least 93%, preferably at least 95%, and 2 o more preferably at least 97%, and in particularly preferred embodiments, at least 98% or more. The degree of homology will vary with the length of th¢ compared segments. Homologous pr~teins or peptides, such as the allelic variants, will share most biological activities with the embodiment described in- Table 1.
2 5 As used herein, the term "biological activity" is used to describe, without limitation, inducing characteristic cell~ stimu~ation, Ig production, cellular differentiation, or cell viability functians, or more structural properties as receptor binding and cross-reactivity with antibodies raised against the same or an allelic variant of the described human IL-13. --The terms ligand, agonist, antagonist, and analog include molecules that modulate the characteristic cellulaE res-ponses to IL-13 or IL-13-like proteins, as well as molecules possessing the more standard structural binding competition features of ligand-3s receptor interactions, e.g., where the receptor is a natural receptor or an antibody. The cellular responses likely are mediated through .

2 l ~ 2 8 B ~ PCI/US93/0764~ .

binding of IL-13 to cellular receptors. Also, a ligand is a molecule which serves either as a natural ligand to which said receptor, or an analogue thereof, binds, or a molecule which is a functional analogue of the natural ligand.
trhe functional analogue may be a ligand with structural modifications, or ma~ be a wholly unrelated molecule which has a molecular shape which interacts with the appropriate ligand binding determinants. The ligands may serve as agonists or antagonists, see, e.g., Goodman e~ al., Eds., The Pharmacological Bases of Ther~peutics, 0 1990, Pergamon Press, New York.
II. Activities The human IL-13 protein has a number of different biQlogical activities. The human IL-13 is homologous to the mouse ~2600 protein, but has structural differences. For example, the human IL-13 gene coding sequence has only about 50% homology with the nucleotide coding sequence of mouse P600. At the amino acid level, ~`
there is about 66% identity.
The mouse P600 molecule had rather minimally defined _ biological activities. In particular, it has the ability to stimulate -~
20 undifferentiated mouse bone marrow cells to undergo early stages of differentiation. The mouse P600 protein appears to activate both mouse cells and human cells in th-i$ assay.
- The~ present disclosure also describes new activities which have boen discovered using the mouse P600 molecule. The 25 ~ difference in ~structure between the human IL-13 and the -homologous mouse P600 protein introduce some uncertainty about whether the two proteins will have identical functional properties.
'i However, the handful of identified activities appear to be sharedbctween the homologues. It is likely that many of the activities of - 30 ~ mouse P600 on mouse cells or human cells will also apply to the human IL-13. In fact, the cross species activides indica~e that many structural features are not critical in the function of the molocules. ¦
ln particular, the human IL-i3 exhibits a number of identified activities when provided to human cells. The Examples section :

WO 94/04680 PCI`/US93/0764i ;

21~286~
below describes procedures used to study the effects of human IL-13 on cell viability, morphology, proliferation, and differentiation. In particular, human IL-13 affects B cells, PBMC, and macrophages. On B cells, the cytokine affects proliferation, 5 alone or in combination with other cytokines; sustains cell viability;
affects survival; causes modification of Ig surface markers; has - ' effects specifically on CD40; and affects IgE switching.
On PBMC or macrophages, it induces morphological changes, causes changes of cell surface markers, affects nitric oxide ' 0 production, affects IL-la and II,-6 expression, and affects antibody `' dependent cell-mediated cytotoxicity (ADCC). Importantly, the similarity between IL-4 and IL- 13 leads to antagonists of IL- 13 whose structures are based upon similar antagonists of IL-4. These -ac~ivities can be useful in treating immunological conditions ' characterized by corresponding disfunctions. See, e.g., Merck Manual, or Paul, Fundamental Immunology.
A. B cells 1. Cofactor/factor Proliferation: Cell Viabilitv Mouse P600 made from E. coli has the ability to stimulate or '`
costimulate proliferation of large in vivo activated mouse B cells.
The combination of IL-4, soluble anti-CD40, and mouse P600 induces proliferation of these cells. However, since the large in vivo activated mouse B cells may contain some monocytes and other cells, -~
- other celIs may be induced to secrete various growth faètor's''which support the proliferation observed. Thus, the mouse P600 stimulates the B cells, either directly or indirectly, alone or in !' i ' i c'dmbination with 'other factors. Human IL-13 should exhibit` similar biological activity. ~
2. Sustaincd Survival of B Cells: Selectivitv~ --IL-13 enhanced the DNA synthesis of B cells activated through their antigen receptor. This induction was dose dependent, and was comparable in effect to IL-10, but less than IL-2 or IL-4. The time ~ 94/04680 1 7 2 i 1 2 ~ 6 o P~/us93/o7645 course of the effect was also different from IL-2 and IL-4.
Similarly, B cells activated through their CD40 receptor were also affected, again in a dose dependent manner, and comparable to IL-4 and IL-10 effects. The kinetics of the effects exhibited a different 5 time course from IL-4 or IL-10 responses. The combination of IL-10 and IL-13 exhibited additive effects, but IL-13 did not appear to increase any IL-4 effects. This suggested, with other data, that `
these two cytokines may share some components in signal transduction, though other data show some independence of effects.
More specifically, IL-13 induced expression of various Ig's, particularly IgE. The target cell population for IL-13 also appears ~
more restricted than for IL-4. Thus, although IL-4 and IL-13 share `-many biological properties, their signal transduction pathways are physiologially and mechanistically distinguishable.

3. Modification of~ Ie Production l;
B cells activated by B21 T cell clone, their membranes, or anti-~CD40 appear to exhibit modified Ig production patterns after exposure to mouse P600 or human IL-13. The levels of production of ~various Ig molecule subtypes when human IL-13 is co-administered to B cells with an inducing agent, e.g., activated B21 T ~'!~., cdls, m`embranes from activated B21 T cells, or anti-CD40 andbodies were increased, partiGularly IgE. The changes in Ig~ production are '`
suggesdve of accelerated differentiation including IgG4 and/or IgE
class switching. Both possibilities are consistent with a ``
2s ~differentiation~effect`caused by mouse P600 or human IL-13. `~;
~` A similar inducdo~i of IgE and IgG4 synthesis by IL-13 was , ~ observed when B cells were stimulated with CD40-L, as described~
below.
,~. ,.

4. E3ffects on CD40-mediated B Cell Proliferation and 30 Differentiation The effccts of anti-CD40 andbody or CD40 ligand on B cell - ~ proliferation were enhanced in the presence of either IL-4 or IL-13. , Both cytokines had comparable effects in spite of their significant WO 94/04680 PCI~tUS93/0764 21428~0 sequ`ence divergence. The B cell proliferation was accompanied by induction of IgM, IgG4, total IgG, and IgE levels. IgA was not stimulated. The two cytokines appeared to act through different structural mechanisms as anti-IL-4 antibodies blocked IL-4 effects,

5 but not IL-13 effects. The effects on IgE suggest that IL-13 contributes to IgE production and is an important factor in controlling IgE mediated allergic reactions.
5. I~E Switchin~
IL-13 induced IgG4 and IgE synthesis by unfractionated 0 peripheral blood mononuclear cells (PBMNC) and highly purified B
cells cultured in the presence of activated CD4+ T cells or their membranes. IL-13-induced IgG4 and IgE synthesis was IL-4-independent, since it was not affected by neutralizing anti-IL-4 monoclonal antibody (mAb). Highly purified sIgD+ B cells could also be induced to produce IgG4 and IgE by IL-13, indicating that the production of these isotypes reflected IgG4 and IgE switching and not a selective outgrowth of eommitted B cells.
IL-4 and IL-13, added together at optimal concentrations, had no additive or synergistic effect, suggesting that common signaling 20 pathways may be involved. This notion is supported by the observation that IL-13, like IL-4, induced CD23 expression on B cells and enhanced CD72, surface IgM (sIgM), and class II MHC amigen expression. In addition, like IL-4, IL-13 induced germline transcription in highly purified B cells. Collectively, these data 25 indicated that IL-13 is another T cell-derived cytokine that, in addition to IL-4, ef~lciently directs naive human B cells to switch to IglG4 and IgE production.
B. PBMC and Macro~hages ~
1. Induction of Mor~holo~ical Chan~e - ~ :-- `
3 o Thè mouse P600 also induced morphological changes in adherent human peripheral blood mononuclear cells. The treated cells exhibited significantly different morphology and clusters of 94J04680 2~ 12 PCI`/US93/07~45 small cells. The generic cells rounded up, and there was evidënce of clonal proliferation, observations which were consistent with induction of cell proliferation.
2. ModificatioQof Cell Surface Markers Mouse P600 induced significant changes in the cell surface markers of adherent cells from peripheral blood. These adherent cells were mostly monocytes, e.g., macrophage precursors, but also included more differentiated cell types, dendritic cells, and some B
cells. ~-Many of the cell surface makers on these adherent cells were up regulated or down regulated, or their dispersion in expression level changed. The following markers tended to increase on a per cell basis: CDllb, CDllc, class II MHC (as measured by binding of monoclonal antibodies Q5/13 or PdVS.2), CD 23, and CD18. In contrast, per cell expression of the following decreased: CD32, CD16, IL-2Ra, and CD14. The homogeneity of per cell expression changed ``
for CD32, and CD14. There was no change for CDlla, CD54, and CD58. Although in one case there was no change for CD44 and class I MHC, other experiments indicated increases in expression levels.
2 o These changes in expression level were detectable also at 10 days, and in certain cases exhibited a more dramatic shift, whereas others showed a lesser shift. Depending on subsets of cells, some features may have been lost by ten days.
In spite of the sequence divergence of the mouse P600 and the human IL-13, the two molecules seemed to cause similar changes in `
the adherent human cells. It is likely that activities found for one of the molecules will be found also by the other. In addition, the molecules appeared to exhibit cross-species activities, e.g., the mouse P600 was active on human cells, and the human IL-13 was -~
30 active on mouse cells.

wo 94/04680 2 1 ~ 2 8 ~ ~ 2 0 P~/US93/0764~

3. Nitric Oxide Synthesis IL- 13 (P600) was assayed by its LPS stimulated inhibi~iory effect on the production of nitric oxide (NO) by GM-CSF-derived bone marrow macrophages. IFN-~ induced NO production, while 5 IL-4 or IL- 13 inhibited NO production.
4. Effect on IL-la. IL-6~ IL-10 and TNF~X Production IL-4 and IL- 13 inhibited the production of IL- 1 a, IL-6, IL- 10, and TNF-a by LPS-activated human monocytes. The inhibitory effects of IL-4 and IL- 13 on cytokine production by LPS activated human monocytes were independent OI IL- 10, since IL-4 and IL- 13 inhibited the production of IL-la, IL-6, and TNF-a in the presence of neutralizing anti-IL- 10 mAb 1 9Fl .
5. Antibodv-dependent_ Cell-mediated CvtotoxicitY
IL- 13 induced significant changes in the phenotype of monocytes. Like IL-4, it cnhanced the expression of CDllb, CDllc, CD18, CD29, CD49e (VLA-5), class II MHC, CD13, and CD23 whereas it decreased the expression of CD64, CD32, CD16, and CD14 in a dose dependent manner. IL-13 induced upregulation of class II MHC
antigens and its downregulatory effects on CD64, CD32, and-CD16 20 expression were prevented by IL-10. --IFN-~ could also partially prevent the IL-13 induced downregulation of CD64, but not that of CD32 and CD16. However, IL-13 strongly inhibited spontaneous and IL-10 or IFN-~y induced antibody dependent cell-mediated cytotoxicity (ADCC) activity of 2 5 hurnan monocytes toward anti-IgD coated Rh+ erythrocytes, indicating that the cytotoxic activity of monocytes was inhibited.
These results indicated that IL-13 has anti-inflammatory and immunoregulatory activities. - _ --94~04680 21 ~ Z 8 6 o PCI/U~i~3/07645 C. IL-4 Anta~onist: Interactions Observations that the hIL-4.Y124D antagonist competitively inhibited the biological action on TF-1 cells of both hIL-4 and IL-13 demonstrated a relationship between IL-4R and IL-13R. The ability of mIL-13 to compete for I~25-hIL-4 binding to T~-l cells confirmed the commonalty of ~-4R and IL-13R. This relatedness may also have been expected from the similar biological responses known to be elicited by hIL-4 and IL-13, and perhaps from the close linkage of the IL-4 and IL-13 genes in both humans and mice.
0 See~ e.g., Morgan et al., Nucleic Acids Res. 20:5173 (1992), and other expenments herein. A straightforward explanation of the above observations would be that IL-4 and IL- 13 act through the same receptor.
D. Bioloei~al Relevanc~ - j The mouse P600 protein can sustain or prom~e the proliferation of large in ~ivo activated B cells. As such, the factor -appears to be either a stimulatory or costimulatory factor useful in_ ~ `
promoting activated B cell growth. Human IL-13 is therefore expected to be a useful factor in cLrcumstances where activated B
cell growth is des*ed. -~
These include genetic, developmental, or acquired immune system deficiencies, e.g., congenital aglobulinemias, immature infants, or chemotherapy patients. In vitro experiments would be performed to deeermine what effects IL- 13 possesses . In particular, 2 5 dose response relationships for various immunological assays will be tested with the compositions of this invention. See, e.g., Coligan et al., Current Protocols in ~mmunology, 1991, Greene/Wiley, New York.
Regarding the proliferative response, mouse P600 induces changes in morphology of the monocyte cells. The monocyte cells ;
consist primarily of macrophage precursors, and similar results should apply to monocyte equivalents found in organs or tissues other than the peripheral blood, e.g., the aveolar, intraperitoneal, or spleen/lymph macrophage precursors. The IL-13 or antagonist, e.g., an~ibody or IL-4 antagonist, would be indicated for conditions `

WO 94~04680 PCI`/US93/0764' ~60 22 where regulation of localized or systemic immune responses is desired and appropriate. The effec~s on class II MHC are especially relevant in these contexts.
Besides a growth factor/cofactor activity, human IL-13 also 5 affects differentiation of various cells of the immune system. For instance, in activated B cells, it accelerates or promotes the differentiation of Ig producing cells. It induces the B cells to produce Ig molecules characteristic of later or faster differentiation.
As such, the human IL-13 and mouse P600 appear to be a 0 differentiation factor for B cells.
Thus, Ig production should be regulatable by IL-13, alone or in combination with other factors. Agonists and antagonists, when provided in appropriate amounts and schedules, will be useful in treating or controlling abnormal B cell conditions, or to accelerate or decelerate B cell differentia~ion when appropriate. j -Peripheral blood monocytes are also sensitive to the presence of both human IL-13 and mouse P600. These cells, consisting primarily of macrophage precursors and more differentiated cell types, exhibit both a proliPerative response and a differentiation 20 response.
In one context, IL-4 is appropriate in antitumor situations, e.g., to stimulate an endogenous response to counter the tumor; IL- 13 should also be a useful therapeutic. In a different context of proliferative disorder, after radiotherapy or chemotherapy, where 2s the immune funtion is typically compromised, IL-13 would be useful to restore function by promoting recovery and differentiation~
of the remaining immune funcdon. See, e.g., Moller (ed), "Fc - -Receptors" in Immunological Reviews 125:1 (1992). Sirnilar problems exist in transplantation contexts, as well as in other 30 genetic or developmental immunodeficiencies, e.g., in newborn -infants. See, e.g., Baker et al., N. Eng. J. Med. 327:213 (1992).
In fact, the role of IL-13 in promoting restoration of imm~ne~ --~-function under these circumstances is supported by the cell marker changes obser~ed. With respect to cell marker differentiation, the -35 general trend is that the class II MHC markers are affected. Also, CD23 is affected. The effects on class II MHC markers indicate that ~94/04680 23 ~1~2~o PCr/uss3/0764s -~

systemic responsiveness to infections can be modulated with IL- 13 ~ ;
or mouse P600, or agonists or antagonists thereof.
The observed decreases in CD32 and CD16 indicated a lowered receptor for IgG Fc, which would be correlated with a lessened 5 response to infections. If so, an IL-13 antagonis~, or mouse P600 antagonist, would be useful in stimulating an immunoglobulin-mediated response. This antagonistic activity could lead to increased Fc~y receptor expression and functional increase in `
opsonization and clearance of infective particles.
0 Antagonists to IL- 13, e.g., antibodies or IL-4 antagonist, would be indicated for modulating B cell growth and proliferation, perhaps reflecting excessive humoral responses. Various autoimmune conditions or hyperimmunoglobulinemias should respond to treatment with appropriate amounts of antagonists administered over defined schedules. IL-4 antagonist will be a preferred antagonist for IE-13 effects. ! ~}
IL-13 mediates changes in CD11 marker expression, which are ; -, associated with cell adhesion, e.g., cell-cell contacts, Thus, increasing j ~
CD11 should facilitate cellular interaction and the functional results I `
therefrom. See also Springer et al., Leukocyte Adhesion Molecules, ! `
1988, Springer-Verlag, New York.
Like IL-4, IL-13 induces IgG4 and IgE switching and IgG4 and IgE synthesis. IgE antibodies are major mediators~of allergic `
reactions. Allergen-specific antibodies of the IgE isotype have the 2s specific ability to bind to high affinity Fc receptors for IgE (FcRI) on ~- - mast cells and basophils. Binding of the relevant allergen to these , - receptor-bound IgE antibodies results in cross-linking of the receptor and activation of the mast cells and basophils. This results in! degranulation of these cells and the release of mediators of allergic reactions such as histamine, prostaglandins and proteases, ! `~
which cause immediate-type hypersensitivity reactions in the ~- - various target organs, e.g., nose airways, lungs, gut and skin.
In addition, IL-13 like IL-4 induces the expression of the low affinity receptor for IgE (FcRII, or CD23) on B cells and monocytes, and the subsequent release of a soluble form of CD23. Soluble CD23 ~`
enhances the production of IgE [see, e.g., Pene et al.7 Eur. J. Immunol. I

wos4/04~so Pcr/uss3/0764s 2142860 24 1 ~

18:929 (1988); Aubry et al., Nature 358:S05 (1992)]. Therefore, downregulation of IgE synthesis and soluble CD23 production reduce or inhibit IgE-mediated allergic diseases. IL-4 and/or IL- 13 antagonists such as antibodies, or IL-4 mutant proteins like Y 1 24D
s or similar mutant IL-13 proteins that compete for IL-4/IL-13 receptor binding, would be useful for blocking IgE production.
III. Nucleic Acids ~-This invention contemplates use of isolated nucleic acid or fragments which encode this or a closely related protein, or 0 fragments thereof, to encode a biologically active corresponding polypeptide. In addition, this invention covers isolated or -recombinant DNA which encodes a biologically active protein ;or polypeptide having characteristic IL- 13 activity. Typically, the - nucleic acid is capable of hybridizing, under appropriate conditions, with a nucleic acid sequence segment shown in Table 1.
Said biologically acdve protein or polypeptide can be a full length protein, or fragment, and will typically have a segment of r amino acid sequence highly homologous to one shown in Table 1.
Further, this invention covers the use of isolated or recombinant nucleic acid, or fragmonts thereof, which encode proteins having fragments which are homologous to the disclosed IL-13 protein. The isolated nucleic acids can have the respective regulatory sequences --in the 5' and 3' flanks, e.g., promoters, enhancers, poly-A addition signals, and others from the natural gene.
An "isolated" nucleic acid is a nucleic acid, e.g., an RNA, DNA, or ~
a mixed polymer, which is substantially pure, e.g., separated from other components which naturally accompany a native sequence, such as ribosomes, polymerases, and flanking genomic sequences --from the originating species. The term embraces a nucleic acid --sequence which has been removed from its naturally occurring environment, and includes recombinant or cloned DNA isolates, - -which are thereby distinguishable from naturally occurring compositions, and chemically synthesized analogues or analogues - -biologically synthesized by heterologous systems. A substantially 94/04680 ~? PCl /US93/07645 25 i~
S~ '~

pure molecule includes isolated forms of t~e molecule, either completely or substantially pure.
An isolated nucleic acid will generally be a homogeneous composition of molecules, but will, in some embodiments, contain heterogeneity, preferably minor. This heterogeneity is typically found at the polymer ends or portions not critical to a des*ed biological function or activity.
A "recombinant" nucleic acid is defined either by its method of production or its struc~ure. In reference to its method of production, 0 e.g., a product made by a process, the process is use of recombinant -nucleic acid techniques, e.g., involving human intervention in the nucleotide sequence. Typically this intervention involves in ~itro `;
manipulation, although under certain circumstances it may involve ``
more classical animal breeding t~chniques. .;
Alternatively, it can be a nucleic acid made by generating a sequence comprising fusion of two fragments which are not ! `
naturally contiguous to each other, but is meant to~xclude products of nature, e.g., naturally occurring mutants as found in their natural state. Thus, for example, products made by transforming cells with 2 o any unnaturally occurring vector is encompassed, as are nucleic acids comprising sequence derived using any synthetic oligonucleotide process. Such a process is often done to replace a codon with a redundant codon encoding the same or a conservative amino acid, while typically introducing or removing a restriction 2s enzyme sequence recognition site.
In still another altern~tive, the process is performed to join - together nucleic acid segments of desired functions to generate a single genetic entity comprising a desired combination of functions - - not found in the commonly available natural forms, e.g., encoding a 30 fusion protein. Restriction enzyme recognition sites are often the _ target of such artificial manipulations, -but other site specific targets, ;-~- ~ ~--~ e.g., promoters, DNA replication sites, regulation sequences, control sequences, or other useful features may be incorporated by design.
A similar concept is intended for a recombinant, e.g., fusion, 35 polypeptide. Specifically included` are synthetic nucleic acids which, by genetic code redundancy, encode similar polypeptides to WO g4/04680 PCl`/US93/~764"

2,3,~2~6~ -fragments of the interleukins, and fusions of sequences from various different interleukin or related molecules, e.g., growth factors.
A "fragment" in a nucleic acid context is a con~iguous segment of at least about 17 nucleotides, generally at least 20 nucleotides, more generally at least 23 nucleotides, ordinarily at least 26 nucleotides, more ordinarily at least 29 nucleotides, often at least 32 nucleotides, more often at least 35 nucleotides, typically at least 38 nucleotides, more typically at least 41 nucleotides, usually at least 44 nucleotides, more usually at least 47 nucleotides, preferably at 0 least 50 nucleotides, more preferably at least 53 nucleotides, and in I ;
particularly preferred embodiments will be at least 56 or more nucleotides. Typically, fragments of different genetic sequences can be compared to one another over appropriate length stretches.
A nucleic acid which codes for an IL-13 will be particularly useful to identify genes, mRNA, and cDNA species which code for the IL-13 or closely related proteins, as well as DNAs which code for allelic or other genetic variants, e.g., from different individuals.
Preferred probes for such screens are those regions of the interleukin which are conseNed between different allelic variants, and will preferably be full length or nearly so. In other situations, allele specific sequences will be more useful.
This invention further covers recombinant nucleic acid `
molecules and fragments ~àving a nucleic acid sequence identical to or highly homologous to the isolated DNA set forth herein. In 2s particular, the sequences will often be operably linked to DNA
segments which control transcription, translation, and DNA
replication. These àdditional segments typically assist in expression -of the desired nucleic acid segment.
!~ i ' I Homologous nucleic acid sequences, when compared to one another or Table 1 sequences, exhibit significant similarity. The standards for homology in nucleic acids are either measures for homology general~y used in the art by sequence comp`arison or based upon hybridization conditions. Comparative hybridization conditions are described in greater detail below.
3s "Substantial homology" in the nucleic acid sequence comparison context means either that the segments, or their 680 ~ PCl'tUS93/07645 ~'`
27 ~8~ ~

,, . . ~
complementary strands, when compared, are identical when optimally aligned, with appropriate nucleotide insertions or ~,-deletions, in at least about 60% of the nucleotides, generally at least ;
66%, ordinarily at least 71%, often at least 76%, more often at least 80%, usually at least 84%, more usually at least 88%, typically at least 91%, more typically at least about 93%, preferably at least about 95%, more preferably at least about 96 to 98% or more, and in ~-~
particular embodiments, as high at about 99% or more of the nucleotides . --0 Alternatively, substantial homology exists when the segments will hybridize under selective hybridizatian conditions, to a strand or its complement, typically using a sequence derived from Table 1.
Typically, selective hybridization will occur when there is at least .
about 55% homology over a stretch of at least about 14 nucleotides, ` --more typically at least about 65%, preferably at least about 7S%, and I -more preferably at least about 90%. See, Kanehisa, Nucleic Acids I `~
Res. 12:203 (1984). . ! ~-The length of homology comparison, as described, may be over longer stretches, and in certain embodiments will be over a stretch 20 of at least about 17 nucleotides, generally at least about 20 nucleotides, ordinarily at least about 24 nucleotides, usually at least about 28 nucleotides, typically at least about 32 nucleotides, more `
-- - typically at least about 40 nucleotides, preferably at least about 50 !.' ~~ nucleotides, and more preferably at least about 75 to 100 or more `
2 5 _nucleotides.
Stringent conditions, in referring to homology in the - --hybridization context, will be stringent combined conditions of salt, -temperature, organic solvents, and other parameters typically controlled inl hybridization reactions. Stringent temperaturel ~ 30 conditions will usually include temperatures in excess of about 30C, -- -- --_- more usually in excess of about 37C, typically in excess of about .
45C, more typically in excess of about 55C, preferably in excess of I:
~- - about 65C, and more preferably in excess of about 70C. Stringent - salt conditions will ordinarily be less than about 1000 mM, usually 35 less than about 500 mM, more usually less than about 400 mM, typically less than about 300 mM, preferably less than about 200 WO 94/04680 Pcr/uss3/o764 21~2860 28 mM, and more preferably less than about 150 mM. However, the combination of parameters is much more important than the ;
measure of any single parameter. See, e.g., Wetmur et al., J. Mol.
Biol. 31:349 (1968).
The isolated DNA can be readily modified by nucleotide substitutions, nucleotide deletions, nucleotide insertions, and inversions of nucleotide stretches. These modifications result in -novel DNA sequences which encode this protein, its derivatives, or proteins having IL- 13 activity. These modified sequences can be 0 used to produce mutant proteins (muteins) or to enhance the -expression of variant species. Enhanced expression may involve gene amplification, increased transcription, increased translation, and other mechanisms. Such mutant I~- 13 derivatives include predetermined or site-specific mutations of the protein or its fragments.
"Mutant IL-13" as used herein encompasses a polypeptide otherwise falling within the homology definition of the human IL-13 as set forth above, but having an amino acid sequence which differs from that of human IL-13 as found in nature, whether by way of deletion, substitution, or insertion. In particular, "site specific mutant lL- 13 " encompasses a protein having substantial homology with a protein of Table 1, and typically shares most of the biological activities of the form disclosed herein.
Although site specific mutation sites are predetermined, 2 5 mutants need not be site specific. Human IL- 13 mutagenesis can be achieved by making amino acid inserdons or deletions in the gene, --coupled with expression. Substitutions, deletions, insertions, or any combinations may be generated to arrive at a final construct.
Insertions include amino- or carboxy- tèrminal fusions. Random mutagenesis can be conducted at a target codon and the expressed- ~
human IL-13 mutants can then be screcned for the desired activity.
Methods for making substitution mutations at predetermined sites ~=- ~~`~
in DNA having a known sequence are well known in the art, e.g., by M13 primer mutagenesis. See also Sambrook et al. (1989) and 3s Ausubel et al. (1987 and periodic Supplements).

