CA1341503C - Molecular cloning and expression of human il-3 - Google Patents

Abstract

A cDNA encoding human interleukin-3, also designated herein hIL-3 or hmulti-CSF, has been disposed in expression systems and caused to produce hIL-3 substantially free of impurities normally accompanying this protein as it is produced inter alia by peripheral blood lymphocytes in nature.
The resulting hIL-3 can be produced in practical amounts and is useful in a variety of therapeutic and diagnostic protocols.

Description

GIRT-BROCADES N. V. - 1 '- ~ 3 4 1 5 0 3 Field of the Invention S
r~~he present invention relates to cDNA encoding human interleukin-3 (hIL-3) and its use, inter ~alia, in the cloning and expression in various organisms, including microorganisms, in part~.cular yeasta, bacteria and fungi, tissue culture cell~ and tranagenic an~.ma7.~ and plants.
Hsokqround of the Invention ._ Hemopoiesis involves the active process of proliferation and differentation of pluripotent progenitor cells into all types of mature blood cells arid some specialized tissue cells. Production of functional blood cells xec~ulated by speaifi.c proteins, the hemopoietic growth factors (HGFB). Some of the HGFs control maturation of a specific maturation lineage, ~rrhereas others stimulate proliferation and differentiation of progenitors along multiple pe~thw$y~s . Much of our kno~aledga~ of the hemopoiet~.c differentiation process has been obtained from mouse studies in vitro and in yiya, using purified growth factors. The murine growth factor interleukin~-3 (mIL-3), also termed Multi-csF, mast cell growth factor, stem cell activating faetox yr several other desigriat~.ons, stimulates the proliferation of developmentally early, multipotent cells (CFU-S) as detected by the spleen colony assay, resulting in the production of progenitor cells along the erythroid, megacaryoeyte, granulocyte/macrophage, osteoblast and several other lineages.
Furthermore, mIL-3 has been implicated in replication of p~.uripotent stem cells, probably in synergism with other FiGFs.
In recent years, several groups have succeeded in cloning mIL-3 cDNA. No results have been reported sofar of ident~.fying homologous sequences i.n human pNA using mrL-3 DNA

a ~ ..

as a probe. Presumably, the human gene has diverged extensively from the mIL-3 gene or has lv$t its function during primate evolution. However, human leukocytes were found to produce a HGF(s) which can rapiac$ mIL-3 in supporting the proliferation of muririg CFU-S. Thus, the exiaxenca of a human F3C~F was postulated, ~rhich aha~res biological properties with mIL-3 and therefore could !~e the human homolog. Yang, x-C, et al, Cell (1986) 47:3-10, dated 10 actober disclo$es cDNA
encoding a prr~tein having IL-3 Like activity from gibbon T-Ig cells, and retrieval of a genQmic DNA w~xich encodes the human counterpart. TYte sequence of a al7NA encoding human ZL-3 can be deduced from the human gene sequenoe published by Yang et al.
However, said article does neither di$elose nor teach a method for isolation of a cnNA enoo~iing human IL--3, nor was the produGtieri ef hIL--3 achieued, This invention de$cribes for the first the isolation of a CDNA comprising the entire coding sequence for human IL-.3.
Human IL-3 protein has never been prepared in purified form, nor have xts characteristics, other than its activity in certa~.n inproliferat~.on assays and deduced primary structure, been disclosed. The present invention permits the recovery of pur~.fied human IL-3, and identification of its charaoteristiCS through recombinant pr~ductivn from a cDNA clone.
Summary of the Invention As stated above, the present invention for the time describes the isolation Of a eL~NA comprising the entire coding sequence for human IL-3. The low degree of homology between the nNA sequences coding for murine and human IL--3 does not permit the retrieval of a cDNA for hIL-3 by hybridization with the mIL-3 coding sequence. Unexpectedly, the hIL-3 cDNA clone could be isolated by exploiting the rather high degree of homology in the 3' noncoding part of the cDNA's. 'IYhe avail-ability of the cDNA clone permits the product~,an of hIL-3 by a wide range of host organisms. Subsequent to large scale pro--duction txie pxOtei.n may be purified and used therapeutically, The present invention permits production of recombinant human IL-3 protein in a wide range of host cells by transcription and translation from a cDNA sequence cncoding the human IL-3 protein. The production of th$ protein is il7.ustrated in several hosts, including E. cola, COB cells, 0127 cells, 8. subtilis and 8. lichenifarmis, S. cerevisiae and K. lactic, hereinbelow. Production in other hosts using appropriate expression aysterns is also made possible by provie~ion of the intronless cDNA. More generally, the a~rail-ability of antihuman IL-3 antibodies wh~.~h permit identific-ation of colonies e~rhibiting successful production of the xeaombinant protein aids in ,production of human IL-3 from any recombinant system.
In one aspect, therefore, the invention is di.reeted to a recombinant, intronless, DNA encoding human IL-3 protein.
In another aspect, it is directed to expression systems capable of effecting the expression of said DNA
sequence encoding hIL-3 in an appropriate host.
In other aspects, the invention is directed to recombinant human IL-3 protein in glyeosylat~d ox unglycosylated form, to human IL-3 free of substances normally accompanying said protein, and to antibodies specifically reactive with these recombinatxt or purified proteins.
Brief Description of the Drawings Figure 1 shows a comparison of DNA and protein sequences of human mufti-CSF and mouse IL-3. The hmulti-CSF
protein and DNA secjuence (clone Dli, top lines) were aligned 3D with the mIL-3 DNA (11, 35) and protein sequence (30).
Identical nucleotides are indicated by a vertical line, identical amino acids are shown in boxes. Black dots indicate a polyadenylation signal sequence and horizontal bars mark ATTTA repeat units.
Figure 2 shows the construction of plasmid pLB4 containing human IL-3 cDNA. E = EcaRI, Sm = SmaI, B = BamHZ, S = SstI, K ~ KpnI.
Figure 3 shows the biological activity of COS/pLB4 CM on human bone marrow progenitors. The mean numbers of erythroid (BFU-E), granuleoyte-macrophage (Cl~u-GM), granulo-cyte ( CFU~-G ) , eosi.nophi l { CFU-Eo ) , macrophage ( CFU--M ) and mixed tCFU-MIX) colonies (+Sp) axe shown for duplicate cultures stimulated with graded volumes of CDS/pLB4 CM.
Figure .4 shows induction of AML prol~.feration by CCS/pLEg CM as assessed in a colony culture a$aay (panel A) and in a DNA synthesis (3H-TdR incorporation) assay (panel 9).
F~.gure 5 shows a vonstruction diagram of the _E.
coli expres8lon vectors pGS/IL-301, GB/IL-302. pGH/IL--303, pGB/IL~-304, pG$/IL-305 and pCB/IL-30~. In this Figure X stands for XhoI, E for EcoRI, $ for ~amHI and A tvr Aval $ite.
Figure 6 shows the sequence of the multioloning site in pTZl$R (Pharmacia) and its derivat~.ve pTl.
Figure "~ shows a schematic presentation of hmul.ti-CSF expression clones. For the eucaryote expression plasmids pLE4 and pLHl only the hmulti-CSF cDNA insert is shown.
Leader peptide (~) and mature hmulti-CSF protein {~) coding regions are indicated in boxes. Bacterial expression Clones of hmulti-CSF (derived from pLHl) contain the lacZ and multi-linker protein coding region (~), the 5' terminal noncoding region of hmulti-CSF (~) and the hmulti-CSF
coding region. The arrow marks the ATG startcodon used in the particular veotor.
Figure $ shows the sequences of fusion regions of lacZ/hmulti-CSF DNA for various bacterial expression vectors.
The sequence of clones is given from the start of the lacZ
protein in either pUCB or pTZlBR (lower case letters) and of hmulti-CSF DNA sequence up to the Clal site at position 3Q 158. Mutations in the hmulti-CSF DNA sequence are underlined, resulting in: trpl3~rgl3 {pGB/IL-302); leug~pro'~ and trpl3---~argl3 (pGH/IL~-303 ) ; met3 -~thr3 and a silent change {pGH/I~,-304). The superscripts denote the amino acid residue number of the mature protein.
Figure 9 shows polyacrylamide gel-electrophoresis of bacterial hmulti-CSF produced from bacteria containing pGB/IL-a01 and pGB/IL-302.
Figure 10 shows the titration of hmulti-CSF fusion - 5 ..

protein an AML bleat cells.
Figure I1 shows a 6Jostern blot demonstrating the IL-3 specific reaction of rabbit antisara raised against the 21 kd protein isolated from a lysate of E. coli transformed with p~s/xL-Col.
Figure 12 shows the effect of the antisera of Figure 11 vn IL-3 activity.
F'ir~mr~ 1~ ~thC~wa a ~tthamati~e repxesetztati~ra vt plasm~.d pGB/IL-307. The box (~) indicates the human IL-3 coding t~~c~upnr~. The N~tarminal amino acids of the ~usiorr protein are depicted below the drawing.
Figure 14 shows a schematic rapre~sntation of plasmid pG8/IL~30$. ~"he nucleotide eec~uence Qf the prompter raglan i~, drF.i.ctCd below the drawing.
Figure 15 shows the construction of plagmid pG~/IL-309. The first box (~) indicates a part of the human IL-3 ~~gusr~ce, viz. the e~ignr~~ s~~y.~Pncv p7.us 20 amino s,eido of the motura ~LIJ~.C111, Tlta ut~ler "~vx ( ~) ~.naicat~s part of the 3' noncoding region of the IL~3 eDNA sequence.
Figure 16 is a schematic representation of plasmid pGB/IL-310.
Figure 17 shows the nucleotide sequence of plasmid pDf lAl .
Figure 18 straws the cpnstruotion of the plasmids pGB/IL-311 end gGB/IL~31?. ThP box (~) indicates the precursor human IL~3 coding region.
Figure 19 bhc~ws the construction of the plasrnid pGB/IL-313. The sequence at the 5' side of the IL-3 sequence is depicted below the drawings.
Figure 20 shows a schematic representation of pla'mi.tl ~rC3H~/ZL-317.
Figure 21 shows a schematic representation of plasmid pGB/IL-31b.
Figure 22 shows the nucleotide sequence of plasmid pGB/IL-3IG between the unique SacII site in the lactase promoter and the HindIII site behind the terminator (residues 4457 to 7204).
Figure 23 shows the nucleotide sequence of plasmid -~- 13 41543 pGB/IL-313 between the unique SacII site in the lacatse gromate~r and the IlixxdIII site behind the terminator (residues 4457 to 7190).
Figure 2~4 shows the nucleotide sequence of the EF-ao~.
promoter, Sall-BglII-Xhol linker and actin termiri$tor as present on plasm~.d pt3B/TEFact.
Detailed Description c~f the Tnvention A. Definitions As used herein, "human IL-3", "hIL-3, "human multi-CSF", arid "hmulti-CSF" ere usa~d interchangeably, and designate a protein preparation wh~.ch exhibits the following aativit~.es:
1, The protein stimulates colony formation by human hemopaietic progenitor cells wherein the colonies formed include erythroids, granulocytes, granulocyte macrophages, arid mixed.

