GB2219799A

GB2219799A - Epidermal growth receptor-binding polypeptides

Info

Publication number: GB2219799A
Application number: GB8906758A
Authority: GB
Inventors: Joseph P Brown; Daniel R Twardzik; Hans Marquardt; George J Todaro; A Gregory Bruce; Timothy M Rose; Anthony F Purchio; Shiu-Lok Hu; Wen Chang; Suo Win Liu; Thomas J Franceschini; Mohammed Shoyab; Greg Plowman
Original assignee: Oncogen LP
Current assignee: Oncogen LP
Priority date: 1988-03-24
Filing date: 1989-03-23
Publication date: 1989-12-20
Anticipated expiration: 2009-03-23
Also published as: JPH0279981A; IL89673A0; FI891308A; ES2013408A6; DK146789D0; LU87489A1; AU2990492A; FR2632321A1; DE3910084A1; SE8901031L; GB8906758D0; SE8901031D0; IE890950L; GR1000876B; NL8900724A; CH681081A5; IT8919917A0; KR890014743A; PT90118B; IT1234961B

Description

DESCRIPTION

POLYPEPTIDES AND PRODUCTION THEREOF INTRODUCTION

Technical Field

This invention relates to novel polypeptide compositions having growth factor activity, including chimeric polypeptides, and methods for their prepara- tion using recombinant DNA techniques.

Background

A significant number of polypeptides secreted by mammalian cells are found to have growth factor ac- tivity The sequence of these polypeptides is substan- tially conserved over a wide range of mammals Much of the interest in these compounds is stimulated by their association with oncogenicity There is also interest in how the production of the growth factors is regu- lated and how they in turn regulate cellular activity.

It has also been noted that infection of mam- malian cells by viruses results in proliferation of the growth of the infected cells For the purpose of this invention, of particular interest are members of the r poxvirus family, such as variola, vaccinia and viruses associated with particular diseases such as Shope fibroma virus, Yaba tumor virus, and Molluscum conta- giosum virus (MCV).

It would be of interest to determine whether viruses causing cellular proliferation upon infection, produced polypeptides involved with growth factors, either acting as the growth factor or the growth factor receptor These compounds could then be used in the studies of viral actions, in assays for the presence of the virus, in nutrient media as mitogens, and in the development of therapeutic agents for treating viral infection It is also of interest to develop compounds which may be agonists or antagonists of growth factors for use in vitro in growing cell cultures, in investi- gating mitotic processes, and in therapy.

Relevant Literature Venkatesan et al, J Virol ( 1982) 44:637-646 describes the DNA sequencing of the structured gene encoding vaccinia virus proteins but there was no recognition that any of these proteins had growth factor activity Cooper et al, ibid ( 1981) 37:284- 294, report the translation of m RNA's encoded within the inverted terminal repetition of the vaccinia virus genome Proliferative diseases for members of the poxvirus family have been reported for Shope fibroma virus (Shope, J Exp Med ( 1932) 56:793-822; Yaba tumor virus (Niven et al, J Path Bacteriol ( 1961) 81:1-14) and Molluscum contagiosum virus (MCV) (Postlethwaite, Arch Environ Health ( 1970) 21:432 452) Descriptions of epidermal growth factor (EGF) may be found in Scott et al, Science ( 1983) 221:236- 240 and Gray et al, Nature ( 1983) 303:722-725 The presence of three disulfide bridges in EGF and trans- forming growth factor (TGF) is reported by Savage et al., J Biol Chem ( 1973) 248: 7669-7692 See also A Doolittle et al, Nature ( 1984) 307:558-560 and Australian Patent Application No 85006288 The receptor binding region of the EGF molecule has been suggested as lying in the loop between the third and fourth cysteine residues (Komoriya et al, Proc Natl.

Acad Sci USA ( 1983) 81:1351-1355).

New descriptions of vaccinia virus growth fac- tor (VGF) may be found in Brown et al, Nature ( 1985) 313:491-492; Reisner, Nature ( 1985) 313:801-803 and Blomquist et al, Proc Natl Acad Sci USA ( 1984) 81:7263-7367 Natural Shope fibroma virus growth factor (SFGF) is described by Chang, W C, et al, in Molecular and Cellular Biol ( 1987) 7:535-540 Shope, J Exp Med ( 1932) 56:793-802 reports that infection of adult rabbits by Shope fibroma virus causes a rapid proliferation of fibroblasts leading to a benign fib- roma Infection of newborn or immunosuppressed adult rabbits causes a rapid proliferation of fibroblasts leading to an invasive atypical fibrosarcoma Allison, A C, J Natl Cancer Inst ( 1966) 36:869-876; Smith, et al, J Natl Cancer Inst ( 1973) 50:1529-1539; Allison, A C J Natl Cancer Inst ( 1966) 35:859-868.

Natural Myxoma virus growth factor (MGF) is described by Upton et al, J Virol ( 1987) 61:1271- 1275 Infection of adult rabbits by Myxoma virus causes a rapidly proliferating invasive myxosarcomatous tumor Strayer, D S and Sell, S, J Natl Cancer Inst ( 1983) 71:105-116 A DNA sequence capable of encoding a conserved growth factor (Natural Molluscum contagiosum virus-growth factor (MCGF)) has been de- tected Porter, C D and Archard, L C, J Gen Virol.

( 1987) 68:673-682 Infections of humans by Molluscum contagiosum virus induces benign skin tumors character- ized by rapidly proliferating cells Brown, et al, Sex Transm Dis ( 1981) 8:227-234 Yaba tumor virus causes subcutaneous histocytomas in monkeys and humans, Bearcroft, et al, Nature (London) ( 1958) 182:195-196, and because of its relationship to Shope fibroma virus, Myxoma virus and Molluscum contagiosum virus is predic- ted to have a related growth factor.

Mouse EGF (m EGF) (Scott et al, 1983, supra; Gray et al, 1983, supra) and human EGF (h EGF) (Gregory et al, Int J Pept Protein Res ( 1977) 9:107-118), are known to be homologous to human, mouse and rat TGF (Marquardt et al, Proc Natl Acad Sci USA ( 1983) 80:4684-4688) and residues 45 to 85 of a 140 residue polypeptide, vaccinia growth factor (VGF) encoded by the vaccinia virus genome (Venkatesan et al, 1982, supra).

Expression of foreign peptides employing a baculovirus vector in insect cells is described by Maeda et al, Nature ( 1985) 315:592-594, and Carbonell et al, J of Virology ( 1985) 56:153-160.

The disclosures of these references are incorporated herein by reference.

SUMMARY OF THE INVENTION

Novel polypeptide compositions are provided finding analogy in fragments of viral proteins, which compositions act as mitogens and can be used in nutri- ent media, as reagents for the detection of growth fac- tor receptors or the presence of growth factors, and as competitors for transforming growth factor and epider- mal growth factor The compositions find therapeutic use, for example, to promote epithelialization and healing of burns and wounds The novel compositions may be synthesized The subject peptides may be formed as oligopeptides or fused proteins employing recom- binant techniqgues in a wide variety of hosts.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is an amino acid sequence comparison of the vaccinia virus protein with known growth fac- tors, where the initial amino acid of the VGF is aspar- 1 r tic acid (D) at position 20 from the methianine, which position corresponds to aa-24 of the disclosed peptides; and Figure 2 shows the proposed structure of ma- ture VGF, with asparagine (N) replacements Potential glycosylation sites in MGF and SFGF are indicated by asterisks.

DESCRIPTION OF THE SPECIFIC EMBODIMENTS

Novel compositions which can act as agonists or antagonists of epidermal growth factor (EGF) or transforming growth factor (TGF) are provided and provide for a wide variety of applications in cell cul- ture, diagnostics, in vivo therapy, and in combination with other peptides in the formation of hybrid poly- peptides as immunogens for production of antibodies.

Polynucleotide sequences may be isolated or prepared which may be introduced into an expression vector for expression of the subject compositions.

The compositions find analogy in fragments of viral proteins, particularly poxvirus proteins As shown in Figure 1, VGF has uncharged and hydrophobic residues near the N-terminus between residues 5 and 15 and near the C-terminus between residues 100 and 124.

These residues may be considered by analogy to integral membrane glycoproteins as having an N-terminal hydro- phobic signal sequence, which is removed proteolytic- ally during or immediately after translation, and a C-terminal transmembranous sequence, which serves to anchor the mature-protein in the membrane Cleavage of the viral polypeptide at arg-43 and arg-90 of the mature peptide would lead to release of a soluble polypeptide The 140 residue VV polypeptide may be assumed to give rise first to a membrane-associated protein of approximately 121 residues after removal of the ig amino acid sequence Following intracellular processing, a soluble growth factor peptide of about 77 a.

residues is obtained Cell ( 1985) 42:383-393.

Comparison of the Myxoma growth factor (MGF) and Shope fibroma growth factor (SFGF) with the other members of the growth factor family (see Figure 1) reveals several noticeable differences MGF and SFGF, synthesized as precursor molecules, have a signal sequence at the N-terminus but unlike the other growth factor members have no transmembrane domain at the C-terminus, indicating that they are secreted mole- cules Sequence analysis shows a significant use of the amino acid asparagine in both MGF and SFGF In six positions, both MGF and SFGF have conserved asparagines where the other members do not have asparagine Of these positions, asparagines replace two highly con- served glycine residues at positions 8 and 31, where the first cysteine is at position 2 Both of these glycine residues occur in combination with highly con- served tyrosine residues at positions 9 and 32 Be- cause of the highly conserved nature of the amino acids in this face of the molecule, it is believed that the cysteine loops present in this face are involved with receptor binding and function of the growth factors in this family.

The amino acids glycine and asparagine are predicted to have the highest Beta turn potential (Chou and Fassman, Ann Rev Biochem ( 1978) 47:251-276).

Thus, it appears that conservation of this turn poten- tial is extremely important for the growth factor ac- tivity The growth factors other than Myxoma and Shope utilize almost exclusively glycine in important turn regions of the molecule In MGF and SFGF, these gly- cines have almost all been converted to asparagines (see Figure 2) This indicates an important biological function due to the replacement of the highly conserved glycine with the asparagine residue The possibility exists that the replacement of glycine in the conserved turn regions in the growth factor structure by aspara- I I gine may lead to increased functionality of the growth factor either by increasing the binding to the EGF re- ceptor or to a related receptor, enhancing the direct effect of the growth factor in the affected cells or by increasing the stability of the molecule.

Two of the novel occurrences of asparagines in MGF and SFGF yield potential N-linked glycosylation sites Furthermore, these sites are present in the two cysteine loops which allow for variable numbers of amino acids Thus, the novel occurrences of aspara- gines may offer multiple advantages to the growth factor beyond increased functionality One of these could be stabilization of the protein in vivo by protection of degradation or immunogenic sites by glycosylation.

The subject polypeptides are characterized by being capable of having at least one loop or circle, usually three loops or circles, as a result of cys- teines which form bridges, where the physiologically active portion of the molecule providing a growth fac- tor related activity will be from about 12 to 65 amino acids, usually 15 to 60 amino acids Of the three loops, two of the loops are of from 12 to 15 amino acids ( 14-17 annular amino acids), exclusive of the cysteine bridge, more particularly one is of 12 or 13 amino acids ( 14 or 15 annular amino acids) and the other of 15 amino acids ( 17 annular amino acids), where the N-terminal proximal loop will usually be of from 12 to 13 amino acids, more usually of 12 amino acids, and the middle loop will be of 15 amino acids.

The third loop or C-proximal loop will be of 8 amino acids ( 10 annular amino acids) and including flanking amino acids will have the following formula:

l I N pp 3 (aa 25-C-aa 27)m C-aa 29 _-aa 30 _Gy I I PP 4-(aa 40-aa 39-aa 38)n-C R aa 35 G-aa 33 wherein:

the one-letter symbols for amino acids have their conventional meaning, wherein C is cysteine, G is glycine, N is asparagine, Y is tyrosine, and R is arginine; aa 25 is a neutral amino acid which may be ali- phatic, particularly of from 3 to 4 carbon atoms, or aromatic, particularly of 9 carbon atoms, and having from 0 to 1 hydroxyl group, e g, alanine, serine, threonine, tyrosine, or phenylalanine; aa 27 is a neutral or basic amino acid, being particularly of from 3 to 6 carbon atoms, and when neutral having from 0 to 1 carboxamide group, e g, arginine, alanine, valine, leucine, isoleucine, asparagine or glutamine; aa 29 is a neutral or basic amino acid, being aliphatic or aromatic, wherein aromatic is exemplified by histidine and aliphatic is of from 2 to 6, more usually of from 3 to 6 carbon atoms, having from 0 to 1 hydroxyl group, e g, serine, threonine, leucine, valine, isoleucine, or arginine; aa 30 is a neutral amino acid, being aliphatic or aromatic, wherein aromatic is exemplified by histidine and aliphatic is of from 3 to 6 carbon atoms, and having from 0 to 1 hydroxyl group, particularly serine, isoleucine and valine; aa 33 intends a neutral aliphatic amino acid of from 2 to 6, more usually of from 3 to 6 carbon atoms having from 0 to 1 hydroxyl group, more particularly serine, threonine, valine, leucine or isoleucine, or an acidic amino acid, as exemplified by glutamic acid and aspartic acid; aa 35 is a neutral or acidic aminc acid, of from 3 to 6 carbon atoms, where neutral is exemplified by alanine, valine, leucine, isoleucine and serine, and where acidic is exemplified by aspartic and glutamic acids; aa 38 is an aliphatic neutral substituted amino acid wherein the substituent is carboxamide and is of from 4 to 5 carbon atoms, or an acidic amino acid, e.g, asparagine, glutamine, aspartic acid or glutamic acid; aa 39 is an aromatic amino acid or a neutral aliphatic amino acid of from 3 to 4 carbon atoms usually having an hydroxyl substituent, preferably aromatic, e g, phenylalanine, histidine, tyrosine, serine or threonine; aa 40 is a neutral substituted or unsubstituted aliphatic amino acid of from 3 to 6 carbon atoms, or a basic amino acid, e g, alanine, valine, isoleucine, glutamine or arginine; m and N are 0 or 1; PP 3 and PP 4 may be the same or different and are either hydrogens, indicating the termination of the polypeptide, or may be polypeptide chains of not grea- ter than a total of 1000 amino acids, usually of not greater than a total of about 500 amino acids, prefer- ably where at least 90 % of the amino acids, more preferably at least about 95 % of the amino acids present are in one of the two polypeptide chains; in some in- stances, the chain may be of only one amino acid and not more than 100-amino acids, frequently of not more than about 50 amino acids, depending upon the use of the polypeptide and the role of the extended chain; the polypeptide chains may be related to the naturally oc- curring polypeptide chains associated with naturally occurring growth factors and poxvirus proteins or may be other than the naturally occurring chains or frag- ments thereof associated with the polypeptide chain specifically set forth in the formula, usually unrelated.

The definitions of the amino acids are set forth below.

Neutral (Ne) aliphatic (Al) unsubstituted substituted oxy thio amido aromatic (Ar) unsubstituted substituted heterocyclic G, A, V, L, I, P S, T C, M N,Q F Y H, Charged (at p H 6 0) basic (Ba) acidic (Ac) W K, R D, E The abbreviations in parentheses refer to the particu- lar amino acid groups By unsubstituted is intended no other heterosubstituents than the carboxy and amino group present in glycine The amino acids are the naturally occurring L-amino acids.

The neutral amino acids may also be described as having non-polar or polar (but uncharged) R groups.

The amino acids which come within these definitions are as follows:

Non-polar amino acids Polar amino acids A, V, L, I, P, M, F, W G, S, T, C, Y, N, Q The loop between the cysteines at positions 28 and 37 in the above formula will have from 0 to 1 acidic amino acid (S) The amino acids immediately flanking (i e amino acids 27 and 38) the cysteines outside the loop may include at least one charged amino acid and at least one amido substituted aliphatic amino acid Of the amino acids in the loop ( 28-37) from 4 to 6, usually 5 amino acids will be neutral aliphatic amino acids and not more than 2, usually 1, will be aromatic amino acids Of particular interest are compounds where the polypeptide is of fewer than 130 amino acids, more par- ticularly of fewer than 55 amino acids, and at least 44 amino acids, preferably at least about 53 amino acids.

Of particular interest are polypeptides including amino acids 44 to 88 and fragments thereof of VGF (aa 1 to aa 47 see Figure 1).

Preferably aa 29 is a substituted or unsubsti- tuted aliphatic amino acid of from 3 to 6 carbon atoms having from 0 to 1 hydroxyl group, particularly serine or valine; or an aromatic amino acid, particularly histidine.

aa 30 is preferably histidine, serine or an un- substituted aliphatic amino acid of from 5 to 6 carbon atoms, particularly isoleucine; aa 33 is preferably a substituted or unsubsti- tuted aliphatic amino acid having from 0 to 1 hydroxyl groups and of from 3 to 6 carbon atoms; aa 35 is preferably an aliphatic amino acid of from 3 to 6 carbon atoms; aa 38 is preferably glutamine or glutamic acid; aa 39 is preferably aromatic, more preferably histidine or phenylalanine; aa 40 is preferably a neutral or basic amino acid.

Of particular interest is the presence of two loops, with the C-proximal loop joined to the second loop, where the amino acids of the second loop may be widely varied The second loop is preferably of from about 14 to 16 amino acids exclusive of the cysteine bridge, more preferably of about 15 amino acids Of the amino acids, from 6 to 9, preferably 7 to 9, more preferably about 8 are aliphatic amino acids either substituted or unsubstituted, preferably not more than about 3 of the amino acids being substituted, more preferably from 1 to 2 amino acids being substituted; there will be from 2 to 4 aromatic amino acids, more particularly 3 aromatic amino acids, desirably histi- dine and tyrosine; there will be from 2 to 4 acidic amino acids, preferably 3 acidic amino acids, more particularly aspartic acid; and there will be from 0 to 2, more usually 1 basic amino acid, particularly arginine.

Desirably, the cysteine forming the subject loop clos- est to the cysteine of the other loop, will be separa- ted by from 0 to 2, more usually from 0 to 1 amino acids, particularly arginine.