94/04~80 21~ 2 8 6 G pcrtus93/0764~

The mutations in the DNA normally should not place eoding sequences out of reading frames and preferably will not create complementary regions that could hybridize to produce secondary mRNA structure such as loops or hairpins.
The phosphoramidite method described by Beaucage et al., Tetra. Letts. 22:1859 (1981), will produee suitable synthetic DNA
fragments. A double stranded fragment will often be obtained either by synthesizing the complementary strand and annealing the strand together under appropriate conditions or by adding the 0 complementary strand using DNA polymerase with an appropriate primer sequence.
Polymerase chain reaction (PCR) techniques can often be applied in mutagenesis. Alternatively, mutagenisis primers are ~-commonly used methods for generating defined mutations at `
predetermined sites. -IV. Proteins~ Pe~tides ~-As described above~ the present invention encompasses the human IL-13 whose sequence is disclosed in Table I and desc~ibed above. Allelic and other variants are also contemplated.
"Substantially pure", in the polypeptide context, typieally means that the protein is free from other c~ntaminating proteins, ~~ nucleic acids, and other biologicals derived from the original source organism. Purity may be assayed by standard methods, and will - - ordinarily be at least about 40~o pure, more ordinarily at least about - --2s- 50% pure, generally at least about 60% pure, more generally at least about 70% pure, often at least about 75% pure, more often at least about 80% pure, typically at least about 85% pure, more typically at ~ ~ least about 90% pure, preferably at least about 95% pure, more - preferably at least about 98% pure, and in most preferred -~ - 30 embodiments, at least 99% pure. The analysis may be weight or --- molar percentages, evaluated, e.g., by gel staining, spectrophotometry, or terminus labelling.
The present invention also provides recombinant proteins, e.g., heterologous fusion proteins using segments from this human WO 94~04680 PCI`/US93/0764' ` `

protein. A he~erologous fusion protein is a fusion~of proteins or segments which are natur~ly not normally fused in the same manner. Thus, the fusion product of a growth factor with an ~ `
interleukin is a continuous protein molecule having sequences fused s in a typical peptide linkage, typically made as a single translation `
product and exhibiting properties derived from each source peptide.
A similar concept applies to heterologous nucleic acid sequences.
In addition, new constructs may be made from combining similar functional or structural domains from other related proteins, 0 e.g., growth factors or other cytokines. For example, receptor-binding or other segments may be " swapped" between different new fusion polypeptides or fragments. See, e.g., Cunningham et al., Science 243:1330 (1989); and O'Dowd et al., J. Biol. Chem. 263:15985 ( 1 988).
Thus, new chimeric polypeptides exhibiting new combinations - ~ of specificities will result from the functional linkage of - receptor-binding specificities. For example, the receptor binding domains from other related ligand molecules may be added or substituted for other domains of this or related proteins. The resulting protein will often have hybrid function and properties. For example, a fusion protein may include a targetting domain which may serve to provide sequestering of the fusion protein to a particular organ, e.g., - ~ a ligand portions which- is specifically bound by spleen cells and would serve to accumulate in the spleen.
Candidate fusion partners and sequences can be selected from various sequence data bases, e.g., GenBank, c/o IntelliGenetics, ^ -= -Mountain View, CA; and BCG, University of Wisconsin Biotechnology Computing Group, Madison, WI.
f' i ' nDerivativès" of the human IL-13 include amino acid sequence mutants, glycosylation variants, metabolic derivatives and covalent or aggregative conjugates with other chemical moieties. Covalent derivatives can be prepared by linkage of functionalities to groups ~
which are found in the II,-13 amino acid side chains or at the N- or ~ ~ ~
C- termini, e.g., by means which are well known in the art. These derivatives can include, without limitation, aliphatic esters or amides of the carboxyl terminus, or of residues containing carboxyl ':

~94/04680 31 21 ~2~G PCI'/US93/0764~

.
:`
side chains, O-acyl derivatives of hydroxyl group-containing residues, and N-acyl derivatives of the amino term inal amino acid or amino-group containing residues, e.g., lysine or a rginine. Acyl groups are selected from the group of alkyl-moieti es including C3 to s C18 normal alkyl, thereby forming alkanoyl aroyl species.
In particular, glycosylation alterations are included, e.g., made ~-by modifying the glycosylation patterns of a polypeptide during its synthesis and processing, or in further processing steps. Particularly preferred means for accomplishing this are by exposing the 0 polypeptide to glycosylating enzymes derived from cells which normally provide such processing, e.g., mammalian glycosylation ¢^
enzymes. Deglycosylation enzymes are also contemplated. Also embraced are versions of the same primary amino acid sequence -~
which have other minor modifications, including phosphorylated amino acid residues, e.g., phosphotyrosine, phosphoserine, or phosphothreonine .
A major group of deri~atives are covalent conjugates of the interle~lkin or fragments thereof with other proteins of polypeptides. These denvatives can be synthesized in recombinant culture such as N- or C-terminal fusions or by the use of agents known in the art for their usefulness in cross-linking proteins through reactive side groups. Preferred derivatization sites with cross-linking agents are at free amino groups, carbohydrate moieties, and cysteine residues.
2s Fusion polypeptides between the interleukin and other ~- homologous or heterologous proteins are also provided. Homologous polypeptides may be fusions between different growth factors, resulting in, for instance, a hybrid protein exhibiting ligand specificity for multiple different receptors, or a ligand which may - 30 have broadened or weakened specificity of binding to its receptor.
- _ Likewise, heterologous fusions may be constructed which would :~` ~ ~ exhibit a combination of properties or activities of the derivative proteins.
Typical examples are fusions of a`reporter polypeptide, e.g., 35 luciferase, with a segment or domain of a receptor, e.g., a ligand-binding segment, so that the presence or location of a desired ligand WO 94~04680 PCr/US93/0764~
2 ~ 4 2 ~

may be easily determined. See, e.g., Dull et al., U.S. Patent No.
4,859,609. Other gene fusion partners include glutathione-S-transferase (GST), bacterial B-galactosidase, trpE, Protein A, ~-lactamase, alpha amylase, alcohol dehydrogenase, and yeast alpha s mating factor. See, e.g., Godowski et al., Science 241:812 (1988).
The phosphoramidite method' described by Beaucage et al., Tetra. Letts. 22:1859 (1981), will produce suitable synthetic DNA
fragments. A double-stranded fragment will often be obtained either by synthesizing the complementary strand and annealing the 0' strand together under appropriate conditions or by adding the complementary strand using DNA polymerase with an appropriate primer sequence.
Such polypeptides may also have amino acid residues which have been chemically modified by phosphorylation, sulfonation, 15 biotinylation, or the addition or removal of other moieties, particularly those which have molecular shapes similar to phosphate groups. In some embodiments, the modifications will be useful labelling reagents, or serve as purification targets, e.g., affinity ligands.
20 ~ Fusion proteins will typically be made by either recombinant ~' nucleic acid methods or by synthetic polypeptide methods.
Techniques for nucleic àcid manipulation and expression are described generally, for example, in Sambrook et al., Molecular Cloning: A Laboratory ~anual (2d ed.), 1989,Vols. 1-3, Cold Spring 25 Harbor Laboratory; and Ausubel et al. (eds), Current Protocols in - Molecular Biology, 1987 and periodic supplements, Greene/Wiley, New York. Techniques for synthesis of polypeptides are described, ~ -~' for example, in Merrifield, J. Am. Chem. Soc. 85:2149 (1963);
Merrifield, Science 232:34i (1986); and Atherton et al., Solid Phase 30 Peptide Synthesis: A Practical Approach, 1989, IRL Press, Oxford.
This invention also contemplates the use of derivatives of the human IL-13 other than variations in amino acid sequence or ~~
- ~ ~ glycosylation. Such derivadves may involve covalent or aggregative associadon with chemical moieties. These derivatives generally fall -'~
35 into three classes: (1) salts, (2) side chain and terminal residue covalent modifica~ions, and (3) adsorption complexes," for example ~' :~ ' .','.
,.

.

~ 94/04680 PCrlUS93/0764~ ~
33 21~286~

= , with cell membranes. Such covalent or aggregative derivatives are useful as immunogens, as reagents in immunoassays, or in purification methods such as for affinity purification of a receptor or other binding molecule, e.g., an antibody. ~- -For example, the human IL- 13 ligand can be immobilized by covalent bonding to a solid support such as cyanogen bromide-activated Sepharose, by methods which are well known in the art, or -;
adsorbed onto polyolefin surfaces, with or without glutaraldehyde cross-linking, for use in the assay or purification of IL- 13 receptor, o antibodies, or other similar molecules. The IL-13 can also be labelled with a detectable group, for example radioiodinated by the chloramine T procedure, covalently bound to rare earth chelates, or I i conjugated to another fluorescent moiety for use in diagnostic assays.
lS The human IL-13 of this invention can be used as an - I
immunogen for the production of antisera or antibodies specific for ¦
the interleukin or any fragments thereof. The purified interleukin can be used to screen monoclonal antibodies or antigen-binding I ;
fragments prepared by immunization with various forms of impure 2 o preparations containing the protein. In particular, the term "antibodies" also encompasses antigen binding fragments of natural antibodies. The purified interleukin can also be used as a reagent to - -detect any antibodies ~ generated in response to the presence ofelevated levels of expression, or immunological disorders which lead 2s to antibody production to the endogenous cytokine.
- Additionally, IL-13 fragments may also serve as immunogens to produce the antibodies of the present invention, as described immediately below. For example, this invention contemplates - --- - antibodies having binding affinity to or being raised against the - --30 amino acid sequence shown in Table 1, fragments thereof, or - homologus peptides. In particular, this invention contemplates ^
antibodies having binding affinity to, or having been raised against, specific fragments which are predicted to be, or actually are, - exposed at the exterior protein surface of the native cytokine.
3s The blocking of physiological response to these interleukins ;
may result from the inhibition of binding of the ligand to the ..
.

Wo 94/04680 Pcr/US93/0764 21~2860 receptor, likely through competitive inhibition. Thus, in vitro assays of the present invention will often use antibodies or ligand binding segments of these antibodies, or fragments attached to solid phase substrates. These assays will also allow for the diagnostic determination of the effects of either binding region mutations and modifications, or ligand mutations and modifications, e.g., ligand analogues.
This invention also contemplates the use of competitive drug screening assays, e.g., where neutralizing antibodies to the 0 interleukin or fragments compete with a test compound for binding to a receptor or antibody. In this manner, the neutralizing antibodies or fragments can be used to detect the presence of any J
polypeptide which shares one or more binding sites to a receptor and can also be used to occupy binding sites on a receptor that 5 might otherwise bind an interleukin.
V. Makin~ Nucleic Acids and Protein DNA which encodes the protein or fragments thereof can be obtained by chemical synthesis, screening cDNA libraries? or by screening genomic libraries prepared from a wide variety of cell 20 lines or tissue samples. Natural sequences can be isolated using standard methods and the sequences provided herein, e.g., in Table l.
This DNA can be expressed irl a wide variety of host cells for - -the synthesis of a full-length human interleukin or fragments which `
can in turn, for example, be used to generate polyclonal or 2s monoclonal antibodies; for binding studies; for construction and ~ -~
expression of modified agonist/antagonist molecules; and for strlucture/function studies. Each variant or its fragments can~be ~expressed in host cells that are transformed or transfected with appropriate expression vectors. These molecules can be substantially free of protein or cellular contaminants, other than ; thosc derived from thc recombinant host, and therefore are particularly useful in pharmaceutical compositions when combined with a pharrnaceutically acceptable carrier and/or diluent. The I

: :.

~94/04680 35 2~8S;~PCr ~;

human protein, or portions thereof, may be expressed as fusions with other proteins.
Expression vectors are typically self-replicating DNA or RNA
constructs containing the desired receptor gene or its fragments, 5 usually operably linked to suitable genetic control elements tha~ are recognized in a suitable host cell. These control elements are capable of effecting expression within a suitable host. The specific type of control elements necessary to ef~ect expression will depend upon the eventual host cell used. -0 Generally, the genetic control elements can include a prokaryotic promoter system or a eukaryotic promoter expression control system, and typically include a transcriptional promoter, an optional operator to control the onset of transcription, transcription enhancers to elevate the level of mRNA expression, a sequence that -encodes a ~uitable ribosome binding site, and sequences that ~ ;;
terminate transcription and translation. Expression vectors also usually contain an origin of replication that allows the vector to replicate independently of the host cell.
The vectors of this invention include those which contain DNA
which encodes a protein, as described, or a fragment thereof encoding a biologically active equivalent polypeptide. The DNA can be under the control of a viral promoter and can encode a selection marker. This invention further contemplates use of~such expression vectors which are capable of expressing eukaryodc cDNA coding for such a protein in a prokaryodc or eukaryotic host, where the vector is compatible with the host and where the eukaryotic cDNA coding for the receptor is inserted into the vector such that growth of the host containing the vector expresses the cDNA in question.
L' ~ Usually, expression vectors are designed for stable replication -- - 30 in their host cells or for amplification to greatly increase the total - _ number of copies of the desirable gene per cell. It is not always~_ ~ necessary to require that an expression vector replicate in a host cell, e.g., it is possible to effect transient expression of the interleukin protein or its fragments in various hosts using vectors 35 that do not contain a replication origin that is recognized by the host .

WO 94/04680 PCI`/US93/~)764 21~2860 cell.- It is also possible to use vectors that cause integration of the human protein or its fragments into the host DNA by reeombination.
Vectors, as used herein, comprise plasmids, viruses, bacteriophage, integratable DNA fragments, and other vehicles - ~`
which enable the integration of DNA fragments into the genome of the host. Expression vectors are specialized vectors which contain genetic control elements that effect expression of operably linked genes. Plasmids are the most commonly used form of vector but all ~
other forms of vectors ~,vhich serve an equivalent function and ~:
which are, or become, known in the art are suitable for use herein.
See, e.g., Pouwels e~ al., Cloning Vectors: A Laboratory Manual, 1985 and Supplements, Elsevier, N.Y.; and Rodriquez et al. (eds), Vectors~
A Survey of Molecular Cloning Vectors and Their Uses, 1988, ~ ~ :
Buttersworth, Boston. ` - ;
. ~ .
Transformcd cells are cells, preferably mammalian, that have been transformed or~ transfected with receptor vectors constructed using recombinant DNA techniques. Transformed host cells usually express; the ~desired protein or its fragments, but for purposes of cloning,~ amplifying, and manipulating its DNA, do not need to 20 ~ express the~ subject protein. This invention further contemplates cul~uling transformed ce!ls in a nutrient medium, thus permitting ,i;
the~ interleukin~ to accumulate in the culture. The protein can be recovered, ~ei~thèr from,the culture or from the culture medium. ,-For;~ purposes of this invendon, nucleic sequences are operably ~`
2s ~linked when~lhey are functionally related to each other. For example, DNA~ for~ a~presequènce or secretory leader is operably linked~ to a~ polypeptide if it is expressed as a preprotein or ~ ~~
participates in directing the polypeptide to the cell membrane or in '`; ~ sécredon of th~ polypcptide. A promoter is operably linked to a - -3 0 ~coding sequence if it controls the transcription of the polypeptide;- a ribosome binding sitc is operably linked to a coding sequence if it is _ .
` positioned to pennit translation. Usually, opcrably linked means ~ ~
~ . .
condguous and in reading frame, howe~er, certain genetic elements such às ropressor gcnes are not contiguously linked but still bind to 3~5 operator sequences that in turn control e~pression.
.~.
,~., ~.
-.

~0 94/04680 ~? PCl`/US93J0764~ ~
~8~

Suitable host cells include prokaryotes, lower eukaryotes, and higher eukaryotes. Prokaryotes include both gram negative and gram positive organisms, e.g., E. coli and B. subtilis Lower eukaryotes include yeasts, e.g., S. cerevisiae and Pichia, and species 5 of the genus Dictyostelium. Higher eukaryotes include established tissue culture cell lines from animal cells, both of non-mammalian origin, e.g., insect cells, and birds, and of mammalian origin, e.g., human, primates, and rodents. -Prokaryotic host-vector systems include a wide variety of 0 vectors for many different species. As used herein, E. coli and its vectors will be used generically to include equivalent vectors used ' `
in other prokaryotes. A representative vector for amplifying DNA is pBR322 or many of its derivatives. Vectors that can be used to express the receptor or its fragments include, but are not limited to, 5 such vectors as those containing the lac promoter (pUC-series); trp promoter (pB~322-trp); Ipp promoter (the pIN-series); lambda-pP
or pR promoters (pOTS); or hybrid promoters such as ptac (pDR540).
See Brosius et al., "Expression Vectors Employing Lambda-, trp-, lac-, and Ipp-derived Promoters", in Vectors: A Sun/ey of Molecular I:
20 Cloning Vectors and Their Uses, (eds. Rodriguez and Denhardt), 1988, Buttersworth, Boston, Chapter 10, pp. 205-236.
Lower eukaryotes, e.g., yeasts and Dictyostelium. may be ---~ transformed with IL- 13 sequence containing vectors . For purposes of this invention, the most common lower eukaryotic host is the - -- 2s - baker's yeast, Saccharomyces cerevisiae. It will be used to generically represent lower eukaryotes although a number of other strains and species are also available. Yeast vectors typically consist of la replication origin (unless of the integrating type~, a selection --- gene, a promoter, DNA encoding the receptor or its fragments, and 30 sequences for translation termination, polyadenylation, and transcription termination.
- - Suitable expression vectors for yeast include such constitutive promoters as 3-phosphoglycerate kinase and various other glycolytic enzyme gene promoters or such inducible promoters as 3 5 the alcohol dehydrogenase 2 promoter or metallothionine promoter.
Suitable vectors include derivatives of the following types: self-wo 94/04680 - Pcr/US93/0764 ~

6~ 38 replicating low copy number (such as the YRp-series), self-replicating high copy number (such as the YEp-series); integrating types (such as the YIp-series), or mini-chromosomes (such as the YCp-series).
s Higher eukaryotic tissue culture cells are normally the preferred host cells for expression of the functionally active interleukin protein. In principle, any higher eukaryotic tissue culture cell line is workable, e.g., insect baculovirus expression systems, whether from an invertebrate or vertebrate source. ~
0 However, mammalian cells are preferred. Transformation or ;-`
transfection and propagation of such cells has become a routine procedure. Examples of useful cell lines include HeLa cells, Chinese hamster ovary (CHO) cell lines, baby rat kidney (BRK) cell lines, insect cell lines, bird cell lines, and monkey (COS) cell lines.
Expression vectors for such cell lines usually include an origin of replication, a promoter, a translation initiadon site, RNA splice -sites (if genomic DNA is used), a polyadenylation site, and a transcnption termination site. These vectors also usually contain a selection gene or amplification gene. Suitable expression vectors may be plasmids, viruses, or retroviruses carrying promoters derived, e.g., from such sources as from adenovirus, SV40, parvoviruses, vaccinia virus, or cytomegalovirus. Representative examples of suitable expression vectors include pCDNA1; pCD, see~
Okayama e~ al., Mol. Cell Biol. 5:1136 (1985); pMClneo PolyA, see Thomas et al., Cell 51:503 (1987); and a baculovirus vector such as _ `--pAC 373 or pAC 610. ~
Por secreted proteins, an open reading frame usually encodes a ~
polypeptide that consists of a mature or secreted product covalently 1~, t linked at istiN-terminus to a signal peptide. The signal pepdde is --cleaved prior to secretion of the mature, or active, polypeptide. The cleavage site can be predicted with a high degree of accuracy from _ cmpirical rules, e.g., von Heijne, Nucleic Acids Research 14:468~
(1986), and the precise amino acid composition of the signal peptide does not~ appear to be critical to its function, e.g., Randall et al.
-~ ~ 35 Science 243:1156 (1989); Kaiser e~ al., Science 235:312 (1987).

. ~ .

~094/04680 39 1~8~o PCI/US93/07645 It will often be desired to express these polypeptides in a system which provides a specific or defined glycosylation pattern.
In this case, the usual pattern will be that provided naturally by the expression system. However, the pattern will be modifiable by 5 exposing the polypeptide, e.g., an unglycosylated form, to appropriate glycosylating proteins introduced into a heterologous expression system. For example, the interleukin gene may be co-transformed with one or more genes encoding mammalian or other -glycosylating enzymes. Using this approach, certain mammalian 0 glycosylation pattems will be achievable in prolcaryote or other cells.
The source of human IL-13 can be a eukaryotic or prokaryotic host expressing recombinant huI~- 13 DNA, such as is described ~-above. The source can also be a cell line such as mouse Swiss 3T3 5 fibroblasts, but other mammalian cell lines are also contemplated by this inven~ion, with the preferred cell line being from the human ~ .
species.
Now that the entire sequence is known, human IL- 13, fragments, or derivatives thereof can be prepared by conventional 2 o processes for synthesizing peptides. These inciude processes such as are described in Stewart et al., Solid Phase Peptide Synthesis, 1984, Pierce Chemical Co., Rockford, IL; Bodanszky e~ al., The Practice of Peptide Synthesis, 1984, Springer-Verlag, New York; and Bodanszky, The Principles of Peptide Synthesis, 1984, Springer-Verlag, New 25 York. For example, an azide process, an acid chloride process, an acid anhydride process, a mixed anhydride process, an active ester process (for example, p-nitrophenyl ester, N-hydroxysuccinimide ester, or cyanomethyl ester), a carbodiimidazole process, an oxidative-reductive process, or a dicyclohexylcarbodiimide 3o (DCCD)/additive process can be used. Solid phase and solution phase syntheses are both applicable to the foregoing processes.
The IL-13 protein, fragments, or derivatives are suitably prepared in accordance with the above processes as typically employed in peptide synthesis, generally either by a so-called 3 5 stepwise process which comprises condensing an amino acid to the terminal amino acid, one by one in 5equence, or by coupling peptide WO 94~04680 . PCI`/US9~/0764 6'~ ~

fragments tO the tenninal amino acid. Amino groups that are not being used in the coupling reaction typically must be protected to prevent coupling at an incorrect location.
If a solid phase synthesis is adopted, the C-terminal amino 5 acid is bound to an insoluble carrier or support through its carboxyl group. The insoluble carrier is not particularly limited as long as it -has a binding capability to a reactive carboxyl group. Examples of such insoluble carriers include halomethyl resins, such as chloromethyl resin or bromomethyl resin, hydroxymethyl resins, 10 phenol resins, tert-alkyloxycarbonylhydrazidated resins, and the like. ` :
An amino group-protected amino acid is bound in sequence through condensation of its activate~ carboxyl group and the reactive amino group of the previously formed peptide or chain? to synthesize the peptide step by step. After synthesi~ing the complete sequence, the peptide is split off from the insoluble carrier to produce the peptide. This solid-phase approach is generally described by Merrifield et al., in J. Am. Chem. Soc. 85:2149 (1963).
The prepared protein and fragments thereof can be isolated ¦ -2 o and purified from the reaction mixture by means of peptide I ~-separation, for example, by extraction, precipitation, electrophoresis, various forms of chromatography, and the like. The interleukin of this invention can be obtained in varying degrees of purity depending upon its desired use. Purification can be accomplished by !:
2 s use of the protein purification techniques disclosed herein or by the use of the antibodies herein described in methods of immunoabsorbant affinity chromatography. This immunoabsorbant affinity chromatography is carried out by firs~ linking the antibodies ' ' to a solid suppdrt and then contacting the linked antibodies with solubilized lysates of appropriate cells, lysates of other cells`
- expressing the interleukin, or lysates or supernatants of cells producing the protein as a result of DNA techniques, see--berow.
Generally, the puriffed protein will be at least about 40% pure, ordinarily at least about 50% pure, usually at least about 60% pure, typically at least about 70% pure, more typically at least about 80%
pure, preferable at least about ~0% pure and more preferably at ~ g4/04680 ~.7 PCI`/US93/07645 41 ~8~ ::
b' least about 95% pure, and in particular embodiments, 97%-99% or more. Purity will usually be on a weight basis, ~ut can also be on a molar basis. Different assays will be applied as appropriate.
VI. Antibodies Antibodies can be raised to the various human IL-13 proteins and fragments thereof, both in natu~ally occurring native fo~ms and -in their recombinant formst the difference being that antibodies to the active ligand are more likely to recogni~e epitopes which are only present in the native conformations. Anti-idiotypic antibodies 0 are also contemplated, which would be useful as agonists or antagonists of a natural receptor or an antibody.
Antibodies, including binding fragments and single cha;in versions, against predetermined *agments of the protein can be raised by immunization of animals with conjugates of the fragments with immunogenic proteins . Monoclonal antibodie~ are prepared from cells secreting the desired antibody. These antibodies can be screened for binding to norrnal or defective protein, or screened for agonistic or antagonistic activity. These monoclonal antibodies will_ usually bind with at least a KD of about 1 mM, more usually at least about 300 IlM, typically at least about 100 ~lM, more typically at least about 30 IlM, preferably at least about 10 ~M, and more preferably at least about 3 ~M or better.
The antibodies, including antigen binding fragments, of this invention can have significant diagnostic or therapeutic value. They 2s can be potent antagonists that bind to the interleukin and inhibit binding to the receptor or inhibit the ability of huamn IL-13 to elicit a,biological response. They also can be useful as non-neutralizing antibodies and can be coupled to toxins or radionuclides to bind producing cells, or cells localized to the source of the interleukin.
Further, these antibodies can be conjugated to dNgs or other therapeudc agents, either directly or indirectly by means of a linker.
The antibodies of this invention can also bç useful in diagnostic applications. As capture or non-neutralizing antibodies, they can bind to the interleukin without inhibiting receptor binding.

WO 94/04680 PCr/US93/0764:

As neutralizing antibodies, they can be useful in competitive binding assays. They will also be useful in detecting or quantifying IL- 1 3 . ~;
- Protein fragments may be joined to other materials, particularly polypeptides, as fused or covalently joined polypeptides to be used as immunogens. The human IL- 13 and its fragments may be fused or covalently linked to a variety of immunogens, such as keyhole limpet hemocyanin, bovine serum albumin, tetanus toxoid, etc. See Microbiology, Hoeber Medical Divisian, Harper and Row, 1969; Landsteiner, Specificity of Serological Re~ctions, 1962, 0 Dover Publications, New York; and Williams et al., Methods in Immunology and Immunochemistry, 1967, Vol. 1, Academic Press, New York, for descriptions of methods of preparing polyclonal antisera. A typical method involves hyperimmunization of an animal with an antigen. The blood of the animal is then collected shortly after the repeated immunizations and the gamma globulin is ~ `
isolated.
In some instances, it iS desirable tO prepare monoclonal antibodies from various mammalian hosts, such as mice, rodents, primates, humans, e~c. Description of techniques for preparing such monoclonal antibodies may be found in, e.g., Stites et al. (eds), Basic .
and Clinical Immunology (4th ed.), Lange Medical Publications, Los Altos, CA, and references cited therein; Harlow et al., Ant~bodies:. A ``
Laboratory Manual, 1988, CSH Press; Goding, Monoclonal An~bodies:
Principles and Practice (2d ed), 1986, Academic Press, New York;
2s and particularly in Kohler and Milstein, Nature 256:495 (1975), which discusses one method of generating monoclonal antibodies. --~
Briefly, this method involves injecting an animal with an - ~-immunogen. The animal is then sacrificed and cells taken from its spleen, which are then fused with myeloma cells. The result is a hybrid cell or "hybridoma" that is capable of reproducing in vitro.
The population of hybridomas is then screened to isolate individual `
clones, each of which secrete a single antibody species to the~~~: ~
immunogen. In this manner, the individual antibody species obtained are th~ produc~s of immortalized and cloned single B cells 3 5 from the immune animal generated in response to a specific site recognized on the immunogenic substance.

.~94/04680 ~8~5'o Pcr/US93/0764s Other suitable techniques involve in Yitro exposure of lymphocytes to the antigenic p~lypeptides or alternatively ~o selection of libraries of antibodies in phage or similar vectors. See, Huse et al., Science 246:1275 (1989); and Ward et al., Natl~re 341:544 (1989). The polypeptides and antibodies of the present invention may be used with or without modification, including chimeric or humanized antibodies.
Frequently, the polypeptides and antibodies will be labelled by joining, either covalently or non-covalently, a substance which 0 provides for a detectable signal. A wide variety of labels and conjugation techniques are known and are reported extensively in both the scientific and patent literature. Suitable labels include radionuclides, enzymes, substrates, cofactors, inhibitors, fluorescent moieties, chemiluminescent moieties, magnetic particles, and the like. Patents teaching the use of such labels include U.S. Patent Nos.
3,817,837; 3,850,752; 3,939,350; 3,996,345; 4,277,437; 4,275,149;
and 4,366,241. Also, recombinant or chimeric immunoglobulins may be produced, see Cabilly, U.S. Patent No. 4,816,567.
The antibodies of this invention can also be used for af~lnity chromatography in isolating the IL-13. Columns can be prepared ;
where the antibodies are linked to a solid support, e.g., particles, such as agarose, Sephadex, or the like, where a cell lysate may be passed through the column, the column~~washed, followed by increasing concentrations of a mild denaturant, whereby the purified protein will be released.
The antibodies may also be used to screen expression libraries for particular expression products. Usually the antibodies used in such a procedure will be labelled with a moiety allowing easy '~ detection of presence of antigen by antibody binding.
Antibodies raised against cach human IL-13 will also be used to raise anti-idiotypic antibodies. These will be useful in detecting or diagnosing various immunological conditions related to expression of the protein or cells which express receptors for the protein. They also will be useful as agonists or antagonists of the interleukin, 3s which may be competitive inhibitors or substitutes for naturally occurring ligands.
I

W0 94t04680 PCI /US93/0764' : ~:4~ :
~4~,~6~

VII. Use~ of IL-13 Compositions Nu~leic_acids Both naturally occurring and recombinant forms of the human Interleukin-13 molecules of this invention are particularly useful in ~
kits and assay methods. For example, these methods would also be ~-s applied to screening for binding activity to these proteins. Several methods of automating assays have been developed in recent years so as to permit screening of tens of thousands of compounds per year. See, e.g, a BIOMEK automated workstation, Beclcman Insauments, Palo Alto, California, and Fodor e~ al., Science 251:767 0 (1991). The latter describes means for testing binding by a plurality of defined polymers synthesized on a solid substrate. The development of suitable assays to screen for a receptor or agonist/antagonis~ homologous proteins can be greatly facilitated by the availability of large amounts of purified, soluble interleukin in ~-~
5 an active state such as is provided by this invention.
Rational dmg design may also be based upon structural s~udies of the molecular shapes of a receptor or antibody and other -~
effectors or ligands. Effectors may be other proteins which mediate other functions in response to ligand binding, or other proteins 2 o which normally interact with the receptor. One means for determining which sites interact with specific other proteins is a physical structure determination, e.g., x-ray crystallography or 2 dimensional NMR techniques. These will provide guidance as to which amino acid residues form the molecular contact regions. For a detailed description of protein structural determination, see,- e.g.,- ` ~~ ~
Blundell et al., Protein Crystallography, (1976) Academic Press, New- ~
York.
Purified interleukin-13 can be coated directly onto platès for use in the aforementioned receptor screening techniques. Ho~,vever, non-neutralizing antibodies to these proteins can be used as capture antibodies to immobilize the respective interleukin on the--soii~
phase, useful, e.g., in diagnostic uses. - ~~ ~
This invention also contemplates use of interleukin-13, ---fragments thereof, peptides, and their fusion produ~ts in a variety of diagnostic kits and methods for detecting the presence of the protein I `

94t04680 ;~,;r PCI/US93/07645 8~o or its receptor. Alternatively, or additionally, antibodies against the-molecules may be incorporated into the kits and methods. Typically the kit will have a compartment containing either a defined IL-13 peptide or gene segment or a reagent which recognizes one or the s other. Typically, recognition reagents, in the case of peptide, would be a receptor or antibody, or in the case of a gene segment, would be -a probe.
A preferred kit for determining the concentration of, for example, IL-13, a sample would typically comprise a labelled o compound, e.g., receptor or antibody, having known binding affiniey for IL-13, a source of IL-13 (naturally occurring or recombinant) as a positive control, and a means for separating the bound from free labelled compound, for example a solid phase for immobilizing the IL-13 in the test sample. Compartments containing reagents, and instructions, will normally be provided.
Antibodies, including antigen binding fragments, specific for lL.-13 or a` peptide fragment, or receptor fragments~re useful in diagnostic applications to detect the presence of elevated levels of IL-13 andjor its fragments. Diagnostic assays may be homogeneous ' 2 0 (without a separation step between free reagent and antibody-antigen complex) or heterogeneous ~with a separation step). Various commercial assays exist, such as radioimmunoassay (RIA), enzyme-linked immunosorbent assay (ELISA), enzyme immunoassay (EIA), enzyme-multiplied immunoassay technique (EMIT?, substrate-2s labelled~fluorescent immunoassay (SLFIA) and the like.
~- For example, unlabelled antibodies can be employed by using a~ second antibody which is labelled and which recognizes the andbody to IL-13 or to a particular fragment thereof. These assays , ihavd à!so been extensively discussed in the literature. See, e.g., 30 Harlow et al., Antibodies: A Laboratory Manual, 1988, CSH; and Coligan ~Ed.), Current Protocols ln Immunology, 1991 and periodic supplements, Greene/Wiley, New York.
Anti-idiotypic antibodies may have similar use to serve as agonists or antagonists of IL-13. These should be useful as 3s therapeutic reagents under appropriate circumstances.