2. The protein stimulates D~1A synthesis by hum2~n acute myelogencus leukemia (AML) blasts, as evidenced, for example. by labeled thymidine uptake.
To fit the definition of hmulti-CSF, the activity in the foregoing assay must not be substantially inhibited by antibodies raised in response to, and immunospecific for, GM-CSF, unless these antibodies also inhibit these activities by the illustrative hmult~.-C5F below.
One illustrative form of hmulti-CSF ~.s shown in Figure 1 as a 133 amino acid mature protein, having a 19 amino ' 30 acid signal sequence. The amine acid sequence of figure 1 is identical with that disclosed by Yang, Y-C., et al., Cell (1986) 47:3-10 (supra) except at position 8 of the mature protein wherein the Ser of the Yang protein is xeplaced by Pro herein. As shown herein, this amino acid sequence is effective iri its rionglycosylated form. However, it contains two glycosylation sites, and the glycosylated form is also included within the scope of the invention. It is also recognized that the protein may exist in acid addition salt -'- 1341503 farm, basic salt form, or may be neutral, depending upon the pH of its surroundings. Der~.vati~ation by phosphorylation, acetylation, and so forth, to the extent that activity is not destroyed, also result$ in a protein included within the scope of they i,rwention.
zt is also recognized that the entire sequence may not be necessary far activity, parts of the amino said soe~usnce may be daletid dr replaced, while retaining biological activity. A$ illustrated herein, the a7.anine at position 1 may be deleted, as may as many as the first fourteen amino acid residues if replaced by a sequence of residues of a fused peptide sequence. In addition, it is believed that the marine form of the protein requires only the first 79 residues for activity= this corresponds approximately to the first 83 reg~.due~s of the human counterpart. According-ly. fragments which cvinprise only the first 83 amino acid residues of the protein, and the N-terminal replaced forms thereof are also included within the scope of the invention.
Furthermore, it should be considered that the N-terminus of mature hIL-3 is formed by the residues ale--pro-met etc. (see Figure 1). It is known that the protein, when secreted by a yeast host, may in same instances be shortened by two amino acids (ale-pro), due to the interaction with a dipeptidyl-aminapeptidase (72). The hIL-3 without the N-terminal alan~.ne and proline still retains its' biological activity. Yeast strains carrying a null mutation of the X-pralyl dipeptidyl-aminopeptidase gene will produce complete h1L-3 (amino acids 1'133 ) . Accordinc~J.y, included in the mufti-CSFs of the invention are those which contain and those which do not contain the N-terminal alanine and praline, produced by X-prolyl dipeptidylaminopeptidase mutants and wild type hosts, respectively.
lyhen produced as a mature protein in a procaryotic host, the ceding sequence for the mature protein will be prefaced by an A'~G start codon. The resulting N-terminal m~fihi nni ne: may t.hpn hø removprl, or part i a l ly rpmnved, ry processing within the bacterial host, depending on the nature of the subsequent amine acid sequence. Again, bath forms of ' ° ' 13 4 1 5 03 h3L-3 are biologically active. :'herefore, included in the hrnulti-OSPs of the invention are those which contain and those which do not contain the N-terminal methionina.
From the above it is char that amino acid changes m$y be introduced into the human IL.~3 protein, without affecting its biological function. It is recognized that miridr changes in azriino acid sequences by chemical modification of the encoded residue, substitution c~f a different residue, or dQletion or addition of one or more, but preferably only one, re$idue re$ults i~a pro~.~ina whi.ah rat~aio a~;tivitY. Aooozding_ ly, these riondeBtl'uctive mutations are also included within the invention, in particular. the naturally ocourrinq allelic variations and other mutations which are nonlethal to the $ ~4't l~~ty a On the other hand it should be considered that amino acid changes in the human IL-3 protein may be beneficial to the therapeutic use of t'tie protein. Ae recognized herein, the mature protein has four conserved domains at residues 15-3~, 54-61, 74-91, and 107-11E. Proteins aont$ining single and multiple amino avid changes in the nonconserved regions, 1-1~
(which are, in any event, replacable by the sequences of host derived fusion proteinsj, 37-53. G2-73, 92-106, and 119-133 are possible. However, it appears that the cysteine residues at positions 16 and 84 may tae necessary for disulfide bridge formation as they are conserved between species. G'hanc~es xn the conserved domains mentioned above may influence biological properties of the protein, such as receptor binding and signal transduGtion. It is envisaged that hIL-3 having altered properties are of therapeutic use. Such derivatives of hIL-3, which may be made by known protein engineering techniques, are to be understood within the scope of the present invention.
The protein preparation may contain the hmulti-CSF
peptides in monomeric or aggregated form, provided the ~ggrPgat,Pg rQt,~i n anti vi t.y as ahnvp-s~Pfi nark.
As u:,ed herein, "expreESion System" refers to a bNli sequence which contains both a coding sequence whose expression is desired arid appropriate control sequences in -~- 1341503 Operabl8 linkage with it which permits its expres$ion when the GCntz'ol sequences are compatible with the ho~at into which the exprc~$iori system is placed. As is generally understood, "control sequences" refers to DNA segments which are required for ar regulate the expression cf the coding saquenca with which they are operably linked.
control sequences for ail hosts include promoters, which may or may not be controllable by regulation of their environment. Typical promoters suitable for procaryotes include, for examp~.e. the trp promoter ( indua~.bla by IO tryptophan deprivation), the lac promoter (induaible with the ga~.actose analog IPTG). the beta-lactamese promoter, and the phage -derived PL promoter (inducible by temperature variat,ion), Additionally, especially fox expression in Bacillus, useful, promoters include those for alpha-amylase.
protease, Spot and synthetic promoter sequences. Suitable promoters for expre8sion in yeast include the 3-phospho~-glycerate kinaBe promoter and those for other glyac~lytie enzymes, as well as promoter regions for alcohol dehydragenase and yeast phosphatase, Also useful a~:e the transcription elongation factor (TEF) and lactase promoters. Mammalian expression generally employs promoters derived from viruses such as the aa~tmvi~:us promoters .aria tl~e St144 promoter systems, but they also include regulatable promoters such as the metailothionexn gromoter, which is controlled by heavy metals or glucoCOrticoia concentration. There are also now available viral-based insect cell expression systems, as well as expression systems based on plant cell promoters such as the nopaline synthetase promoters, rn addition to the promoter DNA sequence, v~th~.ch is ~fi necessary for the transoription of the gene by RNA polymerase, a variety of control sequences, includ~Lng those regulating termination (for example, resulting in polyadenylation sequences in euCaryotic systems) ass also useful in Contxvliing expression. Some systems also Contain enhancer elements which are desirable but not necessarily necessary in effecting expression, Translation Controls include a ribosome binding site _ lfl (RBS) in procaryotic systems, whereas in eucaryotic systems translation may be controlled by the nucleotide sequence around the AUG codon.
As implied above, recombinant protein production can be effected in a wide variety of hosts, including bacteria (predominantly E. coli, Bacillus, and Streptomyces), in yeast and fungi {such as Saecharomycee, Kluyveromyces, and Aspergillue), and in mammalian and other cell cultures such as cOS cells, X127 cells, Chinea~a hamster ovary oells, Spodoptera 1Q frugiperda {Sf9) cells, and so forth. The protein may be produced as an intracellular mature or fusion pratein, or may be secreted when the DNA encoding an appropriate compatible signal is included in the gene.
~rhe present invention for th a first time enables large scale production of recr~mbinant human IL-3, so that thi$
protein - in purified fbrm - can now be used as a therapeutic agent. The methods described harei.n pxov~.da means for producing glycosylated as well as unglycoaylated forma of the prc~te,im, w'~iic:~i c:am b~ YuiiFi~c3 to rsu~,i~l.&YW,1a11y puY~ l'moaY~ IL-3. "Purified" human ZL-3 refers to human IL-3 as defined above which is free of other protein$ which normally accompany it.
~. Retrieval of cDNA Encoding Human IL-3 Human IL-3 was isolated according to 'the foilowi.ng strategy:
1. A prbcedure was developed which aXlowed for reproducible production of hemopaietic growth factors (HGFs) by human leucocytes.
2. mRNA was prepared from such producing cells and transcribed into double-stranded cDNA.
3. The cDNA was screened with a complete mIL--3 cDNA
which contained both the coding and untranslated 3' downstream portions to obtain DII.
g. ihP hybrid~rzin~J cDNA Clnne DII was inserted into an expression vector pL0 to obtain pL~4 which was expressed in CAS cells to confirm the presence of the sequence encoding human IL-3. Conditioned media from these cells showed the biological activity expected of hIL-3.
The human cDNA was retrievable using this procedure because despite considerable lack of homology with the marine coding sequence, a surprising degree of homology was present in the ~~ uratrr~nslatec~ regism. Appl~c:ants aTP nnawarr-_s of any prior disclosure of the use of a 3' untransiated ~se~quence homology to retrieve an alternate species gene.
In more detail, conditioned medium of lymphocytes cultured in the presence of 12-Q-tetradecanoylphorbol-I3 1~ acetate (TPA) and aoncanavalin A (Con A) is a suitanle source for human HGFs as determined by assay of the medium using sti.mulativn of mouse CFU-S in suspens~.on cultures, proliferation of mIh-3 departdant DA-I cells, human hemopoietic progenitor assays by colony formation in vitro, and in vitro etitnulation of acute leukemia blasts. A cDNA library from human lymphocytes was constructed in lambda gt~-10 phage (20) alld SCx'G'~n~c~ llSitlg 1_hC ~3111i~IIS-~CZ3aI rraeJlll~Il~ or IIIIL-~
C:I7NA, for the occurrence df mIL-3 related sequences. No hybridizing clones were identified.
However, whets complete marine IL-3 cDLdA was used as probe, four clones were identified. Restriction enzyme analysis o~f the largest clone (D11) indicated a 910 by insert cantaini.ng an internal EcoRI site (at position 411. Figure 1).
(It was investigated whether this EcoRI site had arisen by ligation of two independent cDNA fragments or was a naturally occurring site, gouthern analysis of restriction enzyme digested human DNA using labeled 5' and 3' EcoRI
fragments of clone D11 as probe, revealed identical DNA
fragments following digestion with HindIII (15 kb) and HamHI
(4.6 kb). Furthermore, the DNA sequence around the EcoRI site does not correspond to linker sequence (pCCGAATTCGG) used far inserting cDNA inns phage DNA, indicating that these EcoRI
fragments are derived from a single mRNA.) From hybridization and sequenc~.ng experiments it was rc~nc~luded that the small clones (II, IV and VI) are identical to the 3' nucleotide sequence of clc~nc D11 and derived from the same mRNA species.
Computer assisted alignment (Figure 1) of the Dli cDNA and the mIL-3 cDNA revea~.ed sequence homology in the 5' terminal 300 bp, between nucleotides 236-269 and between nucleotides 59a--Sp3 in the 3' terminal region (68ro, 71g and 73~ homology, respectively). In particular, the region between S nualec~tidea 706 and 763 is highly conserved (93$ homology) and contain$ repetitive AT-rich s$guencas. The low homology in the 5' terminal 600bp of the human cDNA (52~) precludes detection by hybridization with the HindIII-~Cbal fragment of mIL-3.
Analysis Qf the human aDNA clone for an encoded protein shows an open xeading frame up to the termination codQn TGA at position 495 497 (~'igure 1). The first ATG
triplet is probably the actual initiation codon of the encoded polypeptide. "~he putative encoded protein consists of a hydrophobic leader peptide of 19 amine acids, which is prob-i5 ably cleaved between the glycine and alanine zesidues (22, 23).
The alignment of the predicted amino acid residues of the human and mouse IL--3 (Figure 1) reveals a homology of 50~ fvr the leader peptide (residues -26 to +1) and 28$ for the mature protein (residues 1 to 133). ~Tithin the leader peptide, there are two Conserved regions of four amino acids (residues -13 to -10 and -3 to +1), of which the second one encloses the processing site. The mature protein is 133 amino ac~.ds long and has a molecular weight of 15 kd. The mature protein has faux conserved domains (residues 15-36, 54-61, 74--91 and 7.07-118) and contains two potential glyGasylation sites (residues 15-17 and 70-72). Both cysteine residues present in the human protein (positions 16 and 84) are conserved and may play an essential role in protein folding by disulfide bridge formation.
In order to verify that this human cDNA encodes a functional protein that resembles mIL--3, the D11 cDNA was inserted in an eucaryotic expression vector (pLO, containing a SV40 transcription unit) to obtain the expression vector pLF34 and transfected to COS 1 cells. The C05/pLB4 conditioned medium (CM) was tested for (1) its capacity to stimulate colony formation by 2xuman bone marrow cells, and (2) to stimulate human acute myelogenous leukemia (AML) blasts.
In vitro colony growth of human hemopoietic ..13- 1341503 progenitors depletad of myglomonaaytic (Virn~-2 poa~.tive) and T-lymphooytic (T-3 po$itive) accessory cells, was effioiently stimulated by COS/pL$~ CM. Tha data demonstrate etimulat~.on of progenitors of several hemopvietio differentiation lineages and cf a subpopulation of BFU-E by COS/pLH4 CM.
Iri 8 separate experiment, bCsne marroca was enriched for progenitor cells by dens~.ty centrifugation, E-rosette sedimentation to remove T-lyrnphocyt.es and adherence to remove mononuclear phagocytes and cultured in enriched med~.um containing Fetal calf serum. Under these canditia~rs, the majority of the colonies ab~.ai.ned upon stimulation with COS/pL~4 CM contained two or more hemopoietic dif~erentiatipn lineagest all contained maerophag~s, ~xpgroximately half immature blasts and/or immature erythroid cel~.s and/or neutrophilic granulacytes and a minority, in addition, basophilic or eosinoph~.7.ic granulocytes. These resu7.ts demonstrate the mult~.lineage stimulatory properties of the protein encoded by the human cDNA clone D11 and its action on developmentally early, multipvtent hemopoietio cells.
t~tith respect to AML stimulation, AML blasts of five patients were stimulated with the COS/pLB4 CM ana assayed for a response by measuring ~H-TdR incorporation and colony formation. Thxee o~ the five leukemia Cell samples responded to the CpS/pLB4 CM in both assays; characteristic dose-response rezationships For colony formation and pNA synthesis of AML blasts of different patients were obtained. The responses to GM-CSF demonstrated further phenotypic differences among the leukemias responding to the CoS/pLB4 CM.
~0 These data demonstrate that the D11 cDNA clone contains the complete genetic information for a biologically active protein which is exported into the culture medium in the transformed COS cells. Despzte the apparent lack of homology with respect to the protein sequence 'between the human protein and mIL-3 (only 3p$), the proteins are comparable with respect to their biological function.
Both proteins exert their effect on developmentally early hemopoietic progenitors of various lineages. The law homology 1~~ 1341503 at the amino acid level i3 also reflected by a low homology in the coding nucleotide sequenoe. Hows'ver, very unexpectedly, a rather hi.c~h degree of homology -- suffic~.ent for retrieval of the human eDNA clone .-~- oaourred in the 3' untrangiatcd region.
SOuthexn analysis of human DNA revedJ.ed a single hybridising gene indicating that this oDrdA does not belong to a family of closely related genes.
From the foregoing results we conclude that the iQ human cDNA insert in D11 encodes the human homc~loc~ of mIL-3.
~~e decided to use the operational term hmulti-CSF for the protein encoded by the CDNA Clone D11 in view of its major biological effect and aw ay.
The identification of hmulti-CSF cDNA o~.ones by virtue of hyl~ridizati,on with the 3' terminal region of the mIL-3 aDNA was ~xnexgeated. t7hereas homologous DNA sequences are in general predominantly found in the coding region, the hmulti-CSF sequence has extensively diverged (45%
homology) in this part of the gene. Analysis of the highly 2Q conserved domain in the 3' terminal, nc~n-coning region reveals the occuranee of 5 ATTTA repeat units which are all preserved in the mIL-3 cDNA (Figure 1).
hMulti-CSF and mrL-3 display considerably less protein homology than other murine and human growth factors or lymphokines such as GM-CSF (25), interleukin-2 (25), interleukin~l (26) and interferons (27-29). The biological activity of the mature mZh-3 appears to be contained in the first 79 amint~ acids, including an absolute requirement for the cysteine residue at position 17 (30). This cysteine residue is conserved in hmulti-CSF (Fig. 1, pos. 16) and may play an essent~.al role in protein folding. The occurrence of a potential glycosylation site around this cysteine residue may interfere with disulfide bridge formation.
C. Production and Formulation of hmulti-CSF
Applicants have provided a representative variety of expression systems capable of producing human IL-3 protein in a variety of forms -~- as fusion proteins, as mature intra-cellular proteins, and as Secreted proteins. Applicants axe unwaxe of availability anywhere in the art of recombinant forms of human IL-3, or, indeed, of any human IL-3 in a preparation which ~.s free of proteins normally accompanying this desired protein. Accoxdingly, the invention herein provide, for the first time, the human IL-3 protein in s manner which is capable of adaptation to therapeutic and diagnostic uses.
The human IL-3 can be produced as a fusion protein with sequences heterologoug to the human ZL-3 amino acid sequence, ~y "heterologc~us" is meant a sequence which is not found in human TL-3 itself, but is an unrelated sequence. This heteroloe~ous sequence may be derived from a bacterial protein, a yeast protein, a mammalian protein, or any of variety of miscellaneous fortuitously ericod2d eequenoes such aa, for example, those encoded by polylinkera. It is clear from the results hereinbelow that at least the first 14 amino acids o~
the N-terminus of the human IL-3 sequence Gs,n be replaced by a heterol.ogous sequence, at least if tho fusion protein i8 further extended past the N-terminus.
The protein can also be obtained as a mature intra-cellular protein by constructs in which the ATG start codan is placed immediately upstream of the des~.red N-terminus. These intracellular proteins, whether mature or fusion proteins, Can be recovered by lysing the cel7.s and purifying the human IL--3 using standard protein purification techniques.
Protein purification is simplified if the human IL-3 is secreted into the medium. ~rhen produced in mammalian cells 3Q with which the dative signal sequence is camt~atibLe, this native signal sequence can be used to effect secretion into the medium. In bacterial or yeast systems, signal sequences compatible with these hosts, such as the penicill.~.nase ar alpha-amylase sequence in bacteria ar the alpha-factor signal sequence in yeast can be used.
ty!'ien produced recombinantly, the human rL-3 is free of proteins normally accompanying it, arid can be purified from the proteins and other materials indigenous to the recombinant _ lg _ ..