Of particular interest for some applications for the subject compounds are compounds of the follow- ing formula:

1 6 a 1 Y L ppl-aal-C-aa 3-aa 4-aa 5-aa 6-aa 7-aa-F-C-F N H_(aa 12 a) _G-aa 14-C-aa 16-Arl-(aa 17 a)p-(aal 7 b) p p 18-aa19-aa 20-aa 2 l-aa 22-(aa 22 a)-(aa 22 b)aa 23 a P -(a 22 -aa 2 aa 24-(aa 25-C-aa 27-C-aa 29-aa 30-G-Y-aa 33-G-aa 35 R-C-E-aa 39-aa 40)-aa 41-L-aa 43-aa 44-pp 2 RC Ea wherein:

the symbols aa 25 through aa 40 in the formula have been described; pp 1 and pp 2 are the same or different and may be hydrogens, indicating the terminal portion of the indicated polypeptide or may be polypeptides having a total of up to about 1000 amino acids, more usually of up to about 500 amino acids and may have a total of as few as 1 amino acid, or may individually or separately be polypeptides of from 1 to 100 amino acids, more usu- ally from about 1 to 75 amino acids, more particularly from about 5 to 50 amino acids; these polypeptides will have specific applications in modifying the specifical- ly described sequence for a predetermined purpose; pp 1 and pp 2 may be the same or different from the natural poxvirus polypeptide, usually different; aa 1 may be any amino acid, more particularly an aliphatic amino acid, basic amino acid or acidic amino acid, preferably an unsubstituted aliphatic amino acid of from 2 to 6 carbon atoms, including glycine, leucine, valine, lysine, glutamic acid and aspartic acid; aa 3 is a neutral amino acid, particularly of from 2 to 4 carbon atoms, more particularly asparagine, glycine and proline; aa 4 may be any amino acid, aliphatic or aroma- tic, particularly of from 2 to 6 carbon atoms, which may be neutral or acidic, more particularly of from 3 to 6 carbon atoms, including proline, aspartic acid, histidine, serine and leucine; aa 5 is an acidic or a neutral substituted aliphatic amino acid, particularly glutamic acid or aspartic acid or an hydroxy substituted amino acid of from 3 to 4 carbon atoms, more particularly serine; aa 6 is an unsubstituted aliphatic amino acid of from 2 to 6, usually 2 to 3, carbon atoms, or an aromatic amino acid, particularly glycine, histidine or tyrosine; aa 7 is an acidic, basic or neutral amino acid, particularly of from 3 to 6 carbon atoms, such as glutamic acid, aspartic acid, lysine and threonine; aa 8 is a neutral unsubstituted aliphatic amino acid of from 2 to 3 carbon atoms, or an amino acid sub- stituted with carboxamide of from 4 to 5 carbon atoms, e.g, asparagine, glutamine or glycine; aa 11 is a neutral amino acid, either aliphatic or aromatic, more particularly aliphatic, particularly of from 5 to 6 carbon atoms, more particularly leucine or phenylalanine; aa 12 is an aromatic amino acid or a carbox- amide substituted aliphatic amino acid of from 4 to 5 carbon atoms, particularly histidine or asparagine; aa 112 a is a neutral aliphatic amino acid or acidic amino acid, more particularly glycine, aspara- gine or aspartic acid; aa 14 is an acidic amino acid or neutral ali- phatic amino acid substituted or unsubstituted of from 4 to 5 carbon atoms, having from 0 to 1 hydroxyl group, particularly aspartic acid, threonine or valine; aa 16 is a bond or a neutral or basic aliphatic amino acid, either substituted or unsubstituted, of from 4 to 6 carbon atoms, particularly isoleucine, arginine or methionine; Ar 1 intends an aromatic amino acid, which may have a carbocyclic or heterocyclic ring, and includes histidine, phenylalanine or tyrosine; aa 117 a is a neutral aliphatic amino acid, either substituted or unsubstituted, of from 3 to 6, usually 3 to 5 carbon atoms, having 0 to 1 hydroxyl group, and includes valine, leucine, isoleucine and threonine; aa 117 b is a neutral unsubstituted amino acid of from 5 to 6 carbon atoms, or an acidic amino acid, more particularly valine, isoleucine or glutamic acid.

aa 18 is a neutral aliphatic amino acid, sub- stituted or unsubstituted, of from 2 to 6, preferably of from 3 to 6 carbon atoms having 0 to 1 hydroxy sub- stituent, particularly alanine, serine or glutamine; aa 19 may be any amino acid, particularly ali- phatic acidic or basic amino acids, more particularly I I of from 4 to 6 carbon atoms, more preferably of from 5 to 6 carbon atoms, particularly arginine, leucine, valine and glutamic acid; aa 20 is an acidic amino acid or aliphatic carboxamide-substituted amino acid, and includes aspartic acid, asparagine, glutamic acid or glutamine; aa 21 is a neutral unsubstituted aliphatic amino acid or basic amino acid of from 3 to 6 carbon atoms and having from 0 to 1 hydroxyl or carboxamide group, and includes isoleucine, leucine, asparagine, serine, threonine, lysine, and arginine; aa 22 is a bond, or an acidic amino acid which is aspartic acid or glutamic acid, or a neutral alipha- tic amino acid, particularly valine; aa 22 a is a neutral aliphatic amino acid, having 0 or 1 hydroxyl group, particularly 1 hydroxyl group, more particularly serine; aa 22 b is a neutral aliphatic amino acid having 4 to 6 carbon atoms, particularly leucine or isoleucine; aa 23 is a bond, or a neutral aliphatic amino acid, substituted or unsubstituted, of from 2 to 6 car- bon atoms, having from 0 to 1 hydroxyl substituent, including glycine, serine, threonine and asparagine; aa 24 is an aliphatic amino acid, particularly a thiosubstituted aliphatic amino acid, more particu- larly methionine, proline or an aromatic amino acid, particularly tyrosine; aa 41 is a neutral substituted or unsubstituted aliphatic or acidic amino acid of from 4 to 6 carbon atoms, particularly valine, asparagine or aspartic acid; aa 43 is a neutral aliphatic, acidic or basic amino acid, which may be substituted or unsubstituted of from 4 to 6 carbon atoms, and includes valine, leucine, isoleucine, arginine, lysine and aspartic acid; aa 44 may be any amino acid, particularly other than a basic amino acid, and may be acidic, neutral or aromatic, when other than aromatic, being of from 3 to 6 carbon atoms, particularly aspartic acid, alanine, leucine, tryptophan or threonine; and p is 0 or 1.

Of particular interest is when P Pl has the following sequence:

ppl' aa-24 _aa-23 _aa-22 _aa-21-aa-20 aa-19-aa-18 L -aa-17-aa-16-aa-15-aa-14-aa-13-aa-12 _aa-11 aloaa-10-aa-9 _aa-8-aa-7 _aa-6 aa-5 4 _aa-3-aa-2-aa-1 wherein:

ppl' in combination with the subsequent amino acid symbols in the above sequence is the equivalent of pp 1 The above sequence comes within the definition of i pp 1; ppl may be hydrogen or an amino acid sequence.

One or more of the amino acids, symbolized by aa-x (x is any number), may be a bond, so as to serve to reduce the number of amino acids in the N-terminal chain.

Therefore, all or a portion of the amino acid sequence indicated may be present When a portion of the se- quence is present, preferably the sequence will involve contiguous amino acids, that is, amino acids in their numerical order without deletions Usually, there will be at least one amino acid, more usually at least three, more usually at least six, where the remaining upstream amino acids may be absent, so that P Pl may be joined to aa-1, aa-6, or the like; aa-1 may be any amino acid, particularly ali- phatic, of from about 3 to 6 carbon atoms, preferably neutral or basic, more particularly including lysine, arginine, proline, asparagine or serine; aa-2 may be any amino acid, either aliphatic or aromatic, particularly aliphatic, more particularly of from about 3 to 6 carbon atoms; aa-3 may be aliphatic or aromatic, particu larly aliphatic, either polar or non-polar, of from 2 r to 6, more usually of from 2 to 5 carbon atoms, gen- erally having from 0 to 1 hydroxyl substituent; aa-4 may be any aliphatic amino acid, neutral or basic, generally of from 3 to 6 carbon atoms, usually of from 5 to 6 carbon atoms, particularly asparagine, proline or lysine; aa-5 may be any aliphatic amino acid, particu- larly neutral, generally of 3 to 6 carbon atoms, par- ticularly valine or isoleucine; aa-6 is a neutral or acidic amino acid, when aliphatic of from about 3 to 6, more usually of from about 4 to 6 carbon atoms, more particularly isoleu- cine, valine or aspartic acid; aa-7, aa-8,aa 9 aa 12 aa-14 aa-15 aa-16 aa-17, aa 19, aa-20, aa-21 and aa 23 are aliphatic amino acids of from 3 to 6 carbon atoms, either polar or non-polar, particularly polar having from 0 to 1 hydroxyl substituent or aromatic; aa-10 and aa-25 are non-polar aliphatic amino acids or acidic amino acids of from 2 to 6, more usually of from 2 to 3 carbon atoms; aa-13, aa-22, and aa-24 are polar aliphatic amino acids or acidic amino acids of from 4 to 5 carbon atoms, particularly having a carboxamido substituent; aa-ll and aa-18 are aliphatic amino acids, either polar or non-polar, or acidic amino acids.

Of particular interest is an N-terminal frag- ment of from aa-1 to aa-24 or aa-1 to aa-7, more par- ticularly aa-l to aa 6 ppl' may conveniently be an unrelated amino acid sequence which may serve as a fusion protein, particularly to provide an immunogenic peptide for production of antibodies to the compound.

The primary aspects of the subject composi- tions are the sequence aa 25 through aa 40, particularly the sequence from the cysteine at position 28 to the cysteine at position 37 and the loops generated by the cysteines at positions 2 and 10 and the cysteines at positions 15 and 26 Desirably, the subject composi- tions have the loop created by the cysteines at positions 28 and 37 in conjunction with the loop created by the cysteines at positions 15 and 26.

Chimeric polypeptide sequences may be prepared by combining fragments of various polypeptides having a sequence substantially similar to naturally occurring polypeptide chains associated with naturally occurring growth factors and poxvirus proteins The resulting chimeric polypeptides will befrom about 40 to 65 amino acids, usually about 45 to 60 amino acids, particularly 49 to 53 amino acids In each case, the framework structure of the cysteines is retained with the cysteine bridges defining loops of the sizes described previously Thus, a fragment may be employed from any growth factor having substantial homology with VGF (see, for example, Figure 1), or other growth factor having the same framework structure as the subject compositions and similar physiologic activity This will be regardless of the mammalian source, such as primate, e g, human, rodent, e g, rat and mouse, bovine, avian, porcine, etc.

Up to three junctures may be required, depen- ding upon the source of the fragments and the length of the desired polypeptide Desirably, the junctures will be made at some point between aa-6 and aal; aall and aa 5; aa 25 and aa 29; and aa 42 and aa 53 The sequence between about aa -6 and about aa 15 is referred to as the N-terminal sequence; that between about aa 14 and about aa 29 is referred to as the central fragment; and that between about aa 25 and the end of the peptide (gener- ally aa 44 to aa 47) is referred to as the C-terminal domain The exact point of juncture will vary depen- ding upon the location of restriction sites, etc In addition, an extreme N-terminal sequence comprising about aa-24 to aa-7, more usually aa-23 to aa-7, may also be joined to the chimeric polypeptide.

C The fragments will generally be of from about to 50, usually 15 to 45 amino acids, since aa 20 does not intend the twentieth amino acid of the compound, but only of the specific sequence which has been spe- cifically defined (see Figure 1).

Several convenient restriction sites are de- signed into the synthetic genes used to construct the chimeric polypeptides When possible, the restriction sites leave the amino acid sequence of the growth fac- tor gene unaltered However, in some cases incorpora- tion of the new restriction site yields an altered amino acid sequence Of particular interest is where the coding sequence for VGF is modified to introduce a Kpn I restriction site by changing the amino acid sequence at aa I 3 to aa I 5 from GDC to GTC and to introduce a fph I site at aa 23 to aa 28 by changing the sequence GMYCRC to GYACVC These changes provide convenient sites for linking fragments from the VGF gene to fragments from other growth factors For example, the gene fragment aa 26 to aa 47 can conveniently be joined to a gene fragment of u-TGF encoding amino acids corresponding to residues aa-6 to aa 25 to produce a TGF/VGF hybrid polypeptide Thus, the sequence of the chimeric peptide will comprise aa 6 to aa 13 (N-terminal) and aa 14 to aa 25 (central fragment) derived from a TGF, and aa 26 to aa 47 (C-terminal) derived from VGF, modified as above.

Preparation of plasmids capable of expressing chimeric proteins having the amino acid sequences derived from fragments of more than one growth factor or poxvirus polypeptide with sequence changes where necessary to introduce a convenient restriction site are described in detail in the experimental section.

Fusion proteins may also be prepared where a growth factor or a chimeric polypeptide is expressed joined to the N-terminal amino acids of a signal sequence or the N-terminal amino acids of a highly I expressed bacterial, bacteriophage or eukaryotic gene Additionally, an enzymatic or chemical cleavage site may be provided following the leader sequence to provide for cleavage of the mature polypeptide from the leader sequence Preparation of plasmids capable of expressing these fusion polypeptides and methods for cleavage and isolation of the desired polypeptide are described in detail in the Experimental section.

Preparation of Growth Factors The subject compositions can be prepared in a variety of ways depending on the size of the composi- tion Particularly below about 80, more particularly below about 60 amino acids, the composition can be prepared by synthesis in accordance with conventional ways See, for example, Merrifield, Solid-Phase Pep- tide Synthesis, "The Peptides Analysis, Synthesis Biology," Special Methods in Peptide Synthesis, Part A, Vol 2, Gross and Meinhofer eds, Academic Press, NY, 1980, pp 1-284 See also, U S Patent No 4,127,526.

Alternatively, the use of hybrid DNA technol- ogy can be employed, where DNA sequences can be used which code for the desired polypeptide or precursor thereof DNA sequences encoding the growth factor of interest can be synthesized employing conventional techniques such as overlapping single strands which may be ligated together to define the desired coding sequence The termini can be designed to provide restriction sites or one or both termini may be blunt- ended for ligation-to complementary ends of an expression vector For expression of the sequence an initial methionine is provided Expression vectors are generally available and are amply described in the literature.

Instead of synthesizing the gene of interest, the growth factor gene may be isolated by various techniques For example, where the growth factor is a viral growth factor, m RNA can be isolated from the virus which codes for the polypeptide including the growth factor, the m RNA reverse transcribed, the resulting single-stranded (ss) DNA used as a template to prepare double-stranded (ds) DNA and the ds DNA gene isolated Another technique is to isolate a piece of the viral DNA, and using a probe, appropriately degenerate, comprising a region of the most conserved sequences in a growth factor gene, identify sequences encoding a factor in the viral genome The probe can be considerably shorter than the entire sequence, but should be at least 10, preferably at least 14, more preferably at least 20 nucleotides in length Longer oligonucleotides are also useful, up to the full length of the growth factor gene Both DNA and RNA probes can be used.

In use, the probes are typically labeled in a detectable manner (for example with 32 p or biotinylated nucleotides) and are incubated with single-stranded DNA or RNA from the organism in which a gene is being sought.

Hybridization is detected by means of the label after single-stranded and double-stranded (hybridized) DNA (or DNA/RNA) have been separated, typically using nitrocellulose paper Hybridization techniques suit- able for use with oligonucleotides are well known to those skilled in the art.

Although probes are normally used with a detectable label that allows easy identification, unlabeled oligonucleotides are also useful, both as precursors of labeled probes and for use in methods that provide for direct detection of double-stranded DNA (or DNA/RNA) Accordingly, the term "oligonucleo- tide" refers to both labeled and unlabeled forms.

Once the desired DNA sequence has been ob- tained, it may then be manipulated in a variety of ways to provide for expression; for example, chimeric polypeptide sequences may be prepared by combining gene fragments from at least two polypeptides having sequences which are at least above 30 % homologous to naturally occurring growth factors which bind to the EGF receptor ("natural ligands") It is highly desirable that the three-dimensional structure, especially the loop structure, of the polypeptide be retained, particularly that portion(s) of the structure which may be responsible for binding to the EGF receptor and for biological activity of the naturally occurring growth factors which bind to the EGF receptor.

Depending upon the source of the fragments and the length of the desired polypeptide, convenient restriction sites may be designed into the synthetic genes used to construct the chimeric polypeptides as described above When possible the restriction site(s) leaves the amino acid sequence of the polypeptide unaltered.

However, in some cases, incorporation of a new restriction site(s) may yield an altered amino acid sequence without changing the activity of the protein.

Where the gene is to be expressed in a host which recognizes the wild-type transcriptional and translational regulatory regions of the growth factor, the entire gene with its wild-type 5 ' and 3 '- regulatory regions may be introduced into an appropriate expression vector Various expression vectors exist employing replication systems from mam- malian viruses, such as Simian Virus 40, adenovirus, bovine papilloma virus, vaccinia virus, insect bacu- lovirus, etc These replication systems have been developed to provide for markers which allow for selec- tion of transfectants, as well as providing for conven- ient restriction sites into which the gene may be inserted.

Where the gene is to be expressed in a host which does not recognize the naturally occurring wild- type transcriptional and translational regulatory re- gions, further manipulation will be required Conveni- I ently, a variety of 3 '-transcriptional regulatory re- gions are known and may be inserted downstream from the stop codons The non-coding 5 Y-region upstream from the gene of interest may be removed by endonuclease restriction, Ba 13 l resection, or the like Alternative- ly, where a convenient restriction site is present near the 5 '-terminus of the gene of interest, the gene of interest may be restricted and an adaptor employed for linking the gene of interest to the promoter region, where the adaptor provides for the lost nucleotides of the gene of interest Various strategies may be em- ployed for providing for an expression cassette, which in the 5 '-3 '-direction of transcription has a transcriptional regulatory region and a translational initiation region, which may also include regulatory sequences allowing for the induction of regulation; the gene of interest under the transcriptional and translational control of the initiation region; and a tran- scriptional and translational termination region.

Choice of appropriate regulatory sequences will take into account the following factors which affect expression In terms of transcriptional regu- lation, the amount and stability of messenger RNA are important factors which influence the expression of gene products The amount of m RNA is determined by the copy number of the particular gene, the relative effi- ciency of its promoter and the factors which regulate the promoter, such as enhancers or repressors Initi- ation is believed to occur in the region just upstream of the beginning -of the coding sequence.

The promoter in prokaryotic cells comprises nucleotide sequences which can affect the efficiency of transcription The sequences include the regulatory regions at about -35 and -10 nucleotides from the start of the DNA chain Efficient promoters include those in which the nucleotide sequence of the -35 and -10 regulatory regions is substantially the same as consensus sequences for these regions in bacterial promoters from highly efficient genes Generally these regions are about 5 nucleotides and 6 nucleotides, respectively, in length, and each sequence may vary by about one nucleotide in length and/or in sequence A preferred sequence for the -35 consensus regulatory sequence is from the trp promoter, namely TGACA, and for the -10 consensus regulatory sequence is from the lac promoter, namely TATAAT.

Not only the nucleotide sequences but also the spacing of the consensus sequences of the -35 and -10 regulatory regions, with respect to each other, is important for obtaining optimum transcription of the gene of interest Generally, the consensus sequences of the -35 and -10 regulatory regions are separated by about 16 to 18 nucleotides, preferrably by about 17 nucleotides Each sequence may vary by about one nu- cleotide.

Illustrative transcriptional regulatory re- gions or promoters include, for bacteria, the s-gal promoter, lambda left and right promoters, trp and lac promoters, trp-lac fusion promoter, etc; synthetic promoters having sequences substantially similar to these sequences may also find use A preferred promoter for bacteria is a fusion promoter comprising the -35 regulatory region from the trp promoter and the -10 regulatory region from the lac promoter Most preferrably, the promoter is one in which the -35 trp consensus sequence is located about 17 nucleotides upstream from the 10 lac consensus sequence for yeast, glycolytic enzyme promoters, such as ADH-I and -II promoters, GPK promoter, and-PGI promoter, trp promoter, etc; for mammalian cells, SV 40 early and late promoters, adenovirus major late promoters, I 5 promoter or enhancer sequences and the like.

The transcriptional regulatory region may additionally include regulatory sequences which allow the time of expression of the gene of interest to be modulated, e g by presence or absence of nutrients or expression products in the growth medium, temperature, etc For example, expression of the gene of interest may be regulated by temperature of the host cell growth medium by including using a regulatory sequence comprising the bacteriophage lambda PL promoter, the bacteriophage lambda O L operator and the gene CI 857 which codes for the temperature-sensitive C 1 repressor in the expression vector This will allow regulation of the promoter by interaction between the repressor and the operator at low temperatures, for example about 300 C Increasing the temperature to about 420 C would inactivate the repressor and allow expression of the gene of interest.

As an example of modulation using growth medium nutrients, regulation of the lac or the trp-lac hybrid promoter can be accomplished by use of the gene for the Lac I repressor, which binds to the lac promoter region downstream from the -10 regulatory region The Lac I repressor gene may be present on an episome, preferrably the Lac I enhanced mutant, or can be included in the expression cassette itself Presence of a significant concentration of the repressor molecule in the growth medium inhibits promoter function in the absence of inducers Thus addition of IPTG or lactose to the host cell growth medium enhances promoter function When the bacterial strain is Lac+, lactose may be used as an inducer instead of IPTG In both eukaryotic anid prokaryotic systems, termination regions can also contain sequences or structures which increase the stability of the m RNA species and allow for higher expression Thus, the transcriptional regulatory region may additionally include regulatory sequences which terminate transcription and which provide sequences or structures which inhibit degra- dation of the m RNA and thus increase the stability of the m RNA species and allow for higher expression.

Several examples of prokaryotic sequences are known, for example the Trp terminator, the gene 32 (T 4) terminator, or synthetic terminators which are similar irf sequence to gene 32.

In eukaryotic systems, transcriptional termi- nation, RNA cleavage and polyadenylation regions pro- vide for proper maturation for the m RNA transcripts and are necessary for efficient expression The native 3 ' untranslated region may suffice, but the polyadnylation signal from, for example, SV 40, particularly including a splice site which provides for more efficient expres- sion, could also be used Alternatively, the 3 'untranslated region derived from a gene highly expressed in a particular cell type (for example, Ig in myeloma cells) could be fused with the gene of inter- est Such 3 ' sequences could also be paired with 5 ' regulatory sequences from the same highly expressed gene and could even include coding regions in reading frame to generate fusion proteins as described below.

In terms of translational regulation, given the presence of m RNA, expression can be regulated by influencing the rate of initiation (ribosome binding to the m RNA), the rate of elongation (translocation of the ribosome across the m RNA), the rate of post-translation modifications and the stability of the gene product.

The rate of elongation is probably affected by codon usage; the use of codons for rare t RN As may reduce the translation rate It is therefore preferrable to use codons which frequently appear in genes normally expressed by the host cell to increase the translation rate.