WO 94/04680 PCr/US93/0764' ` ~-~

6 ~ `

Frequently, the reagents for diagnosti~ assays are supplied in lcits, so as to optimize the sensitivity of the assay. For the subject invention, depending upon the nature of the assay, the protocol, and the label, either labelled or unlabelled antibody, or labelled receptor ~;
is provided. This is usually in conjunction with other additives, such `:7''~``~' as buffers, stabilizers, materials necessary for signal production such as substrates for enzymes, and the like. Preferably, the kit will also contain instructions for proper use and disposal of the contents after use. Typically the kit has compartments for each useful reagent. ~;~
0 Desirably, the reagents are provided as a dry lyophilized powder, i .`~
where the reagents may be reconstituted in an aqueous medium 1 "
having appropriate concentrations for performing the assay. i ;
Any of the constituents of the diagnostic assays may be used without modification or may be modified in a variety of ways. For -~-example, labelling may be achieved by covalently or non-covalently joining a moiety which directly or indirectly provides a detectable signal. In any of these assays, a test compound, IL-13, or antibodies thereto can be labelled either directly or indirectly. Possibilities for direct labelling include label groups: radiolabels such as 125I, ~20 enzymes (U.S. Pat. No. 3,645,090) such as peroxidase and alkaline phosphatase, and lluorescent labels (U.S. Pat. No. 3,940,475) capable `~
of monitoring` the change in fluorescence intensity, wavelength shift, or fluorescence polarization. Possibilities for indirect labelling - j ~
- include biotinylation of one constituent followed by binding to `-2s avidin coupled to one of the above l?bel groups.
There are also numerous methods of separating the bound ~
*om the *ee ligand, or alternatively the bound from the free test compound. The IL-13 can be immobilized on various matrixes !; ` followed by washing. Suitable matrices include plastic such as - an -ELISA plate, filters, and beads. Methods of immobilizing the-receptor to a matrix include, without limitation, direct adhesion to ! j,, plastic, use of a capture antibody, chemical coupling, and~~~
biotin-avidin. :
The last step in this approach involves the precipitation of antibody/antigen complex by any of several methods including those udlizing, e.g., an organic solvent such as polyethylene glycol or .

.:) g4/04680 PCI/US93/0764 4 7 ~ !
~ ' a salt such as ammonium sulfate. Other suitable separation techniques include, without limitation, the fluorescein antibody magnetizable particle method described in Rattle et al., Clin. Chem.
30:1457 ~1984), and the double antibody magnetic particle 5 separation as described in U.S. Pat. No. 4,659,678.
The methods for linking protein or fragments to various labels `
have been extensively reported in the literature and do not require `
~detailed discussion here. Many of the techniques involve the use of activated carboxyl groups either through the use of carbodiimide or 0 active esters to form peptide bonds, the formation of thioethers by rcacdon of a mercapto group with an activated halogen such as chloroacetyl, or an activated olefin such as maleimide, for linkage, or -the lil~e. Fusion proteins will also find use in these applicatiqns.
Another diagnostic aspect of this invention involves use of -~
oligonucleotide or polynucleotide sequences taken from the I:
se~uence ~of an IL-13. Thesc sequences can be used as probes for detecting levels of the IL-13 in patients suspected of having a proliferative ccll conditions, e.g., cancer. The preparation of both :
RNA and DNA nucleotide sequences, the labelling of the sequences, 20 ~and ~the preferred size of tho sequences has received ample description and discussion in the literature.
Normally an oligonuc~eotide probe should have àt least about 14 nucleotidos, usually at least about 18 nucleotides, and the polynucleotide pr:obes~ may be up to several kilobases. Various 2s~ labols may be cmployed, most commonly radionuclides, particularly 32p. However, other tochniqucs may also be employed, such as using biodn modified nucleotides for introduction into a polynucleodde. The biotin then serves as the site for binding to avidin or antibodiés, which may be labelled with a wide variety of ~30 labols, such as radionuclides, fluorcscers, enzymes, or the like.
Altcrnatively, antibodies may be employed which can :~ ~ recognize specific duplexes, including DNA duplexes, RNA duplexes,-~ ~; DNA-RNA hybrid~ duplexes, or DNA-protein duplexes. The antibodiesin ~turn may be labelled and the assay carried out where the duplex 35 is~bound to a surface, so that upon the formation of duplex on the - ~ surfacc, thc presence of antibody bound to the duplex can be " ~

:~:

WO 94/04680 . PCTtUS93/0764!
~ ~q ~ 6~

detected. The use of probes to the novel anti-sense RNA may be carried out in any conventional techniques such as nucleic acid hybridization, plus and minus screening, recombinational probing, hybrid released translation (HRT), and hybrid arrested translation -s (HART). This also includes amplification techniques such as polymerase chain reaction (PCR).
Diagnostic kits which also test for the qualitative or quantitative presence of other markers are also contemplated.
Diagnosis or prognosis may depend on the combination of multiple indications used as markers. Thus, kits may test for combinations of , markers. See, e.g., Viallet et al., Progress in Growth Fac~or Res. 1:89 (1989).
VIII. Therapeutic Utilitv This invention provides reagents with significant therapeutic value. The IL-13 (naturally occurring or recombinant), fragments thereof, and antibodies thereto, along with compounds identified as having binding affinity to the interleukin or its receptor or antibodies, should be useful in the treatment of conditions exhibiting abnormal expression of the interleukin. Such abnormality will , 20 typically be manifested by immunological disorders. Additionally, this invention should provide therapeutic value in any disease or disorder associated with abnormal expression or abnormal ~~ ` -~ `
triggering of response to the interleukin.
Recombinant IL-13 or IL-13 antibodies can be purified and - -2 5 then administered to a patient. These reagents can be combined for - - -therapeutic use with additional active ingredients, e.g., in conventional pharmaceutically acceptable carriers or diluents, along with physiologically innocuous stabilizers and excipients. These- - ~
combinations can be sterile filtered and placed into dosage forms as 30 by lyophilization in dosage vials or storage in stabilized aqueous_ preparations. This invention also contemplates use of antibodies~~ or-binding fragments thereof which are not complement binding.
Receptor screening using II,- 13 or fragments thereof can be performed to identify molecules having binding affinity to the I
1 , 94/046~0 PCr/US93/07645 8s~

interleukin. Subsequent biological assays can then be utilized to determine if a receptor can provide competitive binding, which can block intrinsic stimulating activity. Receptor fragments can be used as a blocker or antagonist in that it blocks the activity of IL-13.
5 Likewise, a compound having intrinsic stimulating activity can activate the receptor and is thus an agonist in that it simulates the activity of IL-13. This invention further contemplates the therapeutic use of antibodies to IL-13 as antagonists.
The quantities of reagents necessary for effective therapy will o depend upon many different factors, including means of administration, target site, physiological state of the patient, and other medicants administered. Thus, treatment dosages should be titrated to optimize safety and efficacy. Typically, dosages used i n vitro may provide useful guidance in the amounts useful for in situ administration of these reagents. Animal testing of effective doses i -for treatment of particular disorders will provide further predictive indication of human dosage. Various consideration~ are described, e.g., in Gilman et al. (eds), The Pharmacological Bases of Therapeutics, 8th Ed., 1990, Pergamon Press; and Remington's 20 Pharmaceutical Sc~ences, 17th ed., 1990, Mack Publishing Co., Easton, Penn.
Methods for administration are discussed therein and below, - e.g., for oral, intravenous, intraperitoneal, or intramuscular administration, transdermal diffusion, and others. Pharmaceutically 25 acceptable ca~ners will include water, saline, buffers, and other compounds described, e.g., in the Merck Index, Merck & Co., Rahway, - New Jersey.
Because of the likely high affinity binding between an IL-13 - and its receptors, low dosages of these reagents would be initially 30 expected to be effective. Thus, dosage ranges would ordinarily be - expected to be in amounts lower than 1 mM concentrations, typically less than about 10 ~lM concentrations, usually less than about 100 nM, preferably less than about 10 pM (picomolar), and most preferably less than about 1 fM (femtomolar), with an 35 appropriate carrier. Slow release formulations, or slow release apparatus will often be utilized for continuous administration.

50 ' .

,S6~
IL- 13 or fragments thereof, antibodies or fragments thereof, antagonists, and agonists, may be administered directly to the host to be treated or, depending on the size of the compounds, it may be desirable to conjugate them to carrier proteins such as ovalbumin or `
5 serum albumin prior to their administration. Therapeutic formulations may be administered in any conventional dosage formulation . r '~
While it is possible for the active ingredient to be administered alone, it is preferable to present it as a pharmaceutical 0 formulation. Formulations comprise at least one active ingredient, as defined above, together with one or more acceptable carriers thereof. Each carrier must be both pharmaceutically and physiologically acceptable in the sense of being compatible with the other ingredients and not injurious to the patient. Formulations 5 include those suitable for oral, rectal, nasal, or parenteral (including subcutaneous, intramuscular, intravenous and intradermal) administration.
The formulations may conveniently be presented in unit dosage form and may be prepared by any methods well known in 2 o the art of pharmacy. The therapy of this invention may be combined with or used in association with other immunotherapeutic or immunopreventive agents. -EXAMPLES
The b~oad scope of this invention is best understood with 25 reference to the following examples, which are not intended to limit the inventions in any manner. Unless otherwise specified, ~~
percentages givlen below for solids in solid mixtures, liquids in liquids, and solids in liquids are on a wt/wt, vol/vol and wt/vol basis, respectively. Unless de~med otherwise, all technical and ~
30 scientific terms used herein have the same meaning as commonly ~
understood by one of ordinary skill in the art to which this ~~~
invention belongs.

$ ~ c ' ~C'~46~: ~ ~

2~an~;, te^nrLiques ~clica3ie t-` ~ _A and IL- !0 mz~; ~e a~ d ~o l:L-13, as desc,i~ed~ e.g.~ in U.S; Patent ~-~. 5,~ 91 (L-4) and C ~ 0?~.5;.~51 (IL-10).
~. ~
~ pr~xima~y ~ ? D~.~ rra~Tnent d~n~,ed ~ro~. 2 Ps~ru~ re~tncu~n C!lg~:St 0~ the mous~ P60û cn~iA lon~ [3rowil et a~ ol~ i42:679 (~8~i was iso~ated by po~yacrylam~ac g-l e~e~opnoresis and su~se~ucnt e1u~io~ a~d ~t~anol p~cip~a~.ion. T~i~ fragm_n which encompa~ses m~st of the codmg ,e~io~ of thc mouse ?500 _3~iA. u~ a~ioa~iv~ be11eld by r~nd m primi~1g, iD, ~e pres~n~ o~ ~32PldCTP.
Fi~ter i~ts were prepa~ed following standa~d ?roCe~U~CS Irom t~n ag~ p~tes each w~th appro~ma~y 5000 c~lo~ies of a B21 cDNA ~ his ~ib~Y was made fr~m cloned hum2n T cells, dc~i~atcd B'~ 1, which ha~ becn stimu~a~d with ants-~ for 7 ~ours ~rior t3 the ~so1a~io~ of thc R~. The -onst~ction of this ~i~rary i:s descnb~d i~ S. ap~icadon Se . 37f453.9~1.
The ~ters were hybr~zet overr:Light at 4~ C wi~ thc ~abel~-d mousc P600 ~agment in ~O'i~ fo~d~, 6X S~PE~ 0.19~ S~S, SX
Den~ardt's solut~ , and 100 ~LgJ~l tR.~A T~e filters w~re wa~hed 3 ~i~c~ w~th ~X ~SPE. 0~ o SDS at room tempcrature for 20 m~utes each, ~wice w~ lX SSPE, Q.1% S~:)S at ~C fcsr I ho~r, ~en e~poscd to film o~emi~ht. Eight pcsitiYes wer~ ~den~f~ed ~nd picked ~or fu~er pur~ficauor~. Se~en OI th~se ~rc posiuYe upon ~esore~nmg S~ clones had ~mHl i~e~s of 1.35 kb and one ail insert o~ oniy 0.6 ~ The 1.35 kb in~crt of on- clo~e~ des~ a~ed p~2!.2~f. was subcloned into M13 ~d scquenced by the dide~xy method.
S-quc~ce compas~s~on demons~atcd that ehis 1.15 kb ^D~T.A encodes a hu~ ~omolog cf mouse P60G.
Thc.huIL-13 cD?~A isol~d from thc ~2i lib~-ary was no; ~ull leng~ as compared ~o the mousc P600 cl?NA. Rcpca~ed a~tempts to isolate a full-~cn~:h cDNA fro~n the B~ sary wcrc un~uccessr^ul.
Thus, a d~ffe~nt Library was scrccncd wt~h ~ne pB~1.2~f in~rt Lor a AMEN~ED SY'~T

J 94/04680 PCl`/US93/0764 52 ?~

~. &60 full length clone. A PCR probe was derived from the human cDNA
beginning 50 bp from the 5' end and ending at the stop codon.
A cDNA library was made from a clone of an A10 T cell line.
The same hybridization conditions as described above were used.
The filters were washed once in lX SSPE, 0.05% SDS at room temperature for 15 minutes and then twice at 55C for 30 min to one hour. They were exposed to film overnigh~. Several posi~ives colonies were detected and rescreened. Double s~anded sequence obtained from the 5' end of several of the cDNA inserts from these 0 positives indicated that they were full-length. One of the 1.3 kb cDNA inserts, from a clone designated pA10.66, was subcloned into M13 and sequenced. Its sequence is shown in Table l. The sequence of the full length clone differed from the se~uence shorter clone by a single codon, which is present in the full length clone, see Table 3.
II. l~pression and Pur~ication of Mouse P600 and Human II~-13 Protein The pB21.2Bf clone, containing a 1.16kb cDNA encoding human IL-13, lacked the first 23 N-terminal amino acids. The insert was prepared~ for ligation into an expression vector pGEX-2T by using PCR to provide unique restriction sites at the 5' BamH1 and 3' EcoR 1 ends. The pGEX-2T vector is designed to produce a fusion protein, where the distinct protein segments are separated by a readily cleaveable protease site. The gel purified DNA was ligated into the _ expression vector pGEX-2T so that when the plasmid is expressed in - 25 E. coli the protein encoded by the DNA insert produces a fusion - protein with glutathione-S-transferase with a thrombin cleavage site in between as described in detail by Smith et al., Gene 67:31 (1988).
~ The resuldng plasmid was transformed into E. coli and successful - transformants were grown in the presence of IPTG. Expression ~ 30 products of this construct, when grown under inducing conditions, accumulated in inclusion bodies.
.~ .
:

~94/04680 21~286~ Pcl`/Us93/0764~

Mouse P600 Refoldin~ and Purification Transformed E. coli cells were grown in media at 37 C and induced with IPTG at 0.5 OD. The induced cells were grown until maximal OD was reached either by shake flask or fermentation.
Cells were harvested by centrifugation at 4000 x g at 4C for 30 min and were frozen at -10C.
Cells were resuspended at room temperature in TE buffer (50 mM Tis-HCI, 10 mM EDTA pH 8 with 1 mM Peflobloc, a protease inhibitor). Cells were passed through a microfluidizer at 18,000 psi, collected, and centrifuged at 10,000 x g at 4C for 30 min. Cell pellets were repeatedly washed in TE buffer and centrifuged until supernatant was clear. The pellet was solubilized with 6 M
guanidine-HCl, 10 mM Dl~, ~50 mM Tris-HCI pH 9, and 1 mM
Peflobloc and mixed at 4 C for 2 hours.
~The protein concentration was measured by the Bradford Protein method and typically was approximately 2.~ mg/ml total protein concentration. The mixture was diluted over a period of hours~ to 100 f~1d of its unfolding volume into 50 mM Tris-HCl, 150 mM~NaCl, 2 mM reduced glutathione, 1 mM oxidized glutathione, 0.5 M guanidine-HCl, and 10 mM EDTA at pH 9Ø The solution was mixed at 4C for 24 hours, allowing refolding of disulfide linkages of the molecule. -Precipitates were removed by centrifugation at 4000 x g for 30 min~at 4 C or by filtration with a 0.45 ~lm filter. The supernatant 2s was cQncentrated using a Pel1icon and diafilter against 50 mM TRIS-HCL, 150 mM~ NaCI, 2.5 mM CaCl2. pH 7.5 buffer at 4 C. The glutathione-S-transferase fusion partner was cleaved by adding human thrombin at- 10 ng per 50 ~g of fusion protein. The salution~
was mixed at 4C for 18 to 48 h to allow the fusion protein to be 30 ~ cleaved by thrombin. SDS-PAGE gels and TF-l bioassay was used to ~; characterize P600 conformation and activity.
~- ~ Ammonium sulfate was aded to the refolded material at a 25%
saturation rangc after complete thrombin cleavage was observed.
~G~ e~refolded material was adjusted to pH 8.5 with 6N NaOH and ; 35 loaded onto a butyl Toyo Pearl column, equilibrated at pH 8.5 with wos4/w680 21~286u 54 PCI/US93/0764' 2~% ammonium sulfate, 50 mM TRIS-HCL, 0.05 M NaCl buffer. The column was washed with equilibration buffer until the A 2 8 0 approached baseline. The colmn was eluted with 50 mM TRIS pH
8.5 buffer for 5 bed volumes. An A280 pool was collected and 5 concentrated in a 5000 molecular weight cutoff Amicon stir cell to a volume less than 3% of the bed volume of the S200 gel filtration column.
The S-200 column was equilibrated with depyrogenated buffer 50 mM NaPi, 150 mM NaCl, and 0.01% Tween-20 pH 6Ø The 0 concentrated S pool was loaded onto the gel filtration column and fractions were collected and verified for P600 protein content by SDS-PAGE gels. Practions were pooled based on SDS-PAGE, concentrated, filtered through 0.22 llm filters and tested by tbe TF-1 bioassay for biological activity. Protein concentration was 5 determined by silver stairling and scan using the Molecular Dynamic gel scanner.
Endotoxin was measured using the Whittaker colorimetric Limulus assay and typically was < 10 eu/ml typical preparations resulted in ~ 95% purity by staining, with a bioactivity of about 1 x 20 lo6 units/ml. Variations on the refolding procedure will be ~- effective, e.g., protein concentrations may vary over some range, typically a 5^fold difference will also work. The glutathione concentrations may be varied, and the periods of time for slow -- -~ dilution and overnight incubations may be titrated. Each of the -~; 25 refolding parameters described should be titrated where appropriate. - - - -Similar procedures were used p~epare and to purify human --III. Activities on B Cells -; - 30 A. Cofactor/facto~ Proliferation: Cell Viabil~
. .
Mouse P600 functions as a stimulator/costimulator of cell viability, e.g., mouse P600 made from E. coli can stimulate or costimulate proliferation of large in ~vivo activated mouse B cells.

) 94/~o 2 1 ~ 2 ~ 6 V PCr/US93/~764~

Decreasing amounts of P600 administered to the cells resulted in lessened cell growth as determined by 3 H-thymidine incorporation.
To construct a cDNA encoding the extracellular domain of CD40 (designated "soluble CD40"), the following PCR primers containing an - 5 XhoI site were synthesized on an Applied Biosystems 380A DNA
synthesizer:
sense~ ACAGCTCGAGCCATG-GTGTCTTTGCCTCGGCTGTG-3' and antisense: S'-GTAGCTCGAGCTCACCGGGACTTTAAACCACAGATG-3'.
These primers were used to produce PCR fragments encoding 0 191 amino acids from the start codon of mouse CD40. PCR fragments were digested with XhoI and then ligated into XhoI cleaved mammalian/bacterial expression vector (pME18S). The inserted fragment was sequenced by the dideoxy sequenceing method to confirm the sequence.
; Plasmids carrying the soluble CD40 cDNA were transfected into COS-7 cells by electroporation by standard procedures, See Ausubel e~ al. (1987 and periodic supplements). Briefly, 0.75 ml of COS-7 cell suspension in serum free Dulbecco's Minimal Essential (DME) ~`5~ medium~ at 107 cells per ml were incubated with 50 ~11 of 20 llg 20 ~ plasmid ~ at room temperature for 10 min, and subjected to electroporation ~using a Bio-Rad gene pulser ~60 F, 220 V). Ten minutes after electroporation, COS-7 cells were cultured in four 10 cm~ dishes for 3 days. For the purification of soluble CD40, medium was changed to ~phenol red-free RPMI 1640 supplemented with 2s ~HBlOl~ (HA~A Biologics, Alameda, CA) one day after electroporation.
Soluble CD40 was purified by ion exchange chromatography on anilon exchange columns using standard procedures. The protein was eluted from the column using a linear NaCl gradients and analyzed by Western Blotting using a rabbit antiserum against a 30 ~ CD40 pcptide made by standard methods.
~,... . .
Eight woek old female Lewis rats were obtained from Harlan Sprague-Dawley (Indianapolis, IN). These rats were immunized intaperitoneally with 10 ~lg of soluble CD40 in complete Freund's ~, ~s~,.. .:., ,~ , WO 94/04~80 PCI`/US93/07645 21~86~

adjuvant followed by boosts of 10, 10, 10, and 50 ',lg of soluble CD40 in incomplete Freund's adjuvant at 3, 4.5, 6, and 8.5 weeks, respectively. A final boost in saline was injected at 12 weeks. Test bleeds were evaluated for anti-CD40 antibody content by ELISA.
Small dense B cells from unstimulated mouse spleens were prepared as described in Hodgkin et al., Cell. Immunol. 134:14 (1991). Spleens were teased into complete RPMI ~cRPMI) containing 5% fetal calf serum (FCS; J.R. Scientific, Woodland, CA), 5 x 10-5 M
2-mercaptoethanol (Polysciences, Inc., Warrington, PA), 2 mM
0 glutamine (J.R. Sciendfic), and 25 mM HEPES buffer (Irvine Scientific, Santa Ana, CA), 100 U/ml penicillin, and 100 ~lg/ml streptomycin (Irvine). Red blood cells were Iysed using 0.83%
ammonium chloride, pH 7.4.
T cells were removed using two successive treatments with anti-mouse Thy 1.2 mAb (New England Nuclear, Boston, MA) and anti-L3T4 antibody (RL172.4 hybridoma, a gift from Dr. H.R.
~ ~ MacDonald, Ludwig Insdtute, Epalinges, Switzerland) for 20 min on t'~ , ice fo!Iowed by complement (1:10 dilution of rabbit low-tox complement, Cedarlane Laboratory, Ontario, Canada) for 30 min at 37C. Small dense B cells were then isolated by density gradient ` ~ centrifugation using a discontinuous gradient composed of 75%, 65%, and 50% percoll (Pharmacia' ~ine' Che~nicals, Uppsala, Sweden) at 2500 x g~ for 25 min at 4C.
Cells collected from the interface between 65% and 75% percoll 25 were used in subsequent experiments. Large in vivo activated B
- ~ cells were-collected~ from the 65% and 50% interface. B cells were-cultured in flat bottomed 96 well tissue culture plates (3072, Falcon ; La~ware) at~ variou$ cell densities in cRPMI, plus additional stimulants, as indicated. Proliferation was evaluated via a 4 hr - -30 pulse of 3H-thymidine (Amersham) added at 48 hr after~ culture initiation. - _-.

, ~, , ., i ~ .
``; 35 -2i~?~ Pcr/uss3/0764 B. Sustained Sùrvival of B Çells: Selectivitv Rea~ents The anti-CD40 mAb89 and anti-CD23 mAb25 were produced by standard procedures against the respective antigens, see Vallé e t al., Eur. J. Immunol. 19:1463 (1989); Bonnefoy et al., J. Immunol.
138:2970 (1987). The CDw32/Fcy RII transfected Ltk-cell line (CDw32 L cells) was described by Peltz et al., J. Immunol. 141:1891 ( 1988). Anti-IgM antibodies coupled to beads (anti-M) were purchased from Biorad (Richmond, CA).
Cell phenotype was determined using FITC conjugated mAb 0 originating from Becton Dickinson (Mountain View, CA). The -' neutralizing anti-IL-4 monoclonal antibody was kindly provided by ~Dr. Grassi. The extracellular domain of the 130 kDa IL~ receptor was derived from COS-7 cells transfected with a plasmid containing a truncated IL-4 cDNA described by Garrone et al., Eur. J.~lmmunol.
21:1~3~65 (1991). The recombinant protein was purified from transfected cell culture supernatant by purification nn an IL-4-Affi-gel ~10 ~column. The anti-130 kDa IL-4 receptor antibody was generated~ after immunization of mice with the extracellular domain of ~e IL~ receptor. Cultures were carried out in modified Iscove's medium.
B Cell Pre~arations and Cell Cultures B cells~ were isolated from tonsils as described by Defrance et ;al., J. Immunol. 139:1135 (1987). Briefly, after a rosetting step with sheep red blood ce!ls, non-rosetting cells were further incubated 2s with anti-CD2, anti-CD3, and anti-CD14 mAbs prior to negative $election perf/ormèd ~ with magnetic beads coated with anti-mouse ' ~ IgG (Dynabeads, Dynal, Oslo, Norway). The isolated populadon expressed ? 98% CD19 or CD20 ~B cells) and < 1% CD2 (T cells) or CD14 (monocytes).
30~ ~ Assavs With Anti~en Rece~tor Activated' B Cells '; B lymphocytes, adjusted at 5 x 105 cells/ml, were stimulated for 72 hours with insolubilized anti-IgM (5 ,ug/ml). A 16 hour pulse WO g4/04680 PCI /US93/0764 2 ~ Q

with 1 ~lCi (3H)TdR was usually performed at day 3 and 6. [3H]-TdR
uptake was measured by standard liquid scintillation counting techniques .
CD40 ~ystem For proliferation assays, 2.5 x 104 purified B cells were cultured in the presence of 2.5 x 103 irradiated (7000 rad) CDw32 L
cells and 0~5 ~g/ml of anti-CD40 mAb89 in a final volume of 200 For Ig production, B cells were tested at 2.5 x 105 cells/ml.
Supernatants were harvested after 10 days and Ig levels were 0 determined by EL~SA.
Isolation of ~IgD_ and s~- B Cell Populations `~
Purified B lymphocytes were separated using a preparative magnetic cell separation system (MACS, Becton-Dickinson), according to the experimental procedure described in detail by Miltenyi et al., Cytometry 11:231 (1990). The separation based on sIgD expression has been described e~rlier by Defrance et al., J. Exptl Med. 175:671 (1992). Purity of the sorted cell populations were > 99% for the sIgD+ B cell subpopulation, while < 1% of sIgD- B cell subpopulation .
expressed dIgD, as assessed- by fluorescence analysis using a FACScan.

,CYtokines .
Purified recombinant hIL-2 ~Amgen, Thousand Oaks, CA, 3 x 1 o6 U/ml), recombinant hIL-4 (Schering-Plough Research Institute, Bloom~leld, NJ, 1 x 107 U/mg), recombinant hIL-10 (Schering~Plough 2s Research Institute, Bloomfield, NJ, 1 x 107 U/ml) were respective~y~
;~ used at 10 U/ml, 50 U/ml, and 100 nglml. Il,-13 was expressed- asa fusion protein with glutathione-S-transferase using the pGEX-2T
vector (Pharmacia, Uppsala, Swede). -A DNA fragment encoding hIL-13 residues 24-109 was prepared by polymerase chain reaction (PCR) and cloned into the BamHI/EcoRI site of the vector. A DNA fragment encoding mIL-13 residues 19-109 was also prepared and cloned. The human and 94/04680 ~ PCI`/US93/07645 5~

mouse IL-13 fusion proteins were expressed as insoluble aggregates in Escherichia coli, extracted by centrifugation, solubilized, and subjected to a renaturation step, see van Kimmenade et al.? El~r. J.
Biochem. 173:109 (1988).
The refolded IL- 13 was cleaved from the fusion partner by thrombin, purified by cation exchange (S-Sepharose FPLC, Pharmacia) and gel filtration (Sephacryl s-200 FPLC, Pharmacia) chromatography. The gel ~lltration column was calibrated with protein standards (Bio-Rad). Proteins were quantitated by SDS-0 PAGE, silver staining (ISS), and scanning densitometry (Molecular Dynamics) with normalization to chicken egg Iysozyme (Sigma, St.
Louis~ MO). Endotoxin (determined by the Limulus ameobocyte lysate assay (Whittaker Bioproducts, Inc.) was typically < 1 ey/ml.
IL-13 Enhances the D~A Svnthesis of B Cells Activated Throu~h Their Antigen Rece~tor The DNA synthesis of highly purified human B lymphocytes activated through their antigen receptor with anti-IgM antibody is enhanced by recombinant cytokines, such as IL-2, IL-4, and IL- 10.
Recombinant murine IL-13 also enhanced, in a dose dependent fashion, the day 3 DNA synthesis of human tonsilIar B lymphocytes cultured in the presence of insolubilized anti-IgM antibody. The - ~ maximum stimulation- was obtained for a concentration of 10-25 nglml~ of murine (or human) IL-13.
The stimulatory effect was lower than that of either IL-2 or IL-4 but comparable to that of IL-10. As IL-4, but unlike IL-2, t}le ~- co-stimulatory effect of IL- 13 on anti-IgM activated B cells could ~e observed after 3 days of culture, and decreased to be virtually undetéctable after 6 days.
,, ~
~13 Acts As Growtb Factor for B Cells Stimulated Throu~h Their -~ 30 CD40 Antigen IL-13 was also assayed for its ability to enhance the proliferation of anti-CD40 activated B cells in comparison with IL-4 and IL-10. Thus, 2.5 x 104 purified tonsillar B lymphocytes were ~"~
~ :~
.., ~
,.
.
.,,~
,-~ ~ . . ...