host using, for example, Chromatographic methods, gel filtration, ammonium sulfate precipitation, and so forth, As describ~d hereinbelow, the protein ~.s useful for therapeutic and diagnostic purposes. For therapeutic uses, t'he protein may be formulated in ways standard for pharmaceutical compositions which are usad.for the administration of proteins. Suitable exoipiants inCluBe, for example, physiolog-ical saline, Ringer ~ s solut~.on, arid so forth. Alternate formulations, including solid formulations (e. g. lyophilized), can alto be employeQ.
D, preparation of Antibodies ~1 Y
The availability of recombinant IL-~ protein or 1S parts thereof will pertttit prot3uction of antibodies directed against the prote~.n or parts thereof, as demonstrated herein-below. Such antibodies are useful, inter alia, for in vitro detection of col.anies producing hIL-3. for therapeutical use, and for the purification o~ both ~tatura,l and recombinant hIL~3.
Statement of Utility ~.'he nucleotide sequence of the whole or parts of the ~5 cDNA of human IL~3, or closely-related DNA sequences wzll advantageously enable the detection of genetic abnormalities, including genomic rearrangements, restriction fragment-length polymorphisms, mutations and altered gene expression with the use of such techniques as the analysis of chromosomal DNA
using restriction enzymes, DNA and FtNA blotting as well as hybridization techniques (l~aniati.s et al. 1882) and two-dimensional gel electrophoresis (Fisher and Lerman, 1983).
The recombinant hmulti-CSF as provided by the present invention will facilitate a detailed analysis of its role in human hemopoiesis, ~.n particular the possible synergism of hmuxti-CSF and various other HGFs. Furthermore, hmulti-CSF is of considerable interest because of its applicability for in vitro diagnosis o~ human diseases in rl'' 1341503 which hemopoietic progenitor sells are involved, which include the leukemia, as well. as potential therapeutic applications aimed at expansion of hemap4iesis in vivo. The effect of hmulti-cSF on various hemopoietie malignancies with respect to terminal differentiation of the leukemic ae~.ls also needs to be explored. In addition hMulti-CSF may be required for establishing a pro~.iferative state o~ human stem cells in gene therapy protocols, since stimulation with mIL~3 was shown to be required fox euccesful infection of mouse stQm calls with 1p recombinant, replication defective retroviruses.
IL-3 protein can also advantageously be used for the detection of early hemopoiat~ic praaurs~or cells in standardised in vitro cultures (ZJac~ernaker and Visser, 1980t Metcalf et al.
1982; Merchav and ~~agemaker, 1984, Metcalf, 19$6).
IL-3 protein and variants can further be used for the multiplication of hemopoietic stem cells in vitro, possibly in conjunction With other growth factors, for bona marrow transplantation and the genetic manipulation of stem cells (LOwenberg and Dicke, 1977; Vagemaker and Petem, 1978;
Lemischka et al, 1986).
The IL-3 protein can be used for the determination 'of the response pattern of malignant hemopoietic Gells in in vitro tests (Touw and Lowenberg, 1985; Griffin et al, 1986:
Griffin and Lowenberg, 1986).
The Ih-3 protein can furthex be used for the deteGti.on of remaining leukemic cells by in vitro methods (TOUw and 1'.owenberg, 19$6 t Griffin et al, i9f36; Griffin and Lowenberg, 19$5).
Furthermore, the IL-3 protein can be used in vivo for the ~.reatment and prevention of malignant and non-malignant disorders, either by itself or in combinatidnt in which an obtained specific response by the hemopoi.etic system cars result in a clinical benefit.
These applications include:
- Cytopenias and/or immunosuppression due to infections such as A.11~8 - cytopenias due to chemotherapy and~or irradiation - bone disorders such as bone fractures and osteoporosis -ls-immunodeficienties due to general anaesthetic procedures - recovery following bone maxrow transplr~ntat~.on - adjunct to vaccinations and e~djunctive therapy c~f infections.
The eloncd human IL-3 DNA sequence or vlc~e~ely-related I7NA can ba used for gene therapy in genetic deviations from the normal IL-3 gene.
To facilitate the above-described analysis, a large quantity of human IL-3 i.s required. The easiest way to obtain 8uffieient. amounts of the protein is the production vtith microorganisms, in particular yeast~s, bacteria and fungi, e.g.
SacaharomyeeR, Kluyyeromyces, Asparqillu$, 6treptomycea, Bacillus and E. eoli species. Production in mammalian arid ~wr.r~ r other eucaryoCic systems, such as 0127 cells, Spodoptera cells and transgenic animals and plants, is also possible for skilled persons following the teaching of the present invention. These possibilit~.es are all included within the scope of this ~.nvention.
As an illustration how to obtain living cells that produce the human IL-3 protein by expression of the hIL-3 cDNA, a number of plasmids were constructed and transferred to E. coli, B, subtilis, H. licheniformis, S. eerevisiae, K.
_lactis and 0127 cells. Using these host strains the production of recombinant human IL-3 was achieved. The products were tested for their capacity to stimulate human AML blasts as described above for the COS/pL84 conditioned medium. From these experiments it appeared that the proteins made were biologically active.
Tkte following examples are intended to illustrate but not to limit the invention.