In prokaryotes, downstream from the -35 and -10 regulatory regions is a consensus nucleotide sequence, generally AGGA, termed the Shine-Dalgarno sequence, which is believed to be involved in ribosomal binding Optimum ribosomal binding and initiation of translation can be achieved by using a ribosome binding site functional in the host cell from a highly expressed gene Evidence also points to the presence of nucleotide sequences surrounding the Shine-Dalgarno sequence and sequences within the coding region which can affect ribosome binding, possibly by the formation of structural motifs through which the ribosome recognizes the initiation site Altering nucleotide sequences of the coding region therefore can be used to achieve optimum binding and initiation of transla- tion The sequence of the first about 7 to 30 codons after the initiating codon ATG can also affect binding and expression Preferrably the leader sequence and the Shine-Dalgarno sequence are obtained from the same gene, or where they are obtained from different genes, the codons of the leader sequence can be modified using codon degeneracy to approximate the nucleotide sequence of the natural gene that follows the leader sequence, as described in co-pending U S Patent Application serial number 264,098, filed October 28, 1988, which disclosure is hereby incorporated by reference.

The position of the AGGA sequence with respect to the initiating ATG codon can influence expression.

Generally the Shine-Dalgarno sequence is located from about 5 to 9 nucleotides from the initiating codon, although, unexpectedly, high levels of expression can be achieved using expression cassettes wherein the Shine-Dalgarno sequence is located from about 10 to 13 nucleotides, preferrably 11 to 12 nucleotides from the initiating codon - Stability of the m RNA is governed by the susceptibility of the m RNA to ribonuclease enzymes In general, exonuclease digestion is inhibited by the presence of structural motifs at the ends of the m RNA; palindromic structures, altered nucleotides or specific nucleotides Endonuclease digestion is believed to occur at specific recognition sites within the m RNA and stable m RNA would lack these sites There is also some evidence that m RN As undergoing high levels of translation are also protected from degredation by the presence of ribosomes on the m RNA.

Stability of the expression product may be achieved by providing for synthesis of a fusion protein in which the desired polypeptide is expressed in con- junction with a second polypeptide or fragment thereof Preferrably, stability of the expression product is achieved by providing for synthesis of a fusion protein in which the polypeptide of interest is expressed joined to a leader sequence A DNA sequence encoding an N-terminal amino acid sequence from, for example, a highly expressed bacterial or bacteriophage gene such as the bacteriophage lambda-N protein gene, cro gene or B-galactosidase or a eukaryotic gene is joined upstream from and in reading frame with the gene of interest The leader sequence usually includes from about 8 to about 50, preferrably from about 15 to about 45, more preferably 18 to 25, N-terminal amino acids.

Expression of the polypeptide of interest as a fused protein with a leader sequence from another gene has several advantages in addition to providing for stability For example, the presence of the N-terminal amino acids provides a means for using general purification techniques for purification of any of a variety of polypeptides For example, the N-terminal amino acids of the N-protein are particularly antigenic, and thus specific antibodies raised against the N-terminal amino acids of the N-protein may be used as a means of purifying fusion proteins containing the the N-terminus of the N-protein Furthermore, the N- terminus of the N-protein has a high positive charge, which facilitates purification of the desired protein by ion-exchange chromatography and the like.

The leader sequence can also be a hydrophobic amino acid sequence, which may additionally function as a signal sequence for secretion A DNA sequence encoding the signal sequence is joined upstream from and in reading frame with the gene of interest.

Typically, the signal sequence includes a cleavage site which is recognized by a signal sequence peptidase.

Thus, positioning the polypeptide of interest directly after the signal sequence cleavage site will allow the polypeptide of interest to be specifically cleaved from the signal sequence and secreted as a mature polypep- tide Examples of hydrophobic amino acid sequences include the bacterial alkaline phosphatase signal sequence; the OMP-A-B-C-D-E or F signal sequences; the LPP signal sequence, B-lactamase signal sequence; and toxin signal sequences and mutants thereof For euka- ryotic cells, signal sequences may include the signal sequence from simian TGF-a, P 97 gene (human melanona antigen), K lactis killer toxin sequence, a-mating type factor; and other killer toxin signal sequences or the normal signal sequence associated with the gene of interest together with mutants of the signal sequences.

Other leader sequences which can be used include hydrophilic sequences, for example the N- terminal 41 amino acid residues from amphiregulin which may provide for modification of the function of the polypeptide of interest In addition, a cytotoxic agent such as a toxin A-chain fragment, ricin A-chain, snake venom growth arresting peptide, or a targeting molecule such as a hormone or antibody can be coupled covalently with the leader sequence with, in most cases, minimal effect on the biological activity of the gene product of interest As described for the other leader sequences, a DNA sequence encoding the leader sequence is joined upstream from and in reading frame with the gene of interest.

Where the leader sequence is not a signal sequence or does not contain a convenient natural cleavage site, additional amino acids may be inserted provide an enzymatic or chemical cleavage site for cleavage of the leader peptide, following purification of the fusion protein, to allow for subsequent purification of the mature polypeptide For example, introduction of acid-labile aspartyl-proline linkages between the two segments of the fusion protein facilitates their separation at low p H This method is not suitable if the desired polypeptide is acid- labile As another example, the fusion protein may be cleaved with, for example, cyanogen bromide, which is specific for the carboxy side of methionine residues.

Positioning a methionine between the leader sequence.

and the desired polypeptide allows for the release of the desired polypeptide This method is not suitable when the desired polypeptide contains methionine residues.

Where the leader sequence comprises a signal sequence, genes of interest with secretory leader sequences can be expressed with or without the leader sequence under conditions where the sequence may be retained or cleaved In addition, to obtain a high proportion of the desired polypeptide as a mature, cleaved and refolded peptide secreted into the medium, it is preferrable to use a promoter, such as the tac promoter, which can operate at a lower temperature, for example about 300 C Unexpectedly, higher levels of secretion can be obtained at the lower temperatures.

Extremely high expression levels can prevent full translational modifications of the protein to occur, resulting in aggregation and accumulation of uncleaved precursor (i e structural protein and secretory leader) Similarly, growth at elevated temperatures, for example 420 C, also tends to result in aggregation and accumulation of uncleaved precursor The preparation of fused proteins including methods of cleavage and refolding of the polypeptide of interest are further described in co-pending U S Application serial number 264,098 supra.

After each manipulation of the DNA in the development of the cassette, the plasmid will be cloned and isolated and as required, the particular cassette component analyzed as to its sequence to ensure that the proper sequence has been obzained Depending upon the nature of the manipulation, the desired sequence may be exised from the plasmid and introduced into a different vector or the plasmid may be restricted and the expression cassette component manipulated, as appropriate In some instances a shuttle vector will be employed where the vector is capable of replication in different hosts requiring different replication systems This may or may not require additional mar- kers which are functional in the two hosts Where such markers are required, these can be included in the vector The plasmid containing the cassette, two replication systems and the marker(s) may be transferred from one host to another as required For selection, any useful marker may be used Desirably, resistance to neomycin or tetracycline are of interest However, although a marker for selection is highly desirable for convenience, other procedures for screening transformed cells have been described See for example G Reipin, et al, Current Genetics, 1982, 189-193 Transformed cells may also be screened by the specific products they make, for example, synthesis of the desired product may be determined by immunological or enzymatic methods.

The expression cassette may be included within a replication system for episomal maintenance in an ap- propriate cellular host or may be provided without a replication system, where it may become integrated into the host genome The DNA may be introduced into the host in accordance with known techniques, such as trans- formation, using calcium phosphate-precipitated DNA, transfection by contacting the cells with the virus, microinjection of the DNA into cells or the like Once the gene of interest has been introduced into the appropriate host, the host may be grown to express the gene of interest.

A variety of host cells may be employed.

Examples of prokaryotic host cells include gram nega- tive organisms such as E Coli, for example, JM 109, JM 101, and JM 107; HB 101, DHI or DH 5 Particularly suitable are gram positive organisms such as B subtilis which have no periplasmic space and can directly secrete polypeptides into the growth medium.

Host eukaryotic cells can include yeast cells, insect cells and mammalian cells, for example, COS cells, CHO cells, monkey kidney cells and silkworm cells (sf 9).

The construct containing the gene of interest and flanking regions providing regulation of expression may be introduced into the expression host by any con- venient means, e g, transformation, with for example, calcium phosphate precipitated DNA, transfection, transduction, conjugation, microinjection, etc The host may then be grown to a high density in an appro- priate nutrient medium Where the promoter is inducible, permissive conditions will then be employed, e.g, temperature change, exhaustion or excess of a metabolic product or nutrient, or the like.

For example, where the regulatory sequence comprises the bacteriophage XPL promoter, the bacteriophage x OL operator, and CI 857 temperaturesensitive repressor, the host cells may be grown at the permissive temperature, generally about 300 C, at which temperature transcription from the PL promoter is repressed and the host cells may grow unhindered by the demands of the synthesis of the foreign gene product, which additionally may be toxic to the host organism.

When the host cells have reached an optimal density, the temperature may be increased to a non-permissive temperature, for example, about 421 C, at which time the CI repressor is rendered inactive, permitting transcription from the PL promoter Maximal secretion can obtained by using the lac promoter or a trp-lac promoter, and induction with a metabolic inducer such as lactose for a lac+ host strain and providing lac Iq on a vector Examples of host cells which could be used with such a system include DH 1, DH 5 or HB 101.

Where the product is retained in the host cell, the cells are harvested, lysed and the product isolated and purified by extraction, precipitation,chromatography, electrophoresis, etc Where the pro- duct is secreted, the nutrient medium may be collected and the product isolated by conventional ways, e g, affinity chromatography To produce an active protein it may be necessary to allow the protein to refold If the protein is expressed as a fusion protein with the leader sequence, the leader sequence may be removed by treatment with, for example, formic acid or cyanogen bromide The leader sequence preferably is removed after refolding of the protein.

The recombinant products may be glycosylated or non-glycosylated, having the wild-type or other glycosylation In general, the glycosylation will differ by not more than about 50 % usually by not more than about 20 % from the wild-type glycosylation The amount of glycosylation will depend in part upon the sequence of the particular peptide, as well as the organism in which it is produced Thus expression of the product in E coli cells will result in an ungly- cosylated product, and expression of the product in insect cells generally will result in less glycosyla- tion than expression of the product in mammalian cells.

Use of Growth Factors The subject compositions find a wide variety of applications in vitro and in vivo as agonists or antagonists for growth factors, such as epidermal growth factor (EGF), and transforming growth factor, particularly a-TGF A discussion of growth factors, used by themselves or in combination with other compo- sitions, particularly polypeptide compositions, for regulating the growth of cells, and other activities, may be found, for example, in Handbook of Experimental Pharmacology, Tissue Growth Factors, ed Baseraga, Vol.

57, Springer-Verlag, Berlin, 1981, Chapter 3, particu- larly pages 98-109; and Carpenter, Ann Rev Biochem.

( 1979) 48:193-216.

Human EGF appears to be identical to human urogastrone and exerts a variety of effects on prenatal and neonatal tissue growth Among the effects are pre- cocious eye-opening, wound healing, incisor eruption, and accelerated maturation of the lung EGF receptors are found in a wide variety of adult tissues EGF is found to stimulate phosphorylation of its own receptor.

EGF is also found to be related to increased bone resorption.

Transforming growth factor, particularly a-TGF, has many analogous activities to that of EGF.

TGF binds to the EGF receptor leading to phosphoryla- tion of the receptor, enhancement of its tyrosine- specific kinase activity and to stimulation of cell growth Cohen, in: Biological Response Mediators and Modulators (ed August, J T), Academic, New York, 1983, pp 7-12; Tam et al, Nature ( 1984) 309:376-378; Ibbotson et al, Science ( 1983) 221:1292-1294.

Myxoma and Shope fibroma viruses induce a significant increase in the proliferative potential of cells within the affected area While vaccinia induces a minor proliferative event at the site of infection, Shope fibroma virus induces a major proliferative event leading to tumor formation, which in adult rabbits is benign, but in immunosuppressed adults or in newborns is proliferative and invasive Myxoma virus induces a rapidly disseminating myxosarcoma tumor Although these proliferative events could be due to other viral factors, the viral growth factors appear to be highly implicated in the induced cellular proliferation caused by viral infection.

The subject compounds have particular applica- tion as drugs as agonists for EGF, and for wound heal- ing, such as epithelialization of wounds, such as burns, eye wounds, surgical incisions, and the like The active ingredient may be employed in a convenient vehicle, e g, Silvadene, in an amount sufficient to promote epithiliazation and/or wound healing, generally in amounts ranging from about 0 01 to 10 0,ig/ml, usually from about 0 075 to 05 Oig/ml EFG equivalents prefer- ably 0 5 to 5 pg/ml The formulation is spread over the wound, so as to provide a complete coating of the wound with the formulation Treatments may be as frequent as four times a day or as infrequent as every other day or less, depending upon the nature of the wound, its response to the treatment, the concentration of the active ingredient, and the like.

The subject compositions can be used as re- agents in diagnostic assays or for the preparation of reagents, such as polyclonal or monoclonal antibodies.

As reagents, they may be used for the detection of analogous growth factors or for the detection of anti- bodies to the growth factors in physiological fluids, such as blood.

Depending upon the particular protocol and the purpose of the reagent, the polypeptide may be labeled or unlabeled A -wide variety of labels have been used which provide for, directly or indirectly, a detectable signal These labels include radionuclides, enzymes, fluorescers, particles, chemiluminescers, enzyme sub- strates or cofactors, enzyme inhibitors, magnetic particles, etc See for example, U S Patent Nos.

3,654,090, 3,817,837, 3,935,074, 3,996,345, 4,277,437, 4,374,925, and 4,366,241.

A wide variety of methods exist for linking the labels to the polypeptides, which may involve use of the N-terminal amino group for functionalization to form a pyrazolone, while other free amino groups are protected, where the pyrazolone may then be contacted with various reagents, e g, amino groups, to link to the detectable signal generating moiety By protecting the arginine amino acids associated with the third loop or proximal thereto, other arginines may be function- alized for conjugation to amino groups or thio groups in accordance with known ways Alternatively, the polypeptide may be contacted with an active agent, e.g, an activated carboxylic acid and randomly sub- stituted, where biologically active material may be separated from biologically inactivated material as a result of the random substitution Finally, depending upon the method of synthesis, the polypeptide may be modified to provide for the desired functionality as part of the synthetic procedure.

The subject compositions can also be used for monitoring EGF receptors The subject compositions can also be used for monitoring cellular response to EGF and/or TGF by providing for competition between these naturally occurring materials and a composition accord- ing to the subject invention In this way, changes in the receptor conformation can be monitored.

The subject compositions can also be used for various therapeutic purposes involving growth stimu- lation or control of bone formation These compounds may be administered in appropriate physiological car- riers intraperitoneally, subcutaneously, intravenously, intraarterially, or by application to the site of in- terest In addition, the subject compositions can be introduced into liposomes, which may or may not involve the use of antibodies for site direction Various car- riers include phosphate buffered saline, saline, water, or the like The concentration of the additive will vary widely, depending upon its ultimate use and activity.

Other additives may also be included in the formulations, such as EGF, TGF, other growth factors, bacteriocides e g, antibiotics, bacteriostats, buf- fers, etc.

For preparing antibodies, the subject poly- peptides where pp 1-4 are hydrogen or short oligopeptide chains (fewer than five amino acids), may be joined to antigenic polypeptides or proteins, for injection into mammalian hosts The antigenic protein will have at least about 60 amino acids and will usually be not more than 100 kilodaltons (k Dal) Numerous techniques exist for joining to polypeptides, either at a specific site or randomly, using bifunctional reagents, e g, p-male- imidobenzoic acid, glutaraldehyde, p,p'-benzidine, etc.

Common antigenic proteins include bovine serum albumin, keyhole limpet hemocyanin, tetanus toxoid, etc The subject polypeptides are joined to the antigenic pro- tein in sufficient number to provide the desired immunogenic response Usually there will be two or more booster injections after the initial injection.

For antisera, blood is removed from the immunized host and the immunoglobulin fraction isolated For mono- clonal antibodies, the spleen is isolated and spleno- cytes fused with an appropriate fusion partner in accordance with conventional ways The resulting hy- bridomas are then screened for antibodies binding to the epitopic sites of the subject polypeptide These antibodies may be-used for a variety of purposes, such as diagnostic reagents, therapy, etc The antibodies when used as reagents may be labeled or unlabeled, as described for the polypeptides.

The following examples are offered by way of illustration and not by way of limitation.

EXPERIMENTAL Table of Contents Table of Contents Example I:

Example II:

Preparation of Synthetic Genes of Interest A TGF Synthetic Oligonucleotides B VGF Synthetic Oligonucleotides C EGF Synthetic Oligonucleotides Description of Cloning a

Expression Plasmids A Plasmid p LE Bam B Plasmid p B Mll C Plasmid p B Mll M 4 D Plasmid p B Mll M 5 E Plasmid p B Mll/NDP F Plasmid p B Mll/PAD G Plasmid p B Mll/PAK H Plasmid p TC Pt I Plasmid plac/cro-Bgal J Plasmid ptac/cro-Bgal K Plasmid p RSV L Plasmid p Ac 610 M Plasmids p Toxl and p Tox 2 N Tak Pak Example III:

Assembly of Growth Factor Genes for Expression in Prokarvotic Cells A Preparation of Plasmid p LE Bam/TVV B Preparation of Plasmid p LE Bam/TWVV VGF and Fragments Thereof Example IV:

Example V:

Expression of the Polypeptides of Interest as a Fusion Protein with the N-Protein A Modified Synthetic TGF B Modified Synthetic TGF-VGF Hybrid Preparation of the Polypeptide of Interest as a Fusion Protein with the N-Protein and a Cleavage Site A Modified Synthetic VGF B Modified Synthetic TGF-VGF Hyabrids C Synthetic EGF Example VI:

Preparation of the Polypeptide of Interest as a Fusion Protein with the Modified Alkaline Phosphatase Signal Sequence A Preparation of p B Mll/PAD/EGF B Preparation of p B Mll/PAD/n VG Fa Example VII:

Expression of the Polypeptide of Interest as a Fusion Protein with an Alkaline Phosphatase Signal A Preparation of p B Mll/PAK/n VG Fa (Alkalaine phosphatase signal sequence/n VG Fa with natural VGF Nterminus and sequences GTC and GYACRC) B Preparation of p B Mll/PAK/EGF C Preparation of Tac Pak/EGF (alkaline phosphatase signal sequence/human EGF) Example VIII:

Expression of a Polypeptide of Interest as a Fusion Protein with the Alkaline Phosphatase Signal Sequence Using an Expression Cassette Comprising a Transcriptional Termination Region A Preparation of p TC Pt/EGF (ltrp- 35116 bpllac-10 lllac SDlllbplATCl/ alkaline phosphatase signal/human EGF/trans term -NEO B Preparation of p TC Pt/n VG Fa (ltrp35116 bpllac-10 l llac SDllcro SDl llbplATGl/alkaline phosphatase signal/n-terminal VG Fa with sequence GTC and GYACRC)/trans term -NEO) C Preparation of p TN Pt/EGF (ltrp- 35117 bpllac-10 lln SDl 8 bplATGl/alkaline phosphatase signal/human EGF/trans.

term -NEO) Example IX:

Assembly of Growth Factor Genes for Expression in Eukaryotic Cells A Preparation of Plasmid PRSV/VGF B Preparation of Plasmid p Ac/VGF C Preparation of p Ac/SFGF Example X:

Preparation of Polypeptides A Solid-phase Synthesis of BGF and TGF B Isolation of Recombinant Polypeptides Prepared in Prokaryotic Cells C Isolation of VGF and SFGF Produced by Eukaryotic Expression of the Natural Genes D Natural EGF Example XI:

Activity Assays A Mitogenic Assay B Soft Agar Colony Growth Stimulation Assay C EGF Receptor Binding Inhibition Assay D Radioimmunoassay E Wound Healing Example XII:

Example XIII:

Biological Activity of Recombinant Growth Factors Prepared in Procaryotic Cells A EGF R Eceptor Binding B Mitogenic Activity C Wound Healing Biological Activity of VGF Prepared in Eukaryotic Cells A VGF Prepared in Monkey Kidney (BSC-1) Cells B VGF Prepared in CHO Cells C VGF Prepared in Silkworm D Immunological Comparison of VGF and TGF Biological Deposits The following expression plasmids, all transformed into E Coli HB 101, were deposited on the indicated date with the American Type Culture Collection, 12301 Parklawn Drive, Rockville, MD 20852 and have their identification and ATC designations given below:

Identification ATCC Designation Date of Deposit PB Mll 67366 March 25, 1987 PBM 14 67367 March 25, 1987 PB M 11/PA/VGF 67417 June 3, 1987 PB Mll/DP/VG Fa 67418 June 3, 1987 PBM 11/PA/EGF 67419 June 3, 1987 PB Mll/M 5 67436 June 16, 1987 PBM 11/C 2 67437 June 16, 1987 PBM 11/NDP/EGF 67547 October 23, 1987 Methods General cloning techniques were used as described in Maniatis et al, 1982, "Molecular Cloning: A Laboratory Manual", Cold Spring Harbor Laboratory, CSH, New York All DNA-modifying enzymes were obtained from commercial suppliers They were used according to the manufacturer's instructions.