WO 94~04680 PCr/US93/07645 21~286U 60 cuitured over CDw32 expressing L cells together with 0.5 ,u g/ml of an anti-CD40 antibody Mab 89 with or without increasing concentrations of IL-13. [3H]-TdR incorporation was measured at day 6. Both murine and human IL- 13 strongly enhanced anti-CD40 induced DNA synthesis. Maximum stimulation was reached between 3 and 30 ng/ml IL-13 and plateaued thereafter without demonstrating any inhibitory effect (even at 1000 ng/ml). Under these culture conditions, the half maximal stimulation was observed between 0.03 and 0.3 ng/ml in three independent experiments.
0 The growth stimulatory effects of IL-13 were then compared to those of IL-4 and IL-10. IL-13 activity was comparable to that of IL-4 and IL-10 when assayed early in the culture at day 6. The IL- 13 stimulatory activity was particularly striking at day 9, where it surpassed that of IL-10 and more notably that of IL-4. IL-13 also showed stimulatory effects at day 12 and again was more efficient than either IL-4 or IL-10. Cultures grown in the presence of IL-13 fQrmed extremely tight clumps. Clumps were very difficult to dissociate and thus rendered extremely inaccurate the enumeration of viable lymphocytes during cell cultures. Nevertheless, cell cultures were split every fifth day for up to 25 days, at which time the number of viable B lymphocytes had increased about 12 fold (estimated conservatively).
Whether IL-13 could act in concert with IL-4 or ~L-10- for maximal B cell proliferation was studied. Combination of an optimal 2s concentration of IL-4 and with increasing concentration of IL-13_ resulted in a DNA synthesis which was comparable to that obtained with IL~, thus indicating the lack of synergy and even additivity between these two cytokines. In contrast, combination of IL- 13 and IL-10 resulted in an additivity of their stimulatory effects.--The additive effects of IL-13 and IL-10 on B cell proliferation- were observed at all times tested. Taken together, these results indicated ¦~ that IL-13 is a growth factor for human B lymphocytes. -~ .

-94/~4680 2 PCr/US93/07645 61 1~2(~o IL-13 Induces Anti-CD40 Activ~ted B Cells to Secrete I~E
In view of the powerful effects of IL- 13 on the proliferation of anti-CD40 activated B cells, culture supernatants were analysed for their immunoglobulin content. Like IL-4 but unlike IL-10, IL-13 5 did not stimulate the production of IgM, IgG, and IgA in day 8 cultures of anti-CD40 activated B cells. However, IL-13 was able to induce B cells to secrete IgE~ in a dose dependent fashion (Table 4).
IL-13 was comparable to IL-4 in its capacity to induce IgE. The combination of IL-4 and IL-13 resulted in a production of IgE which 0 was either comparable or slightly enhanced when compared to that obtained with IL-4 alone. IL-13 unable to induce either resting B
cells or anti-,u activated B cells to secrete IgE.

-~- Table 4: IL-13 Induces Anti-CD40 Activated B Cells to Secrete IgE.
s cytokine IgG ~llg/ml ) IgA ~llg~ M ~lg/ml ) IgE (ng/ml ) none 1.2 + 0.09 0.4 + 0.04 0.08 + 0.005 < bg .: IL-4 1. 4 + 0 . 01 0 . 4 + 0 . OOS 0 . 2 + 0 . 005 37 + 1.1 IL--10 lS.0 + 1.9 12.0 + 2 18.0 + 0.6 < bg IL--13 1.6 + 0.4 0.4 + 0.001 0.2 + 0.0425.2 + 3.3 .
S x 104 purified B cells were cultured for 10 days with 2.5 x 103 irradiated CDw32 L cells with the anti-CD40 mAb89 without or with 50 U/ml IL-4, 100 ng/ml IL-10, 30 n~/ml hIL-13. Ig levels 2 0 represent the mean + SD values of quadruplicate determinations.
Representative of three experiments. (< bg = lower than 150 pg/ml of IgE.j , i . .
The IL-13 Tar~et B(~ Sub~ ulation is ~Iore P~icted ~han the II,-4 (~e 2s To further compare the effects of IL-4 and IL-13, B cells cultured in the CD40 system with either cytokine were phenotyped after six days of culture. IL-13 is able to induce CD23 expression on cultured B cells with an intensity comparable to that obtained with IL-4. However, whereas IL-4 induced > 90'rO of the cultured B cells . ' WO 94J0468~ PCI /US93/0764:
~ 62 to express CD23, IL-13 induced CD23 only on 40% of the B cell population. In addition, whereas IL-4 was able to readily induce - CD23 on both resting and anti-M activated B cells, IL-13 was much less efficient under these conditions.
s The transferrin receptor expression was also analyzed on B
cells cultured for six days in the CD4~ system with or without IL-4 or IL-13. All B cells expressed transferrin receptors (TfR), but two populations could be clearly distinguished according to levels of expression which were designated TfR low and TfR high. In the 0 CD40 system alone, 80% of the B cells were TfR low, and 5% TfR high.
- In cultures grown with IL-13, 55% of the cells were Tf~ low and 40%
were TfR high. In cultures performed with IL-4, 10% of the c~lls were TfR low and 85% were TfR high.
As sIgD+ B cells consist of naive B cells, whereas sIgD- B cells consist of a mixture of germinal center B cells and memory B cells, the reactivity to IL-4, IL-10, an~ IL-13 of sIgD+ and sIgD- B cells was tested in the CD40 system. Both sIgD~ and sIgD- B cells ~-- proliferated strQngly in response to IL-4 and IL-40, as measured by (3H)TdR incorporation after six days of culture. In contrast, IL-13 ¦; 2 0 preferentially enhanced the anti-CD40 induced DNA synthesis of sIgD I cells. Furthermore, IL-13 and IL-4 were able to induce both sIgD+ and sIgD- B cells to secrete IgE.
Taken together, these results indicate that IL-13 acts - -preferentially on sIgD+ cells and thus acts on a B cell subpopulation ~s more restricted in size than that stimulated by IL-4.
he IL-13 Biolo~ical Effects Are Independent of IL-4 - `~
I
As Ik-13 displays many of the biological effects of IL-4, it was suspected that it might act either through an induction of IL--4- ^
secretion or through binding to the IL-4 receptor. To address this question, IL-13-induced B cell proliferation was tested in the-presence of three different IL-4 antagonists: (1 ) a neutraliz}ng anti-IL-4 monoclonal antibody; (2) a soluble extracellular domain of the 130 kDa IL-4R [see Garrone e~ al., Eur. J. Immunol. 21:1365 (1991)];
~ and (3) a blocking anti-130 kDa IL-4R monoclonal antibody. These :

.7/O 94/04680 ~ PCI/US93/0764;
63 '~
o three antagonists bloeked by 80-90% the effects of IL-4 on the proliferation of anti-CD40 activated B lymphocytes without affecting the proliferation induced by IL- 13 . These IL-4 antagonists also failed to block IL- 13 induced CD23 expression and IgE production, while they did totally block that induced by IL-4.
Induction of B Cell Proliferation Bv IL-13 or IL-4 and Cos Cells Expressin~ the Human or M ouse Ç D 40-L
Five x 104 highly purified (> 98% CD2û+) negatively sorted splenic B cells were co-cultured with 1.6 x 104 irradiated (7,000 0 rad) Cos cells transfected with human or mouse CD40-L or the empty pJ~ -14 vector as control. IL-13 or IL-4 were added at 400 U/ml. Soluble anti-CD40 mAB89 and the control mAbA4 were used at 50 ~g/ml. The cultures were harv~sted 3 days later, after addition of [3H]-Thymidine in the last 16 hours of culture. B cells were stimulated by any of IL-4, COS supernatants containing either mouse or human IL-l 3, or combinations of supernatants and IL-4 or IL-13 .
Biological effects of IL-13 on human B cell growth and differentiation have been described. IL-13 costimulated with anti-IgM antibody to induce DNA synthesis but its effects were less ~- conspicuous than those of IL-2 or IL-4. IL-13, as other cytokines, -~ failed to greatly induce the multiplication of B cells activated ~- through their antigen receptor. However, IL- 13 displayed striking growth promoting effects on B cells which were cultured in the CD40 2 5 system, which is composed of a fibroblastic cell line expressing the - human Fc receptor CDw32 and monoclonal antibody to CD40. Under these condidons, IL- 13 was at least as active as IL-4 and its effects ~ on B cells were long lasting, thus allowing the multiplication of viable B cells. IL-13 altered the phenotype of activated B cells, as it : 30 induced B cells to express CD23.
- The IL-13 dependent induction of CD23 on CD40 activated B
cells is not mediated by an anti^CD40 activation, since resting and anti-IgM activated B cells can also be induced to express CD23 in response to IL-13. However, the proportion of cells expressing CD23 3s was lower with IL-13 than with IL-4. Likewise, IL-4 induced WO 94/04680 . PCr/US93/07645 ~4~36~

virtually all CD40 activated B cells to express high levels of transferrin receptors. IL- 13 induced only half the cells to express - transferrin receptors at high density. This indicated that IL- 13 was acting on a subpopulation of B cells which was more restncted than s that affected by IL-4.
Accordingly, when cells were separated according to surface IgD (sIgD), which distinguishes naive B cells from germinal center and memory B cells, IL-13 was found to be more effective than IL-4 on sIgD+ B cells. IL-4 was abIe to enhance better than IL-13 the 0 proliferation of sIgD- B cells. The different population target for IL-13 and IL-4 could be explained by differential IL-13 and I~-4 receptors expression, the demonstration of which will await the production of labelled IL-13 or of IL-13 receptor specific antibodies.
The lower response of sIgD- B cells is particularly interesting and 15 warrants further analysis. IL-13, as IL-4, poorly enhanced the -- ~ synthesis of lgG and IgM by B cells cultured in the CD40 system.
Surprisingly, however, IL-13 induced anti-CD40 activated B cells to produce IgE.
-~ The levels of IgE produced in response to IL-13 were20 comparable to those produced in response to IL-4. IL-4, as well as IL-13, was able to induce IgE synthesis by anti-CD40 activated B
cells. As a consequence of isotype switching, IL-13, like IL-4, was able to induce sIgD+ to secrete IgE. As IL-4, IL-13 could not induce - -resdng B cells to secrete IgE. At first, this IL-13 induced IgE
25 production contrasts with other studies indicating IL-4 as being the sole inducer of IgE synthesis. However, reccnt studies have ~ -described assay systems resulting in the secretion of IgE, while IL-4 - ~ was totally blocked by neutralizing antibody. IL-2 has been l reported to induce secretion by Staphylococcus aureus activated B- ~ 30 cells, whereas IL-4 was ineffective. It may be possible that in these - - studies, IL-13 was responsible for these effects.
` ~ While IL-13 causes many functions similar to IL-4 on B cells, the present study demonstrates that these effects are independent of possible inducdon of IL-4 secretion or the use of the 130 kDa 35 IL-4 receptor, e.g., neutralizing anti-IL-4 antibody, blocking soluble IL-4 receptor, and blocking anti-IL-4 receptor antibody were unable , .
, 94/04680 2~ P~/US93/07645 65 ~86'/' to affect the action of IL-13 on B cell proliferation and IgE secretion.
However, it cannot be excluded that IL- 13 may share with IL-4 some common transducer.
Crosslinking studies have shown the binding of IL-4 to 60-70 and 70-80 kDa components unrelated to the 130 kDa molecule. In this context, it is worth noting that both human and murine IL- 13 act on human B cells with similar ef~leiency, while murine IL-4 is species specific. The respertive roles of IL-13 and IL-4 in IgE
production can be determined ' more particularly when mice whose 0 IL-4 genes have been knocked out (e.g., IL-4 knock-out mice) are studied to determine circulating IgE levels.
Finally, it will be important to establish whether human IL-13 is produced only by T cells or if other cell types also produce it.
Furthermore, IL-13 may be involved in abnormal B cell 5 proliferation, as occurs in leukemic and autoimmune diseases. Thus, agonists or antagonists of IL-13 may be useful in therapeutic treatment of such conditions.
C Mod~fication of I~rfaoe Markers on Activated ~man BC~lls j Highly purified B cells were isolated from normal human 2 o spleens obtained from cadaver transplant donors. Splenocytes were -- - obtained by aseptically squ'ashing spleens though a sterile metal mesh and' frozen in aliquots for subsequen~ use. Highly purified B
cells (> 98% CD20+) were obtained by negative FACStar Plus Becton _ Dickinson sorting after staining the splenocytes with the following ~E-conjugated mAb: and-CD3, a~ti-CD4, anti-CD8, anti-CD14, anti-CD16, and anti-CD56. ~Becton Diclcinson).
Human IL-13 was u$ed at 30 ng/ml final concentration.
- -'- '' Récombinant ' IL-4 (used at 400 U/ml) was provided by Schering Research (Bloomfield, NJ).
Five thousand highly purified B cells were co-cultured with an ~ -- equal number of T cells from clones B21 or spA3, harvested five days after stimulation with feeder cells and PHA, in a final volume - of 0.2 ml of Yssel's medium supplemented with 10% FCS, 10 ~Lg/ml ultra pure transferrin (Pierce), and 400 U/ml recombinant IL-4.
Cultures were set up in eight replicates in U-bottom 96 well plates WO 94~04680 PCI`/US93/0764:
~;~ 66 .

(Linbro) and incubated 14 days at 37C in 5% CO2. At the end of the incubation period, the supernatants from each of the eight wells were harvested and pooled for isotype determination. In some cultures, the T cell clones were replaced by 5 ~g of plasma membranes derived from the T cell clones, or by 50 ~g/ml anti-CD40 '~ mAb 89.
Ig content of the supernatants was determined by ELISA as described by Gascan et al., Eur. J. Immunol. 22:1133 (1992). Plasma membranes were prepared from the CD4+ T cell clone B21, also as 0 ~ described therein. The amounts of production of the respective Ig isotypes were typically highest when B cells were stimulated with membranes in the presence of IL^4, but IL-13 typically had similar ~- effects. The amount of IgG4 production was particularly hig4, P`
~, though IL-13 seemed to enhance 'the IL-4 effect on IgE production '15~ after sdmulation w~th T cell clones.
Highly pu~nfied sIgD+ splenic B cells (5000 cells/well) were co-culturéd with pJE;E14 vector transfected or sorted COS-7 cells (250 cells/well) transfected with and expressing the human (h) or mouse (m)~CD40-L. IL-13~ (30 ngiml~ and IL-4 (400 Ulml) were added.
2~0~ Ihe ~epddvities~ of the ELISAs (0.2 ng/ml for IgE and IgM, 0.4 ;`n~nl~ for lgG4 ànd `total IgG) were determined with c,alibrated Ig stànda~ds (Bchring,, Marburg, Germany). Both IgG and IgE levels -were~ mcreased ~by ~-13. - -D. Effccts of CD4Q Li~and 2s~ Thè humàn CD40 ligand (hCD40-L) was cloned from a cDNA
lîbrary constructed from'an activated CD8+ T-cell clone and two ,cDNA's wcre~dctected ,representing a 2.1 kb and a 1.2 kb clone.
5~ Both~ cDNA clones had identical open reading frames of 261 amino acids and~ dîffcred only in the length of their 3' untranslated ends, ~~
30 ~ and probably rcpresent the 2.1 kb and 1.2 kb transiently expressed :' -mRNA ~species~ detected by Northern analysis in an activated CD4+ T- -' --~
~",'~ ccll clone. hCD40-~ transcripts could also be detected in CD4+ and CD8+- T cell receptor ~TCR) a~ T cells, TCR ~ T cells, natural killer -cells, monocytes, small intestine, and fetal thymocytes, but not in .. .... " , _ , 0 94/04680 ~C~ PCI/U~93/0764~
67 ~ 5.~

purified B cells, fetal liver, fetal bone marrow, brain, kidney, or heart.
COS-7 cells transfected with hCD40-L ~COS-7/hCD40-L) induced human B cell activation as judged by the induction of 5 homotypic aggregates of Epstein-Barr Virus (EBV) transformed, and normal B cells. In addition, COS-7~CD40-L induced B cell proliferation, which was further enhanced by IL-4, or IL- 13 . IL- 13, like IL-4, synergized with the mouse and hCD40-L to induce IgM, total IgG, IgG4, and IgE, but not IgA production by highly purified B
0 cells.
Anti-IL-4 antibodies inhibited IL-4 and COS-7/h~D40-L
induced Ig production by 13 cells, but had no effect on IL-13 and COS ~7/hCD40-L induced B cell differentiation, indicating that IL- 13 and hCD40-L induced Ig produc~ion, including isotype switching to IgE, independently of IL-4. hCD40-L induced B cell differentia~ion was blocked by soluble CD40, confirming the requirement for specific engagement of CD40-L. Collectively, these ~ata indicate that CD40-L and IL-13 expressed by human CD4+ T helper cells are important components of T and B cell interactions resulting in B cell 2 o proliferation, differentiation, and IgE switching. However, the distribution of the hCD40-L suggests a broader function of this - molecule.
- Induction of B cell prolifera~ion and differentiation into Ig producing cells requires T cell help. Antigen-specific T cell and B
2s_ cell interactions involving binding of the TCR to peptide class II MHC
complexes in the B cells result in T cell activation. Activated T cells -- deliver both contact and cytokine mediated signals, inducing B cell proliferation ;and differentiation. Once T cells are activated, they can interact with any B cell in an antigen independent class II MHC non--~ 30 restricted fashion. Lymphokines produced by the activated T helper cells do not only determine the amounts of Ig produced, but they .
-- - also direct isotype switching.
IL-4 is a B cell growth factor which induces human B cells to - switch to IgG4 and IgE production, whereas TGF-,B directs IgA35 switching. The human cDNA homologue of P600, a protein produced by mouse Th2 clones following acdvation, has been recently clone~

WO 94/04680 PCr/US93/0764 21~28~0 ~

and expressed, as described herein. The human IL- 13 protein induced human monocyte and B cell growth and differentiation and was designated IL-13. Human IL-13 is a non-glycosylated protein of 132 amino acids with a molecular mass (Mr) of 10,000 and is 5 produced by T cells. IL-13 has no significant homology with other cytokines except IL-4, which is -30% homologous. IL-13, like IL-4, can specifically induce IgG4 and IgE switching in human B cells, independently of IL-4.
The contact mediated signals delivered by activate T helper 0 cells can be replaced by anti-CD40 mAbs. One of the contact T
helper signals is delivered by the CD40 ligand (CD40-L), a 33 kDa molecule expressed on activated CD4+ T cells. CD40-L ~ransfectants induced proliferation of B cells and induced IgE production in the presence of IL-4. Here is described the isolation of human CD40-L
clones from a cDNA library constructed from an activated CD8+ T- ~
cell clone. The distribution of the human CD40-L, its ability to i activate B cells, and its role as a co-activation molecule with IL-13 compared with IL-4 to differen~iate B cells were assessed. Cells transfected with the human CD40-L exhibited induced B cell 2 o aggregation, proliferation, and considerable Ig production, including IgE synthesis, in the presence of IL-13. ;:.^!
`
.
Rea~ents . . .
Human rIL-4 was provided by Schering-Plough Research (Bloomfield, Nl) and human-rIL-13 was provided by W. Dang (DNAX
2s Research Institute, Palo Alto,~CA). The CD40-Ig fusion protein was ~; obtained by fusion of ~the cDNA segments encoding the extracellular domain of CD40 to cDNA fragments encoding the human IgGl. The ... ..
mAb89 was kind~y~ provided by Dr. J. Banchereau (Schering-Plough, Dardilly, France). Streptavidin-PE and all antibodies, unless stated 30 otherwise, were from - Becton-Dickinson (Mountain View, CA).

69 f~60 Cell Purification and Culture B lymphocytes (~98% CD20~) were purified from spleen using density gradient centrifugation over Ficoll-Hypaque (Pharmacia Fine Chemicals, Piscataway, NJ), followed by negative cell sorting using a FACStar Plus (Becton Dickinson). Surface IgD+ positive cells were sorted directly from the negatively sorted B cell population. The CD4+ T-cell clone B21 and the CD8~ T-cell clone A10 have been described by Roncarolo et al., J. Exp. Med. 167:1523 (1988). In co-culture experiments, various numbers of purified B cells were 0 cultured with different concentrations of COS-7 cells in U-bottom 96-well trays in 0.2 ml.
' After 10 days, 50% of the medium was replenished, and after 14 days the supernatants harvested and assayed for Ig's by ELISA.
COS-7 cells were transiently transfected. For CD40-L staining, COS-7 or B21 cells wcre incubated on ice with 1.4 ~g/ml biotinylated CD40-Ig ;n PBS, 1% FCS for 20 min, washed twice in ~BS with l~o FCS, and stained with ll5 dilution of streptavidin-PE, and washed twice again. Cells specifically expressing CD40-L were sorted using a ~ ~ FACStar Plus (Becton Dickinson) before use.
- 20 ~man and Mouse CD40-L cDNA's - - Mouse CD40-L DNA provided as a PCR product- by Dr. N. Harada and Dr; R. Chang (DNAX Research Institute) was subcloned into the ' mammalian expression vector pJFE14. The human CD40-L was ''- ~ cloned by using the mouse CD40-L cDNA as a-probe to screen colony 2s blots of a cDNA library derived from the CD8+ T-cell clone A10. To make the library, 108 A10 cells were activated for 8 hours with 10 ~g/ml con A, harvested, and extracted for RNA. mRNA was purified using a Pharmacia (Uppsala, Sweden) mRNA- Puriffcation Kit. cDNA
was synthesized and cloned esscntially according_to the manufacturers instructions using the SuperSc-r~'p`r'~PI'asmid System (BRL, Grand Island, NY) the only modificadon being the use of pJFE14 as the vector for cloning. The library contained 106 independent clones with an average insert size of 1.4 kb.

WO 94/0468~ PCI`/US93tO7645 2i~286U

Northern and PCR Analysis RNA was isolated using RNAzol B (CNNA: Biotech, Friendswood, TX) according to manufacturer's instructions. RNA from brain, heart, kidney, and small intestine were from Clontech (Palo Alto, CA).
5 cDNA was synthesized using SuperScript (BRL) and PCR reactions performed in a GeneAmp PCR System (Perkin-Elmer Cetus, Emeryville, CA) with 30 cycles of 94C, 55C, and 72C for 0.5 n~in, 0.5 min, and 1 min respectively. Primers for detection of CD40-~transchpts were 5'-ACA GCA TGA` TCG AAA CAT ACA-3', 5'- TGG CTC
0 ACT TGG CTT GGA TCA GTC-3' and for hypoxanthine phosphoribosyltransferase (HPRT) transcripts S-TAT C3G~ (~GG~C
T&~ AC~TCI`1~-3',5' - GAGACA AACAT&A~TG~iA ATCC~GA-3'.
Products of PCR reactions were electrophoresed through 1.2%
agarose and transferred by capillary blotting to GeneScreen nylon 5 membranes ~NEN Research Products, Boston, MA) according to manufacturer's instructions. For Northern analysis RNA was electrophoresed through 0.85% agarose and transferred to BA-S
nitrocellulose (Schleicher and Schuell, Keone, NH). CD40-L 32p cDNA
probes for Nortihern and Southern blots were made using as a 20 template a L.3 kb EcoRI-XhoI fragment of pJFE14-CD40-1, conlaining the CD40-L coding region.
Cloning and Characterization of~the Human CD40-I~
To obtain clones of the human CD40-L, a cDNA library derived from the CD8~ T-cell cIone~ was screened using the mouse C~0-L
25 ~ cDNA as a probe, see Armitage et al., Nature 57:80 (1992).
~-;- Positive clones were present in the library at 0.005% and were rieptcsented prcdominantly by~ a --2.1 kb length clone, but an . ~ additional clone of 1.2 kb was detected. Both cDNA's contained an identical open reading frame of 261 amino acids, which would give 30 riise to a protein with an~lrnniodifiedimolecular mass of 29254. The 2.1 kb and 1.2 kb clones differed only in the length of their 3' untranslated ends, and presumably represent the two mRNA species of that size detected by Northern analysis. The nucleotide and predicted amino acid sequences of the human CD40-L`cloned here -:
"~:

~0 94~04680 ~ ~, PCl /US93/07645

7 1 were identical to those reported by Hollenbaugh et al., EMBO ~.
11:4313 (1992); and Spriggs et al., J. Exp. Med. 176:1543 (1992).
Using a biotinylated human CD40-Fc fusion protein in combination with streptavidin-PE, specific expression could be easily 5 detected of human CD40-L on COS-7 cells transiently transfected with an expression plasmid pJFE14 containing the 2.1 kb human CD40-L cDNA, but not on control cells transfected with empty pJFE14 vector DNA. The human CD40-Fc reagent reacted also with COS-7 cells transfected with the same expression vector containing the 0 mouse CD40-L cDNA, which is consistent with previous studies, indicating cross-species binding of human CD40 to mouse CD40-L.
The CD40-L Induces Homotypic Aggre~ation of B Cells and B Cell Proliferation B cell activation with antibodies to CD40 results in homotypic 5 aggregation. To determine whether the CD40-L had similar effects?
COS-7 cells expressing the CD40-L were purified by FACS and co-cultured with purified B cells or JY cells, an EBV transformed B cell line. Indeed, aggregation of JY cells following incubation with the COS-7 cells expressing human or mouse CD40-L was observed, 2 o whereas mock-transfected COS-7 cells were ineffective. Similarly, purified -B cells co-cultured with cells expressing human or mouse CD40-L displayed marked homotypic aggregations~ whereas B cells cultured with untransfected COS-7 cells remained disperse.
Consistent with the B cell activation observed microscopica~
2 5 significant proliferation was obtained when purified B cells were -co- ~
cultured with COS-7/human CD40-L or COS-7/mouse CD40-L (see Table 5). Thisj proliferation was further enhanced in the presence of IL-4 or IL-13 . The growth promoting effects of IL-4 and IL-13 seem to be comparable under these culture conditions.
~= ~ `

wo 94t04680 PCr/USg3/0764:

Table 5: Inductivn of B Cell Proliferation by IL-13 or IL-4 and COS
Cells Expressing the Human or Mouse CD40-L
3H TdR Incorporation (c.p.m. x 10-3) B 0.1 + 0 B + IL-13 0.1 + 0 B + IL-4 0.2 + 0 B + IL-4 + control mAb O.2 + O
B + IL-4 + anti-CD40 21.2 + 4.1 COS hCD40-L1.1 + 0.2 lS COS mCD40-L1.0 + 0.1 COS 1.4 + 0.2 -B + COS hCD40-L16.9 + 2.4 B + COS mCD40-L17.5 + 2.1 .
B + COS 1.2 + 0.2 B + IL-4 COS hCD40-L30.7 + 3.8 B + IL-4 COS mC~40-L35.4 + 4.5 : B + IL-4 COS 1.3 + 0.2 B + IL-13 COS hCD40-L22.5 + 3.0 B + IL-13 COS mCD40-L33.8 + 2.8 B + IL-13 COS1.2 + 0.4 Five x 104 highly purified (> 98% CD20+) nega~ively s~rted splenic B
- cells were co-cultured with 1.6 x 104 irradiated (7,000 rads) COS
cells transfected with human or mouse~ CD40-L or the empty pJPE-14 vector as control. IL-13 or IL-4 were added at 400 U/ml. Soluble anti-CD40 mAb 89 and the control mAb A4 were used at 50 ~Lg/ml.
The cultures were harvested 3 days later after addition of 3 H
! ~ Thymidine in the last 16 hours- of culture. The values represent ~- means and standard deviations of triplicate cultures.

.

-, i ' .

~94/04~;80 ,?1~2 PCr/US93/0764;

:
;, -13 Induces Ig Production bY COS-7/hCD40-L Stimulated B Cells ~ COS-7 cells expressing human or mouse CI)40-L also induced Ig y production by highly purified naive surface IgD+ human B cells in ~he presence of IL-4 or IL-13 (Table 6). Considerable levels of IgM, IgG4, total IgG and IgE, but no IgA were produced. There was no ; IgA production, compatible with previous observations which indicated tha~ IL-4 specifically inhibits IgA synthesis under these culture conditions. Ig levels induced by IL- 13 were in the same range as those induced by IL-4.
0 No Ig production was obtained in the presence of mock-~ansfected COS-7 cells (Table 6). Induc~ion of all Ig isotypes by COS-7 cells expressing CD40-L was effectively blocked by CD40-Ig (10 )lg/ml), confirming that specific engagement of the CD40-L is necessary for induction of B cell differentiation and Ig production.
Inhibition of total IgG production by CD40-Ig could not be measured, since the Ig portion of the CD40-Ig fusion protein gave a strong signal in the IgG ELISA.
Ig production, including IgG4 and IgE production, induced by IL-13 in the presence of COS-7/CD40-L cells was not blocked by anti-IL-4 mAbs (10 ~g/ml), whereas these mAbs s~ongly blocked IL-4-induced Ig production in the presence of COS-7/CD40-L ~able 6). These results indicate that IL-13 induces Ig production ~
indepeDdently from IL-4. They also indicate that II.- 13 is another cytokine that directs naive surface IgD+ human B cells to swltch to ~5 IgG4 and IgE producing cells in ~he presence of a contact-mediated costimulatory signal delivered by COS-7 cells expressing the mouse or human CD40-L.

_ :.