_ 19 _ Example 1 Retrieval of cDNA Encoding Human mufti.-CSF (hmulti-CSF) Human leukocytes stimulated with TPA (5 ngjml) and ConA (10 ug~ml) produced considerable amounts of HGFa ae measured by the marine stem cell proliferation assay and various other colony assays. Cells were harvested 34 hrs after stimuJ.atior~, because mRNA production is often transient following stimulation with pharbol esters rind lectinrs. Already afte~C 2~ hrs, HGFs were easily detectable in the CM.
mRNA Preparation Cells were harvested, washed with pB5 and homogenized i.n guanidinium isothiocyanate ao~.utior~ ( 36 ) . RNA
was p911ated tx'~rouc~t~ a cesium chloride cushion. 0ligo(dT)-cellulose chromatography wa* a*ed for *eleetion of mRNAS (36?.
eDNA Synthesis cDNA was synthesized essentially according to Gubler and Hoffmsn (37), using oligo(dT) as primer and AMV reverse transcriptase. Second.strand was synthesized with RNa~eH and E. coli pNA polymerase T. Gaps were closed with T4-DNA ligase and ends were Flushed by T4-DNA polymerase. To protect internal ECORI restriction sites, the cDNA was methylated with EcoRI methylase. Subsequently, tile cDNA was ligated to phosphorylated EcoRI linkers with T4-DNA ligase. After digestiotl with EcoRI, excess linkers were removed by Sepharose CL-4S chromatography. The material recovered in the void volume of the column was larger than 250bp and was used for construct~.an, of the libraries.
Construction of the Phage cDNA Library.
The cDL~A was l~,c~ated to lambda gt~.0 phage arms (20) and packaged with commercial packaging extracts (GigapaGk, Vector Cloning Systems). The recombinant phages were propagated in E. coli 0600 hfl.

Screenin of the phac~e Library.
Qf eaoh plate Containing 1-5000 plaques, two nitrocellulose filter repliea$ were made according to standard procedures. Fll.ters were then hybridized with radiolabeled mIL-3 probe from the HiridIII-XbaI fragment o~ mIL-3 ebLJA or with the oomplete mIL-3 CnNA clones radiolabeled with ~'andom primer. The mIL~3 aDNA clone (pL101) was isolated from a «EIiI-3D cDNA library. j7~~lI-38 mRrlA was ioolated using the guanidinium isothiocyaMate CsCl method. size fractionated on 1Q sucrose gradient and i.n jested into ~Cen_~w opus laevia oocytes .
RNA ~ractiona inducing the oocytes to produce a factox capable of supporting muri.ne stem cell proliferation, were used for synthesis 4f cDNl1 as described above, obL~TA way tailed v~rith dC
residues and inserted in the pstI site of pUCg. mIL-3 clones were identified using synthetic oligonucleotides (from published mIL-3 sequence, 11). Insert of pL101 was purified on polyacrylamide gel and used far screening of the human cDNA
~,ibrary. Probe DNA was labeled using the random primer method (38). potential positive plaques were rescreened and plaque purified. In this way four clones were ident~.f~.ed, including phage D11.
Sequencing of cDNA Clones.
Recomba.nant phages were grown at large scale and purified, cbNA insert9 were removed from the phage arms by digestion with EcoRI and purified on polyacrylamide gel. The purified fragments were l~.gated into M13mp18 and pTZl$R DNA
digested with EcoRI and used far transformation Qf E. c4li JM109. Single strand DNA was prepared arid sequenced according to established prricedures (39). Sequence data were analyzed using various computer programs (40-43).
The sequence obtained for the insert in phage D11 is shown in Figure 1. This 910 by sequence contains the ent~.re coding region for hmuiti-CSF and its signal sequence, and exhibits high homology to the murine clone pL101 in the 3' untranslated region. 'she homology upstream in the coding sequence is relatively more limited. As described above, the protein has a putative 19 amino acid signal sequence followed ~.21- 1341503 by a 133 amino acid mature protein containing two glycosylation sites ( 1517 and 70-72 ) and ~,wo cysteir~e residues at 16 and 84.
The deduced amino acid sequence ~,s the same as that encoded by the genomic DNA disclosed by Yang, Y-C, et al, (supra), except for one amino acid -- that at position $ of the putative mature proteinr the Yang DNA encodes Ser, the aDNA herein ancode~ Pro.
The intronlesa sequence obtained in the phage D11 can be used for procaryotiC expression, as well as fax expression in eucaryotic systems, r~s illustrated below.
Example 2 Expression in Mammalian Cslls A. Construction of the eucaryote expression v~actor pLB4 Phage D11 (containing the longest cDNA insert) was digested with H~.nd III and Bgllx and eubcloned in plasmid pTl (a derivative of pTZl8R, containing some additional restriction sites #.n the multiJ.~.nker, see Example 3A). Clones containing the phaga fragment containing the epNA insert were identified by restriction analysis. The cDNA insert was removed from this plarmid by partial digestion with EeoRI and purified by polyacrylamide gal electrophoresis. The appropriate fz'agment was inserted in a euoaryote expression vector (pL0) in an SV40 transcription unit.
pL0 comprises: Ecr~RI (filled in) - Pstl of pBR322 (1-755), Pstl-Aval of pBR329 {756-1849), Ava~-PvuTI adapter {1850-1$68), PvuII-HindTII {Filled in) of SV40 (promoter) (1x69-221i), Pvurl-BamHJ. adapter containing the un~.que EcoRI
site (2211-2251), Mbol "splice fragment" of SV40 (2252-2$61), B~II-~3amHI (filled in) "poly A fragment" of SV40 (2862-3098), PvuIZ-H~.ndIII promoter fragment of SV40 (3099-3440), HindIII-HamHI Eco gpt gene (3441.-4501), Mbol "splice fragment" of Sv40 (4502-5111) and the Bcll-9amIir (filled in) "poly A fragment"
o~ SV40 (5112-5348).
The Eco gpt transcription unit is of no importance .in transient expression of proteins in COs 1 cells. The resultant expression glasmid for hmulti-CSF was tQrmed pLB4 and was purified on CBCl. This pla~cmid in E. cold. was deposited With the Centraal Bureau of schimmelcultures (CBS), Baarn, the Netherlands, under the provisions of the Budapest Treaty on December X2, 1986 under CBS 568.86. The construct is shown in Figure ~.
B. Expression of hmulti-CSF in CO8 1 Cells and Bioassays.
1Q pLB4 DNA wa8 transfected to COS 1 CeJ.ls u;~ing the calcium phosphate copreeipitativn method (45). Cells were cultured fdr 48-72 hours in alpha medium containing 10~ fetal calf serum. The culture medium was recovered, filtered and u$ed in assays for eat$blish~.ng its biologic activity. Human bone marxow progenitor colony assays and acute myeloid blasts colony acrd proliferation assays taere performed as follows.
Bone marrow wa~c obtained from hematologically normal adult volunteers by posterior iliac ars~ct puncture ~ollowing informed consent. The mononueleated Dells were separated by density gradient centrifugation on a ~~icoll gradient (Nijegaard arid Co., Oslo, Norway), washed and resuspended in Hanks balanoed salt solution (HdSS). risyeloid cells and '1'-lymphocytes were then removed. For this purpose, marrow cells were lysed following incubatipn with monoclonal antibodies OKT-3 {Cn3; Ortho, Ravitan, N.Y.) and Vim 2 (myelo-monocytic cells, 46) at saturating concentrations in the presence of rabbit complement (40$; 30 minutes, 25°C) according to established procedures (47). The cells wexe washed two times in HBSS, resuspended in Iscove's modified Dulbecco's medium {IMDrI) and cultured in the presence of autologous plasma according to Fauser and Messner (16), as described before (48), at a concentration of 1.5-3 x 104/ml. Erythropoietin 1 U/ml (sheep, step III, Gonnaught, v,iillowdale, Canada) and COS/pLH4 CP4 were added as growth stimulating activities.
Results of standard cultures with phytohaemagglut~.n~.n stimulated leukocytes CM (PH-LCM) in direct comparison with COS/pLB4 CM are also given. Sixty percent of the colonies were plucked ane~ identified by mieroscopical analysis. The C1~ from -~3" 1341503 COS cells trao,s~ected with the vector w~.thout insert (pLC) failed to stimulate colony formation by itself, The results are shown in Figure 3. As shown in the figure, the mean numbers of erxthroid (BFU-F), granulocyte--macrvphage (CFU-GM), granulocyte {CFU-G), eosinoph~.l (CFU-Eo), macrophage (CFU-M) and mixed (CFU-MI3t) colonies (+gD) are shown of duplicated eultuxes etimulattd with graded volumes of cQ8/pL84 CM.
Znduetion of AML Proliferation (see Figure 4) AML blasto warn purif~.ed using a bovine albumin (9SA) density gradient. ~tesidual T-lymphocytes were removed from the AML samples by E rosette sedimentativrl (17, 49, 50).
AML {patient 1) colony formation was detarmiried ndt onl~r in the sstsblished PHA leukocyte feeder (PHA l.f) system, but also in a modified version of the technique in which the leukocytes were replaced by COS/pLB4 CM, peric~.ittirig assessment of its colony-stimulating activity (17, 18, 49, 5a) as shown 24 in Figure 4A. All experiments were performed in triplicate, DNA synthesis of AML blasts {patient 2) was assayed by thymidine uptake as described {51) with results shown in Figure 4B. Both assays showed a dose dependent relationship to COS/pLB4 CM added. Addition of control CCS medium did not affect AML proli~exatxon in either assay.
C. Construction of eucaryotic expression yector pLB4/BPV
Zn order to establish stable cell lines expressing human IL-3, 0127 cel~.s {ATCC CRL 1616) were transfected with a derivative of pLB4. This derivative was constructed by insertion of the entire BQV-1 genome {69) into pLB4 by the following strategy. The BPV-1 BamHI fragment was excised from the vector pdgPV-MMTneo(342-12) (70). The BamHI st~.cky ends were filled in using Klenow polymerise. 'I'hen the vector pLB4 was cleaved at the unique EcoRV site within the Eco gpt gene.
Subsequently, the blunt-ended BPV-1 fragment was cloned into the EcoRV cleaved pL$4, resulting in the vector pLB4/Bpv which is able to replicate in 0127 cells, pLB4/BPV was transfected to C1~7 cells using the aaJ~cium phosphate precipitation method (45). The transfected cells were cultured far 16 days, after which foci were picked from the culture dishes. Several independent cell lines ware e$tablished. 'fhe pLB4/BpV vector appear$ to be stably maxntainec~ within the cells, as judged by Southern blotting of Hirt extracts (71) of several oell linen.
Conditioned culture medium was tested for IL--3 activity using the AML proliferation assay. The stable cell lines produce active human IL-3.