Materials and apparatus for DNA purification and separation were used according to instructions from the supplier.

Example I

Preparation of Synthetic Genes of Interest Synthetic growth factor genes were designed which use host cell codons optimized for high levels of expression In addition, several convenient restriction sites were designed into the synthetic genes When possible, the new restriction sites left the amino acid sequence of the growth factor gene unaltered, however, in some cases incorporation of the new restriction site yielded an altered amino acid sequence These sites roughly divide the synthetic genes into thirds yielding N-terminal, middle and C-terminal domains.

The natural VGF gene product contains an extreme N-terminal domain which has no counterpart in mature TGF VGF fragments lacking this domain are referred to as truncated The restriction sites were used for initial construction of the final genes from partial synthetic oligonucleotide fragments extending from one restriction site to another The oligonucleotides were synthesized on an Applied Biosystems oligo- nucleotide synthesizer and were purified on an acryl- amide gel The oligonucleotides were phosphorylated at the 5 ' end using T 4 polynucleotide kinase and each oligonucleotide was then annealed to its complement.

A TGF Synthetic Oligonucleotides 1 Human TGF N-terminal domain:

TGF -> Bss HII Nco I M V V S H F N D C P D S H T Q F C ' CGCGCCATGGTTGTTTCTCACTTTAACGACTGCCCGGACTCTCATACTCAGTTTTGC 3 ' GGTACCAACAAAGAGTGAAATTGCTGACGGGCCTGAGAGTATGAGTCAAAACG Kpn I F H G T TTTCATGGTAC 3 ' TGF 104 AAAGTAC 5 ' TGF 103 2 Modified human TGF middle domain with the human sequence QEDK being altered to QEEK, the sequence found in rat TGF:

Kpn I Sph I C R F L V Q E E K P A C ' CTGCCGTTTTCTGGTTCAGGAAGAAAAACCGGCATG 3 ' TGF 101 3 ' CATGGACGGCAAAAGACCAAGTCCTTCTTTTTGGCC 5 ' TGF 102 3 Human TGF C-terminal domain:

Sph I V C H S G Y V G A R C E H A D L L ' CGTTTGCCATTCTGGCTACGTTGGCGCACGTTGCGAACACGCTGACCTGCTG 3 ' GTACGCAAACGGTAAGACCGATGCAACCGCGTGCAACGCTTGTGCGACTGGACGAC Bam HI A Ter GCTTAAG 3 ' TGF 205 CGAATTCCTAG 5 ' TGF 206 B VGF Synthetic Oligonucleotides 1 VGF extreme N-terminal domain:

Hind III E D S G N A I E T T S P E I T N A T T 5 ' AGCTGACTCTGGTAACGCTATCGAAACTACTTCTCCGGAAATCACTAACGCTACTACT 3 ' V 31 CTGAGACCATTGCGATAGCTTTGATGAAGAGGCCTTTAGTGATTGCGATGATGA 5 ' V 2 Modified VGF N-terminal domain including Asp-Pro cleavage site, with the sequence HGT replacing the natural sequence HGD:

Bam HI I D P M D I P A I R L C G P E G D G Y ' GATCGATCCCATGGACATCCCGGCTATCCGTCTGTGCGGCCCGGAAGGCGACGGCTAC 3 ' CTAGGGTACCTGTAGGGCCGATAGGCAGACACGCCGGGCCTTCCGCTGCCGATG Kpn I C L H G T TGCCTGCATGGTAC 3 ' VGF 104 a ACGGACGTAC 5 ' VGF 103 a 3 Modified VGF middle domain having the sequence GYAC replacing the natural sequence GMYC:

Kpn I Sph I T C I H A R D I D G Y A C ' CTGCATCCATGCACGTGACATCGACGGCTACGCATG 3 ' VGF 10 la 3 ' CATGGACGTAGGTACGTGCACTGTAGCTGCCGATGC 5 ' VGF 102 a 4 VGF C-terminal domain, 5 ' end:

Sph I Eco RI C R C S H G Y T G ' CCGTTGCTCTCATGGCTACACTGG 3 ' VGF 1 A 3 ' GTACGGCAACGAGAGTACCGATGTGACCTTAA 5 ' VGF 2 A Modified VGF C-terminal domain, 5 ' end, with the sequence VCS replacing the natural sequence RCS:

Sph I Eco RI V C S H G Y T G ' CGTTTGCTCTCATGGCTACACTGG 3 ' VGF 1 3 ' GTACGCAAACGAGAGTACCGATGTGACCTTAA 5 ' VGF 2 6 VGF C-terminal domain, 3 ' fragment, end- ing at YQR instead of PNT, the deduced C-terminus of natural secreted VGF:

Eco RI Bam HI I R C Q H V V L V D Y Q R Ter ' AATTCGTTGCCAGCATGTTGTTCTGGTCGACTACCAGCGTTAAGGATC 3 ' VGF 3 3 ' GCAACGGTCGTACAACAAGACCAGCTGATGGTCGCAATTC 5 ' VGF 4 C EGF Synthetic Oligonucleotides Three sets of overlapping synthetic oligo- nucleotides l(A,B), 2 (A,B) and 3 (A,B) coding for human EGF were synthesized on an Applied Biosystems oligo- nucleotide synthesizer and purified on an acrylamide gel The oligonucleotides were phosphorylated at the ' end using T 4 polynucleotide kinase Each oligo- nucleotide was annealed to its complement.

1.

Nco I Eco RI M N S D S E C P L S H D G Y ' CATGAATTCTGACTCTGAATGCCCGCTGTCTCATGACGGCTAC 3 ' EGF 1 A 3 ' TTAAGACTGAGACTTACGGGCGACAGAGTACTGCCGATGACGGAC 5 ' EGF 2 A 2.

Nsi I Sph I C L H D G V C M Y I E A L D K Y A C ' TGCCTGCATGACGGCGTATGCATGTACATCGAAGCTCTGGACAAGTACGCATG 3 ' EGF 1 B 3 ' GTACTGCCGCATACGTACATGTAGCTTCGAGACCTGTTCATGC 5 ' EGF 2 B 3.

Sph I N C V V G Y I G E R C Q Y R D L K ' CAACTGCGTTGTTGGCTACATCGGCGAACGTTGCCAGTACCGTGACCTGAAA 3 ' GTACGTTGACGCAACAACCGATGTAGCCGCTTGCAACGGTCATGGCACTGGACTTT Bam HI W W E L R TGGTGGGAACTGCGTTAAG 3 ' EGF 3 ACCACCCTTGACGCAATTCCTAG 5 ' EGF 4 Example II

Description of Cloning and Expression Plasmids

A Plasmid p LE Bam was used to clone synthetic oligo- nucleotide fragments because of its convenient Bss HII and Bam HI restriction sites A plasmid with Nco I and Bam HI restriction sites such as p B Mll or p B Mll/NDP (de- scribed below) can be used for cloning the synthetic nucleotide fragments and other DNA sequences of interest.

B Plasmid p B Mll, described in copending U S Applica- tion Serial No 264,098, filed October 28, 1988, allows cloning of a gene of interest downstream of the DNA sequences coding for the 33 N-terminal amino acids of the bacteriophage I N-gene at a Bam HI restriction site Upon induction of the X PL promoter by inactiva- tion of the C 1857 temperature-sensitive repressor at 42 C, the foreign gene product is expressed as the Cterminal part of a fusion protein whose N-terminal sequence is that of the N-gene.

C Plasmid p B Mll M 4, described in copending U S Appli- cation Serial No 264,098, filed October 28, 1988, is derived from p B Mll and allows a gene of interest to be cloned at a Bam HI restriction site directly after the initiating methionine of the N-gene Plasmid p B Mll/M 4 also contains an Nco I site in the neomycin gene.

D Plasmid p B Mll M 5, described in copending U S Appli- cation Serial No 264,098, filed October 28, 1988, is derived from p B Mll in which an Nco I site present in the neomycin resistance gene has been removed by sitedirected mutagenesis Cloning of a gene of interest into p Bmll/M 5 therefore does not require partial digestion of the vector with Nco I.

E Plasmid p B Mll/NDP, described in copending U S.

Application Serial No 264,098, filed October 28, 1988, is derived from p B Mll and has DNA sequences coding for an acid labile aspartic acid-proline dipeptide inserted between the sequences coding for the N-gene and a gene of interest The plasmid contains an Nco I site and a Cla I site for cloning a gene of interest downstream of the PL promoter.

F Plasmid p B Mll/PAD, described in copending U S.

Application Serial No 264,098, filed October 28, 1988, is derived from plasmid p B Mll M 4 and allows a gene of interest to be cloned at a Hind III, Sma I or Bam HI downstream from a modified alkaline phosphatase signal sequence.

G Plasmid p B Mll/PAK, described in copending U S.

Application Serial No 264,098, filed October 28, 1988, is derived from plasmid p B Mll M 4 and allows a gene of interest to be cloned at a Hind III, Sma I or Bam HI downstream from an alkaline phosphatase signal sequence.

H Plasmid p TC Pt, is designed to have the tac promoter elements and utilize the cro SD to express the gene of interest behind the alkaline phosphatase signal sequence An example of the construction of this plasmid is given below in the construction of p TC Pt/EGF.

Construction of p TN Pt (ltrp-35 l 17 bpllac- 10 lln SDl 8 bplATGl/alkaline phosphatase signal/linker/trans.

term -NEO) This plasmid is designed to have the tac promoter elements and utilize the N-gene SD to express a given gene behind the alkaline phosphatase signal sequence It has a p BR 322 background with the Neomycin resistance gene Plasmid p TN Pt was constructed as follows:

(a) Preparation of the 2 8 kb Eco RI-Bam HI fragment of p BM 16 t/VG Fa lacking the Hind III site Plasmid p BM 16 t/VG Fa was digested with Eco RI and Bam HI and the 2 8 kb fragment was isolated The 2.8 kb fragment was ligated to an Eco RI-Bam HI linker and a correct construct was isolated by restriction analysis and is referred to as Intermediate I.

The unique Hind III site near the Neomycin resistance gene was removed from the Intermediate I plasmid by digestion with Hind III, creating blunt ends using Klenow fragment, and religating This resulted in Intermediate II plasmid which lacked the Hind III site.

The 2 8 kb Eco RI-Bam HI fragment of p BM 16 t/VG Fa lacking the Hind III site was isolated by digesting p Intermediate II with Eco RI and Bam HI The resulting 2.8 kb fragment was isolated by agarose gel electrophoresis.

(b) Preparation of the 150 bp Bam HI-Bsm I fragment of p B Mll/PAK Plasmid p B Mll/PAK is identical to p B Mll/PAK/EGF except that it contains a linker region with Hind III, Sma I and Bam HI sites downstream of the alkaline phosphatase signal sequence instead of the EGF gene p B Mll/PAK was digested with Bsm I and Bam HI and the 150 bp fragment containing the N-gene SD, the alkaline phosphatase signal sequence and the linker region was isolated.

(c) Preparation of oligonucleotides Tac A+ and Tac A- Oligonucleotides Tac A+ and Tac A were synthesized on an Applied Biosystems Oligonucleotide synthesizer and were designed to have an Eco RI overhang at the 5 ' end with the trp-35 consensus sequence separated from the lac-10 consensus sequence by 17 nucleotides within which was positioned a Sst I site.

The sequence also contained the 5 ' end of the lac m RNA, the lac repressor binding site and a Bsm I overhang.

Tac A+ 5 'AATTACTCCCCATCCCCCTGTTGACAATTAATCATCGAGCTC GTATAATGTGTGGAATTGTGAGCGGATAACAATTTCACACAG 3 ' Tac A 5 'GTGTGAAATTGTTATCCGCTCACAATTCCACACATTATA CGAGCTCGATGATTAATTGTCAACAGGGGGATGGGGAGT 3 ' (d) Ligation and Isolation of p TN Pt The 2 8 kb Eco RI-Bam HI fragment, the 150 bp Bsm I-Bam HI fragment and oligonucleotides Tac A+ and Tac A were ligated together using DNA ligase and the DNA was used to transform competent JM 109 (lac Iq) A correct construct was isolated by restriction analysis and DNA sequencing.

(Eco RI site of p B Mll) I GAATTACTCCCCATCC Sst I trp-35 ( 17 bp) lac-10 CCCTG lTTGACAl ATTAATCATCGAGCTCG (TATAATG) W Bsm I 'lac m RNA-> N m RNA-> TGTGG/AATTGTGTGAGCGGATAACAATTTCACACAGCATTCAAAGCAGAAGGCT TTGGGGTGTGTGATACGAAACGAAGCATTGGCCGTAAGTGCGATTCCGGATTAGC TGCCAATGTGCCAATCGCGGGGGGTTTTCGTTCAGGACTACAACTGCCACACACC Pvu I m SD ( 8 bp) Signal Sequence -> ACCAAAGCTAACTGAC {AGGA} GAATCCAG ATGAAACAATCTACGATCGCCC M K Q S T I A L Sma I Hind III Bam HI TCGCACTTCTCCCACTGCTGTTCACTCCAGTGACAAAAGCTTCCCGGGATCCGTG A L L P L L F T P V T K (Bam HI site of p B Mll) Trans Term I ACTAATTGGGGACCCTAGAGGTCCCCTTTTTTATTTTAAAACGATCC I Plasmid plac/cro-B gal consists of the operator- promoter region of E Coli lactose (lac) operon, as well as the ribosomale binding sites of lac and cro.

The plasmid was constructed as described in copending U.S Application Serial No 264,098 filed October 28, 1988 Fusion proteins expressed by this vector consist of the N-terminus of bacteriophage x Cro protein, the amino acid sequence encoded by the inserted DNA, and the C-terminus of B-galactosidase The controlling elements of this vector consist of the operator- promoter region of the E coli lactose (lac) operon, as well as the ribosomal binding sites of lac and cro.

J Plasmid ptac/cro-B-gal allows a gene of interest to be cloned downstream of the N-terminal 21 amino acids of the bacterial Cro protein The plasmid was constructed as described in copending U A Application Serial No 264,098 filed October 28, 1988 Expression vector ptac/cro-B-gal is similar to plac/cro-5-gal, with the exception that the promoter of ptac/cro-6-gal consists of the -35 region from the promoter of the tryptophan operon and the Pribnow box of the lac operon This hybrid promoter allows a higher level of expression than plac/cro-B-gal.

K Plasmid p RSV allows expression of a gene of interest in eukaryotic cells with the RSV LTR promoter, Gorman et al, Proc Natl Acad Sci USA ( 1982) 79:6777-6781.

L Plasmid p Ac 610, Smith et al, Molec Cell Biol.

( 1983) 12: 2156-2165, allows expression of a foreign gene in insect cells.

M Plasmids p Toxl and p Tox 2 allow expression of a gene of interest in yeast cells using the X lactis killer toxin signal sequence (EMBO 6, 229-234 ( 1987); Biochem, Biophys Res Comm 144, 613-619 ( 1987) This construct has the URA 3 gene for selection in yeast and the AMP gene for selection in bacteria Transcription initiates witht he CYC 1 promoter and GAL "upstream activation sequence" (Methods Enzymol, 101, 181-191 ( 1983) and terminates within 3 ' end of the FLP gene (Mol Cell Biol, 5, 2770-2780 ( 1985) The construct has both the 2 micron origin and the p BR 322 origin for replication in yeast and E coli, respectively The killer toxin N-terminal region with the signal sequence and the Kex 2 cleavage site was introduced on ligonucle- otides and contained the A-T rich region immediately preceding the initiator codon to maintain optimum translation initiation and elongation signals Several restiction sites were engineered to allow insertion of genes of interest downstream of both the signal cleavage site and the Kex 2 cleavage site.

1 Preparation of the 3 kb Eco RI-Esp I fragment of p BR 322 containing the origin and the AMP resistance gene.

Plasmid p LGSD 5 (-ATG), a yeast expression vector (Methods Enzymol 101:181-191 ( 1983), was digested with Eco RI and Esp I and the 3 O kb fragment was isolated.

2 Preparation of the 2 1 kb Eco RI-Esp I fragment of the 2 micron plasmid containing the 3 ' end of the FLP gene and the origin.

Plasmid p LGSD 5 (-ATG) was digested with Eco RI and Esp I and the 2 1 kb fragment was isolated.

3 Ligation and Isolation of Intermediatel plasmid.

The 2 1 kb and 3 kb Eco RI-Esp I fragments were ligated together using DNA ligase and the DNA was used to transform competent HB 101 A correct construct was isolated by restriction analysis.

4 Preparation of the 1 8 kb Eco RI-Bam HI fragment of p LGSD 5 (-ATG) containing the URA 3 gene, the GAL UAS and the CYC 1 promoter.

The plasmid p LGSD 5 (-ATG) was digested with Eco RI and Bam HI and the 1 8 kb fragment was isolated.

The Bam HI site is present within a linker inserted 4 bp upstream of the CYC 1 initiator.

Preparation of Oligonucleotides p Toxl A+, 2 A-, l B+, 2 B-.

Oligonucleotides were designed to link the codons for the signal sequence of K lactis killer toxin to the Bam HI site downstream of the CYC 1 initiator To conserve optimal translation initiation, 13 bp of the A- T rich sequence upstream of the killer toxin initiator codon wer included This sequence conforms to the consensus sequence determined for yeast initiating codons A Bgl II site was place directly upstream of the A-T region to allow for mutatgenesis of the initiating region and the signal sequence A Hind III site was placed 10 nucleotides upstream of the signal cleavage site to allow for mutagenesis of downstream sequences and ligation of a gene of interest In p Toxl, an Ava I,Xho I site was located directly downstream of the signal cleavage site, while in p Tox 2, a Nae I site was placed at the signal cleavage site for blunt end cloning directly over the signal cleavage which was changed from Gly-Leu to Ala-GlysResolution of the heteroduplex at the cleavage site should result in generation of clones containing the p Toxl sequence and those containing the p Tox 2 sequence In both constructs, the subsequent sequences code for the rest of the killer toxin precursor sequence down to the Kex 2 Lys-Arg cleavage site, at which point several restric- tion sites (Stu I, Sal I, Acc I, Hinc II, and Bam HI were placed to facilitate cloning of a gene of interest.

The Stu I site allows blunt ended cloning directly after the Arg codon of the Kex 2 cleavage site The oligonucleotides have a Bam HI overhang at the 5 ' end and a Hind III overhang at the 3 ' end to allow for cloning into the intermediatel vector The resulting sequence no longer has either of the two sites.

(Bam HI) Bgl II Hind III M N I F Y I F L F L L S 1 A+ 5 'GATCAGATCTAATAATTATAAAATGAATATATTTTACATATTTTTGTTTTTGCTAAGC 2 A 3 ' TCTAGATTATTAATATTTTACTTATATAAAATGTATAAAAACAAAAACGATTCGAAG Signal cleavage site Kex 2 cleavage site (p Toxl) / Ava I,Xho I / Stu I Sal I,Acci Bam H F V Q G L E H T H R R G S L V K R P L S T D P l B+ 5 'TTCGTTCAGGGCCTCGAGCATACTCATAGAAGAGGCTCCTTAGTCAAAAGGCCTTTGTCGACGGATCC 2 B 3 ' AAGTCCGGCCGCTCGTATGAGTATCTTCTCCGAGGAATCAGTTTTCCGGAAACAGCTGCCTAGGTCG A G (p Tox 2) Nae I These oligonucleotides were synthesized on an Applied Biosystems Synthesizer and were purified by gel electrophoresis Subsequently, they were phosphory- lated using T 4 kinase and the complementary pairs were annealed together, ligated and gel-purified.

6 Preparation of Eco RI-Hind IIT Intermediatel vector.

Intermediatel vector was digested with Eco RI and Hin III and then was treated with calf alkaline phosphatase This treatment removed a small Eco RI- Hind III fragment and left a Hind III site upstream of the FLP 3 ' sequence.

7 Ligation and Isolation of p Toxl and p Tox 2.

The 1 8 kb Eco RI-Bam HI fragment was ligated to the oligonucleotide fragment using DNA ligase and the resulting fragment was gel purified and was subsequently ligated to the Eco RI-Hind III Intermediatel vector and the DNA was used to transform competent HB 101 Correct constructs were identified using restriction analysis and DNA sequencing and a clone having the sequence of p Toxl and one having the sequence of p Tox 2 were isolated Both clones lacked the Bam HI site at the junction of the oligonucleotides and the intermediatel vector.