WO ~4/04680 PCr/US93tO764 21~286u Table 6: Induction of Ig Synthesis by IL-13 or IL-4 and COS Cells Expressing CD40-L
IgM IgG IgG4 IgE
~ng/ml) _ COS hCD40-L 4 t 2 38 + 4 12 ~ I <0.2 COS mCD40-L ~).2 6 + 0 3 + O ~0.2 lL4~COS hCD40-L 87 + 8 195 _ 21 148 + 30 80 + 4 IL-4+COS hCD40-L+CD40-lg 3 + 1 ND* 1.8 + 1.4 2.7 _ 1 lL-4+COS hCD40-~+anti-IL4 8 + 3 28 + 7 5 + 3 4 + 2 IL~+COS mCD40-L 64 + 6 208 + 5 177 + 42 68 + 7 IL ~1 ICOS mCD40-L+CD40-Ig <0.2 ND* ~.4 <0.2 IL-4+COS 0_0 4 + 0 1 + 0 ~0.2 IL-13+COS hCD40-L 51 +-1 151 + 9 127 _9 54 + 7 lL-13+COS mCD4~-L - 31 + 3 100 + 2 55 + ~ 37 + 6 ~-13+COS 3 1 3 5 + 0 1 _ 1 <0.2 IL-13+COS hCD40-L+antiIL4 48 + 8 167 + 12 111 + 7 48 + 4 0.2 8 + 1 2 _ O <0.2 IL- 13 c0.2 7 + 0 1 _ 1 ~0.2 IL-4 c0.2 4 + 1 2 + O <0.2 Highly purified sIgD+ splenic B cells (5,000 cells/well) were co-0 cultured with pJ~k;14 vector transfected or sorted COS-7 cells (2~0 Cells/well) transfected with and expressing the human (h) or mouse (m) CD40-L. IL-13 (30 ng/ml) and IL-4 (400 U/ml) were added as indicated. The sensitivitie-s- of`the ELISAs (0.2 ng/ml for IgE and IgM, 04 ng/ml for IgG4 and total IgG) were determined with 5 calibrated Ig standards (Behring, Marburg, Germany).
~No IgG determination was possibre~as there was detection of the Ig portion of CD40-Ig fusion protein ~ added.
_ i - Expression and Distribution of th_CD40-L
. .
Resting CD4+ T cell clones expressed no, or very low levels of CD40-L, as judged by ~indin~ of PE-labelled streptavidin to biotinylated CD40-Ig bound~~to the T cells. However, significant expression of the CD40-L was observed on the CD4+ T cell clone B21 `
~ 4 h after activation with PHA. Consistent with its presence on the <; ~ 25 surface of B21 cells, hCD40-L mRNA was detected by Northern 94~046~0 ~ PCr/USg3/0764 7 5 ~ 86'~, analysis and by PCR. Low levels of hCD40-L mRNA were expressed in resting B21 cells. Kine~ic studies indicated that the 2.1 kb and 1.2 kb mRNA species were already maximally expressed within 2 hours following activation, irrespective of the mode of activation of the T
5 cells.
The expression was somewhat reduced after 4 h. Considerable reduction in the CD40-L mRNA expression was observed 7 hours after activation, but appreciable levels of CD4û-L mRNA were still visible 48 h after activation. Activation of the B21 cells by Ca2+
10 ionophore plus PMA, Con A, anti-CD3 mAbs plus PMA, or PHA plus PMA, did not result in major quantitative differences, or differences in the kinetics of the hCD40-L mRNA expression, although it seems that activation with Ca2+ ionophore plus PMA is slightly more effective. Distribution of the hCD40-L was analyzed by PCR, using 5 primers complimentary to the coding region of the human CD4~)-L
gene. CD4û-L transcripts were not present in B cells, brain, kidney, heart, fetal liver, or fetal bone marrow, but could be readily detected in CD4+ T cell clones, CD8+ T cell clones, a TCR ~ T-cell clone, purified NK cells, monocytes, fetal thymocytes, and small 20 intestine. Expression of CD40-L in the small intestine may reflect IL- 13 production by infiltrating MNC.
The human CD40-L cDNA, which was cloned and expressed in COS-7 cells is very effective in inducing human B cell activation.
- COS-7/hCD40-L induced homotypic aggregation of EBV-transformed 25 and normal B cells and B cell proliferation, similarly as observed with anti-CD40 mAbs. In addition, differentiation of B ce-lls into Ig - secreting plasma cells was observed in the presence ~ of IL-4 or IL-13. The 2.1 kb hCD40-L cDNA was isolated from a CD8~ T cell cDNA library and appeared to be a full-lsngth clone, which--was by 30 sequence comparison, identical to the 1.8 kb cDNA's ~ dèscribed earlier. An additional 1.2 kb cDNA clone probably represents a second mRNA species of that size which was detected-in~ activated T
cells and which apparently encodes the same protein.
The hCD40-L has 80% homology with the corresponding mouse 3s gene. Interestingly, the hCD40-L has also some degree of homology with TNF-a and TNF-,B. The positioning of the four cysteine residues WO 94/04680 PCI /US93tO7645 214286~
..
and the potential extracellular N-linked glycosylation site in the mouse CD40-L are conserved in the human CD40-L, however the human protein has an additional cysteine substituted at position 194. The CD40-L is reported to be a type I~ membrane anchored 5 protein and there is a hydrophobic region of the human protein (amino acids 22-45) representing a potential signal/anchor domain near the amino terminus. B cell proliferation induced by COS-7/CD40-L was enhanced by IL-4 or IL-13. IL-4 and IL-13 seemed to be equally effective, indicating that IL-13, like IL-4, has B cell 0 growth promoting activity. IL-13, like IL-4, also induced Ig production in cultures of naive surface IgD+ B cells that have been co-stimulated by COS-7/hCD40-L.
Considerable levels of IgM, IgG4, total IgG, and IgE were produced under these culture conditions. The profile of Ig 5~ production induced by IL-4 and IL-13 with hCD40-L is similar to that obtained in the presence of IL-4 and anti-CD40 mAbs. Thus IL- 13 and IL-4 appear equally potent in inducing both proliferation and Ig synthesis in B cells. Furthermore, these results indicate that ~ the hCD40-L provides a co-stimulatory signal for IL-4 or IL-13-~ ~ 2 0 induced B cell differentiation, confirming the important role for ~ ~ D40 in B cell activadon and differentiation. Since these o~ ~ experiments were calTied out with naive sIgD+, these results confilll. previous obserYa~ions that II,-13, in addition to IL-4, is another CD4+ T cell derived lymphokine that can direct B cells to 25 switch to IgG4 and IgE producing cells. Ig production, including ~: IgG4 and IgE production,:indùced by IL-13 in the presence of hCD40-L was not blocked by~anti-IL-4 mAbs, indicating that the effects of IL-13 are mediated independently of IL-4.
Help provided by the CD40-L transfectants, and the specific 30 blocking of this help by ~D40-Ig, indicated that expression of CD40-L
;~ on CD4+ T cells may be an important component of both antigenicand non-specific T-B ceH~in~eractions, leading to B cell activation and 5'~ differentiadon. These data are compatible with blocking studies carried out with mAbs against mouse CD40-L, or CD40-Ig, which 3s indicated that CD40-L and CD40 interaction is critical for T cell help i n the mouse system. It is of importance to note that there is a ,... . .
~''' .
.,.. , ~

;) 94/04C80 ~, PCr/US93/0764~

~0 difference in the consequences of signaling by CD40 and activated CD4+ 1' cells suggesting that additional T cell surface molecules may be involved in productive T-B cell interaction. In fact, the transmembrane form of TNF-a expressed on activated CD4+ T cells is also associated with T cell induced B cell activation and differentiation .
Considering these similarities in functions, it is interesting that the CD40-L and the cell surface form of TNF-a are homologous and share some structural similarities, as do CD40 and the TNF receptor.
0 The substantial help given to human B cells by the cloned human and mouse CI~40-I,'s is consistent with previous studiesr demonstrating that signaling through CD40 is of significant consequence for B cell survival, activation, and differentiation! The mouse CD40-L appears as effective as the human CD40-L in activadng human B cells, which is consistent with the ability of murine CD40-L to bind human CD40. The similarity of the protein sequences and the ability of both mouse and human CD40-L to bind CD40 cross-species indicates this is an important interaction in vi-~o to be so well conserved.
Human IL-l3 cDNA was isolated form the same CD8+ T-cell clone library as CD40-L. However, although IL-13 is expressed in the CD8+ T cells, far more IL-13 is expressed in CD4+ T cells. The potency of the combination of these two novel molecules for induction of IgE synthesis, and their abundant co-expression by CD4+ T cells, together with the prolonged expression of IL-13 mRNA
~ following T cell acdvation may be a mechanism contributing to~ IgE-- ~ production in vivo and IgE mediated allergic reactions.
l These experiments have focused on the function of CD40-L
expressed on CD4+ T cells as it relates to B cell activation. CD40-L
expression on cells other than CD4+ T cells, including CD8+ T cells !, from which the gene was cloned, suggests a broader function for the molecule than in T-B cell interaction. It is likely that CD40-L` ls -expressed on other cell types, and even CD4+ T cells, which will have important consequences for their furiction, rather than just 3s providing a one way stimulus to CD40 positive cells such as B cells.
For example CD40-~ and CD4û expression, in thymocytes and thymic .

WO 94/04680 PCI'/US93/0764 21~2860 epithelium may be indicative of'interactions involved in T cell development.
E. IgE Switchin~ ~
The present series of experiments demonstrated that IL- 13 induced IgG4 and IgE synehesis by human B cells. IL- 13 induced IgG4 and IgE synthesis by unfractionated peripheral blood mononuclear cells (PBMNC) and highly purified B cells cultured in - the presence of activated CD4+ T cells or their membranes. IL-13-induced IgG4 and IgE synthesis was IL-4-independent, since it was 0 not affected by neutralizing anti-IL-4 monoclonal antibody (mAb).
Highly purified sIgD+ B cells could also be induced to produce IgG4 ' and IgE by IL-13, indicating that the production of these isotypes ''`
reflected IgG4 and IgE switching and not a selective outgrowth of commiteed B cells. IL-4 and IL-13, added together at optimal concentrations, had no additive or synergistic effect, suggesting that common signaling pathways may be involved.
` ~ ~ This notion is supported by the observation that IL-13, like IL-4, induced CD23 expression on B cells and enhanced CD72~ surface IgM (sIgM), and class II MHC antigen expression. In addition, like IL-4, IL-13 induced germline transcription in highly purified B `
cells. Collectively, these data indicated that IL-13 is another T cell-derived cytokine that,`in addieion to IL-4, efficiently directs naive human B cells to switch to IgG4 and IgE production.
B cells undergo Ig isoty~e switching and differentiation into 2 5 Ig-secreting cells in response to sIgM mediated signals in the '' presence of coseimulatory factors provided by CD4+ T cells. Antigen-~, specific T-B cell interactions require binding of the T cell receptor to pepdde-class II major histocompatibllity complexes (MHC) on B
~ ~- cells, which rosults in T-cell activation and cytokine synthesis. Once ';~` ~ 30 the T cells are activated they''~à~ctivate B cells in an antigen-independent fashion. ~ ~---Cytokines are essential for B-cell proliferation and 'differentiation; they not onIy determine Ig secretion quantitatively, but they also direct Ig isotype switching. IL-4 induces IgG4 and IgE

., , ~ .

21 ~286~

switching, whereas transforming growth factor-~ (TGF-,B) directs lgA
switching. See, e.g., Van Vlasselaer et~al., J. Immunol. 148:2062 (1992); and Defrance et al., J. Exp. Med. 175:671 (1992). In addition to cytokines, contact-mediated signals delivered by CD4+ T cells are required for B cell proliferation and Ig production. Above, the ligand for CD40, which is expressed on activated CD4+ T cells, was shown to be one such membrane associated molecule that acts as a costimulatory signal for IL-4-dependent IgE production by both murine and human B cells. See, e.g., Armitage et al., Nat~re 357:80 0 ( 1992). Moreover, several cytokines, such as IL-2, IL-5, IL-6, IL-8, I~-10, IL-12, interferon-a (IFN-a), IFN-~, tumor necrosis factor-a (TNF-a), and TGF-~ modulate IL-4-induced IgG4 and IgE synthesis.
IL-4 has been thought to be the only cytokine capable of inducing IgE synthesis. Out of 16 cytokines tested, IL-4 was the only one inducing germline or productive transcripts or IgE
synthesis. In addition, anti-IL-4 mAbs preferentially inhibit IgE
synthesis induced by IL-4 producing T cell clones without significantly affecting IgM, IgG, or IgA synthesis. Also in murine models, anti-IL-4 antibodies strongly inhibit IgE synthesis in vivo 2 0 without affecting the other Ig isotypes. Most importantly, IL-4 deficient mice lack IgE in their sera following nematode infection.
However, a non-IL-4-producing T cell clone induces germline ~
transcription in puriffed B cells indicating that an IL-4-independent pathway of induction of germline E transcription is operational.
IL-13 induces CD23 exprossion and germline ~ mRNA synthesis and IgG4 and IgE switching in human B cells. ~ ~ -Rea~ents Human recombinant IL-13 was purified as described~~ herein.
Recombinant IL-4, IFN-a, and IFN-~ were provided by- Schering-Plough Research (Bloomfield, NJ). Fluorescein isothioc~anate (FITC)-conjugated anti-CD72 mAb and neutralizing anti-TG~ mAb were purchased from R&D Systems, Inc. (Minneapolis, MN). FITC- and phycoerythrin (PE)-conjugated mAbs specific for CD~, CD4, CD8, CD14, CD16, CDl9, CD20, CD23, CD25, (~D56, HLA-DR, and control . .. .. . .. , .... ........ ... -- .. . ... .... . .. . .. .. . ... . . . . . . .. . .. ..... ..

WO 94/046~0 PCI`/US93/0764 ~1~286~

antibodies with irrelevant specificities were obtained from Becton-Dickinson (Mountain View, CA). FITC- or PE-conjugated mAbs specific for LFA-1 (L130), LFA-3, ICAM-1 (LB2), B7 (L307), and class I MHC antigen were kindly provided by Dr. J. Phillips (DNAX).
FITC-conjugated anti-IgD and anti-IgM mAbs were obtained from Nordic Immunological Laboratories (Tilburg, The Netherlands). The purified anti-CD40 mAb 89 ~IgG1) described by Banchereau et al., Science 251:70 (1991), was a gift of Dr. J. Banchereau (Schering-Plough France, Dardilly, France). The neutralizing anti-IL-4 mAb 0 25D2.11 was kindly provided by Dr. J. Ab}ams (DNAX).

Cell Pre~arations Blood samples and spleens were obtained from healthy volunteers or from patients undergoing splenectomy due to trauma, respectively. Mononuclear cells were isolated by centrifugation over Histopaque-1077 (Sigma, St. Louis, MO).
- ~ Purified B cells were obtained by negative sorting using a -~ ~ fluorescence-activated cell sorter FACStar Plus (Becton-Dickinson) or ~; ~ magnetic beads (Dakopatts, Norway). Briefly, splenic MNC were washed twice and PE-conjugated mAbs against CD3, CD4, CD8, CDl4, CD16, and CD56 were added at saturating concentrations and incubated at 4C for 30 min. The cells were washed twice with PBS.
Cells with the light scatter characteristics of lymphocytes were gated, and PE- cells were sorted. Alterna~ively, cells stained with 2s mAbs against CD3, CD4, CD8, C~1~, CD1~, and CD56 were incubated for 30 min at 4C with magnetic beads coated with anti-mouse Ig mAbs.
~ ; The cells bearing murine Ig were removed using a magnetic - field. The remaining cells were washed, counted, and used in further experiments. For isolation of sIgD+ B cells, positive sorting by FACStar Plus was used. Splenic-MNC were stained with PE-conjugated mAbs against CD3, CD4, -CD8, CD14, CDl6, and CD56 and FITC-conjugated anti-IgD mAb, and FITC+, PE- cells were sorted. On reanalysis, purities of the sorted cell populations were >98%, and 3s that of cells isolated using magnetic beads ~95%.

~ 94/04680 . PCr/US93/0764~
~2~ 6 The CD4+ T cell clone B21 and the CD4+ non-IL-4 producing T
cell clone SP-A3 were cultured according tO Roncarolo et al., J. ~xp.
Med. 167:1523 (19~8). The cells were obtained 4-6 days after they had been activated by the feeder cell mixture and PHA. In addition, 5 IL-2 (100 U/ml) was added to maintain the activation state of the T
cell clones.

Pre~aration of T Cell Membranes The membranes of a CD4+ T cell clone were prepared according 0 to Gascan et al., Eur. J. lmmunol. 22:1133 (1992). Briefly, the CD4+ T
cell clone B21 was harvested 12 days after activation with fee~er ' cell mixture and phytohemagglutinin (PHA), the cells were washed and restimulated with 10 llg/ml of Concanavalin A (Con A) for 7-8 h at 37 C. During the last 30 min of the Con A s~imulation, 100 ~g/ml 15- of a-methyl-D-mannoside (Sigma, St. Louis, MO) was added. From ~; these cells, membranes were prepared using the m~thod described by Brian, Proc. Natl. Acad. Sci. USA 85:564 (1988); and Maeda et al., Biochim. Bi~phys. Acta 731:115 (1983); and they were'stored under liquid nitrogen (1 x 108 T cell equivalents/ml = 0.2 mg 2 0 protein/ml membrane preparation) until used.
. C~lture Conditions - ~ Purified B cells were cultured at 5000 cells/well in ~- quadruplicate, in round-bottomed 96-well plates (Linbro, McI,ean,-VA) at 37C in a humidified atmosphere containing 5% CO2 in 0.2ml -25 Yssel's medium supplemented with 10% fetal calf serum (FCSj.
Unfractionated PBMNC were cultured at 105 cells/well in 12 replicates. In'coculture experiments, the CD4+ T cell clone SP-A~ '--- ~
was cultured at 5000 cells/well (T:B cell ratio 1:1). After a culture period of 12 days, Ig levels in the culture supernatants were _---~ 30 measured by ELISA. - -wo 94/04680 Pcr/uss3/0764s ,~ 8 2 Measurement of I~ Production IgM~ total IgG, IgA, and IgE secretion were determined by ELISA as described in Pène et al., Proc. Na~l. Acad. Sci. USA ~5:6880 (1988). IgG4 secretion was determined by ELISA as described in s Punnonen et al., J. lmmunol. 148:3398 (1992). The sensitivities of IgM, total IgG, and IgA ELISAs were 0.~-1 ng/ml, and the sensitivities of IgG4 and IgE ELISAs were 0.2 ng/ml.

PhenQty~ic analysis of cultured cells.
0 Purified B cells were cultured as described above and were harvested and washed twice. FITC- and PE-conjugated mAbs were a;dded at saturating concentrations and incubated at 4C for 30 min.
FITC- and PE-conjuga~ed mAbs with irrelevant specificities were used as negative controls. The cells were washed twice with PBS
and cells with the light scatter characteristics of lymphocytes were analyzed using a FACScan flow cytometer (Becton-Dickinson).
RNA Isolation and Northern AnalYsis Total RNA was isolated using RNAzol B (CNNA: Biotech, Friendswood, TX) according to manufacturer's instructions. RNA was - 20 electrophoresed through 0.85% agarose and transferred to 13A-S
nitTocellulose (Schleicher and Schuell, Keone, NH). 32p cDNA probes were made by random priming using as templates, an EcoRI/HindIII
fragment of pBSIgE1-4 for germline e,~an~-DNA complementary to the BglI/SmaI fragment of pH~gA-1 for actin. See Gauchat et al., J.
2s Exp. Med. l72:463 (1990); and Erba et al., Nucleic Acids Res. 14:5275 (1986).
IL-13 Induces CD23 E3x~ression on Purified B Cells The effect of IL-13 on the expression of a vanety of B cell surface antigens was investigated by FACS analysis~ Incubation of purified B cells with IL-13 (200 U/ml) resulted in strong induction of CD23 expression on a proportion (about 20%) of the B cells. IL-13 also upregulated class II MHC antigen, sIgM, and CD72 expression on ;) 94/04680 ~ - PCI /US93/07645 83 `~
O

B cells. These effects of IL-13 were similar to those observed by IL-4. CD23 expression was already detectable after a culture period of 24 h, but maximal responses were observed after 72 h of culture.
The expression of CDl9, CD20, CD25, CD40~ class I MHC antigen, B7, 5 ICAM-1, LFA-l, and LFA-3 were not significantly modified by L-13.
IL-13 Induces IgE Svnthesis bv PBMNC.
Beeause CD23 expression on B cells has been associated with IgE synthesis, IL-l 3 was tested for induction of IgE synthesis by human PBMNC. As shown in Table 7, IL-13 induced IgE synthesis 0 by unfractionated PBMNC in a dose-dependent manner in the absence of e~cogenous IL-4. In addition, strong IgG4 production in response to IL-13 was observed.
Neutralizing anti-IL-4 mAbs failed to inhibit IL- 1 3-induced IgE synthesis, whereas IL-4-induced IgE production was virtually completely blocked, indicating that IL- 1 3-induced IgE synthesis was not mediated through induction of IL-4 production by PBMNC. See Table 8. Similarly to IL-4, maximal induction of IgE synthesis by 13 was usually obtained at concentrations of 50 U/ml. The mean level of IgE produced in response to IL-13 was somewhat lower (63 20 ng/ml, n=6) than that induced by IL-4 (169 ng/ml, n=6). No additive or synergistic effects were observed when both IL-4 and ~- IL- 13 were used at saturating concentrations. ~ ~
,~ .

Table 7: Induction of IgE and IgG4 Synthesis By IL-13 -- ~~ ~ -~- 2s IgE ~ng/ml) IgG4 (ng/nlll - ! _ . .
Medium <0.2 31:~14 ~ -IL-13 62il 7 413+138 3o -~

, .

:
~, W094/04680 21~28~ 84 PCI/US~3/0764~

.
Table ~: Induction of IgE Synthesis By IL-4 and IL-13; Effect of Anti-IL-4 mAbs I~E $vnthesis (n~/ml) Medium <0.2 IL-13 (05 U/ml~ <0.2 IL-13 (5 U/ml) 13 + 3 0 IL-13 (50 U/ml) 35il 3 IL-13 (500 U/ml~ 3 2+1 0 IL-13 ~500 U/ml)+anti-IL-4 mAb 2 6 + 5 I~E Svnthesis (n~/ml) Medium ~0.2 IL-4 (05 U/ml) <0.2 IL-4 ~5 U/ml) 6 + 3 2 0 IL-4 (50 U/ml) 6 8 +: 13 IL-4 ~500 U/ml) 71+1 1 IL-4 ~500 U/ml)+IL-13 ~SOOU/ml) 7 ~ i 1 0 - : IL-4 (500 U/ml)+anti-IL-4 mAb 4 +1 25 IL-13 Induces I~G4 and I~E Switching in B Cell~
The ability of IL-13 to induce IgG4 and IgE synthesis by purified B cells was also tested. - It- was thereby found that IL 13 ~:induced IgG4 and IgE synthesis b-y highly purified B cells cultured in -the presence of membranes of an activated CD4+ T cell clone. Also in ~o ~is culture system the levels of IL-13-induced IgG4 and IgE!
production were generally lou~er than those induced by IL-4. The .~. difference was in the same rangë as observed in the cultures of unfracdonated PBMNC. IL-13- al~o induced significant levels of IgM
~;and total IgG producdon, but~~no- IgA synthesis was observed~
3s . In this aspect IL-13 has properties similar to IL-4, which generally inhibits IgA synthesis. See van Vlasselaer et al., J.
Immunol. 148:1674 (1992). These results show that IL-13 induces

8 5 PCr/US93/0764~

IgG4 and IgE synthesis by human B cells in the absence of IL-4, and indicate that IL-13 acts directly on B cells to induce IgG4 and IgE
synthesis. Furthermore, these results strongly suggest that IL- 13 induces Ig isotype switching to IgG4 and IgE in an IL-4-independent t manner.
To confirm that IgE synthesis observed in above experiments was due to Ig isotype switching and not to an outgrowth of a few IgE-committed B cells, the effects of IL-13 on naive sIgD+ B cells were studied. Culturing of highly purified sIgD+ B cells with the 0 activated, non-IL-4 producing T cell clone SP-A3 in the presence of IL-13 resulted in induction of IgE synthesis. In addition, IL-13 enhanced IgG4. synthesis induced by this non-IL-4 prodùcing T cell clone alone. As was demonstrated for PBMNC, IL-13-induced IgG4 and IgE synthesis could not be inhibited by anti-I~-4 mAbs.
Inducdon of Germline Transcription bv IL-13 ~ . , So far, IL-4 has been the only cytokine known to induce germline transcription in B cells. Since switching to E by IL-4 is preceded by the induction of germline RNA synthesis, it was `
hypothesized that IL-13 would induce germline transcription as well. Indeed, when highly purified B cells were cultured in the :-presence of IL-13 and anti-CD40 mAbs, germline mR~A synthesis, at levels comparable to that in the presence of I~-4 and anti-CD40 mAbs, was detected after a culture period of five days ~see Tables 9 and 10). Since and-CD40 mAbs alone did not induce germline transcription in B cells, these results indicated that IL-13 is another T cell-derived cytokine that, like IL~, could induce germline E
transcripts in B cells. In addition, these results confirmed the correlation between germline transcription and subs`equent~
switching to IgE synthesis. ` ~

.

WO 94~4680 PCI'/US~3/0764 2l~æ8~D

Table 9~ 13 Induces Ig Synthesis By Fetal BM Cells Cultured In The Presence of Anti-CD40 mAbs IgM I~ G4 I~E
Medium <1 ~1 <0.2 <0.2 anti-CD40 (10 ~g/ml) <1 d c().2 <0.2 0 anti-CD40 (10 ~lg/ml) +IL- 13 (400 U/ml) 5 + 2 1 0 + 3 2 + 2 l + 1 Table 10: IL-13 Induces Ig Synthesis By CD19+, sIgM+ Immature B
Cells and CDl9+, sIgM- Pre-B Cells Sorted CD19+, sIgM+ fetal B cells:
2 0 ~ ~ 4 Ig~
Medium <1 <1 <0.2 <0.2 B 21 <1 <1 <0.2 <0.2 B21~IL-13 <l <1 ~0.2 <0.2 B21+IL-7 <1- <1 <0.2 <0.2 B21~IL-7+IL-13 8-+ 2 - --- 2 3 + 6 6 i 2 2 + 1 B21+IL-7+IL-13 ~anti-IL-4 mAb 7 + 2 2 1 ~: 4 2 + 1 2 + 1 Sorted CD19~, sIgM+ fetal B cells: ~ -I~M ~ I~G4 I~E
Medium <1- - -- <1 <0.2 <0.2 B21 <1 <1 ~0.2 <0.2 B21+IL-13 <1 . <1 <0.2 <0.2 B2 1 +IL-7 <1 <1 <0.2 <0.2 B21~IL-7+IL-13 ~ + 1 6 + 1 2 + 1 1 + 13s B21+IL-7+IL-13 +anti-IL-4 mAb 6_1 3 +2 l + 1 2 1 -O 94~04680 2 1 'I 2 8 6 U Pcr/US93~0764;

Effect of IL-12 on IL-13 Plus Anti-CD40-induced I~E Svnthesis.
Ten thousand highly purified B cells were cultured in the presence of anti-CD40 monoclonal antibodies (20 ~ug/ml) and IL- 13 5 (400 U/ml) or IL-4 (400 U/ml). COS supernatant containing IL-12 or mock COS supernatant was added, and IgE was measured by ELISA after 14 days of culture. IL-12 decreased the IL-13 effect, while increasing the IL-4 effect.
IL-4 has been considered the only cytokine to induce IgE
0 switching in human or murine B cells. This was based on studies showing that anti-IL-4 mAbs preferentially block IgE synthesis both in vitro and in ~ivo, and on the observation that no circulatory IgE
could be detected in mice, in which the IL-4 gene had been disrupted. It was found, however, that IL- 13 -induced IgE synthesi~
is ir~dependent of lL-4, since IL-13 induced IgG4 and IgE synthesis in cultures of highly purified B cells in the absence of exogenous IL-4. In addition, anti-IL-4 rnAbs, which efficiently blocked IL-4-induced IgE synthesis, failed to affect IL-13-induced IgE production.
MoreovEr, IL-13-induced IgG4 and IgE synthesis, like that induced ' 20 by IL-4, reflects Ig isotype switching and is not due to a selective outgrowth of a few B cells committed to IgG4 or IgE synthesis, since IL-13 also induced IgG4 and IgE synthesis by naive,-sorted sIgD+ B
.
cells.
~;~ Switching to IgE by IL-13 was preceded by induction of 2s germline E mRNA synthesis, but costimulatory signals provide~ by activated T cells were required for induction of IgE production.- This is consistent with studies showing that IL-4-induced switching to -~ ;jin' both murine and human B cells is preceded by the induction of germline RNA synthesis, and that co-stimulatory signals~~provided 30 by activated CD4+ T-cell clones or anti-CD40 mAbs are required for the inducdon of praductive mRNA transcripts and IgE-~ynthesis by 4. Although their exact role remains ~o be determined,- it- has been suggested that` germline transcripts play an- important role in the -switch process.

,.-wo g4/04680 Pcr/uss3/o764 ~a 8 8 Despite the fact that IL-4 has been considered to be the only cytokine tO induce germline transcription in B cells, an IL-4-independent pathway of induction of germline tranSrIiptiOn iS
operational, since a non-IL-4 producing T cell clone was also capable 5 of inducing strong germline E RNA synthesis. It is likely that IL-13 produced by the non-IL-4 producing T cell clones is responsible for the IL-4-independent induction of ge~nline ~ mRNA in B cells. The present findings may also explain why induction of IgE synthesis by IL-4 producing T-cell clones was never completely inhibited by 0 anti-IL-4 mAbs. A combination of IL-4 and IL-13 antagonists may be quite effective in blocking the switching process, each present at lower levels, e.g., below threshold levels for adverse side effects.
No additive or synergistic effects on IgE synthesis were observed when IL-4 and IL-l 3 were added together at optimal 5 concentrations, suggesting that IL-4 and IL-13 may use common signaling pathways for induction of IgG4 and IgE switching. Indeed, recent studies have shown that receptors for IL- 13 and IL^4 share a common subunit that functions in signal transduction. However, IL-13 did not bind to cells bearing the 130 kDa IL-4 receptor 2 o indicating that IL-13 does not act through this IL-4 binding protein.
The commonality between IL- 13 and IL-4 was further supported by the observation that IL- 13, like IL-4, induced CD23 expression on purified~ B~cells. Similarly to IL-4, IL-13 also upregulated expression of class II MHC antigen, sIgM, and CD72, 25 which is the ligand for CD5. Although the exact role of CI)23 in the regulation of IgE synthesis-remain`s to be determined, a strong correlation between CD23- exprëssion and induction of IgE synthesis was observed and soluble forms of CD23 were found to enhance IgE
synthesis. Since IL-13 induced significant expression of CD23 within 30 24 h, these data also indicated that CD23 expression preceded IL- 1 3-induced e switching, thereby confirming the correlation between induction of C~23 expression and subsequent IgE synthesis.
Despite the similarities between IL-4 and IL- 13 in their effects on B cells, the functions of IL-4 and IL-13 are not identical.
35 The levels of IgG4 and IgE produced in response to IL-13 were generally lower than those induced by IL-4. Moreover, preliminary 94/04680 8 9 ~ PCI`/US93/0764~

results indicated that IL- 13, ~in contrast to IL-4, does not act on T
cells or T-cell clones. IL-13 has no obvious T cell growth promoting activity and appears not to induce CD8a expression on CD4+ T-cell clones, which may be due to lack of functional IL-13 receptors on T
5 cells. The activation state of T cells was essential for their ability to deliver co-stimulatory signals required for B cell proliferation and differentiation. Therefore, the lack of T cell activation inducing effect of IL-13 may partially explain why maximal IgG4 and IgE
synthesis by PBMNC in response to IL- 13 was lower than that 0 induced by IL-4.
These data seem to be incompatible with the finding that IL-4 deficient mice have no detectable circulatory IgE following nematode infections. However, it is not clear whether IL- 13 also induces IgE synthesis by murine B cells. Preliminary data showed 5 that IL- 13 was produced for much longer periods than IL-4 following T cell activation, suggesting an important role for IL-13 in the regulation of enhanced IgE synthesis in allergic individuals.
IV. Activities on PBMC and Macro~ha~es A. Induction of Mor~hological Change in Non-adberent Human - Peripheral blood mononuclear cells (PBMNC) ~ were isolated from normal health human donors by centrifugation over Ficoll-Hypaque. Total PBMNC (1 x 108 cells) were incubated for 30 minutes at 37C in 10 mm tissue culture dishes. Nonadherent cells 25 were removed by extensive washing of the dish with phosphate buf~ered saline (PBS). Adherent cells were incubated in Yssels's medium Yssel et al., J. Immunol. Me~hods 72:219 (1974) with 1%
human AB serum alone, or with mouse P600 derived from E. coli (lot - 560-137-1; used at a concentration of 30 ng/ml), as described 30 above. Alternatively, COS-7 denved mouse P600 -or-- human IL-13 ~ was used at a final dilution of 1/20. Cells were`o~served at regular - intervals .