Example 3 Construction of E, coli Expression Vectors A. Construction of p(~B~IL-3~1 (see Figures 5, 6, 7 arid 8) For construction at E. coli expression vectors, tree following modifications were performed according to standard procedures (3~).
1. The 3'-terminal noncoding sequences between the AvaI site (position 541) and the XhoZ site {position $56) in pLB4 were deleted by fusion of the DNA fragments following filling of the sticky ends with Klenaw enzyme (Figure 5).
2, For introduction of the hmulti-CSF insert into a bacterial expression veotor, the ~oliowing steps were performed. '~'he pLHl vector was digested with AvaII and the recessed ends filled with Klenow polymerase. Following ligation of a BglZI linker (CAGATCTG), the I7NA was digested with BgIII and BarnHI. The BglII-BamHI hmulti-CSF fragment was purified on polyacrylamide gel and subcloned in tile BglIZ site ~0 of pTl, a derivate of pTZl$R (Pharmacia) modified in the multiple cloning site {see Figure 6). 2'wo clones were obtained, which had the insert in the opposite orientation with respect to the lacZ promoter (see Figure 5). Inserts of these two clones were isolated on polyacrylamide gel following digestion with $glII and EcoRV and subcloned in pTl digested with BglII and HindII. The ~unetion of the BglZI linker and the hmulti-GSF DNA was verified by sequence analysis and showed a fusion of the linker to the AvaII site located at v -25- 134150 nt I of the CDNA clone (this Avalz site had rerisen by ligation of the EceRI linker t0 the cnNA molecule). Sinoe this construct (pG8/Ib-300) was not in phase with the ~.acZ protein, the BglII-EcoRV insert was subcloned into HamHi and Hi.ndIZ
digested pUCB (52). 'ihe resulting construct (pGH/iL-301r see Figures 5, 7 and 8) was tested for produetion of a lacZ/hmulti--NSF fusion protein.
e. Construction of pGB/IL-302, pGB/IL-303. pGB/IL-90~ and pG$/IL-305 {Figures 5, 7 and 8) Several base changes were introduced into the codzng sequenc~a for the N-terminal part of the fusion proteins by introduction Qf Synthetic oligo nucleotides into pC~B/IL-300.
The riew raxpression vectors, called pGB/iL-302, pvs/IL-3o3 and pGB/IL-3024 were constructed ass follows: the HindII-$indIII
fragment of pGB/IL-300 was isolated on agarase gel and liga~.ed to a synthetic oligonucleotides comprising the nucleotides 99--. 7.37 Qf hmulti-CSF and a 5' terminal SaII reGOgnition sequence and inserted into pTZl8R digested with Salt and fiindIII. ~'he sequence of several clones was established, Indeed, several base changes Were observed, resulting in modifications of the hmulti-CSF protein. Inserts of several clones were transferred to pUC~ for expression of the lacZ fusion protein (pGB/IL-30Z, pGB/IL-303). Clone pGH/IL-304 was made in Ease with laeZ by ligation of the Sail site following filling of recessed ends with Klenow. Construction was verified by Fvul digestion, Several clones lacked a synthet~.G oligonucleotide and were found to be fused in Frame to the lacZ protein. One example of these clones was called pGB/IL-305.
C. Construction of pG8/IL-3a6 (see Figures 5, T and 8) Arr expression vector coding for a protein lacking the lac'Z N-terminal amino acids was made from pG6/IL-30(7 by deletion looping as described in (53). The synthetic oligonucleotide comprised 22 nucleotides upstream of the pTZ
lacZ gene including the ATG start codon and the first 24 nucleotides coding for mature IL-3. This plasmid was called pGH/IL-306 (Figures 5, 7 and 8).

E_. coli strains containing the plasmids pGB/IL-300, pGB/IL-301 and pGB/IL-302 were deposited with CBS on July 13, 1987 under CBS 377.87, CBS 379.87 and CBS 378.87, respec-tively.
Figure 8 shows the sequence of fusion regions for the various plasmids constructed. The sequence of the clones is given from the start of the lacZ protein coding region in either pUC8 or pTZl8R (lower case letters) and of the hmulti-CSF coding region (upper case letters) up to the ClaI site at position 158. Mutations in the hmulti-CSF DNA sequence are underlined, resulting in trpl3~arg13 (pGB/IL-302); leu9~pro9 and trpl3~arg13 (pGB/IL-303); met3~thr3 and a silent change (pGB/IL-304).
In the priority application, other designations were used for these plasmids as follows:
pGB/IL-300 = pT-hIL3;
pGB/IL-301 = pUC/hmulti;
pGB/IL-302 = pUC/hmulti~lA;
pGB/IL-303 - pUC/hmulti0lB;
pUC/hmulti0lC;
pGB/IL-304 =
pGB/IL-305 = pUC/hmulti02;
pGB/IL-306 = pTZ/hmulti.
D. Expression of lacZ/hmulti-CSF Fusion Proteins and Mature hmulti-CSF in E. coli E_. coli strains (JM 109) carrying various expres-sion vectors were grown in LB medium containing 50 ~g/ml of ampicillin at 37°C until an optical density of 0.5 at 550 nm was reached. Subsequently IPTG (isopropyl beta-D-thiogalacto-side, Pharmacia) was added to the culture to a final concen-tration of 1 mM and incubation was continued for 3-4 hours.
Plasmids pGB/IL-306 and pGB/IL-302 were also trans-formed to E. coli DH1 (wild type lacz operon). Those strains were grown in LB medium or 2 x TY medium containing 50 ~,g/ml of ampicillin at 37°C for 16 hours.

-2~- 1341503 Bacteria were collected by centrifugation and soni-cated in buffer containing 0.1 M Tris/HC1, pH 8.0; 5 mM EDTA
0.20 *Nonidet P40 (NP-40) and 1 mM phenylmethylsulfonyl fluo-ride (PMSF) and centrifuged for 30 min. at 20,000 x g. Poly-acrylamide gel electrophoresis of the pellet and supernatant fractions showed that the bulk of the hmulti-CSF proteins is stored in the bacteria in an insoluble form.
The pellet was re-extracted with 0.5% NP-40 buffer and finally solubilized with 8 M urea 0.1 M Tris/HCl, pH 8.0 and 5 mM dithiothreitol. Thus, an extensive purification of the fusion proteins was achieved (Figure 9).
As shown in the figure, inclusion bodies from bac-teria (E. coli) containing pGB/IL-301 and pGB/IL-302 were iso-lated as described. Lanes 1 show the 0.2o NP40 supernatant (sample corresponds to 0.1 ml of the original bacterial cul-ture). Lanes 2 show the 0.5% NP40 supernatant (0.2 ml) and lanes 3 the pellet solubilized in 8 M urea buffer (A: 0.05 ml; B: 0.2 ml). The proteins were separated on a 13.5% SDS-polyacrylamide gel and stained with Coomassie Brilliant Blue.
Molecular weights (in kd) of marker proteins (lane M) are denoted on the right. The human multi-CSF fusion proteins are indicated by arrows. The fusion protein encoded by pGB/IL-301 has a MW as expected of about 20 kd; that produced from pGB/IL-302, of about 16 kd.
E. Determination of Biological Activity of Bacterial hmulti-CSF Preparations Bacterial protein preparations were diluted in alpha medium containing 1% bovine serum albumin, filter sterilized and assayed in the AML blast proliferation assay. Diluted samples were added to purified AML blasts and cultured for four days. DNA synthesis was measured using 3H thymidine as described (51). One unit per ml is defined as the amount of hmulti-CSF required for half maximal proliferation of AML
blasts. Figure 10 shows this titration. Various dilutions of the urea extracted protein preparation of bacteria containing the plasmid pGB/IL-302, were assayed for the stimulation of AML blast proliferation using 3H-thymidine. The fusion protein concentration of this protein preparation was 33 wg/ml. Based * Trade-mark an the presented titration euxve, the activity Of this preparation is 16,000 unitsjml:
'Erie amount of bacterial fusion protein in the preparations was estimated from polyacrylamide gel-s electrophoresis and used for calculating apeaifie activities.
The results are shown in the (allowing table;
Table 1 Biolo_qical Activity of 9acterial hmulti-C6F Preparat~.ons ._ Mr (x la-3) ug protezn units specific lt~cZ/hmulti per ml per ml activity ( 1 ) ( 2 ) ( 3 ? un~.ts per mg IL-3 pcB/zz~-3ol za 20 ~s ~, goo p~B/zL-soZ~~o~ ~.s ~ ~gao 4so,aoa pG8/IL-304 18 ND(4) 1$ -pC~B jIL-305 16 1 300 300, 000 pOBjIL-306 I5 ND 70 ND

1. Approximate molecular weights are estimated from the bNA
z5 sequence of the fusion protein (Figure a).
2. IL-3 concentrat~,ons were estimated on SDS--polyacrylamide gel and calculated per ml of starting culture.
3. Activity 4f urea solubilized protein was determined in the AML proliferation assay and is expressed per ml of startzng culture.

4. Not determined.
From these results it was concluded that human mufti-CSF expressed as a fusion protein in E. coli was obtained in biologically active form. The results show that changes introduced into the 13-terminus of the fusion proteins may influence the specific activity of these proteins.

~. 29 _ Example 4 Preparation of Antibody Pre.~arations Capable of Immunospacific Reaction with Human IL-3 Protein A. PelyClonal Rabbit Anti~Human xh-3 Anti$erum.
A preparative gel was made from a lysate of E. aoli eanta~.ning the glasmid pQ8/IL-301. The 20 7cd band with the IL-3 fusion protein was sliced out, minced in saline With a mortsr and emulsified in a l:l ratio in Complete Fxeund'a Adjuvant containing 1 mg of Mycobacterium tuberculosis H37ItA
per ml. New Zealand iThite rabbits (spf) were immunized with 1 ml of the emulsion (with + 100 ~.tg IL-3 fusion pxotein) divided over a injection aitQa (2 x i.m. in the thighs, 3 x I5 s.c. on the back). 8oos~ter injections of the same ant~.qen in Incomplete Freund's Adjuvsnt were given at week 2, 4 and 6.
Serum waa collected at week 8 by venafruncture from the eas.
One volume of serum was absor~aed with 9 volumes of .sonicated pUCB containing E. coli (overnight at 4°C) to remove nonspecific antibodies. Immunoblotting o~ all IL-3 constructs made in E. coli.. H, lioheniformis. B, subtilis, S. cerevisiae and K. lactic showed immunospecifzc reaction with the absorbed sera at a dilution of 1 in 6500.
Some of these results are shown in Figure 11. The proteins were isolated fx'om the recombinant hosts as described above and were separated on a 13.5$ polyaCrylamide gel and blotted onto a nitrocellulose membrane, Lane 1: E. coli containing p'~ZlaR (control); Lane 2: pGB/IL-301; Lane 3:
pGB/IL-301; Lane 4: pGB/IL-302; Lane 5: pUCl9 (control);
Lane 6: pGB/IL-301; Lane 7: pGB-IL-302. Lanes 6 and 7 show proteins present in the pellet after the sonification of the bacteria. Lanes 3. 4 and 5 show proteins present in the pellet after the fixst washing step. Lanes 1 arid 2 show the final urea-solubilized protein fractions.
The arrows show the fusion prote~.ns (of the expected size) expressed from pGB/IL-301 and pG8/IL-302.
figure 12A shows the inhibition of IL-3 dependent proliferation of AML, blast cells by anti-ZL-3 antiserum.