Example III

Assembly of Growth Factor Genes for Expression In Prokaryotic Cells A Preparation of Plasmid p LE Bam/TTV The synthetic chimeric growth factor, denoted TTV or (TGF/TGF/VGF) was assembled in the cloning vec- tor p LE Bam This hybrid growth factor contained the amino acid sequence of human TGF in the amino terminal two-thirds of the gene with the exception of the sequence QEEK which was altered from the natural human sequence QEDK The carboxy terminus was derived from the amino acid sequence of VGF and terminated with the sequence YQR upstream of the natural sequence PNT.

Plasmid p LE Bam was digested with Bss HII and Bam HI.

Bss HII-Bam HI p LE Bam was then ligated to oligonucleo- tides TGF 101, 102, 103, and 104, and VGF 1, 2, 3, and 4 usirg DNA ligase and the resulting plasmids used to transform competent HB 101 The transformants were selected on ampicillin and screened by restriction analysis using Eco RI, Nco I and Bam HI and by nucleotide sequencing using the Maxam-Gilbert protocol A correct construction was isolated and denoted p LE Bam/TTV.

TGF Nco I CCATGGTTGTTTCTCACTTTAACGACTGCCCGGACTCTCATACTCAGTTTTGCTT M V V S H F N D C P D S H T Q F C F VGF - Kpn I Sph I TCATGGTACCTGCCGTTTTCTGGTTCAGGAAGAAAAACCGGCATGCGTTTGCTCT H G T C R F L V Q E E K P A C V C S Eco RI CATGGCTACACTGGAATTCGTTGCCAGCATGTTGTTCTGGTCGACTACCAGCGT H G Y T G I R C Q H V V L V D Y Q R Bam HI TAAGGATCC Ter B Preparation of Plasmid p LE Bam/TWVV The synthetic chimeric growth factor, denoted TVV or (TGF/VGF/VGF) was assembled in the cloning vec- tor p LE Bam This hybrid growth factor contained the amino acid sequence of human TGF in the N-terminal do- main of the gene The middle and C-terminal domains are derived from the truncated VGF sequence and end with the sequence YQR upstream of the natural sequence PNT In addition, the synthetic gene has the modifica- tion, GYACVC for GMYCRC.

(a) Preparation of a 4 3 kb Kpn I-Sph I fragment of p LE Bam/TTV Plasmid p LE Bam/TTV was digested with Kpn I and Sph I and the 4 3 kb Kpn I-Sph I fragment was gel purified This digestion removes the middle TGF domain from the synthetic gene TTV in the cloning plasmid p LE Bam.

(b) Ligation and isolation of p LE Bam/TVV Oligonucleotides VG Fl Ola and 102 a were ligated to the 4 3 kb Kpn I-Sph I fragment of p LE Bam/TTV using DNA ligase and the resulting mixture was used to transform competent HB 101 The transformants were selected on ampicillin and were screened by nucleotide sequencing using the Sanger-dideoxy method A correct construct was isolated and denoted p LE Bam/TVV.

TGF - Nco I CCATGGTTGTTTCTCACTTTAACGACTGCCCGGACTCTCATACTCAGTTTTGCTT M V V S H F N D C P D S H T Q F C F VGF Kpn I Sph I TCATGGTACCTGCATCCATGCACGTGACATCGACGGCTACGCATGCGTTTGCTCT H G T C I H A R D I D G Y A C V C S Eco RI CATGGCTACACTGGAATTCGTTGCCAGCATGTTGTTCTGCTCGACTACCAGCGT H G Y T G I -R C Q H V V L V D Y Q R Bam HI TAAGGATCC Ter Example IV.

Expression of the Polypeptide of Interest as a Fusion Protein with the N-Protein A Modified Synthetic TGF l.Preparation of p B Mll/N/TGF The modified human TGF was expressed in this system as part of a fusion with the 33 N-terminal amino acids of the N-gene and has the sequence QEEK replacing the human sequence QEDK.

(a) Preparation of a 780 bp Sphl-Pvu I fragment of p B Mll/N/TTV Plasmid p B Mll/N/TTV was digested with Sph I and Pvu I and the 780 bp Sph I-Pvu I fragment was gel puri- fied This fragment contains part of the p B Mll plasmid at the Pvu I end and at the Sph I end, the N-gene and N- terminal two-thirds of the human TGF gene.

(b) Preparation of the 5 kb Bam HI-Pvu I fragment of p B Mll/N/TTV Plasmid p B Mll/N/TTV was digested with Bam HI and Pvu I and the 5 kb Bam HI-Pvu I fragment was gel purified.

(c) Ligation and isolation of p B Mll/N/TGF:

Oligonucleotides TGF 205 and 206, the 780 bp Sph I-Pvu I fragment and the 5 kb Bam HI-Pvu I fragment of p B Mll/N/TTV were ligated together and used to transform competent HB 101 The transformants were selected on neomycin and were screened by restriction analysis using Eco RI and nucleotide sequencing following the Sanger-dideoxy method A correct construction was isolated and denoted p B Mll/N/TGF.

* N-gene ATGGATGCACAAACACGCCGCCGCGAACGTCGCGCAGAGAAACAGGCTCAATGGA M D A Q T R R R E R R A E K Q A Q W K Bam HI AAGCAGCAAATCCCCTGTTGGTTGGGGTAAGCGCAAAACCAGTTCGGATCCGCAT A A N P L L V G V S A K P V R I R M TGF GGTTGTTTCTCACTTTAACGACTGCCCGGACTCTCATACTCAGTTTTGCTTTCAT V V S H F N D C P D S H T Q F C F H Kpn I Sph I GGTACCTGCCGTTTTCTGGTTCAGGAAGAAAAACCGGCATGCGTTTGCCATTCTG G T C R F L V Q E E K P A C V C H S G Bam HI GCTAGGTTGGCGCACGTTGCGAACACGCTGACCTGCTGGCTTAAGGATCC Y V G A R C E H A D L L A Ter B Modified Synthetic TGF-VGF Hybrid 1 Preparation of p B Mll/N/TTV In this construct, a synthetic modified TTV chimeric gene was expressed as the C-terminal portion of a fusion protein having the first 33 amino acids of the N-gene at the N-terminus This hybrid growth fac- tor contained the amino acid sequence of human TGF in the amino terminal two-thirds of the gene with the exception of the sequence QEEK which was altered from the natural human sequence QEDK The carboxy terminus was derived from the amino acid sequence of VGF and terminated with the sequence YQR upstream of the na- tural sequence PNT.

(a) Preparation of Nco I(blunt)-Bam HI TTV synthetic gene Plasmid p LE Bam/TTV was digested with Nco I and the ends were made blunt by filling in the overhangs using the Klenow fragment of DNA polymerase The DNA was then digested with Bam HI and the 170 bp Nco I(blunt)- Bam HI TTV fragment was gel purified.

(b) Preparation of Bam HI digested p B Mll Plasmid p B Mll was digested with Bam HI.

(c) Ligation and isolation of p B Mll/N/TTV Bam HI digested p B Mll, the Nco I(blunt)-Bam HI TTV fragment, and Bam HI linkers ( 5 'GATCCG 3 ') were ligated together using DNA ligase and the resulting mixture was used to transform competent HB 101 The transformants were selected on neomycin and were screened using restriction analysis and nucleotide sequencing using the Sanger-dideoxy method A correct construct was isolated and denoted p B Mll/N/TTV.

N-gene ATGGATGCACAAACACGCCGCCGCGAACGTCGCGCAGAGAAACAGGCTCAATGGA M D A Q T R R R E R R A E K Q A Q W K Bam HI AAGCAGCAAATCCCCTGTTGGTTGGGGTAAGCGCAAAACCAGTTCGGATCCGCAT A A N P L L V G V S A K P V R I R M TGF GGTTGTTTCTCACTTTAACGACTGCCCGGACTCTCATACTCAGTTTTGCTTTCAT V V S H F N D C P D S H T Q F C F H Kpn I Sph I VGF GGTACCTGCCGTTTTCTGGTTCAGGAAGAAAAACCGGCATGCGTTTGCTCTCATG G T C R F L V Q E E K P A C V C S H G Eco RI GCTACACTGGAATTCGTTGCCAGCATGTTGTTCTGGTCGACTACCAGCGTTAAG Y T G I R C Q H V V L V D Y Q R Ter Bam HI GATCC Example V.

Preparation of the Polypeptide of Interest as a Fusion Protein with the N-Protein and a Cleavage Site A Modified Synthetic VGF 1 Preparation of p B Mll/NDP/VGFA:

The N-terminal sequence of the synthetic VGFA gene is a truncated version of the natural VGF sequence and begins with the sequence DIPAIR In this plasmid the VGFA fragment is located downstream of 32 amino acids of the lambda N-protein and the dipeptide aspartic acid-proline In order to preserve the Kpn I cloning site, the synthetic sequence was altered to code for CLHCGTC instead of the natural VGF sequence CLHGDC and terminates with the sequence YQR upstream of the natural sequence PNT In addition, the VGFA gene codes for the sequence GYACVC which replaces the natural sequence GMYCRC.

a Preparation of a Kpn I-Bam HI 80 bp C-ter- minal fragment of the synthetic VGF gene Plasmid p LE Bam/TWVV was digested with Kpn I and Bam HI and the 80 bp Kpn I-Bam HI fragment was gel purified This fragment contains the C-terminal two- thirds of the synthetic VGF gene with the Kpn I site at the 5 ' end.

b Preparation of Bam HI digested dephosphorylated p B Mll Plasmid p B Mll/N/TTV was digested with Bam HI and the 5 ' phosphates were removed by treatment with calf intestinal alkaline phosphatase The 5 6 kp Bam HI plasmid fragment was gel purified.

c Ligation and isolation of p B Mll/NDP/VGFA Oligonucleotides VGF 103 a, 104 a, the 5 6 kb Bam HI fragment of p B Mll and the 80 bp Kpn I-Bam HI fragment of p LE Bam/TVV were ligated together using DNA ligase and then used to transform competent HB 101 The transformants were selected on neomycin and were screened by restriction analysis using Cla I and nucleotide sequencing following the Sanger-dideoxy technique A correct construction was isolated and denoted p B Mll/NDP/VGFA This construction has the sequences GTC and GYACVC instead of the authentic VGF sequences GDC and GMYCRC.

N-gene ATGGATGCACAAACACGCCGCCGCGAACGTCGCGCAGAGAAACAGGCTCAATGGA M D A Q T R R R E R R A E K Q A Q W K Cla I AAGCAGCAAATCCCCTGTTGGTTGGGGTAAGCGCAAAACCAGTTCGGATCGATC A A N P L L V G V S A K P V R I D P Nco I VGF CCATGGACATCCCGGCTATCCGTCTGTGCGGCCCGGAAGGCGACGGCTACTGCCT M D I P A I R L C G P E G D G Y C L Kpn I Sph I GCATGGTACCTGCATCCATGCACGTGACATCGACGGCTACGCATGCGTTTGCTCT H G T C I H A R D I D G Y A C V C S Eco RI CATGGCTACACTGGAATTCGTTGCCAGCATGTTGTTCTGGTCGACTACCAGCGT H G Y T G I R C Q H V V L V D Y Q R Bam HI TAAGGATCC Ter 2 Preparation of p BM 11/NDP/VG Fa The N-terminal sequence of VG Fa is a truncated version of the natural VGF sequence and starts with the sequence DIPAIR In addition, the VG Fa sequence con- tains the altered sequences GTC and GYACRC instead of the natural VGF sequences GDC and GMYCRC In this plasmid the VG Fa gene is located downstream of 32 amino acids of the lambda N-protein and the dipeptide aspartic acid-proline Treatment of the purified fusion protein with formic acid results in cleavage at the acid labile aspartic acid-proline peptide bond allowing separation of the VG Fa protein from the lambda N-protein amino-terminus Cleavage is such that the VG Fa protein is left with the proline residue at the amino terminus.

a Preparation of Sph I digested, dephosphorylated p B Mll/DP/VGFA Plasmid p B Mll/DP/VGFA ( 10 jg) was digested with 30 units of Sph I and the 5 ' phosphates were removed by treatment with calf intestinal alkaline phosphatase The 5 kb plasmid fragment was recovered after electrophoresis on an agarose gel.

b Preparation of an Eco RI-Sph I 70 bp fragment of p B Mll/DP/VGFA Plasmid p B Mll/DP/VGFA ( 10 pg) was digested with 30 units of Eco RI and then 30 units of Sph I The bp fragment was recovered after electrophoresis on an agarose gel.

3 Ligation and isolation of p B Mll/NDP/VG Fa The 24 bp fragment containing oligonucleotides VGF 1 A and 2 A, the 5 kb Sph I fragment and the 70 bp Eco RI-Sph I fragment of p B Mll/DP/VGFA were ligated together and the mixture was used to transform competent E coli HB 101 cells The transformants were screened by nucleotide sequencing using the Sangerdideoxy nucleotide method A correct clone was isolated and denoted p B Mll/NDP/VG Fa.

N-gene ATGGATGCACAAACACGCCGCCGCGAACGTCGCGCAGAGAAACAGGCTCAATGGA M D A Q T R R R E R R A E K Q A Q W K Cla I AAGCAGCAAATCCCCTGTTGGTTGGGGTAAGCGCAAAACCAGTTCGGATCGATCC A A N P L L V G V S A K P V R I D P Nco I VGF CCATGGACATCCCGGCTATCCGTCTGTGCGGCCCGGAAGGCGACGGCTACTGCCT M D I P A I R L C G P E G D G Y C L Kpn I Sph I GCATGGTACCTGCATCCATGCACGTGACATCGACGGCTACGCATGCCGTTGCTCT H G T C I H A R D I D G Y A C R C S Eco RI CATGGCTACACTGGAATTCGTTGCCAGCATGTTGTTCTGGTCGACTACCAGCGT H G Y T G I R C Q H V V L V D Y Q R Bam HI TAAGGATCC Ter B Modified Synthetic TGF-VGF Hybrids 1 Preparation of p B Mll/NDP/TTV In this construct, the synthetic modified TTV chimeric gene is expressed as the C-terminal portion of a fusion protein having the first 32 amino acids of the N-gene at the N-terminus An acid labile aspartic acid-proline dipeptide separates the two parts of the fusion The hybrid growth factor contains the amino acid sequence of human TGF in the amino terminal two- thirds of the gene with the exception of the sequence QEEK which was altered from the natural human TMR se- quence QEDK The carboxy terminus was derived from the amino acid sequence of VGF and terminated with the se- quence YQR upstream of the natural sequence PNT.

A I, a Preparation of 5 kb Nco I p B Mll plasmid fragment Plasmid p B Mll/NDP/VGFA was digested with Nco I and the 5 kb Nco I plasmid fragment was gel puri- fied This fragment has one Nco I overhang at the aspartic acid-proline cleavage site downstream of the sequences coding for the first 32 amino acids of the N-gene The other Nco I site is in the neomycin re- sistance gene.

b Preparation of 0 6 kb Nco I-Bam HI p B Mll fragment Plasmid p B Mll/N/TTV was digested with Nco I and Bam HI and the 0 6 kb Nco I-Bam HI plasmid fragment was gel purified This fragment has the Nco I overhang in the neomycin resistance gene.

c Preparation of the 170 bp synthetic TGF/TGF/VGF fragment Plasmid p LE Bam/TTV was digested with Nco I and Bam HI and the Nco I-Bam HI 170 bp fragment containing the TGF/TGF/VGF synthetic gene was gel purified This fragment has the Nco I overhang at the 5 ' end of the gene and the Bam HI overhang at the 3 ' end of the gene.

d Ligation and isolation of p B Mll/NDP/TTV The 5 kb Nco I and the 0 6 kb Nco I-Bam HI plasmid fragments were ligated with the 170 bp Nco I-Bam HI TTV gene using DNA ligase and the resulting mixture was used to transform competent HB 101 The transformants were selected on neomycin such that only colonies with correctly reconstructed neomycin resistance genes would survive Transformants were screened using restriction analysis with Nco I and nucleotide sequencing using the Sanger-dideoxy technique A correct construction was isolated and denoted p B Mll/NDP/TTV.

N-gene ATGGATGCACAAACACGCCGCCGCGAACGTCGCGCAGAGAAACAGGCTCAATGGA M D A Q T R R R E R R A E QAQ K QA Q W K Cla I AAGCAGCAAATCCCCTGTTGGTTGGGGTAAGCGCAAAACCAGTTCGGATCGATC A A N P L L V G V S A K P V R I D) P Nco I TGF CCATGGTTGTTTCTCACTTTAACGACTGCCCGGACTCTCATACTCAGTTTTGCTT M V V S H F N D C P D S H T Q F C F Kpn I Sph I VGF TCATGGTACCTGCCGTTTTCTGGTTCAGGAAGAAAAACCGGCATGCGTTTGCTCT H G T C R F L V Q E E K P A C V C S Eco RI CATGGCTACACTGGAATTCGTTGCCAGCATGTTGTTCTGGTCGACTACCAGCGT H G Y T G I R C Q H V V L V D Y Q R Bam HI TAAGGATCC Ter 2 Preparation of p B Mll/NDP/VTV In this construct, the synthetic modified VTV chimeric gene was expressed as the C-terminal portion of a fusion protein having the first 32 amino acids of the N-gene at the N-terminus An acid labile aspartic acid-proline dipeptide separates the two parts of the fusion The hybrid growth factor contained the amino acid sequence of human TGF in the middle domain with the amino acid sequence QEEK replacing the natural sequence QEDK The N-terminal and C-terminal domains were derived from the truncated VGF sequence and begin with the sequence DIPAIR and end with the sequence YQR which is upstream of the natural sequence PNT.

I a Preparation of a 5 kb Bam HI-Nco I fragment of p B Mll Plasmid p B Mll/N/TTV was digested with Bam HI and Nco I and the 5 kb Bam HI-Nco I fragment was gel puri- fied This fragment contains a Bam HI overhang at the 3 ' end of the sequences coding for the first 32 amino acids of the N-gene and a Nco I site in the neomycin resistence gene.

b Preparation of a 700 bp Kpn I-Nco I fragment of p B Mll/N/TTV Plasmid p B Mll/N/TTV was digested with Kpn I and Nco I and the 700 bp Kpn I-Nco I fragment was gel purified This fragment is made up of part of plasmid p B Mll containing part of the neomycin resistence gene at the Nco I overhang, and the C-terminal VGF domain of the TTV synthetic gene at the Kpn I overhang.

c Ligation and isolation of p B Mll/NDP/VTV Oligonucleotides VGF 103 a and 104 a, the 5 kb Bam HI-Nco I fragment of p B Mll and the 700 bp Kpn I-Nco I fragment of p B Mll/N/TTV were ligated together using DNA ligase and then used to transform competent HB 101 The transformants were selected on neomycin and were screened by restriction analysis using Cla I and nucleotide sequencing following the Sanger-dideoxy technique.

N-gene ATGGATGCACAAACACGCCGCCGCGAACGTCGCGCAGAGAAACAGGCTCAATGGA M D A Q T R R R E R R A E K Q A Q W K Cla I AAGCAGCAAATCCCCTGTTGGTTGGGGTAAGCGCAAAACCAGTTCGGATCGATC A A N P L L V G V S A K P V R I D P Nco I VGF P CCATGGACATCCCGGCTATCCGTCTGTGCGGCCCGGAAGGCGACGGCTACTGCCT M D I P A I R L C G P E G D G Y C L Kpn I TGF Sph I VGF GCATGGTACCTGCCGTTTTCTGGTTCAGGAAGAAAAACCGGCATGCGTTTGCTCT H G T C R F L V QE E K P A C V C S Eco RI CATGGCTACACTGGAATTCGTTCGCAGCATGTTGTTCTGGTCGACTACCAGCGT H G Y T G I R C Q H V V L V D Y Q R Bam HI TAAGGATCC Ter 3 Preparation of p BM 16/NDP/TVV In this construct, the synthetic modified TVV chimeric gene was expressed as the C-terminal portion of a fusion protein having the first 32 amino acids of the N-gene at the N-terminus An acid labile aspartic acid-proline dipeptide separates the two parts of the fusion The hybrid growth factor contained the amino acid sequence of human TGF in the N-terminal domain.

The middle and C-terminal domains were derived from the truncated VGF sequence and end with the sequence YQR.