:

WO 94/04680 PCI`/US93/0764 s~

B. Modification of Cell Surface Mark'ers on Non-adherent Cells Five or ten days after nonadherence selection as described above~ the resulting cells were analyzed for expression of cell surface markers by fluorescence activated cell sorting (FACS), e.g., as described in Shapiro, Practical Flow Cytometry (2nd Ed.), 1988, Alan Liss, New York. Exemplary antibodies for recognizing each marker' are: CDlla ~LFA-1; SFN-L7, from DNAX, Palo Alto, CA], CDllb ~Bear1, see Spits et al., Eur. J. Immunol. 14:229 (1984)1, CDllc [plS0; NGH
93, see Visser et al., Blood 74:320 (1989)]; CD54 [ICAM; LB2, see 0 Azuma et al.~ J. Expt'l Med. 175:353 (1992)], Class I MHC [W6/32, from Sera Labs, see also Barnstable et al., Cell 14:9 (1978)]; Class II
MHC [Q5/13, see Quaranta et al., J. Immrlnol. 125:1421 (1980)]; Class II MHC [PdV5.2, see Koning et al., Human Immunol. 9:221 (1984)];
Class II MHC (DQ; SPV-L3), CD58 [LFA-3; TS 2/9, see Krensky et al., J.
Immunol. 132:2180 (1984)], CD32 [IV.3, see Looney e~ al., J.
Immunol. 136:1641 (1986)], CD16 [granulocyte-l, see Huizinga et al., Nature 333:667 (1988)]; or Leu lla, Becton Dickinson, Mountain View, CA); CD23 (gp25, from DNAX, Palo Alto, CA), IL-2Ra ~7G7; or '~ BB10, see Herve et al., Blood 75:1017 (1990)], CD44 [NkI-Pl; see ' 20 Vennegoor et al., J. Immunol. 148:1093 (1992)], CD14 (LeuM3, Becton Dickinson), and CD18 and B7 [L130 and L307; both described in Azuma et al., J. lmmunol. 149:1115 (1992)3.
Mouse P600 matèrial''from either COS-7 supernatants or E. coli - ~ inclusion bodies were cornpared to COS-7 supernatants of human IL-13.
2 s C. Nitric Oxide Svnthèsis :
IL- 13 (P600) was assayed by its inhibitiory effect on the ` production of nitric oxid~ ~NO) by GM-CSF-derived bone marrow macrophages. '`The ~macrophages were derived by 9- 12 days culture in RPMI containing GM-CSF and purified by retention of adherent, : GM-CSF-responsive~fractior~. Cells were 99+% pure, as determined -' by FACS analysis using two color staining.
Macrophages were activated to produce NO by stimulation with LPS at 3 ,ug/ml in the appropriate experiments, either with or without prior stimulation with cytokines, as indicated. The , ~ 94~04680 ~ P~/US93/07645 1~860 macrophages were incubated for 16 h with the cytokines (if used) 16 h prior to treatment with I,PS. Supernatants were taken at the indicated times relative to LPS addition, i.e., 0 h is the time of addition of LPS.
Supernatants were assayed for NO production by the standard Griess assay for nitrites. See, e.g., Coligan, Current Protocols in Immunology, (1991 and periodic supplements) GreenelWiley, New York. Addition of cytokines after addition of LPS to the macrophage cultures or at the time of LPS addition has been tested; under these 0 conditions, none of the cytokines tested (including IL-13) had significant effects. Other macrophages were also tested, but since they generally produced lower levels of NO, they were not used as extensively for bioassay.
Table 11, part A shows NO production from GM-CSF-treated bone marrow derived macrophages after treatment for 16 h with designated cytokines. Note that I~N-y induced NO production, while IL-4 or IL-13 inhibited NO production. L-NMMA is a specific ~ inhibitor of NO production. Parts B and C are similar experiments- ~ titrated over different ranges of P600 amounts. In each case, the IL-13 decreased the producdon of NO.

~:

.' .

, .

WO 94~04680 ~ PCT/US93/0764' .
Table 11: NO from GM-CSF-de~ved Macrophages (nM) Treatment 0 hours 24 hours 48 hours mëdium 1.551 3.638 3.103 + LPS ~ ~N-~ 3.852 10.057 9.57t;
+ LPS + ~-4 1.016 1.016 3.798 + LPS ~ L-N~A 0.963 0.802 0.000 + LPS ~ nL-13 0.856 0.802 0.32:L
+ ~PS 0.963 4.708 5.189 B _ medium 0.000 0.144 0.190 LPS 0.193 4.826 8.253 LPS + L-N~A 0.144 0.482 0.145 LPS + n~N7 0.917 11.872 14.189 LPS + IlL-4 0.000 4.633 7.095 LPS + P600 (25 ng/ml)- - 0.000 2.992 4.537 LPS + P600 (50 ng/ml) - 0.000 3.282 4.730 LPS + P600 (100 ng/ml) 0.261 3.137 4.488 C - . _. . , medium -0.705 - 2.512 2.757 LPS 0.766 2.390 2.665 LPS + D~N-y 2.359 10.509 15.319 LPS +L-NM~A -- 0-.643 0.735 0.950 LPS ~ IL-13 (400 pg/ml) 0.306 1.225 1.562 LPS + ~-13 (2 ng/ml) 0.674 1.225 1.593 LPS ~ DL-13 (10 n~n~ - -0.643 1.164 1.532 LPS + rL-13 (50 ng/mI~ - - 0.613 1.072 1.409 LPS + IL-13 (250 ng/ml) 0.613 1.134 1.409 .

~) 94/0'~6X0 PCI`/US93/0764~
S

D. ~LI~
s II~-13 inhibi~s the production of IL-la. -6~IL-10. and TNFa by LPS activated human monocvtes Peripheral blood mononuclear cells were isolated from normal healthy donors by centrifugation over ~icoll-Hypa~ue. Total PBMNC
~100 x 106 cells/100 mm tissue culture dish) were incubated for 30 ~
min at 37 C and subsequently nonadherent cells were removed by extensive washing of the tissue cultu~e dish with PBS. Adherent cells were incubated in Yssels medium with 1% human AB serum in the absence or presence LPS (E. coli 0127:B8, I)ifco, Detroit, MI) in combination with IL-4 (50 ng/ml), IL-13 (50 ng/ml), or IL-10 (100 - U/ml). In addition, cells were activated by LPS in combination with IL-4 or IL-13 in the presence. of neutralizing anti-IL-10 mAb 19Fl (10 ~g/ml). Supernatants were collec~ed after 12 hrs and production of IL-la, IL-6, IL-10, and TNF-a was measured by cytokine-specific ELISA. Table 12 shows the- results of these studies.

- ~ - -- ., W094/04680 'b~"~ PCI/U593/0764' Table 12: Ef~ect of IL-13 on the Produ~tion of IL-la, IL-6, IL-10, and TN~-a by LPS Activated Human Monocytes.
IL-la IL-6 IL-10 TNF-a s fnglml) (ng/ml) (n~/ml! (ng/ml Medium 0 0 0 0 LPS 8.1 54.4 35.4 2.2 LPS + rL-4 1.7 33.1 30 0.7 LPS + IL-13 2 35.4 22 0.5 LPS + IL-10 - O 8.6 ND O
LPS + aIL-10 mAb 12 101 ND 10.6 LPS + aIL-10 mAb + IL-4 3.7 ~9.5 ND 1.2 LPS + aIL-10 mAb + II,-4 5.4 79.5 ND 1.5 These results indicated that IL-4 and IL-13 inhibited the production of IL-la, IL-6, IL-10, and INl~-a by LPS activated human monocytes. IL-10 also inhibited the production of IL-l~, IL-6, and TNF-a by LPS activated human monocytes. IL-10 was produced by ~; human monocytes and inhibited IL-la, IL-6, and TNP-a in an 20 autoregulatory fashion. Addition of IL-10 neutralizing mAb l9F1 showed that endogenously produced IL-10 also inhibited the producdon of IL-la, IL-6, and~ TNF-a. The inhibitory effects of IL-4 and IL- 13 on cytokine production by LPS activated human monocytes were inde~endent of IL-10 since IL-4 and IL-13 25 inhibited the producdon of II~-la-,-IL-6, and TNF-a in the presence of neutralizing anti-IL-10 mAb l9F1.

_,,, E. Antibodv De~end~nt Cell-mediated Cvtotoxicity tADCC!
The present experiments investigate the effects of IL- 13 alone 30 or in combination wi-th IL-4, IFN-y, or IL-10 on human monocytes.
IL- 13 induced significant changes in the phenotype of monocytes.
Lilce IL-4, i~ enhanced the expression of CDllb, CDllc, CD18, CD29, CD49e (VLA-5), class II MHC, CD13, and CD23 whereas it decreased J 94/04680 ~? PCI /US93/07645 95 , ~S~ ' the expression of CD64, CD32, CD16, and CD14 in~a dose-dependent manner. IL- 13 induced upregulation of class II MHC antigens and its downregulatory effects on CD64, CD32, and CD16 expression were prevented by IL- 10. IFN-y could also partially prevent the IL- 13 5 induced downregulation of CD64, but not that of CD32 and CD16.
However, IL-13 strongly inhibited spontaneous and IL-10 or IF'N-~induced antibody dependent cell-mediated cytotoxicity (ADCC) activity of human monocytes toward anti-IgD coated Rh+
erythrocytes, indicating that the cytotoxic activity of monocytes was 0 inhibited.
Furthermore, IL-13 inhibited production of IL-la, IL-l~, IL-6, IL-8, IL-10, IL-12 P35, IL-12 P40j GM-CSF, G-CSF, IFN-a and TNF-a by monocytes activated with LPS. In contrast, IL-13 enhanced the production of IL-lRA by these cells. Similar results on cytokine 5 production were observed or have been obtained for IL-4. Thus IL-13 shares most of its activities on human monocytes with IL-4, but no additive or synergistic effects of IL-4 and IL-13 on human monocytes were observed suggesting that these cytokines may share common receptor components. Taken together, these results 20 indicate that IL- 13 has anti-inflammatory and immunoregulatory activities.
Activated T cells secrete a number of biologically active poly~eptides, which regulate the proliferation, differentiation and function of cells participating in immune responses against antigens.
2s T cells producing IL-2, IL-3, IL-4, IL-5, IL-6, IL-10, IFN-y, GM-CSF, and TNP/LT simultaneously following antigenic or polyclonal- -stimulation have been described in both mouse and man. --These T
helper cells were designated ThO cells in order to distinguish them from the more specialized Thl and Th2 subsets. Murine Th1 cells 30 produce IL-2, I~N-y, TNF/LT, IL-3, and GM-CS~ which supports their function as regulatory and effector cells in cellular imrnune - responses such as delayed type hypersensitivity (Dl~-~)~--whereas Th2 cells produce IL-4, IL-5, IL-6, IL-10, IL-3, and GM-CSP- which makes them suitable for providing help to B cells in the production of 35 immunoglobulins of different isotypes.
, -:
." ~ .
,,,.,, I
, I

WO 94/04680 - PCI /US93/0764' 5~ 96 In man, T cell clones with restricted cytokine production profiles have also been isolated from patients with inflammatory or allergic diseases. Although these types of clones resembled murine Thl and Th2 clones, there were some differences. Depending on s their mode of activation, Thl clones generally could still produce low quantities of IL-4 whereas Th2 clones were able to produce low to normal quantities of IFN-~. However, a clear imbalance in the production ratios of IL-4 and IFN-~ by Th2 clones was observed following antigenic stimulation. Therefore human T cell clones were 0 defined which produce high levels of I~ and no, or low levels of IL-4 Thl "like" cells and T cell clones which produce no, or low levels of IFN-y and high levels of IL-4 Th2 "like" cells.
Furthermore, IL-10 which is exclusively produced by ThO and Th2 T
cell subsets in the mouse, is produced by ThO, Thl "like", and Th2 "like!' subsets in man.
The present invention makes available a new cytokine, human IL-13, which is related to the mouse P600 protein. Both human IL-13 and mouse P600 proteins were biologically active and affected ~human monocyte and B cell functions.
The biological activities of mouse and human IL-13 on human monocytes were further characterized and compared to those of IL-4, IL-lO, and lFN-~, other cytokines with stimulatory of inhibitory actions on human- monocytes. IL-13 induced dramatic c hanges` in the phenotype ~of- human monocytes and inhibited the 25 ~ production of IL-la, IL-l~, IL-6, IL-8, IL-lO, GM-CSF, G-CSF, and TNF-a following acdvatron by-LPS, whereas it induced the production of IL-lRA. ~l~ese results indicate that IL 13 has anti-inflammatory activities and may play an important regulatory role in ! immune responses. -- -i; ~ .
- ~; 30 Isolation and Culture of-Human Monocvtes i-;~ Human monocytes were isolated from peripheral blood of healthy donors by centrifugation over Ficolll-Hypaque and -~ adherence to plastic. -Briefly, lOO x 106 PBMNC were plated on a 100 mm tissue culture dish in Yssel's medium sùpplemented with . , ~ ~4/04680 ~,~ PCr/US93/0764~
86,~

human serum albumin (HSA) and 1% pooled human AB+ serum and incubated at 37 C for 30 min. This culture medium was endo~oxin free as determined by the Limulus arnoebocyte lysate assay ~< 0.2 ng/ml of endotoxin). Subsequently, nonadheren~ cells were 5 removed by extensive washing and cultured in Yssel's medium with HSA and 1% pooled human AB serum as indicated. Alternatively, highly purified human peripheral blood monocytes were obtained from 500 ml blood of normal donors by centrifugal elutriation.
Mononuclear cells were isolated by density centrifugation in a 0 blood component separator, followed by fractionation into lymphocytes and monocytes. The monocyte preparation was > 95%
pure, as judged by nonspecific esterase staining and contained more than 98% viable cells. These monocytes were cultured in Yssel's medium with HSA and 1% pooled human AB+ serum at a concen~ation of 4 x 106 cellslml in teflon bags (Jansen MNL, St Niklaas, Belgium), which prevented adhesion of these cells. After culture for the times indicated, monocytes were collected and analyzed for cell surface expression by indirect immunofluorescence or analyzed for Iymphokine gene expression by Northern and PCR
2 o analysis. In addition, monocyte culture supernatants were collected for determination of IL-la, IL-1~, IL-6, IL-8, IL-10~ TN~-a, GM-CSF, G-CSF, and IL-lRA production following activation of these cells by LPS (E. coli 0127:B8) (Difco, Detroit, MI) at 1 ~lg/ml. The viability of the cells after culture always exceeded 95% as determined by 25 trypan blue exclusion.
Rea~en~ ~ ~
Recombinant human and mouse IL-13 were expressed in E.
coli as insoluble aggregates of glutathione-S-transferase`-fusion proteins, extracted by centrifugation, solubilized, and subjected to 30 renaturation prior to digestion with thrombin to remo~ the N-terminal fusion part. Subsequently, proteins were purif-ied by cation exchange and gel filtration chromatography, which resulted in active human and mouse IL-13. Purified human r-IL-l~, r-IL-4, and r-WO 94~M680 ~ PCI/US93/0764:
~ 1 98 2~42860 IFN-~ were provided by Schering-Plough Research Institute (Bloomfield, NJ).
The neutralizing anti-IL-4 mAb 25D2 [Chretien et al., J.
~mmunol. Metfiods, 117:67 (1989)], and anti-IL-10 mAb l9F1 [Abrams et al., Immunol. Rev. 125:5 (1992)], were described previously. The following mAbs were used for immunofluorescence studies on the expression of cell surface markers: SPV-L7 [CDlla;
Spits et al., Hybridoma 2:423 (1983)], Bear-l [CDllb; Keizer et al., Eur. J. Immunol. 15:1142 (1985)], CLB FcR gran-1 [CD16; Klaassen et al., J. Immunol. 144:599 (1990)], gp25 [CD23; Bonnefoy et al., J.
Immunol. 138:29?0 (1987)], IV.3 [CD32; Looney et al., J. Immunol.
136:1641 (1986)], 32.2 ~CD64; Anderson et al., J. Biol. Chem.
261:12856 (1986)], Q5/13 [HLA-DR/DP; Quaranta et al., J. Immunol.
125:1421 (1980)~, PdV5.2 [HLA-DR/DP/DQ; Koning et al., Hum.
Immunol. 9:221 (1984)], SAM-l [VLA-5, CD49e; Keizer et al., ~ur. J.
Immunol. 17:1317 (1987)], CD29 (Ts2/16; a kind gift of C. Figdor, Amsterdam), L307 [B7; Azuma et al., J. Immunol. 149:1115 (1992)], IOM13 (CD13; purchased from AMAC, Inc., Westbrook, ME); Leu-M3 (CD14), LeulS (CDllc), and L130 (CD18) were obtained from Becton-20 Dickinson (San Jose, CA).
Probes Oligonucleotides used~for Southem analysis of IL-la, IL-1~
IL-6, IL-8, IL-10, TNP-a, GM-CSF, G-CSF, and ~-actin PCR products have been described by de Waal Malefyt et al., J. Exp. Med.
2s 174:1209 (1991). ~~ -- -The following oligonuclotides were used to detect ' ! j ' ,, IFN-a: 5'-TTCTGGCTGTGAGGAAATACT-3' (nt 360-378), ~L-lRA: 5'-GTCAATTTAGAAGAAAAGATAGATGTGG-3' (nt 207-234), IL-12 P35: 5'-AATGGGAGTTGCCTGGCCTC-3' (nt 488-507), , ~
30 IL-12 P40: 5'-TAAGACClTrCTAAGATGCGAGGCC-3' (nt 417-441), and 94/04680 ~ PCI'/US93J07645 TGF~ S-CGAGCCTGAGGCCGACrACTACGCCAAGGAGGTCACC-CGC-3 (nt 1131-1170).
mRNA isolation and Northern analvsis Total RNA was isolated from 20 x 106 monocytes by a - 5 guanidinium thiocya~iate-CsCl procedure. For northern analysis, 10 llg total RNA per san~ple was separated according to size on l~o agarose gels containing 6.6% formaldehyde, transferred to Nytran nylon membranes (Schleicher 8c Schuell, Keene, NH), and hybridized ~; with probes, labelled to high specific activity (> 108 cpm/mg) by a 10 hexamer labelling technique. Filters were hybridized, washed under stringent conditions, and developed.
PCR Analvsis ~ ~ One microgram of total RNA was reverse transcribed using - oligo (dT)12 18 as primer (Boehringer Mannheim, Indianapolis, IN) 5 and~ AMV reverse transcriptase (Boehringer Mannhei~) in a 20 ~11 reaclion volume. Two microliters of reverse transcript (equivalent to ~100~ng of total RNA) was used directly for each amplification reaction.~ ~ Conditions for PCR were as follows: in a 50 ~Ll reaction volume, 25 nmol of each primer, 125 IlM each of dGTP, dATP, dCTP, 20 ` `and dl~P (Ph~armacia, UpQsala, Sweden), 50 mM KCl, 10 mM Tris--~ ~ HC~, pH 8.3, 1.5 mM Mg12, 1 mglml gelatin, 100 ~lg/ml non-` acetylated~B-SA, and l unit Vent D~A polymerase (New England Biolabs, Beverly, MA).
Primers used~to amplify IL-la, IL-l~, IL-6, I~-8~ , ~a, 2s ~ GM-CSF, G-CSF, and ,g-acdn hàve been described previoùsly by de Waal Malefyt e~ al., J. Exp. Med. 174:1209 (1991). The following '''~!' '' ~p~mers werefalso~used: I~N-a sense primer: 5'- ~` i -- ~ GCTGAAACCATCCCTGTC-3' (nt 161-178), IFN-a andsense primer: !
ç-- ~ ~ 5'-CTGCTCTGACAACCTCCCAG-3' (nt 450-430), IL lRA sense primer:
~- 30 5'-GCAAGCClTCAGAATCTGGGATG-3' (nt 118-14~ IL--lRA antisense f" ~
primer: 5'-GATGTTAACTGCCTCCAGC~GGAGTC-3' (nt 3i4-319), IL-12 P35 sense primer: 5'-CTTCACCACTCCCAAAACCTG-3' (nt 281-302), IL-12 P35 antisense primer: 5'-AGCTCGTCACTCTGTCAATAG-3' (nt 813-792), IL-12 P40 sense primer: 5'-CATTCGCTCCTGCTGCTTCAC-3' .:
~, WO 94/04680 PCI`/US93/0764 10~
2 1 ~ 2 8 6 ~ !

(nt 337-358),IL-12 P40 antisense primer: 5'-TACTCCTTGTTGTCCCCTCTG-3' (nt 603-582), TGF-~l sense pnmer 5'-ACCGGGTGGCCGGGGAGAGTGC-3' (nt 1097-1118), TGF-~l ~tisense primer: 5'-GCCGGTTGCTGAGGTATCGCCAGG-3' (nt 1399-1376).
Reactions were incubated in a Perkin-Elmer/Cetus DNA
Thermal cycler 9600 for 25 cycles (denaturation 30s at 94C, annealing 30 s at 55C, extension 60 s at 72C). Forty microliter of each reaction was loaded on 1% agarose gels in TAE buffer and PCR
products. were visualized by ethidium bromide staining.
Subsequently, gels were denatured in 0.5 M NaOH, 1.5 M NaCl, neutralized in 1 M ammonium acetate, and transferred to Nytran nylon membranes. Membranes were pre-hybridized in 6 x SSC, 1%
SDS, 10 x Denhardt's solution (0.2% Ficoll, 0.2% polyvinylpy~olidone, 0.2% BSA, pentax fraction V), and 200 llg/ml E. coli tRNA
(Boehringer, Mannheim, I:RG) for 4 h at 55C.
Oligonucleotide probes (200 ng), speci~lc for a sequence internal to the primers used in the amplification, were labelled at the 5' end by T4 polynucleotide kinase (New England Biolabs) and ~-32P-ATP (Amersham, Arlington Heights, IL). Probes were separated from non-incorporated nucleotides by passage over a Nick column (Pharmacia, Uppsala, Sweden) and added to the hybridization mix. Following hybridization for 12 hrs at 55C, fflters were washed in 0.1 x.SSC (1 x SSC: 150 mM NaCI, 15 mM Na-citrate pH = 7.0), 1% SDS at room temperature, and exposed to Kodak XAR-5 films for 1-2 hrs. In addition, signals were quanti~led on a Molecular Dynamics phos~2ho~-i'mager (Molecular Dynamics, Sunnyvale,CA) . -Lymphokine Determinations The production of :cytokines by monocytes was determined inculture supernatants by - ~ytokine specific ELISA's. The cytokine specific ELISA's'and~'thëir'sensitivities wcre the following:.IL-la, Endogen (Boston, MA) (50 pg/ml); TNF-a, Endogen (Boston, MA) (10 pg/ml); IL-1~,-Cistron (Pine Brook, NJ) (20 pg/ml); IL-6, Genzyme (Boston, MA) (0.313 ng/ml); IL-8, R&D Systems O 94/04680 ~? ~ PCI/US93/07645 lol ~6 (Minneapolis, MN) (4.7 pg/ml); G-CSF, R&D Systems (Minneapolis, MN) (7.2 pg/ml); IL-lRA, R&D Systems (Minneapolis, MN) (12.5 pg/ml); GM-CSF, Bacchetta et al., J. Immunol. 144:902 (1990), (~0 pg/ml); and IL- 10 (75 pgJml) . '~

s Immunofluorescence_AnalYsis Cells ( 105 ) were incubated in V bottom microtiter plates (Flow Laboratories, McLean, VA) with 10 ~l of purifiled mAb (1 mg/ml) for 30 min at 4 C. After two washes with PBS con~aining 0.02 mM
sodium azide and 1% BSA (Sigma, St Louis, MO), the cells were 0 incubated with 1/40 dilution of FITC labelled F(ab')2 fragments of goat anti-mouse antibody (TAGO, Inc. Burlingame, CA) for 30 min at 4 C. After three additional washes, the labelled cell samples; were analyzed by flow microfluometry on a FACScan (Becton Dickinson, Sunnyvale, CA.).
5 Antibody De~endent Cell-mediated Cvtotoxicitv fADCC~
ADCC activity of cultured human monocytes against antibody coated rhesus positive human erythrocytes was performed as previously described by de Velde e~ al., J. Immunol. 149:4048 ( 1 992).
20 IL-13 and II~-4 Induce Identical Cbanges in the Expression-of Cel~
~face Anti~ens bv Human Monocvtes Both mouse and human IL-13 induced expression of-C~3~
- (FcRII) and upregulated the expression of class II MHC antigens on human monocytes. The effects of IL-13 on the expression of a 25 larger panel of cell surface antigens was investigated. IL-13 affected the expression of multiple cell surface molecules belonging to different supergene families. IL-13 enhanced the expression of several members of the integrin superfamily of adhesion- mohecules.
The expression of a subunits CDllb (C3bi receptor, Mac-1), CDll`c 30 (gplS0,95), and VLA-5 (FNR), as well as their respective ~ subunits CD18 (~2) and CD29 (~1, VLA-b) were upregulated by IL-13. The expression of other members of this family, including CDlla ~FA-1).

WO 94/04680 PCl`/US93tO764`
: 102 2~ 6~) ' VLA-2 (CD49b), VLA-3, VLA-4 (CD49d), VLA-6 (CD49f), ~3 (CD61), and ~4 was not significantly affected by IL- 13 .
II,-13 enhanced the expression of class II MHC antigens. The expression of HLA-DR, HLA-DP and HLA-DQ was upregulated by s IL-13. Expression of other members of the Immunoglobulin superfamily including class I MHC, CDl la (LFA-l), CD54 (ICAM-1), ICAM-2, and CD58 (LFA-3) was not dramatically affected by IL-13.
IL-13 modulated the expression of the various Fc receptors on monoeytes. The expression of CD64 (Fc~RI), CD32 ~Fc~RII), and CD16 o (FcyRIII) on human monocytes was strongly downregulated by IL-13. In conaast, IL-13 induced the expression of CD23 (FcRII).
In addition, IL- 13 upregulated the expression of CD 13 (Aminopeptidase N) and downregulated the expression of CD14. No major effect of IL-13 was detected on the expression of CD25, CD33, and CD44.
IL-4 induced upregulation of CDllb, CDllc, CD18, VLA-5, CD29, class II MHC, CD13, and CD23, and inhibited the expression of CD16, CD32, CD64, and CD14 on human monocytes to the same extent as did IL-13. Taken together, these results indicated that the IL-13 2 o induced changes in the expression of cell surface molecules were similar to those induced by IL-4. Incubation of monocytes with ~;
saturating concentrations of both IL-4 and IL-13 did not result in changes in the phenotype a~ compared to those induced by either ~; cytokine alone.
No additive or synergistic activities of IL-13 and IL-4 on the expression of the various ce~l- surface molecules were detected under these conditions. There is no evidence that monocytes are able to produce IL^4. However, to exclude the possibility that IL-13 acted through the induction of IL-4 b~ monocytes or by a few contaminating T cells, monocytes were incubated in the presence of IL-13 and a neutralizing anti-IL-4 mAb.
As shown in Table l~, the induction of CD23, downregulation of CD14, and upregulation of class II MHC by IL-13 was not affected by the anti-IL-4 mAb. The anti-I~-4 mAb, however, was effective since it completely inhibited the effects of IL-4 in control ; experiments. Thus, IL- 13 acts independeritly of IL-4.

3 94~04680 ~;? PCI`/US93/07~45 b~

. _ _ Table 13 ~ 13 Acts Independently of IL-4.
mAb Medium IL-13 IL-13 IL-4 II.-4 + aIL-4 + aIL-4 control 3 * 5 6 5 3 MHC class II 443 1904 1845 2084 220 5 Monocytes were incubated with medium, IL- 13 (50 ng/ml) or IL-4 (4Q0 U/ml) in the absence or presence of neutralizing anti-IL-4 mAb 25D2 (10 ~Lg/ml) at 37 C for 120 h and expression of HLA-DR/DP (QS/13), CD23 (gp2~)9 and CD14 (Leu-M3) was determined by indirect immunofluorescence.
0 * Mean Fluorescence Intensity (channel num~er) IL-13 induced changes in expression of cell ~face markers were dose-dependent as shown for the modulation of CDllb, CD18, CD16, CD32, CD64, CD23, class II MHC, CD13, and CD14 expression 5 (Table 14). Generally, incubation of human monocytes with 5 pg/ml IL-13 was insufficient to induce changes in the expression of these cell surface markers, whereas 0.5 ng/ml IL- 13 resulted in significa~t changes in phenotype, comparable to those induced- by 0.~ nglml IL-4. Maximal responses were. induced by S0 nglml 20 IL~13, which were again in the sarne range as those induced by 50 ng/ml of IL-4, indicating that IL-4 and IL-13 were equally effective.

.