-3°' 1341503 Figure 128 show$ that the preimmune,serum does not affect the action of ZL-3 on Ar~ir blast cell proliferation. In both panels, ~ - IL-3 at 10 U/ml~~ - IL-3 at lU/ml~ ~ -aontrol,,no addition.
Figure 12A Shows ZL-3 c~~peridcnt growth in the AMI, blast prolifer8nt assay (5I) was inhibited by the sera i.n a dose dependent manner; Figure 12g shows preimmune sera do not have this effect. As eontral, GM-CSF dependent growth was unaffected by these sera in the same assay (Figure 12A where 1Q ~~ GM-GSF at 100 U/ml.) H. Monoclonal Mouse Anti-Human IL-3 Antibc~di.es 8e~lb/C mice were immunized with 3 x O.i mi (s.c.~
of the same emulsion as used for the rabbits. A booster (0.1 ml i.p.) of antigen in Incomplete ~'reund's Adjuvant was given at weeds 2 rrnd three days later spleen lymphocytes were fused with SP2/0 myelom8 cells according to standard proce~dure~s (65). Hybridoma eupernat$$ were eare~eried iri the Enzyme Linked xmmuno$ark~ent Assay, using a lysats of E. Coli pGB/IL.-3Q~
(containing the 17 kd IL-3 fus~,on product) as a positive control and a lysate of E. coli pUCa as negative control. In total, 29 IL-3 hybridoma cultures secreting antibodies specific for IL-3 were selected and stabilized.
z5 Example 5 Construction of Bacillus expression vectors General cloning techniques were used (36).
A. Construction of pGB/IL-30'T (Figure 13) Far construction of pG$/=L-307 the SmaI fragment of pLB4 carrying the hmulti-GSF gene, was ligated into pvull digested pUB110 (54). After transformation to competent cells (56) of bf3105 (a spo- derivative of the protease deficient strain DH104 (55)), two clones were obtained. as expected: the fragment was cloned in both orientations. 'I~he plasmid that harbored the Fragment in the correct orientation with respect -.31- 1341503 to the so-called "~~pg II promoter" (59) was called pGH/IL-307.
In this case a fusion protein w~.J.l. be made (see Figure 13 ) .
. Const~uctian of pGH/IL-310 A hmulti-c6F expression plasmid was prepared as described below.
1. Promoter cloning (Figure 14)~
For expression in Bacillus a synthetic G43 promoter as described (58) is used (the promoter used to be called X55) , Plasmids pPROM55s (58), the promoter containing plasmid, and pc3PA14 {59) were digested with EcaRI and Xbal_ The proma~.er ~rac3ment was ligated into the vector fragment, which ha$ been purii~ied on an agarose gel. After transformation to E. coli {JM 101), the correct plasmid was obtained and called pc38/IL-308 (Fig. 14).
2. Introduction of a synthetic oligonucleatide into pGBjIL-308 (Figure 15).
A synthetic oligonucleotide comprising tile nucleotides 39-158 and 484--546 of hmulti-GSF, a 5' terminal Ball recognition sequence and a 3' terminal, XmaIII site was l~.gated into Sell--XmaIII digested pGB/IL-308. The ligation mixture was introduced into JMlUl. After analysis of a number of transformants, the correct plasmid was found, pG8/IL-309.
3. Introduction of hIL3 {Figure 16).
After transformation to and isolation from 8, subtilis DB105, the plasmid pGB/IL--309 was digested with XmaZrl. The recessed ends were filled in w~.th Klenow polymerase, and the plasmid was cleaved with ClaI. The ~lasmid pGn/IL-307 was digested with Aval, the ends filled in with Klenow and then digested with ClaI. Subsec~,uent~.y, the hmulti-CSF containing fragment was ligated into the pGB/IL-309 fragment and transformed to JM101. The resulting plasmid was called pGB/IL-310 (Figux-e 16). This plasmid harbored the hIL-3 gene with its own signal sequence. After isolation of the correct plasmid, it was also introduced into $, subtilis D8105.
.,.
C. Construction of pGB/IL-311 and pGB/IL-312 (Figures 17, 18}
_ ..
pGB/IL~y3lQ was partially digested with H~.rrdIII and totally with PvuII. The two hmulti--CSF containing PvuII-digested with F3indIII sad SmaI.
Figure 17 shows the nucleotide sequence of plasmid pBHAl. mhe piasmid consists o~ positions ~,X-105 and 121-2151 baateriophage FD terminator (double)s positions 221-307; x part pf plasmid pHR322 (vii, positions 2069-2153)= positions 313-768: bacteriophage FI, origin of replicxtic~n (viz.
positions 5482-5943) positions 772--2571; part of plasmid p8R322, viz. the origin of replication and the beta-lactamaae clans: positions 2572-26$5: transposon Tn903, complete genome:
positions 2719-2772; tryptophan terminator (double): positions 2773-3'29; transposon Tn9, the ehlcramphen~.colacetyltrans-ferase gene. 'fhe nucleotides at poei.tion 3005 (A), 3038 (C}, 3302 (A) and 3409 (A) differ from the wild type cat Boding sequence. These mutations were introduced so as to eliminate the Ncvl, Hail, EcoRI and PvuII sites: positions 3730-3$0.4;
multiple cloning site: positzans 3807-7264; part of plasmid pUBlIO, viz, the replication function and kanamyoin resistance gene (EcoRi-pvulI ~xagment) (66, 67): positions 7207-7331;
multiple clon~.ng site. The fragments war put together by known Clorxing techniques, e.g. filling in of sticky ends with Klenow, adapter cloning, etc. All data were derived ~rom GenbankR, National Nucleic Acid Sequence Data $ank, NZH, USA.
After transformation to JI43101 and analysis of a number of ampicillin resistant colonies, two different plasmids were found: pG$/1L-312, which harbored the complete gene with complete control sequences, and pGB/IL-311, which contained the complete gene and the promoter lacking the -35 region in the other orientation (see Figure 18).
pGB/IL-311 has been transformed to H. subtilis DB105 and B. licheniformis strain T399 (~ amy, spo-, exo-protease negative, rift, see ref. 68, where this strain is named T9) .
D. Construction of pGB/IL-313 (Figure 19).
In order to obtain a smaller plasmid, with the hmulti-CSF gene behind the "HpaII promoter", pGB/IL-312 was digested with BamHI and religated. The ligaton mixture was transformed into DB105 competent cells. A number of neomycin resistant colonies were analysed and the correct plasmid was obtained. The plasmid was called pGB/IL-313.
E. Construction of pGB/IL-317 (Figure 20) In order to clone the hmulti-CSF gene behind the B.
licheniformis alpha-amylase transcriptional and translational initiation region and signal sequence, one of the earlier described pOLS-delta vectors (68) was used, viz. pOLS-2 delta.
Besides the alpha-amylase signal sequence (29 amino acids long) this plasmid harbors one amino acid of the alpha-amylase mature sequence (an Ala) followed by a multiple cloning site:
FcoRI-XmaIII-XmaI-SalI-HindIII (68).
The Sall-PvuII fragment of plasmid pG8/IL-310 containing the hmulti-CSF gene was ligated into the SalI-PvuII
digested pOLS-2 delta vector and transformed to DB105. The resulting plasmid was called pGB/IL-317 (Figure 20). The hIL-3 gene still harbors its own signal sequence on this plasmid.
The plasmid was also introduced into B. licheniformis T399.
F. Expression of Five Expression Plasmids in Bacillus Strains B. subtilis and B. licheniformis strains carrying the expression plasmids mentioned below were grown in TSB
medium containing 20 ~ag/ml neomycin or 10 ~g/ml erythromycin at 37°C (for 16-24 hours); 300 ~9/ml of the culture was centrifuged. The pellet was resuspended in sample buffer and analyzed using polyacrylamide gel-electrophoresis followed by S~estern blotting. The supernatant was TCA precipitated, and the pellet was resuspended in sample buffer. Both supernatant and pellet were analyzed for IL-3 protein (see Table 2).
To determine the biological activity of the produced proteins, the following steps were carried out: The - ~ 3 4 1 5 0 3 cellpellete wexe resuspended in a buffer containing 0.1 M
Tris/HC~. pII 4.0 $na 1Q mM MgCl2- Lysozyme waa added to a final oancentratien of 1 mg/ml and PHISF to a final concentration of 1 tttM. ~'he solution was incubated for 30 min. at 37°C.
Subsequently DNase (final concentration 20 rag/ml) was added and the solution was incubated for 15 min. at 20°C. Finally, the blOlOC~ICal activity of this preparation as well as of the supernatant of the cultured cells was determined as described.
The results are shown in Table 2.
Table 2 Expression of the Bacillus Vectors Plasmid Strain MtJ IL--3 Biological Pellet supernatant activity (kd) (kd) pellet supernatant pGB/1L'307 DB1Q5 21 - + --pGB/zL-310 DD105 15;17 15;17 - -pGH/IL-311 DB105 12.5;15 - + -T399 - ~ - + -pGB/IL-313 DB105 15;17 12.5;15 + -T399 - - + _ pGB/IL-317 DS105 12.5;15 12.5;15 + +

17;20 17 T399 12.5;15 12.5;15 + +

17; 20 17 It can be eor~cluded, that in B. subtilis, using pGB/IL--307, a fusion protein is made that has IL-3 activity.
ZThen the human IL-3 gene only contains its own signal sequence rio significant secretion of human IL-3 is obtained. All IL-3 activity is found ~.ntraeellularly. In those cases it seems that besides precursor IL-3 mature ZL-3 (15 kd) has been formed in the cell. '~hus, some transport across the membrane ma.ght have taken place, but the protein zs not transported across the cell wall. However, using the alpha-amylase regula-tion and secretion signals (pGB/IL-317) most of the IL-3 activ-ity appeared to be secreted into the culture medium. Besides a degradation product, two proteins are detected in the superna-tant, one of about 15 kd and one of about 17 kd, most probably mature IL-3 and precursor IL-3, respectively. These data indi-cate that both processing sites, viz. the alpha-amylase and the hmulti-CSF processing site, are used. In the cell the most abundant product is precursor IL-3 containing the alpha-amylase signal sequence (the 20 kd protein) as shown by Western blot-ting. Sometimes a degradation product is detected.
Example 6 Construction of Kluvveromvces lactic expression vectors A. Construction of pGB/IL-316 A DNA fragment comprising the Tn5 gene (61) confer-ring resistance to gentamycin 6418, under the direction of the alcohol dehydrogenase I (ADHI) promoter from S_. cerevisiae, similar to that described by Bennetzen and Hall (62), was in-serted into the Smal site of pUCl9 (63). An E_. coli strain containing the obtained plasmid, pUC-6418, was deposited with CBS on December 4, 1987 under CBS 872.87.
Into the XbaI-HindIII cleaved pUC-6418 vector a Xbal-HindIII fragment from plasmid pGB903 (64) containing the K. lac-tic lactase promoter and calf prochymosin DNA was inserted, re-sulting in plasmid pGB/IL-314.
The Sall-HindIII fragment from this plasmid was re-placed by a synthetic DNA fragment containing a small multiple cloning site and the lactase terminator (see Figures 21, 22).
The resulting plasmid is designated pGB/IL-315.
In the SacII-Xhol cleaved pGB/IL-315 vector the fol-lowing fragments were ligated:
1. The SacII-Xbal fragment from pKS105 (Canadian Pat. Appln. No. 573,343 filed July 28, 1988), carrying the 3' part of the lactase promoter and the 5' part of the alpha-factor signal sequence -~~- 1341503 of S. cerevisiae.
2. A synthetic oligonucleotide comprising the 3' part of the alpha-factor signal sequence etasting at the xbaI
site and the 5' part of the mature hrL-3 eDrtA sequence upto the 5' half of the HpaI site (aa--residue 14 ) .
3. The Hpal-XhoI fragment carrying most part of the hIL-3 cDNA sequence (residue 15-133 plus the 3' nan-coding region). The resulting glasmid, designated pGH/IL-316, is depicted schematically in Figure 21. The complete vector sequence from the SacxI site in the lactase promoter sequence up to the HindIII $~.te at the end of the synthetic terminator is given in Figure 22.
Figure 22 shows the nucleotide sequence of plasmid pGB/IL-316 between the unique Sac II site in the lacta~sa promoter and the Hind IxI site behind the terminator treaidues 4457 to 7204). Residues 4457 to 6100 comprise the lactase pxomator sequence. Residues 6141 to 6355 comprise the alpha factor signal sequence. Residues 6356 to 7115 comprise the sequence for mature human IL-3 plus the 3' noncoding cDNA
sequence. Residues 711.6 to 7204 comprise the synthetic terminator sequence.
B. Construction of pGB/IL-31$
An expression vector similar td pG$/IL-316 was constructed in which the coding information for the alpha factor signal sequence o~ S. cerev~.sia~e was replaced by the alpha-factor s~.gnal sequence of K. lactic (64). Trie remaining part of the plasmid is identical. to pGB/IL-316. The sequence of pGB/IL-318 between the SacII. site in the lactase promoter and the F3indIII site behind the terminator (residues 4457 tc~
7190) is given in Figure 23.
Residues A.457 to 60$7 comprise the sequence of the lactase promoter and a small linker sequence. Residues 6088 to 6342 comprise the K. lactic alpha factor signal sequence.
Residues 6343 t4 7102 comprise the sequence for mature human IL-3 plus the 3' rioneoding cANA sequence. Residues 7143 t4 7190 Comprise the synthetic terminator sequence.