In addition, the synthetic gene has the modification GYACVC for GMYCRC.

a Preparation of 4 3 kb Nco I-Bgl II fragment of p B Mll/NDP/VG Fa Plasmid p B Mll/NDP/VG Fa was digested with Nco I and Bg 191 II and the-4 3 kb fragment was gel purified The Nco I overhang is positioned at the aspartic acid- proline cleavage site just downstream of the first 32 amino acids of the N-gene.

b Preparation of the 1 2 kb Bam HI-Bgl II fragment of p B Mll M 5 Plasmid p B Mll M 5 was digested with Bam HI and Bgl II and the 1 2 kb fragment was gel purified This fragment differs from the normal p B Mll fragment in that the Nco I site in the neomycin resistence gene has been removed, and all subsequent vectors lacking this Nco I site are referred to as p BM 16.

c Preparation of the 170 bp Nco I-Bam HI TWVV synthetic gene Plasmid p LE Bam/TVV was digested with Nco I and Bam HI and the 170 bp Nco I-Bam HI fragment was gel purified This synthetic gene fragment has the Nco I site at the 5 '-end and the Bam HI site at the 3 '-end.

d Ligation and isolation of p BM 16/NDP/TVV The 4 3 kb Nco I-Bgl II fragment of p B Mll/NDP/VG Fa and 1 2 kb Bam HI-Bql II fragment of p B Mll M 5, and the 170 bp Nco I-Bam HI TVV synthetic gene fragment were ligated together using DNA ligase and the resulting mixture was used to transform competent HB 101 The transformants were selected on neomycin and screened by restriction analysis and nucleotide sequencing using the Sanger-dideoxy technique The plasmid is denoted p BM 16 to indicate the loss of the Nco I restriction site in the neomycin resistance gene.

N-gene ATGGATGCACAAACACGCCGCCGCGAACGTCGCGCAGAGAAACAGGCTCAATGGA M D A Q T R R R E R R A E K QA Q W K Cla I AAGCAGCAAATCCCCTGTTGGTTGGGGTAAGCGCAAAACCAGTTCGGATCGATC A A N P L L V G V S A K P V R I D P Nco I TGF CCATGGTTGTTTCTCACTTTAACGACTGCCCGGACTCTCATACTCAGTTTTGCTT M V V SH F N D C P D S H T Q F C F Kpn I VGF Sph I TCATGGTACCTGCATCCATGCACGTGACATCGACGGCTACGCATGCGTTTGCTCT H G T C I H A R D I D G Y A C V C S Eco RI CATGGCTACACTGGAATTCGTTGCCAGCATGTTGTTCTGGTCGACTACCAGCGT H G Y T G I R C QH V V L V D Y Q R Bam HI 0 TAAGGATCC Ter C Synthetic EGF 1 Preparation of p B Mll/NDP/EGF In this construct the human EGF gene is ex- pressed as part of a fusion with the 32 N-terminal amino acids of the N-gene which is downstream of an Asp-Pro cleavage site.

a Preparation of a 5 kb Nco I fragment of p B Mll Plasmid p B Mll/DP/VGFA was digested with Nco I and the 5 ' phosphates were removed by treatment with calf alkaline intestinal phosphatase The 5 kb plasmid fragment was gel purified This fragment has one Nco I overhang at the Asp-Pro cleavage site downstream of the sequences coding for the first 32 amino acids of the N-gene The other Nco I site is in the Neomycin resis- tance gene.

b Preparation of a 0 6 kb Nco I-Bam HI fragment of p B Mll Plasmid p B Mll/N/TTV was digested with Nco I and Bam HI and the 0 6 kb Nco I-Bam HI plasmid fragment was gel purified This fragment has the Nco I overhang in the Neomycin resistance gene.

c Ligation and Isolation of p B Mll/DP/EGF The three sets of annealed EGF oligonucleotides with an Nco I overhang at the 5 ' end and a Bam HI overhang at the 3 ' end, the 5 kb Nco I fragment of p B Mll and the 0 6 kb Nco I-Bam HI fragment of p B Mll were ligated together using T 4 DNA Ligase and the resulting mixture was used to transform competent E.

coli HB 101 The transformants were selected on Neomycin such that only colonies with a correctly reconstructed Neomycin resistance gene would survive.

The transformants were screened by restriction analysis using Eco RI and Bam HI and by DNA sequencing, as described above.

N-gene ATGGATGCACAAACACGCCGCCGCGAACGTCGCGCAGAGAAACAGCGTCAATGGA M D A Q T R R R E R R A E K Q A Q W K AAGCAGCAAATCCCCTGTTGGTTGGGGTAAGCGCAAAACCAGTTCGGATCGATCC A A N P L L V G V S A K P V R I D P EGF 1 Eco RI EGF 2 CATGAATTCTGACTCTGAATGCCCGCTGTCTCATGACGGCTACTGCCTGCATGAC M N S D S E C P L S H D G Y C L H D EGF 3 Nsi I Sph I GGCGTATGCATGTACATCGAAGCTCTGGACAAGTACGCATGCAACTGCGTTGTTG G V C M Y I E A L D K Y A C N C V V G GCTACATCGGCGAACGTTGCCAGTACCGTGACCTGAAATGGTGGGAACTGCGTTA Y I G E R CQ Y R D L K W W E L R Bam HI AGGATCC Example VI.

Preparation of the Polypeptide of Interest as a Fusion Protein with a Modified Alkaline Phosphatase Signal Sequence A Preparation of p B Mll/PAD/EGF Synthetic oligonucleotides were designed to allow insertion of DNA coding for a modified alkaline phosphatase signal peptide and a linker region with 3 cloning sites (Hind III, Sma I and Bam HI) into the p B Mll expression vector downstream of the PL promoter and N gene ribosomal binding site The nucleotide sequence was optimized to be as similar as possible to the nucleotide sequence of the amino terminus of the lambda N gene as the lambda N gene sequence has evolved with that of its ribosomal binding site for efficient ribosome initiation and translation In addition, the second amino acid of the alkaline phosphatase signal sequence, the basic amino acid lysine was changed to an acidic amino acid, aspartic acid.

1 Preparation of 0 17 kb Eco R I-Bam HI fragment of EGF Plasmid p B Mll/NDP/EGF ( 30 ug) was digested with 30 units of Eco RI and then treated with 4 units of Klenow fragment of DNA polymerase to create blunt ends The DNA was finally digested with 30 units of Bam HI and the 0 17 kb fragment of the EGF gene was recovered after electrophoresis on an agarose gel The DNA so purified has a blunted Eco RI site at the 5 ' end and a Bam HI overhang at the 3 ' end.

2 Preparation of 0 5 kb Pvu I-Hind III fragment of p B Mll/PAD Plasmid p B Mll/PAD ( 18 ug) was digested with 30 units of Hind III and then treated with Klenow fragment to blunt the ends The DNA was then digested with Pvu I and the 0 5 kb Pvu I-Hind III (blunt) fragment was recovered after electrophoresis on an agarose gel.

3 Preparation of the 5 2 kb Pvu I-Bam HI fragment of p B Mll/PAD Plasmid p B Mll/PAD ( 18 ug) was digested with 30 units of Pvu I followed by 30 units of Bam HI The 5 2 kb fragment was recovered after electrophoresis on an agarose gel.

4 Ligation and isolation of p B Mll/PAD/EGF The 0 17 kb Eco RI (blunt)-Bam HI fragment, the 0.5 kb Pvu I-Hind III (blunt) fragment, and the 5 2 kb Pvu I-Bam HI fragment were ligated together and the resulting mixture was used to transform competent E.

coli HB 101 The transformants were screned using DNA sequencing, as described above The desired signal sequence/EGF region had the following sequence:

Signal Sequence ATGGATCAATCTACAATCGCCCTCGCACTTCTCCCACTGCTGTTCACT M D Q S T I A L A L L P L L F T EGF CCAGTGACAAAAGCTAATTCTGACTCTGAATGCCCGCTGTCTCATGAC P V T K A N S D S E C P L S H D Nsi I GGCTACTGCCTGCATGACGGCGTATGCATGTACATCGAAGCTCTG G Y C L H D G V C M Y I E A L Sph I GACAAGTACGCATGCAACTGCGTTGTTGGCTACATCGGCGAACGT D K Y A C N C V V G Y I G E R Bam HI TGCCAGTACCGTGACCTGAAATGGTGGGAACTGCGTTAAGGATCC C Q Y R D L K W W E L R I The efficacy of the production of foreign protein in the p B Mll/PAD expression system and the ability to purify functionally active foreign proteins from the fusion product has been show using p B Mll/PAD/EGF as an example After size exclusion chromatography (TSK-250), 10 3 mg of equivalents of active EGF fusion polypeptide was recovered from 23 g ( 8 liters) of E coli derepressed to express the EGF gene Forty percent of the EGF activity was derived from EDF cleaved from the signal sequence.

B Preparation of p B Mll/PAD/n VG Fa Synthetic oligonucleotides were designed to link the VG Fa synthetic gene with an alkaline phospha- tase modified signal sequence to provide for an optimal signal sequence cleavage site by coding for the addi- tional N-terminal residues occurring immediately down- stream of the signal sequence cleavage site in the natural VGF, denoted extreme N-terminus above The n VG Fa sequence contains the altered sequences GTC and GYACRC instead of the natural VGF sequences GDC and GMYCRC and terminates with the sequence YQR upstream of the natural sequence PNT In this expression system, for the majority of the molecules, the signal sequence remains attached to the n VG Fa forming a fusion protein with n VG Fa at the C-terminus.

1 Preparation of 0 5 kb Hind III-Pvu I digested p B Mll/PAD Plasmid p B Mll/PAD was digested with Hind III and Pvu I and the 0 5 kb fragment was gel purified The Hind III site is located at the C-terminus of the modified alkaline phosphatase signal sequence.

2 Preparation of the 5 2 kb Pvu I-Bam HI p B Mll plasmid fragment Plasmid p B Mll/NDP/VG Fa was digested with Pvu I and Bam HI and the 5 2 kb plasmid fragment was gel purified.

3 Preparation of the 170 bp Nco I(blunt)-Bam HI synthetic VG Fa gene Plasmid p B Mll/NDP/VG Fa was digested with Nco I and the 5 ' overhangs were removed by treatment with 51- nuclease This created a blunt end at the first codon of the VG Fa truncated synthetic gene The DNA was thendigested with Bam HI and the 170 bp Nco I(blunt)-Bam HI fragment was gel purified.

4 Ligation and isolation of p B Mll/PAD/n VG Fa Oligonucleotides VGF 105 and 106, the 0 5 kb Hind III-Pvu I fragment of p B Mll/PAD, the 5 2 kb Pvu I- Bam HI p B Mll fragment and the 170 bp Nco I(blunt)-Bam HI synthetic VG Fa gene were ligated together using DNA ligase and the resulting mixture was used to transform competent HB 101 The transformants were selected on neomycin and screened by restriction analysis and nucleotide sequencing using the Sanger-dideoxy technique A correct construct was isolated containing the modified alkaline phosphatase signal sequence in frame with the n VG Fa gene.

Signal Sequence - ATGGATCAATCTACAATCGCCCTCGCACTTCTCCCACTGCTGTTCACTCCAGTGA M D Q S T I A L A L L P L L F T P V T n VGF CAAAAGCTGACTCTGGTAACGCTATCGAAACTACTTCTCCGGAAATCACTAACGC K A D S G N A I E T T S P E I T N A TACTACTGACATCCCGGCTATCCGTCTGTGCGGCCCGGAAGGCGACGGCTACTGC T T D I P A I R L C G P E G D G Y C Kpn I Sph I CTGCATGGTACCTGCATCCATGCACGTGACATCGACGGCTACGCATGCCGTTGCT L H G T C I H A R D I D G Y A C R C S Eco RI CTCATGGCTACACTGGAATTCGTTGCCAGCATGTTGTTCTGGTCGACTACCAGCG H G Y T G I R C QH V V L V D Y Q R Bam HI TTAAGGATCC -= Ter Example VII.

Expression of the Polypeptide of Interest as a Fusion Protein with an Alkaline Phosphatase Signal Sequence A Preparation of p B Mll/PAK/n VG Fa (Alkaline pho sphatase signal sequence/n VG Fa with natural VGF N-t erminus and sequences GTC and GYACRC) Plasmid p B Mll/PAD/n VGF was mutagenized in vitro (Morinaga et al, Biotechnology ( 1984) 2:636-643) to alter the codons coding for the second amino acid in the signal sequence, namely to change the Asp (D) codon back to that for Lys (K), the residue found in the natural sequence This mutagenesis also introduced a Pvu I site into the signal sequence.

Signal Sequence ATGGATCAATCTACAATCGCCCTCGCACTTCTCCCACTGCTGTTCACTCCAGTGACAAAA M D Q S T I A L A L L P L L F T P V T K n VGF GCTGACTCTGGTAACGCTATCGAAACTACTTCTCCGGAAATCACTAACGCTACTACT A D S G N A I E T T S P E I T N A T T Kpn I GACATCCCGGCTATCCGTCTGTGCGGCCCGGAAGGCGACGGCTACTGCCTGCATGGT D I P A I R L C G P E G D G Y C L H G Sph I ACCTGCATCCATGCACGTGACATCGACGGCTACGCATGCCGTTGCTCTCATGGCTACACT T C I H A R D I D G Y A C R C S H G Y T Eco RI Bam HI GGAATTCGTTGCCAGCATGTTGTTCTGGTCGACTACCAGCGTTAAGGATCC G I R C Q H V V L V D Y Q R Ter B Preparation of p B Mll/PAK/EGF In this expression cassette the EGF gene is part of a fusion with the alkaline phosphatase signal sequence.

Plasmid p B Mll/PAD/EGF was mutagenized in vitro to alter the codons coding for the second amino acid in the signal sequence, to change the Asp (D) codon to that for Lys (K) the residue found in the natural se- quence This mutagenesis also introduced a Pvu I site into the signal sequence.

Pvu I Signal Sequence ATGAAACAATCTACGATCGCCCTCGCACTTCTCCCACTGCTGTTCACTCCAGTGA M K Q S T I A L A L L P L L F T P V T EGF CAAAAGCTAATTCTGACTCTGAATGCCCGCTGTCTCATGACGGCTACTGCCTGCA K A N S D S E C P L S H D G Y C L H Nsi I Sph I TGACGGCGTATGCATGTACATCGAAGCTCTGGACAAGTACGCATGCAACTGCGTT D G V C M Y I E A L D K Y A C N C V GTTGGCTACATCGGCGAACGTTGCCAGTACCGTGACCTGAAATGGTGGGAACTGC V G Y I G E R C Q Y R D L K W W E L R Bam HI GTTAAGGATCC 35, C Preparation of Tac Pak/EGF (alkaline phosphatase signal sequence/human EGF) 1 Preparation of Plasmid Fragments Plasmid p 135-1 was derived from plasmid p DR 540 (Pharmacia) and contained the Cro gene SD and a Bgl II site downstream of the lac SD p DR 540 is an expression vector containing the trp-lac hybrid promoter p 135-1 was digested with B 91 II and Bam HI and treated with bacterial alkaline phosphatase.

Plasmid p B Mll/PAK/EGF was digested with Pvu II and Bam HI and the -230 bp fragment coding for part of the alkaline phosphatase signal sequence and human EGF was isolated.

2 Preparation of Tac Pakl and Tac Pak 2 Oligonucleotides Synthetic oligonucleotides Tac Pakl and Tac Pak 2 were designed with an overhang, compatible with the Bgl II site of p 135-1 and a Pvu II overhand, compatible with the Pvu II site in the alkaline phosphatase/EGF Pvu II/Bam HI fragment The oligonucleotides were synthesized on an Applied Biosystems Oligonucleotide Synthesizer.

B II Pvu I T T Tac Pakl 5 ' GATCTATGAAACAATCTACGAT 3 ' Tac Pak 23 ' ATACTTTGTTAGATGC 5 ' 3 Ligation and Isolation of Tac Pak/EGF Clone The Bgl II-Bam HI digested p 135-1, the 230 bp PAK/EGF fragment and oligonucleotides Tac Pakl and Tac Pak 2 were ligated using DNA ligase, transformed into competent HB 101 and a correct construct was isolated by DNA sequencing.

(Hind III site of p DR 540) I AAGCTTACTCCC trp-35 ( 16 bp) lac-10 CATCCCCCTG lTTGACAl ATTAATCATCGGCTCG (TATAATG) m RNA 5 ' lac I binding site lac SD TGTGG/AATTGTG AGCGGATAACAATTTCACAC {AGGA} AACAGGATCACTA Pvu I cro SD (llbp)Bgl II {AGGA) GGTTCAGATCT Signal Sequence -> ATGAAACAATCTACGATCGCCCTCGCACTTCTCCCACTGCTGTTCACTCCAGTGA M K Q S T I A L A L L P L L F T P V T EGF -> CAAAAGCTAATTCTGACTCTGAATGCCCGCTGTCTCATGACGGCTACTGCCTGCA K A N S D S E C P L S H D G Y C L H Nsi I Sph I TGACGGCGTATGCATGTACATCGAAGCTCTGGACAAGTACGCATGCAACTGCGTT D G V C M Y I E A L D K Y A C N C V GTTGGCTACATCGGCGAACGTTGCCAGTACCGTGACCTGAAATGGTGGGAACTGC V G Y I G E R C Q Y R D L K W W E L R Bam HI GTTAAGGATCC Example VIII.

Expression of a Polypeptide of Interest as a Fusion Protein with the Alkaline Phosphatase Signal Sequence Using an Expression Cassette Comprising a Transcriptional Termination Region A Preparation of p TC Pt/EGF (ltrp-35 l 16 bpllac 10 l llac SDlllbplATGl/alkaline phosphatase sig nal/human EGF/trans term -NEO) This plasmid is designed to have the tac promoter elements and utilize the cro SD to express human EGF behind the alkaline phosphatase signal sequence It has a p BR 322 background with the Neomycin resistance gene.

1 Preparation of the 420 bp Hind III(blunt)-Bam HI fragment of Tac Pak/EGF Tac Pak/EGF was digested with Hind III and then treated with the Klenow fragment of DNA polymerase to create blunt ends The DNA was then digested with Bam HI and the 420 bp fragment containing the tac promoter elements and the coding region for the alkaline phosphatase signal sequence and human EGF was isolated by agarose gel electrophoresis.

2 Preparation of the 2 8 kb Eco RI(blunt)-Bam HI fragment of p BM 16 t/NDP/VG Fa p BM 16 t/NDP/VG Fa was digested with Eco RI and then treated with Klenow to create blunt ends The DNA was then digested with Bam HI and the 2 8 kb fragment was isolated This DNA fragment contains the p BR 322 origin, the neomycin resistance gene with its Nco I site removed, and the gene 32-like transcription terminator downstream of the Bam HI site.

3 Ligation and Isolation of p TC Pt/EGF The 2 8 kb Eco RI(blunt)-Bam HI fragment of p BM 16 t/NDP/VG Fa was ligated to the 420 bp Hind III(blunt)-Bam HI fragment of Tac Pak/EGF and the resulting DNA was used to transform competent J Ml O 9 (lac Iq) A correct construct was isolated by its resistance to neomycin and by DNA sequencing.

(Hind III site of p DR 540) AAGCTTACTCCC trp-35 ( 16 bp) lac-10 CATCCCCCTG lTTGACAl ATTAATCATCGGCTCG (TATAATG) m RNA 5 ' lac I binding site lac SD TGTGG/AATTGTG AGCGGATAACAATTTCACAC {AGGA} AACAGGATCACTA Pvu I cro SD (llbp)Bgl II Signal Sequence -> {AGGA) GGTTCAGATCT ATGAAACAATCTACGATCGCCCTCGCACTTCTCC M K Q S T I A L A L L P EGF -> CACTGCTGTTCACTCCAGTGACAAAAGCTAATTCTGACTCTGAATGCCCGCTGTC L L F T P V T K A N S D S E C P L S Nsi I TCATGACGGCTACTGCCTGCATGACGGCGTATGCATGTACATCGAAGCTCTGGAC H D G Y C L H D G V C M Y I E A L D Sph I AAGTACGCATGCAACTGCGTTGTTGGCTACATCGGC K Y A C N C V V G Y I G Bam HI GAACGTTGCCAGTACCGTGACCTGAAATGGTGGGAACTGCGTTAAGGATCCGTGA E R C Q Y R D L K W W E L R Trans Term.

CTAATTGGGGACCCTAGAGGTCCCCTTTTTTATTTTAAAACGATC B Preparation of p TC Pt/n VG Fa (ltrp-35116 bpllac-10 l llac SDllcro SDlllbplATGl/ alkaline phosphatase signal/n-terminal VG Fa with sequence GTC and GYACRC)/trans term -NEO) This plasmid has the tac promoter elements and uses the cro SD to express the modified VGF gene with the N-terminal extension downstream of the alkaline phosphatase signal sequence The plasmid has a p BR 322 background with the neomycin resistance gene.

1 Preparation of the 350 bp Pvu I-Bam HI fragment of p B Mll/PAK/n VG Fa Plasmid p B Mll/PAK/n VG Fa was digested with Pvu I and Bam HI and the 350 bp fragment was isolated by gel electrophoresis This fragment contains most of the alkaline phosphatase signal sequence and the n VG Fa gene.