WO 94/04680 PCI`/US93/0764 2 1 ~ 2 8 10 4 Table 14~ 13 Induces Changes in Cell Surface Phenotype of Monocytes in a Dose Dependent Manner.

mAB IL- 13 (pg/ml) IL-4 (U/ml) control 3 * 3 3 4 3 3 CDllb 59 54 102 139 102 168 CD16 25 20 20 lS 20 13 MHC class II 355 386 586 607 609 908 ~D13 26 26 113 121 57 102 S
Monocytes were incubated with medium, IL- 13 (5 pg/ml, ~00 pg/ml, or 50,000 pg/ml) or IL-4 (4 U/ml, or 400 U/ml) at 37C for 120 h and ~he expression of cell surface antigens was determined by indirect immunofluorescence.
* Mean Fluorescence Intensity (channel number) .
10 Downre~ulates IL-13 Induced Class II MHC Expression on Human MonocYtes .
To compare the effects of IL-13 with those of other cytokines which modulate the cell surface phenotype, monocytes were f iincubated with IL-10 or IFN-y in the absence or presence of IL-13 and the expression of cell surface antigens was analyzed. IL-lO or IFN-~ alone did not dramatically affect the expression of CDllb, CDl lc, CD18, CD13, CD23, C~ 9,- and VLA-5. In addition, IL-10 or IFN-~ did not noticeably affect the IL-13 induced increase in expression of these markers. IL-10 or I~ had also no observed effect on the expression of CDl4 and the IL-13 induced inhibition of CD14 expression.

.

04680 Pcr/US93/0764 However, IL-10 downregulated not nly the constitutive class II expression on monocytes~ but also inhibited strongly ~he IL- 13 induced class II MHC expression. Similar data were obtained when highly purified monocytes isolated by elutriation and cultured in 5 teflon bags were used (Table 15) Increased expression of class II
MHC antigens was observed following incubation of monocytes in medium alone, which was completely prevented by IL- 10. m-IL- 13, h-IL-13, IL-4, and II:N-y all induced high levels of class II MHC
expression which were blocked by IL-10 (Table 1~). Class II MHC
0 expression induced by IFN-~ was further enhanced by IL-13. IFN-y slightly upregulated expression of B7. Taken together, these results indicate that IL-13, IL-10, and I~N-^y independently modulate the expression of cell monocyte surface antigens.
. - ~
Table 15: IL-10 Inhibits Constitutive and IL-13, IL-4, and IFN-y Induced MHC Class II Expression on Human Monocytes. -Incubation IL-10 (200 U/ml) control 4C 69* nd**
medium 37C 1 5 0 4 6 mIL-13 212 73 hIL-13 ~ 197 8 1 IFN-y 3 4 7 3 6 .
Elutriated monocytes were incubated in medium at 4C or 37C, mIL-13 (50 ng/ml), hIL-13 (50 ng/ml), IL-4 (400 U/ml) or IFN-~y (100 U/ml) in the absence or presence of IL-10 (200 U/ml) in teflon bags for 48 h and expression of HLA-DR/DP was determined-by indirect immunoflourescence.
2s * Mean ~luorescence Intensity (channel number) ~--** Not done WO 94/04680 PCI'/US93/07645 ~,~.42~6~ 1~6 IL-13 Inhibits Monoçvte Fc~R Cell Surface Expression and Cvtotoxicitv IFN-y, IL-4, and IL-10 are able to modulate the expression of Fc~RI ~CD64), Fc~RII (CD32), and Fc~RIII (CD16) on human s monocytes. I~ and IL-10 enhanced the expression of CD64 whereas IL-4 downregulates the expression of CD64, CD32, and CD16. Adding combinations of these cytokines to monocytes showed that I~-10 was able to prevent the IL-4 induced downregulation in cell surface expression of all three Fc~R and that IFN-y partially 0 restored the downregulatory effects of IL-4 on CD64 expression.
IL-10 prevented IL-13 induced downregulation of CD64, CD32, and CD16. In addition, IFN-~ could partially rescue IL-13 induced downregulation of CD64, but did not affect the IL-13 induced downregulation of CD32 and CD16.
` 15 The level of ADCC activity of human monocytes has been shown to ~correlate with the expression of FcyRI. The effects of IL-13 on the functional activity of Fc~RI on monocytes was determined by their ability to lyse anti-D opsonized human Rh+ erythrocytes~ Both human and mouse IL-13 were able to inhibit ADCC activity of 20 monocytes cultured in medium alone. On the other hand, ADCC
activity was enhanced when monocytes were cultured in the presence of IPN-y or IL-10. IL-13 significantly inhibited these effects of IPN-y and IL-10 despite the~fact that~ y and IL-10 partially or completely reversed the inhibition of Fc~RI expression, ~2s indicating that IL-13 affected the Fc7R mediate~ cytotoxicity also by othcr mechanisms. -- -- ~ IL-13 Inhibits Production of Pro-inflammatorv Cvtokines and ~' Hemo~oietic Growth Factors But InducqS l~ R~,.
~ ~ .
To determine the effects of IL-13 on the production of 30 cytokines by human monocytes, monocytes- w-ere activated by LPS
and cytokine producdon was determined` in the culture supernatants ~ ~ after 6 and 24 hours by cytokine specific ELISA's. Activation of '5''~:',~''; ~ monocytes by LPS resulted in the production of IL-la, IL-1,B, IL-6,

9' ~ 8, IL-10, GM-CSF, G-CSF, TNF-a, and I~-lRA. Significant levels of ~:) 94f04680 PCl'/US93/07645 107 ~?~

IL-la, IL-l~, IL-6, IL-8, TNF-a, and IL-lRA were present at 6 h after activation, whereas the production of IL-10, G-CSF, and GM-CSF
was detected at 24 h. At 6 and 24 h after activation, IL-13, IL-4, and IL-10 inhibited the production of IL-la, IL-l~, IL-6, IL-8, 5 IL-10, TNF-a, G-CSF, and GM-CSF, but enhanced the production of IL-lRA.
IL- 13 affected the morphology, phenotype, function, and cytokine production of monocytes. Incubation of monocytes with IL- 13 induced strong adherence of these cells to plastic substrates 0 and their morphology changed to a dendritic appearance. In addition, homot~pic aggregates of cells were observed. The finding that IL-13 upregulated the expression of CDllb, CDllc, CD18, YLA-5, and CD29, which are members of the integrin superfamiliy, is compatible with the observed aggregation and ehanges in morphology, since CDllb/CD18 and CDllc/CD18 heterodimers are involved in cell-cell interactions, homotypic aggregation, adhesion to artificial substrates, and bind fibrinogen.
In addition, the aS~l integrin VLA-5/CI)29 is the receptor for fibronectin, which is an abundant extracellular matrix protein 20 involYed in adhesion processes. IL-13 did not induce changes in the expression of other molecules involved in adhesion or cell-cell interaction, e.g., CDlla, VLA-2, VLA3, VLA-4, VLA-6, ~3, ~4, ICAM-1, ICAM-2, LPA-3? MEL-14, and CD44 but it remains possible that other cell surface structures are involved in the IL- 13 induced 2s changes in morphology and adherence.
IL- 13 upregulated the expression of class II MHC antigens on human monocytes. The expression of HLA-DR, HLA-DP, and ~A-DQ
was significantly increased by IL-13. IL-10 inhibited constitutive and IL-4 and I~N-^y induced class II MHC expression on human 30 monocytes. IL-10 thus inhibits IL-13 induced class II MHC
expression, which further supports the general immunosuppressive activities of IL-10. ~~
The expression of the various Fc receptors for IgG and IgE on monocytes was influenced by several cytokines. CD64 (Fc~RI) 35 expression was upregulated by IFN-y and IL-10 and inhibited by IL-4. Furthermore, IFN-y and IL-10 ~ere able to prevent the WO 94~04680 PCI~/US93/0764i ~ 4~ 60 108 downregulation of CD64 induced by IL-4. ~Iere it was shown that IL-13 inhibited the constitutive e~pression of CD64 and that this inhibition could also be prevented by IL-10 and IFN-y. The expression of CD54 has been shown to correlate with ADCC activity of monocytes.
Spontaneous or IL-10- or IFN-~-induced FcyRI mediated cytotoxicity of monocytes towards IgD coated rhesus positive erythrocytes was strongly inhibited by IL-13 indicating that IL-13 not only affected the phenotype but also the function of human 0 monocytes. Although IL-10 could prevent the IL-13 induced downregulation of CD64 expression, ADCC activity was still inhibited.
This supports the notion that ADCC activity is determined by factors other than just the levels of CD64 expression.
IL-13 also affected the expression of ~c~RII and Fc~RIII. IL-L3 lS downregulated the expression of CD32 and CD16 in a dose dependent manner. However, IL-10, but not IFN-y, could block the IL-13 induced downregulation of CD32 and CD16 on monocytes.
¦ These results indicated that the level of Fc receptor expression was highly regulated by cytokines.
The only cytokine known to induce the low affinity Fc receptor for IgE (CD23) on monocytes was IL-4. However, IL-13 also induced the expression of CD23 on monocytes. It was demonstrated j~ ~` that the IL-13 induced~expression of CD23 could be partially suppressed by I~N-y. It was also shown that IL-13 could induce 25 ~production of ~IgE by PBMC. In addition, IL-13 could initiate germline e transcription in purified sIgM+ B c~lls and switching to IgE production when a second signal provided by T cell clones, T cell membranes, or CD40 ligand was present. The production of IgE is regulàted by number of cytokines, including solubh CD23, which n have either enhancing or inhibitory effects. The effects` of IL-13 and IFN-~ on the expression of CD23 by human monocytes fit well within this concept.
r ,: ., ,, ~ , ~ ~ I

r~./
~ .''';

~ 94/04680 PCl'/US93/0764~

~' o VI. Activities of IL-4 Anta~eonist: Interactions IL-4 and IL- 13 are two cytokines secreted by activated T cells which have similar effects on monocytes and B cells. A mutant form of human interleukin-4 (hIL-4) competitively antagonizes both hIL-s 4 and human interleukin- 13 (hIL- 13). The amino acid sequences of IL-4 and IL-13 are about 30% homologous and circular dichroism spectroscopy (CD) shows that both proteins have a highly a-helical structure. IL-l 3 competitively inhibits hIL-4-binding to functional human IL-4 receptors (called hIL-4R) expressed oo a hIL-4-0 responsive cell line, but not to the cloned IL-4R ligand-binding protein expressed on heterologous cells. hIL-4 has about a S0-fold lower affinity for the IL-4R ligand-binding protein than for the functional IL-4R, while the mutant hIL-4 antagonist protein binds to `
both receptor types with the lower affinity. ~he above results demonstra~e that IL-4 and I~-13 share a receptor component that is ~`~ important for signal transduction. In addition, these data establish that IL-4R is a complex of at least two components one of which is a novel affinity-converting subunit that is critical for cellular signal transduction.
~ IL-13 is one of a number of protein hormones called cytokines-- - that are secreted by activated T cells. Human Il,-13 elicits morphological and cell-surface phenotype changes on human -monocytes and also facilitates growth and immunoglobulin (Ig) ~~
producdon by human B cells. All these biological effects are also 25 ~ elicited by hIL-4, another protein hormone secreted by activated T
- ~ ~ cells.
The biological actions of IL-4 are mediated by a cell surface ~, Ireceptor that binds IL-4 with high specificity and affinity.
- ~Dissociation constant or Kd ~ 10-1 M. See Harada et aa. in Spits et al. (eds), IL-4: Structure and Function, 1992, CRC Press, Boca Raton, - - .
pp. 33-54.] Human and mouse IL-4R have been characterized by _ -:
cDNA cloning which defined a 130 kDa glycoprotein (herein referred ----o as IL-4R ligand-binding protein) with a single transmembrane span. The extracellular domain sequence of IL-4R ligand-binding WO 94/046~0 PCI'/US93/0764i 8 llo protein is structurally homologous to the extracellular domains of other cytokine receptor proteins.
Several of these other proteins participate in heteromeric interactions where one subunit by itself binds the ligand at a s relatively low affinity and the other subunit(s) contribute additionalbinding affinity and often are important for signaling. However, the extracellular domain of the IL-4R ligand-binding protein alone appears to bind IL-4 at the high affinity that characterizes IL-4R on various IL-4-responsive cells. Although the intracellular domain is 0 unimportant for binding, it is important for signal transduction.
The structural homologies between many of the cytolcine receptors is mirrored by structural homologies between their ligands. Por example, IL-4, interleukin-2 (IL-2), Growth Hormone, Macrophage Colony-Simulating Factor (M-CSF), and Granulocyte Macrophage Colony-Stimulaeing Factor (GM-CSF) are not related at - the sequence level, yet all have a similar compact-core-bundle structure of four antiparallel a-helices. See, e.g., Diederichs et al., Science 254:1779 (1991); Bazan, Science S7:410 (1992); McKay, Sclence 257:412 (1992); and Powers et al., Science 256:1673 (1992).
2 0 In mouse IL-2, exhaustive mutational analyses led èo the discovery that substitution of a residue (Glnl41 to Asp) at the C-terminus of the fourth a-helix results in loss of receptor-activation, but retention of~most receptor-binding. This mutant protein is a potent and specific competitive antagonist of IL-2 biological actioni 2 5 Based on the structural homology between IL-2 and IL-4, the importance of residues of hIL-4 that might be analogous to~~nIL-2 Glnl41 was investigated. In these experiments, substitution of a - residue (Tyrl24 with Asp) at the C-terminus of the fourth a-helix of human IL-4 (hIL-4) specifically abrogates IL-4R activation and renders the protein a competitive antagonist of IL-4 biological action. This property of hIL-4.Tyrl24 to Asp (called hIL-4.Y124D) has been described independently. See Kruse et a~;-,-EMBQ J. 11:3237 (1992). This mutant hIL-4 antagonist is defectivë in interaction with a previously unknown second subunit of the functional IL-4R.
In addition, the hIL-4 antagonist blocks many IL-13 biological ' ~ actions.
;, .
, ' ' , ~, .
. ~,.: .
,.~
~..

94/04680 PCl`/US93/0764 111 ~?

~0 Muta~enesis of hIL-4 Based on mutagenesis studies of mIL-2 [Zurawski et al., EMBO
J. 9:3899 (1990); Zurawski et al., EMBO J. 11:3905 (1992)], the shared structural frameworks of hI~-2 and hIL-4 [Bazan, Science 257:410 (1992); McKay, Science 257:412 (1992); Powers et al., : Science 256: 1673 (1992)], and assu~ning evolutionary conservation of functionally important residues, hIL-4 residues E114, K117, and Y124 were selected as those most likely to be speci~lcally involved in receptor activation. Substitution mutagenesis a~ these positions

- 10 used a synthetically reconstructed hIL-4 coding region inserted in the pTacRBS Escherichia coli expression plasmid [Zurawski et al., J. `
Immunol. 137:3354 (1986)]. Double-stranded synthetic oligonucleotides (synthesizer and reagents, Applied Biosystems) corresponding to the sequence between SalI and HindIII -recognition sites in the C-terminal coding region and containing equimolar amounts of each deoxynucleotide at the codon selected for ~; ra~ndomized substitution were ligated to SalI and HindIII digested pTac-hIL-4 plasmid.
Recombinant plasmids were recovered by transformation and the DNA sequence (Sequenase 2.0 kit, US Biochemical Corp.) of their SalI-HindlII intervals were determined. Partially pure mutant hlL-4 proteins were preparod~ as described for mIL-2 proteins ~Zurawski et al., EMBO J. 8:2583 (1989)], except that the refolding buffer contained~ reduced and oxidized glutathione, and assayed using TF-1 ~ ~ 2s cells. It was found that substitutions at Y124 resulted in proteins -~ ~ that were; partial agonists and that the Y124D substitution had the most drastic defect in cellular activation. During the course of this work similar ob$ervations ~ere made by Kruse et al., supra, who ~ also showed that hIL-4.Y124D and hIL-4 have similar affinities for - - 30 hIL-4R.
For production of pure hIL-4.Y124D, the pTrpC11-hIL-4 E. coli expression plasmid was subjected to PCR (Geneamp kit, Perkin Elmer Cetus) using the oligonucleotides:
~; CTCCAAGAACACAACTGAGAAGGAAACCIT (proximal to the single;~ 35 PstI res~ctionsiteinthecodingregionjand "j: ~:
, .,~, `', WO 94/04680 PCr/US93/0764 ~4~8~

TTGAl~AAGCTlTCAaCI'CGAACAClTI`GAAT~ CI`C (a Hindm recognition site precedes the underlined part which corresponds to the C-terminal coding region containing a GAT codon for residue 124~. The PCR product and pTrpCl 1-hIL-4 plasmid were cleaved ~ -with PstI and ~indIII, ligated, and pTrpC11-hIL-4Y124D plasmid was recovered and validated by transformation and sequence analysis using previously described methods [Zurawski et al., EMBO
J. 8:2583 (1989)]. Corresponding changes in hIL-13 at those positions should also have IL- 13 antagonistic effects.

Purification of Proteins `~`
E. coli-derived hIL-4 ~van Kimmenade e~ al., Eur. J. Biochem.
-~ 15 173:109 (1988)], human interleukin-la (hIL-la; Kronheim et al., Bio/Technology 4:1078 (1986), and mIL-13 were purified as described above. hII,-4Y124D was prepared from E. coli K12 cells (strain CQ21) harboring the pTrpC1 1-hIL-4.Y142I) plasmid grown overnight at 37C in 12 liters of L-Broth containing 50 ,ug/ml ampicillin in a G S3 rotatory shaker (New Brunswick Scientific) at 200 rpm. The cells were ~iarvested ~y centrifugation in a RC-3 centrifuge (all rotors SoNall) at 4,500 rpm, 10 min, 4C. The pellets- -- were resuspended in 450 ml of TE buffer (50 mM Tris-HCl pH 8, 1 mM EDTA) by shaking at 200 rpm for 15 min. Cells were ruptured 2s by 4 passes through an ice-cooled Microfluidizer model 110 cell ~ ~ ~--disrupter (Microfluidics).
Inclusion bodies were collected by centrifugation in a GS-3 ~, i rotor at 9,000 rpm, 40 min, 4C. The pellet was then washed by - -resuspension in 450 ml of TE and Triton X-100 was added to a ~lnal ` 30 concentration of 0.5%. Samples were kept at room temperature for- _ ~ 30 min and were then pelleted in a GS~ rotor at 8,500 rpm, 10 min,; ~
- ~ ~ 4 C. The inclusion bodies were resuspended in 60 ml 5 M
~- ~ guanidine-HCl in PBS (120 mM NaCl, 2.7 mM KCl, 10 mM NaPi pH
7.4), 2 mM reduced glutathione, 0.2 mM oxidized glutathione and ,,.,-, ~ i .

.

.

~:) 94~0468~ PCl`/US93J07645 ~86o any remaining insoluble material was removed by centrifugation in a SS-34 rotor at 20,000 rpm, 30 min, 4C.
The supernatant was diluted 10-fold into the same buffer without guanidine hydrochloride and stirred gently overnight at 4C
to permit refolding and oxidization. Concentration and exchange into 100 ml 50 mM Na Acetate pH 5.0 was then performed using a Millipore Pellicon apparatus (Millipore) equipped with a tangential flow ultrafiltration cassette with a size exclusion of 10 kDa. The sample was subjected to anion exchange chromatography (CM
o sepharose 16/100 column, Pharmacia) in the same buffer with elution via a 0-0.7 M NaCl gradient. Fractions containing hIL-4 protein were pooled and subjected to reverse phase chromatography (Poros R 10/100 column, Perseptive Biosystems) with elution via a gradient of 0-50% acetonitrile in 0.1% trifluoroacetic acid/water.
Fractions containing hIL-4 were lyophilized, dissolved in 50 mM Na `
Acetate pH 5.0, and quantified by densitometry (Molecular ~- ~ Dynamics) of stained SDS-PAGE with chicken egg lysozyme (Sigma) as a standard.
;; ~ Cell Proliferation Assays -~
Colorimetric cell proliferation assays used the human TF-l cell - line at 30,000 cells per well for 3 days and were performed as described by Mosmann, J. Immunol~ Methods 65:SS (1983). Cells were assayed in RPMI medium with L-glutamine and 10% fetal bovine serum (JRH Biosciences), 0.5 mM ,B-mercaptoethanol (Sigma).
; ~~ ~- 2s Cells were maintained in the above medium containing 1 nM hGM-- -~ CSF (Schering-Plough).
PHA blasts were prepared by incubation of 106 peripheral --i - blood mononuclear cells per ml with o.i mg/ml s~--- phytohaemagglutinin (Wellcome Diagnostics) in Yssel's medium ~see - ~ 30 Yssel e~ al., J. Immunol. Me~hods 72:219 (1984)~, supplemented with - ~ 1% human AB+ serum in 24-well Linbro plates (Flow Laboratories) - and were used in the proliferative assay after six days of incubation.
SP-B21 is a CD4+ cloned T cell line with unknown antigen specificity and was cultured as previously described [Spits et al., J. Immunol.
. ~

"
:

WO 94/04680 PCl'/US93/0764' 2~2~'6 128:95 (1982)]. Proliferative responses of both PHA blasts and ~: SP-B21 cells were determined at 5 x 104 cells per well and were pe*ormed and developed colorimetrically after three days as described for TF-1 cells.
Li~and E~in~i~
Procedures for preparation of cells, separation of bound from free ligand, ~computer analysis, and quantitation have been described in Zurawski et :al., Eh~BO J. 11:3905 (1992). Ba/F3 cells expressing ~ ~ surface hIL-4R-S protein (hIL-4R ligand-binding protein 0 deleted for most of the~intracellular domain) were grown as for TF-l cells~ except that mouse interleukin-3 (IL-3, 100 U/ml) replaced hGM-CSF and. 50 ~Lg/ml gentamycin sulphate (Sigma) and 800 ~lg/ml ~; Neomycin G418 (Schering-Plough) were added. .
l2~5I-:radiolabelling of E. :coli-derived hIL-4 and binding s~ conditions~were~as~ dos~ribed in Harada et al., J. Biol. Chem.
.2,6, ~:~752 ~ (199~
Circular~ Dichroism S~ectrosco~y Secondary ~structural features of hIL-4, hIL-la, and mIL-13 ~"~ proteins were~e`x.amined: on a J720 spectrophotometer with the 450 20~ W~ xeno.n ~l~np~ data analysis software (Jasco). The sample`s ~ , a~inst~'~20~:mM~-~NaPi, pH 7. Proteid concentrations of ~~
.~ ~ were re-determined by UV absorptIoD scanning on a Lambda: :~6~ spec:trophotometer ~ Perkin-Elmer). ~ Tho absorption -~
mal~imum ~a`t 280~ nm~:was ~used to calculate the amount of protein ~; , s us1ng~ theoret}ZZZcal extinction coeffictents based on known molecular ~ ~' ,wçigh.ts~and ~Zéxpccted~,~residue abs.orption contributions. Samples Z `
were diluted t o 0.2~ ~mg/ml in a 0.2 mm path length cell. - ~' lypical: scan paran~eters for the near UV range were a '' ~"~ conlinoous~ wavelength~ scan at 10 mdeg sensitivity, 0.1 nm step = ~ ' -30 -~rcsolution~at a~scan:speed of 50 nm/min with a time constant of 2 s.~ - ~~~
Z~
Four;: ~accumulations/,scàn were averaged for an increased signal to noisc ratio. Phosphate~ buffer blanks were Nn and subtracted out~

04680 ~ Pc~r/US93/07645 from subsequent protein scans and the spectra were noise-reduced using J700 data analysis software.
Mutant hIL-4 Anta~onist Bloçks IL 13 Action on TF 1 Cells !~.
In a search for mutant hIL-4 antagonists, it was noted that an s Asp substitution at residue Tyrl24 of hIL-4 resulted in loss of receptor-activation wiehout significant loss of receptor-binding. As expected from these properties, hIL-4.Y124D was a competitive antagonist of the action of native hIL-4 on TF-1 cells. TF-1 is a human pre-myeloid erythroleukemic cell line that shows a growth 0 response to vanous human protein hormones, such as GM-CSF, interleukin-3 (IL-3), interleukin-6 (IL-6), IL-4, and both human and mouse IL-13. The maximal responses of T~-l cells to these factors varies widely, but the maximal biological responses of IL-4 and IL-13 are similar. hIL~.Y124D had no effect on the TF-1 responses 15 to GM-CSF, IL-3, or IL-6. In contrast, hIL-4.Y124D was a potent antagonist of both mIL-13 and hIL-13 action on TF-~ cells.
hIL-4.Y12AD was equipotent against hIL-4, mIL-13, and hI~-13 activities on TF-1 cells and inhibited in a dose-dependent manner.
IL-13_Com~etitivelv InhibitsllIL-4 Bindin~ to 1~-1 Cells Since hIL-4.Y124D antagonizes hIL-4 via competitive inhibition of hIL-4 binding to IL-4R, a similar mechanism was hypothesized for its action against IL-13. Such a mode of hIL-4.Y124D action against IL-13 would imply commonalty between ~ ~ ~ IL~R and IL-13R. This was tested by comparing the abilities of - ~ 25 hIL-4 and mIL-13 to competitively displace 125I-hIL-4 binding to TFrl cells. hIL-4 fully competed 125I-hIL-4 binding to T~-l cells ~~ ~ with the concentration required for 50% inhibition (or IC50) - 2 x ~ 10-12 M. mIL-13 also competed 125I-hIL-4 binding. However, compared to hIL-4, it could not completely displace 125I-hIL-4 ~ ~ - 30 binding (~ 70% of the binding was displaced) and its IC50 value (2 x 10-1 M) was higher.

WO 94~04680 PCI/US93/0764 9 ~ ~2'~

IL-13 Does Not Bind to the IL-4R Ligand-bindin~ Protein A possible basis for the commonalty between IL-4R and IL-13R is that they are the same. This was tested by comparing the abilities of hIL-4 and mI~-13 to competitively displace Il 25-hIL-4 5 binding to a derivative of the cloned hIL-4R ligand-binding protein expressed on mouse pro-B Ba/F3 cells. Ba/~3 hIL-4R-S cells were used, which have a large number of binding sites/cell ~- 2000) in the form of a hIL-4R ligand-binding protein deleted for most of the cytoplasmic domain. See Harada et al., J. Biol. C~em. 267:22752 0 (1992). Although hIL-4 fully competed 125I-hIL-4 binding to Ba/P3 hIL-4R-S cells with ICso z 2 x 10-1 M, even high levels of mIL-13 (10-6 M) did not compete.
Some hIB-4-responsive Cell T~es Do Not Respond to IL-13 The earliest characterizations of the biological àctivities of 5 IL- 13 have shown concordance between cellular responses to IL-4 and IL-13, as described herein. Human peripheral blood mononuclear cells ~PBMNC) activated with phytohaemagglutinin (PHA) and certain human T cell cloned cell lines such as SP-B21 proliferate in response to hIL-4. Both these hIL-4-responsive cell 20 types did not proliferate in response to hIL-13.
The Binding Properties of hIL-4.Y124D and hIL-4 The binding of hIL-4 to Ba/F3 hIL-4R-S cells ~=1.6x10~10M~ - -~` carresponded closely to that previously characterized for the high ~ ~
affinity IL-4R (Kd z 10-1 M). Human lymphoma Raji cells have -~ ~
~- ~ 25 high affinity binding sites for hIL-4 (Kd z 10-1 M; see ~ruse et al., supra, and hIL-4.Y124D protein binds to these cells with only a 3-` ~ ~ fold reduced affinity compared to hIL-4. hIL-4.Y124D bound ~
-- ~ 3.5-fold less avidly to the hIL-4 binding sites expressed on Ba/~3 = ~
hIL-4R-S cells. ~ ~ i T~-l cells bound hIL-4 with an apparent affinity that was ~50-fold higher than the "high affinity" binding of hIL4 to Ba/F3 ~
hIL-4R-S cells. This is surprising beca~se although comparisons . .
, ". ~

~) 94/04680 PCl`/US93/0764 were done in parallel and used identical conditians and reagents, these two cell types have been reported to have similar numbers of - binding sites and affinities for hIL-4 as defined by equilibrium binding studies. In contrast to the different binding affinities of hIL-4 seen by competitive displacement binding studies, hIL-4.Y124D bound equally to both TF-l and Ba/F3 hIL-4R-S cells.
In other experiments, hIL-4.Y124D was.used as the labelled ligand and the results were analogous IL-4 and IL-13 are Structural Homolo~ues 0 The commonalty between IL-4R and IL-13R prompted a close examination of the sequence relatedness of IL-4 and IL-13. Only the sequences of the mature human and mouse IL-4 and IL- 13 proteins were examined, while assuming that known disulfide linkages for IL-4 are preserved for IL-13. There was significant, although low (-30%) sequence homology between IL-4 and IL-13.
~he significance of this observation was increased when the known structural features of hIL-4 were considered. All of the 25 residues that contribute to the hIL-4 hydrophobic structural core were conseNed or had conservative hydrophobic replacements in IL- 13 .
2 o Extensive insertion/deletion differences between IL-4 and IL- 13 were, with one exception~ confined to loops that connect the four -~ - a-helices or two short ~-strands. The exception was a shortened a-helîx C, although all the a-helix C residues that contribute to the structural core were retained in IL-13.
Mouse IL-13, unlike the ~-stranded hIL-la, had a CD
- - - absorption spectrum characteristic of a highly a-helical protein such as hIL-4 ~see Johnson, Ann. Rev. Biophys. Chem. 17:146 (1988)].
- The similarity in the two cytokines allow for modifications to -; either cytokine to effect similar properties on the other. Thus, -~: 30 insight into the mechanism of IL-4 antagonist with its receptor will -- likely be useful in modulating IL-13 with its receptor. In particular, -~ the present study provides locations in the IL- 13 molecule which would be expected to lead to IL-13 antagonists~ Moreover, the described IL-4 receptor would be expected to be modifiable while WO 94/04680 PCr/US93/0764 retaining its IL- 13 antagonist activity. This would suggest that shortening of the IL-4 antagonist would be useful while retaining its antagonist function. In particular, specific regions of the cytokine are suggested as useful to modify to achieve the desired biological activity.
IL-13 and IL-4 Rece~tors_ are Functionallv Related The observation that the hIL-4.Y124D antagonist competitively inhibits the biological action on TF-l cells of both hIL-4 and IL-13 demonstrates a relationship between IL-4R and 0 IL-13R. The ability of mIL-13 to compete for 125I-hIL-4 binding to TF-l cells confirmed the commonalty of IL-4R and IL-13R. This relatedness may also have been expected from the similar biological responses known to be elicited by hIL-4 and IL- 13 and perhaps ~; from the close linkage of the IL-4 and IL-13 genes in both humans and mice. See, e.g., Morgan et al., Nucleic Acids Res. 20:5173 (1992), - and other experiments herein. A straightforward explanation ` of the above observations would be that IL-4 and IL-13 act through the same receptor.
~, , However, some cell types responded to IL-4 and not to IL-13.
20 - ~ This is prima facie evidence that IL-4R and IL-13R are different entitie~s.; ~This does not exclude the possibility that IL-13 is a weak partial agonist of IL-4 and that only a subset of IL~R-bearing cells --can ~ efficiently amplify signals generated by IL-13 binding to IL-4R.
- ~; Three lines of evidence helped in resolving the conundrum involving Y ~ 2s IL4R and ~ IL-13R.
Firs~ly, IL-13 failed to compete for I125-hlL-4 binding to cells ~ ~ bearing only the hIL-4R ligand-binding protein. This result - ~ ~ demonstrated that the hIL-4R ligand-binding protein is, itself, not ~-~
the IL-13R. Secondly, two T cell types responded to hIL-4, but not - -~-~
to hIL-13. If hIL-13 is a partial agonist acting via hIL~R, then ~ ----hIL-13 should competitively antagonize the action of hIL-4 on these ~- -cells. This is not the case in one hIL-4-responsive T-cell system ~ ~ that has been tested. Thirdly, if hL-13 is à partial agonist acting 5 ~ via hIL-4R, then hIL-13 should be capable of fully competing hIL-4 .,., ~ .