C. Transformation of Kluyveromyces Lactic and Analysis of Secreted hIL-3 plasmids pGH/IL-316 and p~BJIL-318 were digested at the unique SacII site in the lactase promoter region, and used to transform K. lacti~5 strain CF38 2360 (see 64). Integration of the plasmids is thus targeted to the chramoaomal lactase gene promoter region. The resulting 6418 resistant traneformants were grown to saturation ~.n liquid YEPD medium, and the culture supernatants and cell lysr~tes were assayed for 1Q IL~3 activity using the AML cell DNA synthesis assay.
Virtually all IL-3 appeared to be secreted into the culture medium, and to be aot~.ve. Tha proteins from the culture supernatant were precipitated ur>ing ethanol and analyzed using denaturing polyacrylarnide gel-electrophoresis followed by ~7estern blotting. The pr~dominant product has an apparent Myl of about ~l kd, whereas also a distinct band at about IS kd is observed. The latter product moat probably corresponds to the matuxe unglycosy7.ated IL~3, whereas the 21 kd product is the product carrying core glycasylation at the two potential glycosylatian sites. Incubation with Endoglycosidase H results in a protein migrating in the 1~ kd range, suggesting that all IL-3 is processed correctly during the secretion process and that the bulk of the protein is being glycosylated.
Example 7 Construction of a Saccharomyces Cerevisiae Expression Vector A. Construction of pGB/IL-319 First an expression vector called pGB/TEFact was constructed. On this pTZl8R (Pharmac~.a) derived plasmid the S.
cerevisiae translation elongation factor (EF-lalpha) promoter sequence, which was cloned and sequenced as described (73,74), is coupled by means of a small SaII-HglII-XhoI linker to the S. cerevisiae actin transcription terminator sequence (75), which was synthesized using an Applied Biosystems I?NA
synthesizer. The sequence of the expression cassette is given in Figure 24. Residues 1 to 949 comprise the Ef-lzxJ.pha promoter. Residues 950 tv 967 comprzse the sequence of the Sa7.I~-BglII-XhoZ linker. Residues 968 to 1113 cQrnprise the actin terminator sequence.
The unique Smal site in pGB/TEFact was used tQ
introduce the 6418 resis-~ance cassette d,ascri'bed in Example 6.
'rhe resulting ~lasm3e~ was call eri p~R/TEFactG~l$, Finally, the hIL--3 expression vector pG$/IL-318 was constructec7 by intrcduatiQn of the fo7.~.t~wing DNA s~quences into the Sall-Xhol cleaved pGB/TEFactG91~3 plasmid:
- The Sail-NruI fragment from pG~/IL-316 carrying the 8.
careviaiae alpha factor signal sequence and the hIL-3 coding sequence upto the NruI e~.ta.
- A synthetic NruI-~hol DNA fragment comprising the remaining nucleotides coding far hIL~-3 and the Xhol recognition sequence immediately following the TGA atopcadan.
E. Transformation of sacaharomyaes ~erevisiae and Analysis. of Secreted hIL-3 Plasmid pGB/IL-319 was cleaved at the unique EcoRI
site in the EF-iaC promoter. Integration of the p~.asm~.d is thus targeted to the chromosomal EF-ioC region. S. cerevisiae wild type strain p273-103 (alpha; ATCC 25657) was transformed as described for K. lactic (64). The G41~3-resistant colonies were picked and transformants were given tca saturation in liquid YEPD medium, The culture supernatant was assayed for ~iIL-3 activity using the AML assay. The protein produced by S.
cerevisiae was found biologically active.
The proteins from the supernatant were precipitated using ethanol and subsequently analyzed by polyacrulamide gel-eleetrophoresis followed by Western blotting. Two prominent products could be dist~.nguished on the ~Jesterr~ blot, a 21 kd glycosylated product and an unglycolysed product of about 15 kd .

I Metcalf D., Blood 67, 257--267 (19$6y.
2 ~Jhettan A.D. and Dexter T.M. TI$S 11, 207-211 0.986).
3 «agamaker d., In "Bone Marrow Transplantation" (eds. Van 8ekkum D.tJ. and Lowenberg $.) Marvel Dekker rnc. New York 1-72 (1985).
4 Dorsser$ L. et. al.. Exp. Hematol. i2. 35'7. 1984.

5 Till J.E. and MeCulloeh E.A., Radiat. Rep. 14, 213-222 (1961).

6 Hapel A.J., et. al., Blood 65, 1453-1459 (1985).

7 Scheven B.A.A., Nature 321, 79-81 (1986).
$ Garland J.M. and Crontpton s. Exp. Hematol. 11, 757-761 (1983).
9 Stanley E.R. et, al., Cell 45. 667-674 (1986).
10 Kreiglar A.B. et. al., Blood 60. 503-508 11982).
11 Fung M.C. et. al., Nature 307, 233-237 (1984).
12 Yokata T. et. al., Pros. Natl. Acad. Sci. USA, $i, 1070 -1074 (1984).
13 Lowenberg B. arid Dicke K.A., Exp. Hematol. 5, 319-331 (1977).
14 Z7agemak~,r G. and Peters M.F., Cell. Ties. Kinet. 11, 45-56 (1978).
15 Ihle J.N., et, al., In "advances in viral oncology", vol. 4 (ed Klein G.) 95-137, Raven press, New Yoxk 1984.
16 Fauser A.A. and Messner H.A. BlQVd 52. 1243-1248 (1978).
17 Lowenberg B, et. al., Leak. Res. 4, 143--149 (1980).
18 Lowenberg $. et. al.. Blood 59, 64-645 (1982).
19 Buick R.N, et. al., Blood 54, 95-I04 (1979).
20 I~uynh T.V. et. al., In "DNA cloning", ~trol. 1 (Ed. Glover D.M.) IRL press, Oxford 45-78 (19$5).
Z1 Kozak M. Cell 44, 283-292 (19Fs6).
22 Van Heijne G. Eur J. B~.ochem 133, 17-21 (1983).
23 Perlman D. and Halvorson H.O., J. Mol. Biol. 167, 391-409 (19$3).
24 Shaw G. arid Kamen R. Cell 46, 659-667 (19:36).
25 Schrader J.ZJ., et. al., Pros. Natl. Acad. Sci. USA 83, 2458-2462 (1986?.

-~40- 1341503 26 March C . J . , et . a 1. , Nnt~l~e 315 , 641-64'7 ( 1935 ) .
27 Ilse~nehi Y. e~. al. , J. giol. them. 358, 9533.~953fl ( 1fl83 ) _ 28 ri jkema R. et, al. , ~MBO J. 4, 761--767 ( 1985 ) .
29 Zwarthoff E.C. et. al., Nucleic Acid Res. 13, 791-804 (1965).
30 Clark-l~ewig T. et. al., Ecicnec ~31, 134-139 (1986).
31 Kindler V. et. al., Proc. Natl. Aaad. Sci. USA 83, 1001-1005 (1986).
32 DeLamarter J.F. et. 81., MHO J. 10, 2575-2581 (1985).
33 Lemisahka ~.R. et. al., Cell 45, 917-927 (19x6).
3c1 Yu--Chung Yang et. al., Cell 47, 3-10 (1986).
35 Miyatake S. et. al., Proc. Netl. Acad. Sci. USA 82, 316-320 (1985).
36 Maniatis T. et. al., Its "Molecular Cloning, A. laboratory manual". Cold Spring ~l~rbo~ L~~QratQries, New York (1983).
37 Gubler U. arid ~Iofmann H.J., Gene 25, 263-269 (19$3).
38 Feinberg A.P. and Vogelstein B. Anal. Bioehem. 132. 6-13 (1983).
39 Banger F. et. al., Proe. Natl. Acad. Sci. USA 74, 5463 (1977).
40 Queen C. and horn L.J. Nucleic Acid Res. 12, 58 1599 (1984).
41 Staden R. Nucleic AGZd Res. 10, 2951-2961 (19$2).
42 Devereux J., rat. al., Nucleic Acid Res. 12, 387-395 {1984).
43 Li~man t7.,T. and Pearson W.R. S~i~~~~ 227, 143-1441 (1985).
44 Subramani S. and Southern P.J., Anal. Hioch. 135, 1-15 (1983).
45 6Jigler M., et. al., Cell 14, 725-731 (197$).
46 Majdic O., et. al., Int. J. Cancer 33, 67.7-623 (1984).
47 Lowent~erg H. and l3auman ,7 .t;.J. Hlood ut~, 1Z25-i232 ( 19$4) .
Q$ Delwel R., et. al., Blood 68, 41-45 (1986) 49 Swart K. et. al., Dlvod 53, f316-821 (1382).
50 Swart K. and Lowenberg B. Cancer Res. 44, 657-660 (19$4).
51 Touw I. et. al., Hlood 68, 108$-1094 {1986).
52 Vieira, J. and Messing J., Gene 19, 259-268 {1982) 53 Osinga, K.A. et al., Nucleic Acids Res. 11, 8595--8608 (19$3) 54. Gryczan, T.C. ~ ~1., J. Bacteriology 134, 318-329 (1978).

55. Kawamura, F. and Doi, R.H., J. Bacteriology 160, 442-444 (1984).

56. Bron, S. and Venema, G., Mutat. Res. 15, 1-10 (1972).

57. Zyprian, E. and Matzura, H., DNA 5, 219-225 (1986).

58. EPA 0224294, published June 3, 1987.

59. EPA 0244042, published November 4, 1987.

60. Stanssens P. et ~1., In uProtein Engineering and Site-Directed Mutagenesis". Twenty-Fourth Harden Conference.

Program and Abstracts (1985) (Fersht, A.R, and inlinter, G., edts).