2 Preparation of the 2 8 kb Pvu I-Bam HI fragment of p TC Pt/EGF Plasmid p TC Pt/EGF was digested with Pvu I and Bam HI and the 2 8 kb fragment was isolated by gel electrophoresis.

Ligation and Isolation of p TC Pt/n VG Fa The 2 8 kb fragment and the 350 bp fragment were ligated using DNA ligase and the DNA was used to transform competent J Ml O 9 (lac Iq) A correct construct was isolated using restriction analysis.

3 Litigation and Isolation of p TC Pt/n VG Fa The 2 8 kb fragment and the 350 bp fragment were ligated using DNA ligase aand the DNA was used to transform competent JM 109 (lac Iq) A correct construct was isolated using restriction analysis.

(Hind III site of p DR 540) I AAGCTTACTCCC trp-35 ( 16 bp) lac-10 CATCCCCCTG lTTGACAl ATTAATCATCGGCTCG (TATAATG) m RNA 5 ' lac I binding site lac SD TGTGG/AATTGTG AGCGGATAACAATTTCACAC {AGGA) AACAGGATCACTA Pvu I cro SD (llbp)Bgl II Signal Sequence -> {AGGA) GGTTCAGATCT ATGAAACAATCTACGATCGCCCTCGCACTTCTCCC M K Q S T I A L A L L P n VGF -> ACTGCTGTTCACTCCAGTGACAAAAGCTGACTCTGGTAACGCTATCGAAACTACT L L F T P V T K A D S G N A I E T T TCTCCGGAAATCACTAACGCTACTACTGACATCCCGGCTATCCGTCTGTGCGGCC S P E I T N A T T D I P A I R L C G P Kpn I CGGAAGGCGACGGCTACTGCCTGCATGGTACCTGCATCCATGCACGTGACATCGA E G D G Y C L H G T C I H A R D I D Sph I Eco RI CGGCTACGCATGCCGTTGCTCTCATGGCTACACTGGAATTCGTTGCCAGCATGTT G Y A C R C S H G Y T G I R C Q H V Bam HI Trans.

GTTCTGGTCGACTACCAGCGTTAAGGATCCGTGACTAATTGGGGACCCTAGAGGT V L V D Y Q R Term.

CCCCTTTTTTATTTTAAAACGATC C Preparation of p TN Pt/EGF (ltrp-35117 bpllacl ln SDl 8 bplATGl/alkaline phosphatase signal/human EGF/trans term -NEO) This plasmid is designed to have the tac promoter elements and utilize the N-gene SD to express human EGF behind the alkaline phosphatase signal sequence It has a p BR 322 background with the Neomycin resistance gene.

1 Preparation of 2 8 kb Pvu I-Bam HI p TN Pt Plasmid p TN Pt was digested with Pvu I and Bam HI and the 2 8 kb fragment was isolated by gel electrophoresis.

2 Preparation of 300 bp Pvu I-Bam HI fragment of p B Mll/PAK/EGF Plasmid p B Mll/PAK/EGF was digested with Pvu I and Bam HI and the 300 bp fragment was isolated.

3 Ligation and Isolation of p TN Pt/EGF The 2 8 kb fragment and the 300 bp fragment were ligated using DNA ligase and the DNA was transformed into competent JM 109 (Lac Iq) A correct construct was isolated by restriction analysis and by DNA sequencing.

(Eco RI site of p B Mll) I GAATTACTCCCCATCC Sst I trp-35 ( 17 bp) lac-10 5 'lac CCCTG lTTGACAl ATTAATCATCGAGCTCG (TATAATG) TGTGG/AATTG Bsm I m RNA-> N m RNA-> TGTGAGCGGATAACAATTTCACACAGCATTCAAAGCAGAAGGCTTTGGGGTGTGT GATACGAAACGAAGCATTGGCCGTAAGTGCGATTCCGGATTAGCTGCCAATGTGC CAATCGCGGGGGGTTTTCGTTCAGGACTACAACTGCCACACACCACCAAAGCTAA Pvu I n SD ( 8 bp) Signal Sequence -> CTGAC {AGGA) GAATCCAG ATGAAACAATCTACGATCGCCCTCGCACTTCTC M K Q S T I A L A L L EGF -> CCACTGCTGTTCACTCCAGTGACAAAAGCTAATTCTGACTCTGAATGCCCGCTGT P L L F T P V T K A N S D S E C P L S Nsi I CTCATGACGGCTACTGCCTGCATGACGGCGTATGCATGTACATCGAAGCTCTGGA H D G Y C L H D G V C M Y I E A L D Sph I CAAGTACGCATGCAACTGCGTTGTTGGCTACATCGGCGAACGTTGCCAGTACCGT K Y A C N C V V G Y I G E R C Q Y R Bam HI Bam HI GACCTGAAATGGTGGGAACTGCGTTAAGGATCCGTGACTAATTGGGGA D L K W W E L R (Bam HI site of p B Mll) Trans Term I CCCTAGAGGTCCCCTTTTTTATTTTAAAACGATCC Example IX

Assembly of Growth Factor Genes for Expression in Eukaryotic Cells A Preparation of Plasmid p RSV/VGF:

Transfection of this plasmid into Chinese hamster ovary cells resulted in the production of natural VGF.

(a) Preparation of Hind III(blunt)-Bgl II(blunt) p RSV Plasmid p RSV was digested with Hind III and Bg 1 II and the 4 kb fragment was gel purified The protruding ends were made blunt with the Klenow fragment of DNA polymerase and then the 5 ' phosphates were removed with calf alkaline intestinal phosphatase.

(b) Preparation of a Dde I fragment containing the VGF gene A subcloned genomic fragment of vaccinia DNA was digested with Dde I, the protruding ends were made blunt using the Klenow fragment, and a 550 bp Dde I fragment was gel purified.

(c) Ligation and isolation of p RSV/VGF The 4 kb Hind III (blunt)-Bgl II(blunt) fragment of p RSV and the 550 bp Dde I(blunt) fragment of vaccinia DNA were ligated together using DNA ligase and the transformants were selected with ampicillin and screened by restric- tion analysis A correct construction was isolated and denoted p RSV/VGF.

(d) Transfection of CHO cells by p RSV/VGF.

Plasmid p RSV/VGF was cotransfected with a plasmid p RSV by calcium phosphate precipitation into Chinese hamster ovary (CHO) cells Transfectants were screened by Southern dot blot hybridization and a positive clone designated p RSV/VGF 52 was isolated.

B Preparation of Plasmid p Ac/VGF:

An additional expression system which is used to express the VGF recombinant protein is an insect system See Maeda et al and Carbonell et al, supra.

In one such system, Autographa californica nuclear polyhedrosis virus (Ac NPV) is used as a vector to express foreign genes The virus grows in Spodoptera frugiperda cells The envelope gene can be cloned into non-essential regions (for example, the polyhedrin gene) of the virus and is placed under control of an Ac NPV promoter (for example, the polyhedrin promoter).

Successful insertion of the VGF gene construct results in inactivation of the polyhedrin gene and production of non-occluded recombinant virus (i e, virus lacking the proteinaceous coat coded for by the polyhedrin gene) These recombinant viruses are then used to in- fect Spodoptera frugiperda cells in which the inserted gene is expressed.

The VGF gene of interest was placed under the control of a suitable promoter for an insect cell sys- tem Plasmid p Ac 610 contains the polyhedrin gene cloned into a plasmid vector possessing an ampicillin resistance marker Polylinkers are inserted into this gene which are 50 bases downstream from the transcrip- tional start site of the polyhedrin gene and 7 bases before the first ATG The VGF gene is cloned into a convenient polylinker site, so that it is under the control of the polyhedrin promoter The ATG initiation methionine codon and the translational termination codons are those of the VGF gene; the transcriptional start sites and polyadenylation signals are those of the polyhedrin gene.

(a) Preparation of Sma I-Bam HI p Ac 610Plasmid p Ac 610 was digested with Sma I and Bam HI.

(b) Preparation of Hind III(blunt)-Bgl II frag- ment of the VGF gene Plasmid p RSV/VGF was digested with Hind III and the protruding ends were made blunt using the Klenow fragment The DNA was then digested with Bgl II and the 560 bp fragment was gel purified.

(c) Ligation and isolation of p Ac/VGF The Sma I-Bam HI fragment of p Ac 610 and the 560 bp fragment of p RSV/VGF containing the VGF gene were ligated to- gether using DNA ligase and the transformants were screened by restriction analysis A correct construct was isolated and termed p Ac/VGF.

(d) Transfection of Sf 9 cells with p Ac/VGF.

Plasmid p Ac/VGF was cotransfected with Ac NPV wild-type baculovirus DNA into Sf 9 insect cells (Spodoptera frugiperda) The transfectants were screened for occlusion negative phenotype in plaque assays One occlusion negative plaque was isolated and after five rounds of successive plaque purification, a high titer recombinant virus stock was prepared.

C Preparation of p Ac/SFGF Baculovirus expression of the natural Shope fibroma virus growth factor.

(a) Preparation of Bam HI digested p Ac 610.

Plasmid p Ac 610 was digested with Bam HI and Sma I The Sma I site in this plasmid is located downstream of the baculovirus polyhedrin gene promoter.

(b) Preparation of a 430 bp Hinc II-Sau 3 a frag- ment of the Shope fibroma virus genomic DNA The 3 7 kb Bgl II fragment of the 19 kb Bam HI C-fragment of Shope fibroma virus genomic DNA was cloned into a Bam HI digested SP 64 plasmid and denoted SP 64/3 7 Plasmid SP 64/3 7 was digested with Hinc II and the lkb Hinc II fragment was gel purified the lkb fragment was then digested with Sau 3 a and the 430 bp Hinc II-Sau 3 a fragment was gel purified.

(c) Ligation and isolation of p Ac/SFGF The Bam HI-Sma I digested p Ac 610 was ligated to the 430 bp Hinc II-Sau 3 a fragment containing the SFGF gene and the resulting mixture was transformed into competent E.

coli HB 101 The transformants were screened by re- striction analysis and a correct construction was isolated and denoted p Ac/SFGF.

Example X

Preparation of Polypeptides A Solid-phase Synthesis of VGF and TGF 1 Synthetic VGF fragment The sequence corre- sponding to truncated VGF, beginning with the sequence DIPAIR and ending with the sequence LVDY was assembled on an Applied Biosystems Model 430 A Peptide Synthesizer using the standard protocol Treatment of the final peptide resin with liquid HF under standard "low-high" conditions gave the crude deprotected peptide The peptide was subjected to chromatofocusing on PBE 94 (Pharmacia) The partially purified peptide was eluted at p H 6 5 Brief treatment with 0 2 N sodium hydroxide removed the cysteine protecting groups (ethylcarbamoyl) and the peptide was oxidized in the presence of oxi- dized and reduced glutathione Purification by gel filtration and HPLC gave highly purified VGF with the following sequence:

DIPAIRLCGPEGDGYCLHGDCIHARDIDGMYCRCSHGYTGIRCQHVVLVDY 2 Synthetic human and rat TGF-a Human and rat TGF-a were chemically synthesized, essentially as de- scribed above for VGF, and were provided by Penninsula Labs.

B Isolation of Recombinant Polypeptides Prepared in Prokaryotic Cells 1 Growth Factors Produced in p BM-Based Vectors Using the PL Promoter and the ts CI Repressor E coli B (HB 101) containing the p B Mll/NDP/ growth factor plasmids were grown in Luria Broth at C The density of the culture was measured at 550 nm and when the density reached an absorbance of 0.7 to 0 9, synthesis of the growth factor fusion protein was induced by increasing the temperature to 42 C The culture was incubated at this temperature for 5-20 hrs, then the bacteria were isolated by centrifugation and frozen at -70 C until use.

For isolation of the recombinant protein, the cells were thawed-into buffer containing 0 05 M Na H 2 PO 4 p H 7 2, 0 5 M Na C 1, 0 01 M EDTA One hundred fifty ml of buffer was used for a preparation from 50 g bacteria The cells were disrupted by sonication on ice for 15 min using a 1/4-inch probe, 50 % pulse at 60 watts of power Following disruption of the cells, the insoluble protein was collected by centrifugation at 12,000 rpm in a GSA rotor for 90 min The pellet f l containing the insoluble protein was then resuspended in 50 ml of 6 M guanidine hydrochloride The insoluble material was collected by centrifugation for 2 hrs at 25,000 rpm in a Beckman-type 30 ultracentrifuge rotor.

The supernatant was collected and stored at -200 C until fu:rther use.

Purification of the fusion protein was carried out on either a Sephacryl 5300 or Fractogel HW-55 column equilibrated with 1 M guanidine hydrochloride Fractions containing the fusion protein were identified as those fractions containing a polypeptide having a molecular weight consistent with the molecular weight of the poly- peptide encoded by the synthetic gene as determined on a % polyacrylamide-urea gel.

To obtain an active form of the recombinant growth factor, the fusion protein was allowed to refold by incubating it in 50 m M Tris-H Cl buffer, p H 8 7, con- taining 1 M guanidine hydrochloride, 1 25 m M reduced glutathione, and 0 25 m M oxidized glutathione at 40 C for 3-10 days The biological activity of the growth factor was monitored by a competitive receptor binding assay as described above (see Example I C) When a maximum level of activity was obtained, the protein was dialyzed against distilled water and lyophilized to dryness.

If it was desired to remove the leader sequence, the protein was cleaved either by resuspending in 70 % formic acid and incubating at 400 C for 3 days or by incubating overnight at room temperature in a 100-fold molar excess of cyanogen bromide The cleaved product was dialyzed against distilled water and lyophilized to dryness.

To further purify the recombinant growth factor, the growth factor was resuspended in 40 % acetonitrile, 0 1 % TFA and purified by HPLC using a Bio Rad TSK-250 column Fractions containing the growth factor were pooled and further purified using reversed- phase HPLC, either Waters V Bondapak C-18 or Rainin Dynamax C-8 The eluant was a linear gradient of 20- % acetonitrile containing 0 1 % TFA Fractions containing receptor binding activity were pooled, lyophilized and stored at -20 C until use.

(a) TGF and Modified TGF (i) N/TGF Recombinant modified human TGF was produced from plasmid p B Mll/N/TGF and contained 33 amino acids of the N-gene at the N-terminus and the sequence modification QEEK instead of the natural human sequence QEDK.

(b) Modified and Truncated VGF (i) PAD/n VG Fa Recombinant modified VGF was produced from plasmid p B Mll/PAD/n VG Fa containing the extreme Nterminal sequence of VGF and the modified sequences GTC and GYACRC instead of the natural VGF sequence GDC and GMYCRC The n VGFA fragment was expressed as a fusion protein with a modified alkaline phosphatase signal sequence at the N-terminus and was truncated at the sequence YQR at the C-terminus.

(ii) NDP/VG Fa Recombinant modified VGF was produced from plasmid p B Mll/NDP/VG Fa beginning at the DIPAIR sequence and ending at the YKQR sequence in VGF It has the modified sequences GTC and GYACRC instead of the natural VGF sequence GDC and GMYCRC The VG Fa fragment was expressed as a fusion protein with 32 amino acids of the N-gene at the N-terminus and the acid labile di- peptide aspartic acid-proline.

(iii) VG Fa The VGF fragment was prepared as described in 2.b above and, after cleavage from the fusion protein by acid treatment, was subsequently further purified by HPLC.

(iv) NDP/VGFA Recombinant modified VGF was produced from plasmid p B Mll/NDP/VGFA beginning at the DIPAIR sequence and ending at the YKQR sequence in VGF and having the modified sequences GTC and GYACVC instead of the natural VGF sequence GDC and GMYCRC The VGFA fragment was expressed as a fusion protein with 32 amino acids of the N-gene at the N-terminus and the acid labile dipeptide aspartic acid-proline.

(c) Chimeric TGF/VGF Hybrids (i) N/TTV (TGF/TGF/VGF) Recombinant modified TTV was produced from p B Mll/N/TTV and contained the amino acid sequence of human TGF in the amino terminal two-thirds of the gene with the exception of the sequence QEEK which was altered from the natural human sequence QEDK The carboxy terminus was derived from the amino acid sequence of VGF and terminated with the sequence YQR upstream of the natural sequence PNT The TTV fragment was expressed as a fusion protein with 33 amino acids, of the N-gene at the N-terminus.

(ii) NDP/TTV Recombinant TTV was produced from plasmid p B Mll/N/TTV and modified as described in (a) except that the TTV fragment was expressed as a fusion protein with 32 amino acids of the N-gene at the N-terminus and the acid labile dipeptide aspartic acid-proline.

(iii) NDP/VTV Recombinant modified VTV was produced from plasmid p B Mll/NDP/VTV and contained the amino acid sequence of human TGF in the middle domain with the amino acid sequence QEEK replacing the natural sequence QEDK The N-terminal and C-terminal domains were derived from the truncated VGF sequence and begin with the sequence DIPAIR and end with the sequence YQR The VTV fragment was expressed as a fusion protein with 32 % IZ.

amino acids of the N-gene at the N-terminus and the acid labile dipeptide aspartic acid-proline.

(iv) NDP/TVV Recombinant modified VTV was produced from plasmid p B Mll/NDP/TVV and contained the amino acid sequence of human TGF in the N-terminal domain of the gene The middle and C-terminal domains were derived from the truncated VGF sequence and end with the sequence YQR -In addition, the synthetic gene has the modification GYACVC for GMYCRC The TVV fragment was expressed as a fusion protein with 32 amino acids of the N-gene at the N-terminus and the acid labile dipeptide aspartic acid-proline.

2 Growth Factors Produced in Vectors Comprising the tac or lac Promoters Bacterial hosts containing expression cassettes which comprise the tac or lac promoters were grown at 30 to 37 C to an optical density of A 600 = 0 2 to 0 8 in either LB broth or a chemcially defined medium such as M 9 medium supplemented with thiamine and glucose An appropriate antibiotic was included in the growth medium to select for hosts containing the expression cassette The bacterial cultures were induced with a concentration of from 100 to 1000 m M IPTG and grown at 300 C for 16 to 24 hours For the expression cassettes lacking a lac I gene the bacterial hosts carried an F-factor with the lacq I gene, such as JM 109, XL 1, JM 103, etc For expression cassettes which carry the lac I gene, examples of bacterial hosts are HB 101, DH 1, DH 5, etc In the case where the bacterial host has a functional lac operon (lac+), the expression cassette can be induced with 1 % lactose After the induction period, growth factors can be isolated from either the medium or the cell pellet.

% # A (a) PAK/EGF Human EGF produced from the expression cassettes Tac Pak/EGF and p Tc C Pt/EGF was isolated from the medium in an active form with the alkaline phosphatase signal sequence removed Approximately 85 % of the active EGF was found in the medium, with the remainder associated with the cell pellet These expression cassettes yielded 4 mg/l of active EGF The cells are removed from the medium by centrifugation and the medium is passed through an Amicon SY 30, 30,000 Mr cutoff spiral filter and then passed through a Q- Sepharose column and the highly purified human EGF was eluted in 20 m M Na PO 4 p H 7 with a 0 to 0 5 M Na Cl gradient Alternatively, the growth factors can be isolated from the cell pellet by osmotic shock or sonication and purified by essentially the same procedure as described above.

(b) PAK/n VG Fa Recombinant n VG Fa was produced from the expression cassette p TC Pt/n VG Fa The n VG Fa was isolated from the cell pellet by sonication and was shown to constitute approximately 40 % of the total bacterial protein.

C Isolation of VGF and SFGF Produced by Eukaryotic Expression of the Natural Genes 1 Preparation of VGF in monkey kidney cells.

Natural VGF was produced by infection of mon- key kidney cells with vaccinia virus Cercopithecus monkey kidney (BSC-1) cell monolayers were infected with 15 plaque-forming units (pfu) per cell of purified VV (strain WR grown in Hela cells and purified by suc- rose density gradient sedimentation (Moss, B, ( 1981) in Gene Amplification and Analysis, eds Chirickjian, J G and Papis, T S (Elsevier/North-Holland, NY), Vol.

2, pp 253-266)) The infected cells were incubated at 370 C with approximately lml of Eagle's basal medium supplemented with 2 % fetal calf serum per 2 x 106 cells.

Mock-infected cells were treated in an identical manner.

Cell culture supernatants were clarified by low speed centrifugation and lyophilized The residue was then resuspended in 1 M acetic acid and dialyzed extensively against 0 2 M acetic acid Insoluble material was removed by centrifugation and the supernatant was lyo- philized and resuspended in 1/100th of the original volume of 1 M acetic acid and stored at 4 C.