,,;"~
, ,, ~ .
.,.
,, ~ 94/04680 ~? PCI /US93/0764~
1 1 9 ir5~
6'o binding to all cell types bearing hIL-4R. However, IL-13 only partially competed for binding of 125I^hIL-4 to TF-1 cells.
The conclusion from the above three lines of evidence is that IL-13 is not a partial agonist of IL-4 and that IL-4R and IL-13R a~e 5 different. On TF-1 cells, mIL-13 competed hIL-4 binding and hIL-A.Y124D antagonized the action of IL-13. These data compel a further conclusion that IL-4R and IL-13R have a functionally important receptor component in common. This conclusion is contrary to a consensus that the IL-4R ligand-binding protein itself 0 has all of the functional properties of IL-4R. IL-4R complexity is suggested by studies that find proteins associated with the IL-4R
1 igand-binding protein. Also, kinetic studies of soluble natural mouse IL-4R ligand-binding proteins indicated that membrane-bound functional IL-4R/IL-4 complexes were more stable than 5 soluble IL-4R/IL-4.
Functional IL-4R Contains an Additional Subunit(s) that Enhances Affinitv.~ Hel~s Transduce the Signal. and is Shared with IL-13R
Two results from receptor-binding analyses show that IL-41~ i on TF-l cells are complex and can exist in a higher affinity state 2 o t han thought ~ previously. Firstly, the apparent affinity of hIL-4 for 4R on 'rF-1 cclls was -50-fold greater than for the cloned - hIL-4R lîgand-binding protcin on Ba/F3 IL-4R-S cells. The~hIL-4 inding sitcs- on BalE:3 IL-4R-S cells were the typical 'high affinity' IL-4R that are presont on many cell types. Dissociation constant 25 ~--estimates for binding of hIL-4 to hIL-4R commonly vary somewhat, bot--fall within a 5-fold range of Kd-10-1 M.
Because the experiments were parallel, replicated independently,l used the same reagents and cells with similar IL-4R 3 : numbers, and gave analogous results using a different labelled .. , ~ , ..
3Q_ ligand, the "higher affinity" hIL-4 binding detected on IF-l cells should be significant. Secondly, while it bound with only a slightly - reduced affinity to IL-4R ligand-binding protein expressed on Ba/F3 ~- celIs, hIL-4.YI24D bound to IL-4R on TF-l cells with an affinity ~50-fold less than did hIL-4. In essence, this result provided an internal 35 control to confirm the "higher af~lnity" hIL-4 binding detected on ` ~

, ~

WO 94J04680 PCI`/US93/0764 U

TF-l cells. Since the hIL-4R ligand-binding protein cDNA was cloned from T~- 1 cells, it is unlikely that an unusual IL-4R ligand-binding protein accounts for the above results.
A model that does account for the above observations is that 5 functional IL-4R on TF-l cells are a complex between the IL-4R
ligand-binding protein and an additional component (or components) that enhances the affinity of the IL-4R ligand-binding protein for IL-4. This additional component(s) also associates with an IL-13 ligand-binding protein present only on a subset of IL-4-0 responsive cells to form IL-13R. Furthermore, interaction of hIL-4 Tyrl24 residue with this component is essential for productive signal transduction.
Nevertheless, hIL-4.Y124D, which fails to elicit this productive signal transduction, maintains an association between the IL-4 ;5 ligand-binding protein and this additional component(s); In this mo~el. hIL-4.Y124D antagonizes hIL-4 action by competing for IL-4 binding~ sités, but antagonizes IL-13 action by sequestering the addidonaI component(s) from the IL-13R complex by forming a non-prodyctive hIL-4R/hlL-4.Y124D complex.
20 Past Failures tO Co~rectly Define IL-4R and Pro~osed Tests of the New Model . , ;
TWQ: factors may have contributed to past failures to recognize `, t he ~'higher af&ity' state of IL-4R that were detected on TF-l cells.
Firstly, the integrity of the hIL-4 Tyrl24 residue is now known to 25 ~be vîtal~for this 'higher affinity' binding (this is also the case for the -: ~ analogous Tyrll9 residue of mIL-4 where substitution with Asp -~ resulted in a potent competitive antagonist of mIL-4 biological'action~. Thei standard procedure for radio-labelling IL-4; was`via s~-~ iodination of Tyr residues. There are only two Tyr residues in 30 hIL-4, thus it is probable that labelling hIL-4 converted Tyrl24 to iodotyrosine.
Indeed, hlL-4.Y124D labelled with the Bolton-Hunter reagent bound about 3-fold less efficiently than hIL-4. It is possible that hIL-4.Y124iodoTyr has a reduced affinity for functional IL4R and 35 that affinity constants derived using this reagent in direct-binding ~, ~

94/04680 ~ PCI /US93/0764 ~0 experiments have underestimated the actual affinity of IL-4R for IL-4. This was not an issue in the experiments which used hIL-4.Y 1 24iodoTyr as the labelled-ligand and native hIL-4 as a 'cold' competitor. A second factor that may have hindered the discovery of two affinity states for IL-4R is that the difference between the two affinities is only -50-fold. Thus, if cells have a mixture of IL-4R in both states, or if the 'lower affinity' state predominates, then two affinities may be impossible to recognize separately using conventional methods. The hIL-4R subunit-specific - 10 defect of hIL-4.Y124D is a powerful new reagent for dissection of hIL-4R complexity.
The notion that IL-4-responsive cell types vary in IL-4R
composition is being tested. Other direct tests of this model will require molecular characterization of the IL- 1 3R ligand binding s protein by binding analyses, cross-linking studies, and cloning.
However, the reagents to permit direct characterization of IL- 1 3R
. have not yet been developed. The very low affinity (~- 3x10-8M) IL-4-binding sites that have been detected on human lymphocytes may be a property of an additional IL-4R component.
¦~ 20 Common Subunits in Other Cytokine Receptors ~ , , -- - The molecular nature of the functionally important receptor c~mponent in common bet~veen IL-4R and IL-13R is unclear. The - above model to account for our data is based on the existence of - other affinity-modulating proteins that are obligatory components 2s share~~between several functional cytokine' receptors. Such shared components have been discovered in receptors for IL-6, alcostatin-M, leukemia inhibitory factor, and ciliary neurotrophic factor, which all ' share'~~gp130 [see Kishimoto et al., Science 258:593 (1992)], as well as - for ' human IL-3, interleulcin-5 (IL-5), and GM-CSF receptors, which all_share the ~c protein [Miyajima et al., Trends ln Biochemical : Sciences -17:378 (1992)].
This shared ~c receptor subunit accounts for the observed ~ ~ - cross-competition of IL-3, IL-~, and GM-CSF binding to certain cell ;: types. When assayed on TF-l cells, hIL-4.Y124D did not antagonize WO 94/04680 PCI'/US93J0764' ,6~

the biological activities of hIL-6, mouse leukemia inhibitory factor, hIL-3, or hGM-CSF, and neither hI~-6 or hGM-CSF competed for hIL-4 binding. Therefore, gpl30 or the ~c protein are not likely candidates for the additional IL-4R component, nor the component 5 shared between IL-4R and IL- 1 3R.
On the basis of their common genetic locations/structures and relation in protein structure,`it has been proposed that IL-4, IL-3, IL-S, and GM-CSF form a protein family. See Boulay et al., J. Biol.
Chem. 267:20525 (1992). The biological data regarding commonalty 0 between IL-4 and IL-13 show that IL-13 also belongs to this family.
However, the available data support a clear functional separation between the receptors for IL-4/IL-13 and IL-3/IL-5/GM-CSF. For example, no effects of hIL-4.Y124D on IL-3 or GM-CSF responses on .!' TF-l were noted. Also, in T~-l cells the pattern of intracellular lS tyrosine-phosphorylation that is elicited by - IL-~/GM-C~F is different from that elicited by IL-4.
Im~cations of Jointb Anta~onia ing IL-4 and IL- 13 Res~onses I n Vrvo The ability of the hIL-4.Y124D antagonist to act against both hIL-4 and hIL-13 biological responses should provoke a reappraisal of the therapeutic potential of hIL-4.Y124D. These results show that, unlike soluble IL-4R ligand-binding protein or anti-IL-4 antibodies, hIL-4.Y124D is not a specific antagonist of hIL-4 action.
Inhibitory IL-4 variants have been suggested as potentially useful drugs in the treatment of IgE-mediated discases. The possibility ~-; 25 exists for antagonizing~ both hIL-4 and hIL-13 responses by hIL-4.Y124D treatmeni for various disease states. The structural homology between IL-4 and IL-13 and sharing of receptor subunit(s) between IL-4R and IL- 1 3R suggest that particular IL- 13 residues within a-helix D are specifically important for receptor signaling and that substitutions in these residues may result in :
- IL-13 variants that are antagonists. These results also predict that such IL- 13 antagonists will be effective antagonists against IL-4-responses on cell types that also respond to IL-13.
~, ~
~; Antagonistic Effect on Other IL-13 Activities . ~ ~
, - . I

94/04680 PCrtUS93/07645 123 ~
~0 Results with cocultures of highly purified B cells and activated T cell clones with 400 U/ml IL-4 showed inhibition of IgE synthesis by the IL-4 antagonist used at 10 ~g/ml. See Tables 16 and 17. The assay was as described abo~e for IgE synthesis. '-0 Table 16: Induction of IgESynthesis by Ik4, I~- 13 and IL~4-Mutant Protein 1 .
IgE SYnthesis (n~/ml) .

~Medium <0.2 -~ 15 IL-4 (200 U/ml) 173+45 13 (200 U/ml) 110+42 IL-4-Antagonist (Y. 1~4, 1 mg/ml) 13+6 Mock-control c0.2 2 ~
Table 17~ 4-Mutant Protein Inhibits IL-4-Induoed IgE~ynthesis by PBMC
~, . .
E Svnthesis (n~/ml~
, :. - - - -2s Medium <0.2 - II~-4- (50 U/ml) 265_49 -- - -- IL-4 (50 U/ml)+Y.124 (0.003 mg/ml) 108+60 - ~ ~ IL-4 (50 U/ml)+Y.124 (0.03 mg/ml) 12+54 (50 U/ml)+Y.124 (0.3 mg/ml) 12+3 o---IL~ (50 U/ml)+Y.124 (3 mg/ml) 5+2 - ~IL-4 (50 U/ml)+Mock-control 194_46 ~" -Il,-4 antagonists like Y124 also effectively inhibited the proliferation of purified human B cells stimulated by anti--CD40 WO 94/1)4680 PCI /US93/0764' ~Q~ 124 mAbs in the presence of either II,-4 or IL- 13 . IL- 13 antagonists may have similar effects. Thus, administration of IL-4 and IL- 13 antagonists may provide the preferred mehtod not only to inhibit IgE syn~hesis, but also to prevent the expansion of IgE producing cells.

VII. Antibodies tQ Human IL- 13 Ra~ polyclonal antiserum was raised against E. coli-derived human IL-13 by standard procedures. See, e.g., Harlow & Lane 0 (1989) or Coligan (1991 and supplements). Serum from these rats was useful in immunoprecipitating 3 5 S-methionine labelled supernatants of C0S7 cells expressing IL-13.
Monoclonal antibodies against hIL-13 were produced using standard methods. Rats were immunized with E. coli-produced hIL-13. The neutralizing capability of four different monoclonal antibodies were tested on TF- 1 cells stimulated with COS-producèd hIL-13 or E. coli-produced hIL-13. The TF-1 cells (5,000 cellslwell~
were incubated for 72 hours with 1:100 dilution of COS-produced hIL-13 or 5 ng/ml E. coli-produced hII.-13, with dilutions for supernatants containing rat anti-hIL-13 Monoclonal antibody. After 72 hours, cell viability was determined by alamar blue staining.
To determine which monoclonal antibodies from the above spleen-hybridoma fusions bind to hIL-13 produced from COS or E.
coli, an ind*ect ELISA was pe~formed. PVC microtier plates were coated for 2 hours with either 0.5 ~g/ml E. coli-produced hI~-13 or a 1:15 dilution of COS produced hIL-13 in PBS at 37 C. A standa~d ELISA protocol was used. Specific binding was observed in both cases.
Many modifications and variations of this invention can be made without departing from its spirit and scope, as will become apparent to those skilled in the art. The specific embodiments described herein are offered by way of example only, and the invention is to be limited only by the terms of the appended claims.

~) 94/04680 PCr/US93/07645 125 ~

~0 SE~QUENCE LISTIN~

( 1 ) GENERAL INFORMATION:
~.
5 (ij APPLICANT:
(A) NAME: Schering Corporation (B) STREET: One Giralda Farms (C) CITY: Madison (D) STATE: New Jersey - 10 (E)COUNTRY: U.S.A.
(F) POSTAL CODE (ZIP): 07940-1000 (G) TELEPHONE: 201-82~-7375 (H) TELEFAX: 201-822-7039 (I) TELEX: 219165 (ii) TITLE OF INVENTION: Human Interleukin-13 (iii~ NUMBER OF SEQUENOES: 6 2 o (iv) COMPUTER READABLE ~ORM:
(A) MEDIUM TYPE: Floppy disk (B) COMPUTER: Apple Macintosh (C~ OPERA~G SYSTEM: Macintosh 6Ø5 - j (D) SOPTWARE: Microsoft Word 5.1a 2s (v) CURRENT APPLICATION DATA:
(A) APPLICATION NUMBER:
(B) FILING DATE:

. . .
- 30 (vi) PRIOR APPLICATION DATA:
(A) APPLICATION NUMBER: US 08/012543 ~- ~ (B) FILING DATE: 01-FEB-1993 (vi) P}~IOR APPLICATION DATA:
(A) APPLICATION NUMBER: US 08/010977 (B) FILING DATE: 29-JAN-1993 :

WO 94/04680 PCr/US93/~764 6~

(vi) PRIOR APPLICATION DATA:
(A) APPLICATION NUMBER: US 07/933416 (B) EILING DATE: 21-AIJG-1992 (2) ~ORMATION FOR SEQ ID NO. 1:
(i) SEQUENCE CHARACI'ERISTICS:
(A) ~ F~NGTH: 1290 base pairs 0 (B) TYPE: nucleic acid (C) STRA~EDNESS: double (D) TOPOLOGY: linear (xi) SEQVENOE DESCRIPIION: SEQ ID N.O:l:
:.
TCC GCrCCTC~ ccTcTccrGT T~Gc~crGGG CCTC ~rG GCG cTr TTG 56 . M~t Ala Leu Leu TTG ACC ACG GTC ATT GCT CTC A~r TGC CTT GGC GGC T~ GCC TCC CCA 104 Leu Thr mr Val Ile Ala Leu mr Cys Leu Gly Gly Phe Ala Ser Pro . 5 10 15 20 GCC CCT GTG CCT CCC TCT A~ GCC CTC AGG GAG CTC ATT GAG GAG CTG 152 ~- Gly Pro Val Pro Pro Ser Thr Ala Leu Arg Glu Leu Ile Glu Glu Leu 25 30 35 GTC AAC AIC ACC CAG A~C CAG AAG GCr CCG crc TGC A~r GGC Af~C ATG 200 Val Asn Ile Thr Gln Asn Gln Lys Ala Pro Leu Cys Asn Gly Ser ~et 40 -. 45 50 GT~ TGG A~;C ATC AAC CTG ACA GCT GGC ATG TAC TGr GCA GCC crG G~A 248 - Val Trp Ser Ile Asn Leu rhr Ala Gly Met Tyr Cys Ala Ala Leu Glu 55 60 6S
TCC CTG ATC AAC GTG TCA GGC TGC AGT GCC ArC GAG AAG ACC CAG AGG 296 Ser Leu Ile Asn Val Ser Gly Cys Ser Ala Ile Glu Lys Thr Gln Arg ATG CTG AGC GGA TTC TGC CCG CAC AA~; GrC TCA GCr GSG CAG m TCC 344 ~bt Leu Ser Gly Phe Cys Pro His Lys Val Ser Ala Gly Gln Phe Ser ~- A~;C TTG CAT GTC CGA GAC ACC AAA ATC GAG GTG GCC CAG m GTA A~G 392 ~- I
- Ser Leu His Val Arg Asp Ihr Lys Ile Glu Val Ala Gln Phe Val Lys 1~5 110 115 GAC CTG CTC TTA QT TTA AAG A~A CTT m CGC G~G GGA CGG TTC A~C 440 Asp Leu Leu Leu His Leu Lys Lys Leu Phe Arg Glu Gly Arg Phe Asn ~94~04680 ~ P~T/US93tO764 127 ~ ~

TGPaACTTCG PP2~CA3C~ TAITTGCaGA GACPGGACCT GACTATTGAA GTTGCAEPIT 500 CAIITTTCTT TCTGAIGTCA AAaA~GTCTT GGGTAGGCGG GAAGGA~GGT TA{GGAGGGG 560 TAAAAITCCT TA~CTTAGAC CTCACCCTGT GCTGCCCGTC TTCh~CCTAC CCGACCTCAG 620 CCTTCCCCTT GCC ~ CASCCTGGTG GGCCTCCTCT GTCCAGGGCC CTGAECTCGG 680 TGGACCCAGG GalGaC~GT CCCTACACCC CTCCCCTGCC CTh~ASCACA CTGT ~ T 740 A~AGTGGGTG CCCCTTGC CAGACA5GTG GTGGGACAGG G~CCCACTTC ACACP~AGGC 800 AaCTGaGGCA GACaGCAGCT CAGGCACACT TCTTCTTGGT CTTATTTATT ATTGTGTGTT 860 A m AA~G~ GTGTGTTTGT CACCGTT~&G G~IIGGGC-AA GACTGTa~CT GCTGGCACTT 920 GGA~CCAAGG GTTChGAGAC TCAGGGCCCC AGCACTAAAG C~GTGGACCC CAGCi~GTCC~ 980 TGGTA~TAAG ThCTGTGTAC h-~AATTCTGC TACCTCACTG GGGTCCTGGG GCCTCGGA~C 1040 CIC~ICCG~G GCAGGGTCAG GaCa3GGGC~ GAA~AGCCGC TCCTGTCTGC CAGCCAGCAG 1100 CCAGCTCTC~ GCCA~CG~GT A~TTTA~TGT TTTTCCTCGT ATTTA~ATAT TAAATATGTT 1160 hGCaAAG~GT TAA~ATATAG AAGGGTACCT TGAA~C~GG GGG~GGGGAC ~TGA~AAG 1220 TlGTll Q IT GACTa~CAAA ~ CCAG AAATAAAGTT GGTGACPGaT APaAA~AAAA 1280 ~ 1290 (2) INFORMATION FOR SEQ ID NO:2 - 35 (i) SEQIJENOE OE~AFL~C~ERISTICS:
(A) LENGTH: 132 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TY~E: protein , - ~ ~ (xi) SEQUENC]E D~SCRIPIION: SEQ ID NO:2:
- Met Ala Leu Leu Leu Thr Thr Val Ile Ala Leu Thr Cys Leu Gly Gly . . 45 1 5 10 15 ~:~ Phe Ala Ser Pro Gly Pro Val Pro Pro Ser Thr Ala Leu Arg Glu Leu Ile Glu Glu Leu Val Asn Ile m r Gln Asn Gln Lys Ala Pro Leu Cys W O 94/04680 PCT/US93/0764j ~ 6~ 128 Asn Gly Ser Met Val Trp Ser Ile Asn Leu Thr Ala Gly M~t Tyr Cys Ala Ala Leu Glu Ser Leu Ile Asn Val Ser Gly Cys Ser Ala Ile Glu Lys Thr Gln Arg M~t Leu Ser Gly Phe Cys Pro His Lys Val Ser Ala Gly Gln Phe Ser Ser Leu His Val Arg Asp Thr Lys Ile Glu Val Ala Gln Phe Val Lys Asp Leu Leu Leu His Leu Lys Lys Leu Phe Axg Glu 115 120 i25 Gly Arg Phe Asn (2) INFORMATION FOR SEQ ID NO:3:
~i) SEQUENOE CHARACI~ERISTICS:
~A) LENGTH: 1212 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: double (D) TOPOLOGY: linear (xi) SEQ~CE DESCRIPIION: SEQ ID NO:3:
G~CC~G C~GCCTA~;GC CAGCCCACAG TT~TACAGCT CC~ C5~ 60 ~GGGClSC ATG GCG CTC TGG GTG ACT GC~ OEC CTG GCI CTT GCT TGC 108 Mbt Ala Leu Trp Val Thr Ala Val Leu Ala Leu Ala Cys CTT GGT GGT CTC GCC GCC CCA GGG CCG GIY~ cca AE;A TCT GrG TCT crc 156 Leu Gly Gly Leu Ala Ala Pro Gly Pro Val Pro Arg Ser Val Ser Leu c~r CTG ACC CTT AAG G~G CTT ATT GAG GAG CTG AGC AAC ATC A~A QA204 Pro Leu Thr Leu Lys Glu Leu Ile Glu Glu Leu Ser Asn Ile Thr Gln --:

G~C C~; ACT CCC CTG TGC A~C GGC AGC ATG GTA TGG AGT GTG. GAC CTC, 252 _ .
Asp Gln Thr Pro Leu Cys Asn Gly Ser Met Val Trp Ser Val Asp Leu GCC GCr GGC GGG ITC TGT GTA GCC CrG GAT TCC CTG ACC AhC ATC TCC300 50 Ala Ala Gly Gly Phe Cys Val Ala Leu Asp Ser Leu ThL Asn Ile Ser . 75 94/~4680 PCT/US93/0764 ~ o AAT TGC AAT GCC ATC TAC A~G ACC CAG AGG A~A TTG CA~ GGC CTC TGT 348 Asn Cys Asn Ala Ile Tyr Arg Thr Gln Arg Ile Leu His Gly Leu Cys AhC CGC AAG GCC CCC ACT ACG GTC TCC A~C CTC CCC GAT ACC AAA ATC 396 Asn Arg Lys Ala Pro Thr Thr Val Ser Ser Leu Pro Asp mr Lys Ile GAA GIA GCC CAC m ATA A~ AAA CTG CTC AGC TAC A~A A~G Q A CTG 444 Glu Val Ala His Phe Ile Thr Lys Leu Leu Ser Tyr Thr Lys Gln Leu m CGC CAC GGC CCC TTC TAaI~AGGAG AG~CC~TCCC TGGGCA~CI~ 492Phe Axg His Gly Pro Phe .

AGCTGTGGaC TCATTTTCCT TTCTCALAIC AG~CTTTGCT GGGGAA~GC AGGGAEGaGG 552 ;~
GTTGaGG~GG A~GGG~GAIG CCTCA~CTTT GGCCTCAGCC TGCA¢TGCCT GCCTACTGCT 612 CAGGr~DCTC~ GCCTG-~CAAC A~CCCCACCC CACCCCCA~C CCCGCCGCCC CA~CCCATCC 672.
CTA~AGAAA~ CTGCAGCAAG A~CGTGAGTC CA~CCT&TGG CCTGGTCCAC A~AGGGCAAC 732 : 25 TGAGGCAGGC AGChGCTDGA GrACAlllCT TCTTGATCTT A m ATTAIG ~'ll~l~l~ll 792 A~ITAAA~GA GTCTGTCAGT AICCCGGTGG Gr~AC~TGGTT TGCTGCC~T GCCCTGGGGG 852 ~-30 CTCC~CA~T GAAGCAGTGG GCTCTGGGGT CCCTGGCAAT A~TAfTGTAT ACATAA2TCT 912 : GCTACCTCAS TGTACCCTCC AGGTCTCACC CCAGGrAGGA GATGr~GAGGG GAGGCC~G~G 972 CAACACTCCT GICTGCCA~G GCaGCAACCA GCCCTC~GCC AlYaAATAAC TTAII~ T-1032 ~--GTl~-lATAT TTAAAGTATT AAAI~GC~TA GCaAAGaG~T AATAATA~AT GGAAEAATGG 1092 CCTGTTaC~C TCA~GGTG~T &~GTA~IG~A TGGGGG~-aGG GTGGTGr~GTT TGTCACTGAA-li52-CAAACTITTC ATTGACTGTC AAACTA~AAA CCGG~AA~AA AGATGGTGAC A~A~AAAAAA 1212 (2) INFORMATION FOR SEQ ID NO:4:
(i) SEQUENOE CHARACTERISTICS: i;
(A) LENGTH: 131 amino acids (B) TYPE: amino acid ~- - ~~~~~ ~~
(D) TOPOLOGY: linear : (ii) MOLECULE TYPE: protein ~' 50 (xi3 SEQUENCE DESCRIPIION: SEQ II) N~:4:

W O 94/04680 PCT/US93/~764 ~ 4 ~6~

Met Ala Leu Trp Val Thr Ala Val Leu Ala Leu Ala Cys Leu Gly Gly `.
Leu Ala Ala Pro Gly Pro Val Pro Arg Ser Val Ser Leu Pro Leu Thr Leu Lys Glu Leu Ile Glu Glu Leu Ser Asn Ile Thr Gln Asp Gln mr Pro Leu Cys Asn Gly Ser Met Val Trp Ser Val Asp Leu Ala Ala Gly Gly Phe Cys Val Ala Leu Asp Ser Leu Thr Asn Ile Ser Asn Cys Asn Ala Ile Tyr Arg Thr Gln Arg Ile Leu His Gly Leu Cys Asn Arg I.ys - Ala Pro Thr Thr Val Ser Ser Leu Pro Asp Thr Lys Ile Glu Val Ala Hi~ Phe Ile Thr Lys Leu Leu Ser Tyr Thr Lys Gln Leu Phe Arg His ~5 115 120 125 Gly Pro Phe (2) INFORMATION FOR SEQ ID NO:5:
(i) SEQU~CE C~ERISTICS:
- ~ (A) LENGTH: 35 base pairs (B) TYPE: nucleic acid (C~ STRANDEDNESS: single (D) TOPOLOGY: linear (xi) SEQUENOE DESCRIPTION: SEQ ID NO:5:

(2) INFORMATION FOR SEQ ID NO:6:
(i) SEQU~OE CHARACIERISTICS:
4s (A) LENGTH: 36 base pairs (B) TYPE: nucleic acid ~C~ STRANDEDNESS: single (D) TOPOLOGY: linear - - ~o (~i) SE~ ~D~SC~P~O~-: S~Q ~) YO:c:

~v~^~t~ ~ 3 6 ;

. ~ .
.:

' r.. ~

' ', ~ ~ ' ' ,., ~ , - : - .

.
. ,~ ~ . .
:
.,; ~.
~ ~ , _ .. ...

'.5'~

~j "": ' A~lENDED S'rlE~
~' .,~

Claims (20)

WHAT IS CLAIMED IS:

1. Human IL-13 having an amino acid sequence defined by SEQ ID NO: 2.

2. A pharmaceutical composition comprising a pharmaceutically acceptable carrier and an effective amount of human IL-13 having an amino acid sequence defined by SEQ ID NO: 2.

3. A method for making a pharmaceutical composition comprising admixing a pharmaceutically acceptable carrier and an effective amount of human IL-13 having an amino acid sequence defined by SEQ ID NO: 2.

4. An antibody or a binding fragment thereof against human IL-13 having an amino acid sequence defined by SEQ ID NO: 2.

5. A pharmaceutical composition for inhibiting antibody isotype switching to IgE or IgG4 comprising pharmaceutically acceptable carrier and an effective amount of an antagonist of human IL-13.

6. The pharmaceutical composition of claim 5 in which the antagonist is an antibody or a binding fragment thereof.

7. A method for making a pharmaceutical composition for inhibiting antibody isotype switching to IgE or IgG4 comprising admixing a pharmaceutically acceptable carrier and an effective amount of an antagonist of human IL-13.

8. The method of claim 7 in which the antagonist is an antibody or a binding fragment thereof.

9. Use of an antagonist or human IL-13 for inhibiting antibody isotype switching to IgE orIgG4..

10. Use of an antagonist of human IL-13 for the manufacture of a pharmaceutical composition for inhibiting antibody isotype switching to IgE or IgG4.

11. The use of either claim 9 or 10 in which the antagonist is an antibody or a binding fragment thereof.

12. A purified human IL-13.

13. A purified human IL-13 of claim 12 which as an amino acid sequence defined by SEQ ID NO: 2 or SEQ ID NO: 4.

14. A pharmaceutical composition for inhibiting antibody isotype switching to IgE or IgG4 comprising a pharmaceutically acceptable carrier and a purified human IL-13 of either claim 12 or 13.

15. An antibody or a binding fragment thereof against a human IL-13 of either claim 12 or 13.

16. An isolated nucleic acid encoding a human IL-13 of any one of claims 1, 12 or 13.

17. A recombinant vector comprising a nucleic acid of claim 16.

18. A host cell comprising a recombinant vector of claim 17.

19. A host cell of claim 18 which is an E. coli bacterium.

20. A method for making human IL-13 comprising culturing a host cell of either claim 18 or 19 under conditions in which the nucleic acid is expressed.