61. Reiss, B. ,g~ al., EMBO J. 3, 3317-3322 (1984).

62. Bennetzen, J.L. and Hall, B.D., J. Biol. Chem. 257, 3018-3025 (1982).

63. Yanisch-Perron, C. ,e~ ~1., Gene 33, 103-119 (1985).

64. Canadian Appl. No. 573,343, filed July 28, 1988.

65. Salfre, S. and Milstein, C., Meth Enz 73, 3-75 (1981).

66. McKenzie, T. et ~1., Plasmid 15, 93-103 (1986).

67. MCKenzie, T. et ~l_., Plasmid 17, 83-85 (1987).

68. Canadian Pat. Appl. No. 542,396, filed July 17, 1987.

69. Chen, E.Y. ~t al., Nature 299, 529-534 (1982).

70. Law, M-F. g~ al., Mol. Cell Biol. 3, 2110-2115 (1983).

71. Hirt, B., J. Mol. Biol. 2~, 365-367 (1967).

72. Suarez Rendueles, M.P. ~ ~1., FEBS Lett. 131, 296-300 (1981).

73. Najata, S. g~ ~., EMBO J. 3, 1825-1830 (1984).

74. Nagashima, K. ~ al., Gene 45, 265-273 (1986).

75. Gallwitz, D. and Sures, I., Proc. Natl. Acad. Sci. USA
77, 2546-2550 (1980).

Claims (38)

1. An expression system, operable in a host cell, comprising control sequences operably associated with a DNA
sequence which encodes human interleukin 3 having the amino acid sequence Ala Pro Met Thr Gln Thr Thr Pro Leu Lys Thr Ser Trp Val Asn Cys Ser Asn Met Ile Asp Glu Ile Ile Thr His Leu Lys Gln Pro Pro Leu Pro Leu Leu Asp Phe Asn Asn Leu Asn Gly Glu Asp Gln Asp Ile Leu Met Glu Asn Asn Leu Arg Arg Pro Asn Leu Glu Ala Phe Asn Arg Ala Val Lys Ser Leu Gln Asn Ala Ser Ala Ile Glu Ser Ile Leu Lys Asn Leu Leu Pro Cys Leu Pro Leu Ala Thr Ala Ala Pro Thr Arg His Pro Ile His Ile Lys Asp Gly Asp Trp Asn Glu Phe Arg Arg Lys Leu Thr Phe Tyr Leu Lys Thr Leu Glu Asn Ala Gln Ala Gln Gln Thr Thr Leu Ser Leu Ala Ile Phe.

2. A method for producing human interleukin 3 having the amino acid sequence Ala Pro Met Thr Gln Thr Thr Pro Leu Lys Thr Ser Trp Val Asn Cys Ser Asn Met Ile Asp Glu Ile Ile Thr His Leu Lys Gln Pro Pro Leu Pro Leu Leu Asp Phe Asn Asn Leu Asn Gly Glu Asp Gln Asp Ile Leu Met Glu Asn Asn Leu Arg Arg Pro Asn Leu Glu Ala Phe Asn Arg Ala Val Lys Ser Leu Gln Asn Ala Ser Ala Ile Glu Ser Ile Leu Lys Asn Leu Leu Pro Cys Leu Pro Leu Ala Thr Ala Ala Pro Thr Arg His Pro Ile His Ile Lys Asp Gly Asp Trp Asn Glu Phe Arg Arg Lys Leu Thr Phe Tyr Leu Lys Thr Leu Glu Asn Ala Gln Ala Gln Gln Thr Thr Leu Ser Leu Ala Ile Phe, comprising:
(i) introducing into a host cell, an expression system comprising control sequences operatively associated with a DNA sequence encoding said interleukin 3;

(ii) growing said host cell in a nutrient medium under suitable conditions whereby said interleukin 3 is produced; and (iii) recovering said interleukin 3.

3. Human interleukin 3 having the amino acid sequence Ala Pro Met Thr Gln Thr Thr Pro Leu Lys Thr Ser Trp Val Asn Cys Ser Asn Met Ile Asp Glu Ile Ile Thr His Leu Lys Gln Pro Pro Leu Pro Leu Leu Asp Phe Asn Asn Leu Asn Gly Glu Asp Gln Asp Ile Leu Met Glu Asn Asn Leu Arg Arg Pro Asn Leu Glu Ala Phe Asn Arg Ala Val Lys Ser Leu Gln Asn Ala Ser Ala Ile Glu Ser Ile Leu Lys Asn Leu Leu Pro Cys Leu Pro Leu Ala Thr Ala Ala Pro Thr Arg His Pro Ile His Ile Lys Asp Gly Asp Trp Asn Glu Phe Arg Arg Lys Leu Thr Phe Tyr Leu Lys Thr Leu Glu Asn Ala Gln Ala Gln Gln Thr Thr Leu Ser Leu Ala Ile Phe.

4. Use of human interleukin 3 having the amino acid sequence Ala Pro Met Thr Gln Thr Thr Pro Leu Lys Thr Ser Trp Val Asn Cys Ser Asn Met Ile Asp Glu Ile Ile Thr His Leu Lys Gln Pro Pro Leu Pro Leu Leu Asp Phe Asn Asn Leu Asn Gly Glu Asp Gln Asp Ile Leu Met Glu Asn Asn Leu Arg Arg Pro Asn Leu Glu Ala Phe Asn Arg Ala Val Lys Ser Leu Gln Asn Ala Ser Ala Ile Glu Ser Ile Leu Lys Asn Leu Leu Pro Cys Leu Pro Leu Ala Thr Ala Ala Pro Thr Arg His Pro Ile His Ile Lys Asp Gly Asp Trp Asn Glu Phe Arg Arg Lys Leu Thr Phe Tyr Leu Lys Thr Leu Glu Asn Ala Gln Ala Gln Gln Thr Thr Leu Ser Leu Ala Ile Phe, for treating a diseased state characterized by a deficiency in the level of hematopoietic cells.

5. An expression system operable in a recombinant host which expression system consists essentially of a DNA
sequence encoding purified human IL-3, having the amino acid sequence numbered 1 to 133 in sequence "H" of Figure 1, operably linked to control sequences effective in said host.

6. The expression system of claim 5, wherein the DNA encoding human IL-3 contains no introns.

7. The expression system of claim 5, wherein the control sequences comprise a promoter selected from the group consisting of the lac promoter, the HpaII promoter, the .alpha.43 promoter, the alpha-amylase promoter, the EF-1 alpha promoter and the SV40 promoter.

8. The expression system of claim 6, wherein the control sequences comprise a promoter selected from the group consisting of the lac promoter, the HpaII promoter, the .alpha.43 promoter, the alpha-amylase promoter, the EF-1 alpha promoter and the SV40 promoter.

9. A vector which is pGB/IL-300, 301, 302, 303, 304, 305, 306, 307, 308, 309, 310, 311, 312, 313, 314, 315, 316, 317, 318 or 319.

10. A vector selected from the group of pLB4 and pLB4/BPV.

11. A method for producing purified human IL-3, having the amino acid sequence numbered 1 to 133 in sequence "H" of Figure 1, by a host cell, said method comprising:
introducing into said host cell a DNA construct comprising an expression cassette which comprises in the direction of transcription a transcriptional initiation regulatory region functional in said host cell;
a DNA sequence encoding human IL-3; and a transcriptional termination regulatory region functional in said host cell;
growing said host cell comprising said DNA construct in a nutrient medium under suitable culture conditions, whereby human IL-3 is produced; and recovering the human IL-3 product.

12. A method according to claim 11, wherein said host cell is a transformed living host cell containing genetic material derived from recombinant DNA material and coding for purified human IL-3 having the amino acid sequence numbered 1 to 133 in sequence "H" of Figure 1.

13. A method according to claim 12, wherein said transformed living host cell is selected from the group consisting of yeasts, bacteria, fungi and tissue culture cells.

14. A method according to claim 13, wherein said transformed yeast host cell is selected from the group of Saccharomyces and Kluyveromyces.

15. A method according to claim 13, wherein the transformed bacteria host cell is selected from the group of E. coli and Bacillus.

16. A method according to claim 13, wherein the transformed tissue culture host cell is selected from the group COS, C127 and insect cells.

17. A method according to claim 11, wherein said host cell is a transformed living host cell containing genetic material derived from recombinant DNA material and coding for purified human IL-3 having the amino acid sequence numbered 1 to 133 in sequence "H" of Figure 1, transformed with the expression system of claim 5, 6, 7 or 8.

18. A method according to claim 17, wherein said transformed living host cell is selected from the group consisting of yeasts, bacteria, fungi and tissue culture cells.

19. A method according to claim 18, wherein said transformed yeast host cell is selected from the group of Saccharomyces and Kluyveromyces.

20. A method according to claim 18, wherein said transformed bacteria host cell is selected from the group of E. coli and Bacillus.

21. A method according to claim 18, wherein the transformed tissue culture host cell is selected from the group COS, C127 and insect cells.

22. A purified protein having purified human IL-3 activity, substantially free of other substances accompanying said protein and said human IL-3 having the amino acid sequence numbered 1 to 133 in sequence "H" of Figure 1.

23. The protein of claim 22, which is produced by the recombinant expression of the DNA sequence shown as encoding amino acids 1 to 133 in sequence "H" of Figure 1 or the naturally occurring mutants thereof.

24. The protein of claim 22, in glycosylated or unglycosylated form.

25. The protein of claim 23, in glycosylated or unglycosylated form.

26. The protein of claim 22, 23 or 24 produced in a transformed living host cell containing genetic material derived from recombinant DNA material and coding for purified human IL-3 having the amino acid sequence numbered 1 to 133 in sequence "H" of Figure 1.

27. The protein of claim 26, wherein the transformed living host cell is selected from the group consisting of yeasts, bacteria, fungi and tissue culture cells.

28. The protein of claim 27, wherein the transformed yeast host cell is selected from the group of Saccharomyces and Kluyveromyces.

29. The protein of claim 27, wherein the transformed bacteria host cell is selected form the group of E. coli and Bacillus.

30. The protein of claim 27, wherein the transformed tissue culture host cell is selected from the group COS, C127 and insect cells.

31. The protein of claim 22, 23 or 24 produced in a host cell wherein said host cell is a transformed living host cell containing genetic material derived from recombinant DNA
material and coding for purified human IL-3 having the amino acid sequence numbered 1 to 133 in sequence "H" of Figure 1, transformed with the expression system of claim 5, 6, 7 or 8.

32. The protein of claim 31, wherein said transformed living host cell is selected from the group consisting of yeasts, bacteria, fungi and tissue culture cells.

33. The protein of claim 32, wherein said transformed yeast host cell is selected from the group of Saccharomyces and Kluyveromyces.

34. The protein of claim 32, wherein said transformed bacteria host cell is selected from the group of E. coli and Bacillus.

35. The protein of claim 32, wherein the transformed tissue culture host cell is selected from the group COS, C127 and insect cells.

36. An antibody preparation capable of immunospecific reaction with human IL-3 having the amino acid sequence numbered 1 to 133 in sequence "H" of Figure 1.

37. A method to produce an antibody preparation capable of immunospecific reaction with human IL-3 which comprises injecting a vertebrate host with the purified human IL-3 of claim 22, 23, 24 or 25.

38. A method to produce an antibody preparation capable of immunospecific reaction with human IL-3 which comprises injecting a vertebrate host with the purified human IL-3 of any one of claims 26, 27, 28, 29 or 30.