2 Preparation of VGF in CHO cells.

Natural VGF was also produced by transfection of the plasmid p RSV/VGF containing the VGF gene frag- ment into Chinese hamster ovary cells Thetransfected cells were expanded and the media and cell pellet were assayed for the presence of VGF by immunoprecipitation using an antisera against an N-terminal VGF peptide.

3 Preparation of VGF in Silkworm Using Baculovirus Promoter.

(a) Transformation The host cell for Ac NPV is Spodoptera frugiperda (sf 9) In order to produce a recombinant virus stock, the VGF containing plasmid is mixed with Ac NPV DNA and transfected into sf 9 cells using the calcium phosphate technique Recombinant viruses are isolated from the medium and plaque puri- fied on sf 9 cells Recombinant plaques are identified by hybridization using radiolabeled VGF DNA as a probe.

(b) Expression Once recombinant virus is identified, it is then expanded on sf 9 cells The re- combinant virus stock is then used to infect sf cells, the cells lysed, and supernatants screened for produc- tion of VGF protein, which is purified as described above for vaccinia virus-infected mammalian cells.

The predicted sequence of VGF produced in insect cells is:

DSGNAIETTSPEITNATTDIPAIRLCGPEGDGYCLHGDCIHARDIDGMYCRCSHGYTGIR CQHVVLMYQRSENPNTTTSYIPSPGIMLVLVGIIIITCCLLSVYRFTRRTKLPIQDMVVP 1 k I A A potential glycosylation site occurs at the asparagine in position 15 from the N-terminal end of the polypep- tide.

This 121-residue sequence starts with the N-terminal sequence that has been determined directly for VGF purified from VV-infected monkey cells (that is, at residue 20 of the VGF open reading frame) and continues to the last residue of the open reading frame It thus lacks the signal peptide (residues 1-19 of the open reading frame) but includes the transmembranous sequence (YIPSPGIMLVLVGIIIITCCLLSVY).

The predicted MW of the 121-residue peptide is 13,304, not counting carbohydrate The apparent MW of the baculovirus-produced VGF is 17,000, significantly smaller than that obtained from VV-infected cells, indicating that the two forms have probably been pro- cessed differently The baculovirus-produced VGF may lack an additional portion of the N-terminal sequence, or a portion of the C-terminal sequence, or both, and/ or could differ in the type and extent of glycosylation.

4 Preparation of SFGF in Silkworm Using Baculovirus Promoter.

(a) Transfection of Sf 9 cells with p Ac/SFGF.

Plasmid p Ac/SFGF was cotransfected with Ac NPV wild-type baculovirus DNA into Sf 9 insect cells (Spodoptera fru- giperda) The transfectants were screened for an occlusion negative phenotype in plaque assays One occlusion negative plaque was isolated and after 5 rounds of successive plaque purification a high titer recombinant virus stock was prepared The recombinant virus has been shown by Southern analysis to contain the SFGF gene.

D Natural EGF 1 Natural EGF was isolated from mouse salivary glands or was purchased from Collaborative Research.

Example XI

Activity Assays A Mitogenic Assay The mitogenesis assays were performed as follows:

diploid human fibroblasts obtained from explants of newborn foreskin were seeded at a density of 3 x 104 cells/well ( 96-well plates, Nunclon, Roskilde, Denmark) and were grown to confluency in Dulbecco's modified Eagle's medium (GIBCO)/10 % newborn calf serum Cul- tures were then placed in medium containing 0 2 % new- born calf serum, and two days later EGF ( 10 ng/ml) or the growth factor to be tested ( 10 ng equivalents of EGF/ml) was added After 8 hrs, cultures were labeled with 5-l 125 Iliodo-2 '-deoxyuridine (Amersham, 10 p Ci/ml, 5 Ci/mg; 1 Ci = 37 G Bq), and the amount of isotope in- corporated into TCA insoluble material was determined as described (Twardzik et al, Proc Natl Acad Sci.

USA ( 1985) 182:5300-5304).

B Soft Agar Colony Growth Stimulation Assay A 0 5 ml base layer of 0 5 % agar (Agar Noble; Difco Laboratories, Detroit, Michigan) in growth medium was added to 24-well Costar tissue culture plates One- half ml 0 3 % agar in growth medium containing 1 to 1 5 x 104 cells/ml NRK cells or other cell line of in- terest and various concentrations of the factor to be tested was overlaid on the base layer of agar The plates were incubated at 37 C in a humidified atmos- phere of 5 % CO 2 in air and refed after 7 days by addition of 0 5 ml of 0 3 % agar in growth medium and containing the same concentration of the factor to be tested Colonies were enumerated unfixed and un- stained The number of colonies with greater than 6 cells were scored.

C EGF Receptor Binding Inhibition Assay The radioreceptor assays were performed as fol- lows: The binding of 125 I-labeled EGF ( 125 I-EGF) to its receptor on monolayers of A 431 cells was modified from the procedure described by Cohen and Carpenter, Proc Natl Acad Sci USA ( 1975) 72:1317-1321 Cells (lx 103 per well) were fixed on 24-well plates (Linbro, Flow Laboratories) with 10 % formalin in phosphate- buffered saline prior to assay Formalin-fixed cells do not slough off plates as easily as do unfixed cells, and replicate values were thus more consistent Under these assay conditions, 125 I-EGF (lxl O O ocpm/nmol) saturates the binding assay at 3 n M; assays were performed at 10 % of the saturation value Growth factor concentrations are expressed as ng equivalents of EGF/ml, i e, the amount required to produce an inhibition of 125 I-EGF binding equivalent to that produced by a known amount of EGF.

D Radioimmunoassay Each 50 01 reaction contained the following: 20 m M sodium phosphate at p H 7 4, 200 m M Na Cl, 40 m M dithio- threitol, 0 1 % bovine serum albumin, 0 1 % Na N 3, 125 s I- labeled peptide ( 2 xl O 4 cpm) corresponding to the 17 carboxyl-terminal residues of a TGF (Linsley et al, Proc Natl Acad Sci USA ( 1985) 82:356-369), anti- serum at a final dilution of 1:5000, and other addi- tions as specified The reaction was initiated by the addition of antiserum and was continued at 23 C for min An equal volume of 10 % formalin-fixed S aureus (Pansorbin, Calbiochem) was then added, and incubation was continued for an additional 30 min at 23 C The immunoadsorbant was removed by sedimentation, and the amount of bound 125 I-labeled peptide was measured The amount of bound peptide as corrected for nonspecific binding measured in the absence of antibody (less than % of the total) and expressed as a percentage of maxi- mal binding.

E Wound Healing 1 Mid-dermal Thermal Injuries Mid-dermal thermal injuries were made on the dorsal thorax of anesthetized female Yorkshire pigs ( 30 lbs) whose backs had been shaved and depilatated with commercial hair cream remover A brass template ( 3 x 3 cm, 147 gm) was equilibrated in a 700 C water bath and placed in firm contact with the skin for exactly 10 seconds The resulting blister was then removed Five mid-dermal burns were placed on each side of the spine and were separated from each other by approximately one inch Burns were treated twice a day with approximate- ly 3 ml of vehicle cream (Silvadene R) alone or con- taining growth factor or were untreated After 9 days or 10 days of treatment with natural VGF, the pigs were anesthetized and eschar was removed from the burns.

Biopsies were taken of each burn from re-epithelialized areas.

2 Mid-dermal Donor Graft Injuries A 5-month-old 20 5 kg micropig was anesthe- tized with 20 mg/kg ketamine and 2 mg/kg Rompum The dorsal thorax was shaved, prepped with betadine and thoroughly rinsed with saline A series of six 5 x 5 cm donor sites were made on each side of the dorsal thorax with a Padgett dermatome at 60/1000 inch by taking two swipes at 30/1000 inch Topical therapy included 1 ml of saline in 20 grams of Silvadene distributed evenly between the six wounds on the left side The right side was treated with 1 ml of the growth factor to be tested, in 20 grams of Silvadene divided evenly between the six wounds All wounds were covered with a large burn dressing, chux, ace wrap and gerkin The animal I a,.

'A was anesthetized as described above on post operative days l, 2, 3, 4, 7, 8, 9, 10 and 11 The dressings were removed The wounds were gently wiped with beta- dine and thoroughly rinsed with saline The appropri- ate agent was applied and the wounds were recovered as described above.

Example XII

Biological Activity of Recombinant Growth Factors Prepared in Procaryotic Cells A EGP Receptor Binding This assay determines the ability of a molecule to bind to the EGF receptor as measured by its ability to inhibit the binding of EGF to its recep- tor All growth factors and chimeric growth factors, whether modified or truncated, isolated to date were active in the M'F receptor binding inhibition assay A summary of these results is shown below:

Table

EGF Receptor Bindina of Recombinant Growth Factors Expression Binds to Peptide Cassette Purity EGF Receptor N/TGF modified truncated fusion p B Mll/N/TGF 95 % Yes PAD/n VG Fa - modified truncated fusion p B Mll/PAD/n VG Fa95 % Yes NDP/VG Fa - modified truncated fusion p B Mll/NDP/VGP Fa95 % Yes NDP/VGPA modified I t truncated fusion N/TTV modified truncated chimeric fusion NDP/TTV modified truncated chimeric fusion NDP/VTV modified truncated chimeric fusion NDP/TVV modified truncated chimeric fusion PAK/SGF p B Mll/NDP/VGFA p BMII/N/TTV p BM 11/NDP/TTV p BMII/NDP/VTV p B Mll/NDP/TVV p 3 B Ml 11/PAK/EGF p B M 11/PAK/EGF PAD/EGF EGF PAK/EGF EGF PAK/EGF EGF p Mll/PAD/BGF p BM 11/PAD/EGF Tao Pak/EGF Tac Pak/EF p TC Pt/EGF p TC Pt/EGF % % t % % Yes Yes Yes Yes Yes Yes Yes A comparison of the binding inhibition curves for natural mouse EOP and the bacterially expressed re- combinant chimerio growth factor N/TV (a polypeptide fusion of the 32 N-terminal amino acids of the lambda N-gene and the modified and truncated TGF/VGF hybrid) suggested that there were no differences in the binding activity.

B Mitopenic Activity The activity of several of the purified growth factors was tested and the activity determined in all % % % t t % Yes Yes Yes Yes Yes Yes SEO 101 cases was comparable to the effect caused by EGF, The compounds tested are as indicated below:

Table

Mitogenic Activity of Growth Factor.

Mitogenic Peptide Activity TTV modified truncated chimeric Yes N/TTV modified truncated chimeric fusion Yes As measured by 3 R-thymidine or 125 s I-Id U incorporation.

C Wound Realing 1 Mid-dermal Injuries The effect of natural or synthetic TGF, EGP and VGF, as well as recombinant growth factors on mid- dermal injuries was assessed as described in Example VE 1 The percent of the original burn area which had healed was measured by computer-assisted telemetry, and the percent wound re-epithelialization was determined Untreated wounds were approximately 15 % reepithelialized Treatment with Silvadene alone or Silvadene with ZGF resulted in approximately 50 % reepithelialization, while treatment with synthetic TGF or N 4 tural VGF resulted in approximately 90 % re- epithelialization: The optimal concentration to promote re-epithellalization of EGF was 1-10 Ug/ml, while synthetic TGF and natural VGF produced a maximal response at 0 1 ug/ml.

Experiments simllar to those described above were done to test the effect in wound healing of either 102 TGF, a modified, truncated form of VGF (VG Fa), or a modified, truncated chimeric fusion of TGF and VGF (TTV), all of which were produced by recombinant technology in bacteria These recombinant growth factors and hybrid growth factors accelerated wound healing to the same extent as either synthetic TOF or natural VGF, with an optimal concentration at 0 1 vg/ml.

2 Mid-dermal Donor Graft Injuries Modified truncated VG Fa was assayed for its ability to accelerate wound healing in a donor graft model The treatment regimen was as described above (see Example I) using 1 ml VG Fa, S pg/ml, in 20 g Silvadene Photographs were taken on a daily basis A summary of the results is provided below:

Table

Effect of Recombinant VG Fa on Wound Healinq Treatment Saline VG Fa modified, truncated POD 7 very open open; apparent epitheli- a ization Wound Condition POD 8 POD 9 POD 10 open mostly healed with healed some healing mostly healed healed healed Silvadene is the vehicle POD a Post Operative Day The modified truncated VG Pa accelerated heal- ing of the wound as compared to the carrier control.

a 103 Photographs (not provided) at POD 8 showed substantial differences between saline and VG Fa.

Example XIII

Biological Activity of VGF Prepared in Eukaryotic Cells A VOF Prepared in Monkey Kidney (BSC-1) Cells 1 The medium derived from BSC-1 cells 24 hr after infection with WVV was tested for the presence of mate- rial that could compete with 125 I-EGF for binding to EGF receptor-rich human epidermoid carcinoma cells (A 431) W-infected cells released a potent activity that competed with EGF, which activity is designated VGF Control medium volume from mock-infected BSC-1 control cultures contained minimal activity in competi- tion with EGF.

Monitoring VGF production, at the earliest time examined, 2 hr after infection, enhanced levels of VGF were observed in the culture medium By 12 hr, max- imum amounts of this activity were found in culture supernatant with only a slight increase noted at 24 hr.

The level of VGF production was found to be a function of the multiplicity of virus lnfection as dem- onstrated by the following Table.

I 104 Table

Effect of Multiplicity of Infection on VGF Release 4.

Virus Multiplicity VGF Released pfu per cell nq e Q of EGF per ml o- l O 2 7 4 5 6 1 10 1 Cultures of BSC-l cells were infected at the pfu- to-cell ratio indicated, and the supernatants (approximately 1 ml/2 xl Os cells) were harvested at 24 hr after infection, Samples of each were acidified, lyophilised, and tested in duplicate radioreceptor assays for EGF as described (Delarco and Todaro, Proc Natl Aced Sci USA ( 1978) S:401-405) ng q, nanogram equivalents.

a 2 Partial purification of VGF.

The activity competing with ZOF found in W- infected BSC-1 cells was partially purified from aciod- extracted culture supernatants at 24 hr after infection as described above Acid-solubilised polypeptides ( 10 5 mg) from the supernatants were applied to a Bio- Gel PO column equillbrated with 1 M acetic acid and samples of each fraction tested for activity competing with EGF The major active peak (fraction 42) eluted slightly after the Mr 29,000 carbonic anhydrase marker.

with an apparent Mr of 25,000 The molecular weight 3 was confirmed by utilizing tandemly linked Bio-Sil TSK 250 HPLC sizing columns, with all of the activity elu- ting as a major peak in the region of the Mr 25,000 protein marker One microgram of partially purified VGF was equivalent to 90 ng of EGF in the radioreceptor com- petition assay.

3 Further purification of VGF.

Pooled fractions from the gel filtration col- umn ( 25-35) were concentrated by vacuum centrifugation, resuspended in 0 05 % trifluoroacetic acid (TFA), clari- fied and injected into a 3 9 mm x 30 cm u Bondapak C 18 column (Waters, Milford, MA) Peptides were eluted with a linear 20-60 % gradient of acetonitrile in 0 05 % TFA at a flow rate of 1 O ml/min at 22 C Aliquots of each fraction were assayed in a radioreceptor assay for EGF-competing activity The peptide corresponding to the peak activity was collected and diluted with 0 05 % TFA and reinjected into a p Bondapak column and eluted utilizing isocratic conditions The acetonitrile con- centration was about 22 to 25 %.

A comparison of levels of VGF produced by BSC-1 cells infected with 15 pfu of WVV per cell with that of a TGF produced by retrovirus-transformed cells is shown in the following Table.

Table

Comparison of the Biological Activity -of VGF with a TGF and EGF Stimulation of EGF Receptor Induction of Anchorage-independent Binding Z DNA Synthesis 3 Cell Growth A Growth ng eq of EGF/ml t'2 'IlId U incor Soft-agar Factor 1 of medium porated (cpm/dlsh) colonies/plate None 1,779 < 20 VGF 2 3 3,760 108 TWF-u O 2 Nr 294 EGF 8,482 346 NT a not tested.

1 VGF ( 90 ng equivalents (eq) of EGF/ug of protein) was purified by gel filtration followed by elution from a C 18 pldndapak (Waters Associ- ates) column a TGF was purified from Syer-Theilen feline sarcoma virus-transformed Fisher rat embryo cells as described in Marquardt et al., Proc Natl Acad Sci USA ( 1983) 80:4684-4688; EGF was purified fron mouse submaxllary gland (Cohen et al, J Biol Chem ( 1980) 255:41834-41842).

2 Quantltatia of EF equivalents was based on a standard "r I-E O F binding competition curve as described.

3 Mltogenesis assays were as described; quiescent cultures of diploid human fibroblasts were treated with O l O ng of purified EGF/ml or the same number of EGF equivalent of VGF/aul Values for tl" 2 liodo-deoxyuridine ( "a IlId U) incorporated represent the average of triplicate determinaticns.

4 The number of soft-agar colonies represent the average number of colonies containing a minimum of 20 NRK cells/six random low-power fields days after seeding ( 1 5 x 10 cell/ml) with purified EGF ( 5 ng/ml) or the same number of EGF equivalents of VGF/ml and 2 0 ng of TGF-/ml- purified f rom human platelets as described in Assolaan et al J Blol.

Chem ( 1983) 258:7155-T 7160 Plates or NC eells treated with TGF alone above did not form colonies.

107 B VGF Prepared in CHO Cells Biological activity of VGF prepared in CHO cells was evaluated using the EGF receptor binding assay Culture growth medium was used without further purification The growth factor bound to the EGF receptor.

C VGF Prepared in Silkworm Biological activity of partially purified (approxi- mately 50 %) VGF prepared in silkworm was evaluated us- ing the EGF receptor binding assay and was shown to bind to the EGF receptor.

Table

Percent Epithelialization of Wounds Following Treatment with Growth Factor Pig 1 ( 9 Days Post-Burn) Right Side S Left Side S Pig 2 ( 10 Days Post-Burn Right Side S Left Side S ilvadene Untreated % O % ilvadene Untreated % 55 % ) ilvadene Untreated % 0 % ilvadene Untreated % 0 % Natural VGF ug/ml 0.1 0 1 % 65 % h-EGF ug/ml 0.1 0 1 0 1 % 70 % 30 % VGF ug/ml 0.1 0 5 1 0 % 95 60 % 0.1 0 5 1 0 % 75 % 40 % e -log Table (cont'd) Percent Epithelialization of Wounds Following Treatment with Growth Factor Pig 3 ( 9 Days Post-Burn) Rat TGF ug/ml Right Side Silvadene Untreated 0 1 1 0 10 % 15 % 90 % 65 % 90 % Human TGF ug/ml Left Side Silvadene Untreated 0 1 1 0 10 25 % 5 % 90 % 85 % 65 % VGF: Natural VGF prepared from vaccinia virus infected cells D Immunological Comparison of VGF and TGF In the radioimmunoassay described above, a 50 % displacement of antigen from antibody to the carboxyl- terminal 17 amino acids of rat TGF-a molecule was observed at an antigen concentration of approximately 0.2 to 0 3 ng equivalents of EGF, where 125 I-labeled a TGF competes with a TGF When VGF was tested at equivalent concentrations, no competition was observed In a competitive radioimmunoassay for native EGF, VGF preparations, even when tested at 50 ng of EGF equivalents/ml, exhibited a minimal displacement (< 10 %) of 1251-EGF from a polyclonal antibody to native EGF.

It is evident from the above results, that the VGF is a potent epithelializing agent comparing satis- factorily with other growth factors which have been previously tested and demonstrated to have mitogenic activity The subject polypeptides may be used with various hosts for inducing an immunogenic response, enhancing cellular proliferation, wound healing or the like For wounds, epithelialization and vasculariza- tion is observed, as well as rapid restoration of the strength of the wound, that is, resistance to separa- Oc tion or tearing Hosts include mammals, such as rodents, domestic animals, primates and humans.

In accordance with the subject invention, novel compositions are provided having a wide ranging capability in a variety of fields, such as diagnostics, ir vitro and in vivo effects on cells, acting as mitogens, additives in nutrient media, use as agonists and antagonists to EGF and TGF, acting as immunogens, acting as therapeutics, enhancing wound healing, and the like Furthermore, by providing for a small oligo- peptide having binding activity, the oligopeptide can be used by itself or in combination with other poly- peptides, fused to the other polypeptides to vary the properties of the other polypeptides, resulting in binding of the polypeptide to cells having growth fac- tor receptors Thus, one can reversibly bind various polypeptides to cells having growth factor receptors, affecting the properties of the cells in predetermined ways.

All publications and patent applications mentioned in this specification are indicative of the level of skill of those skilled in the art to which this invention pertains All publications and patent applications are herein incorporated by reference to the same extent as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference.

The invention now being fully described, it will be apparent to one of ordinary skill in the art that many changes and modifications can be made thereto without departing from the spirit or scope of the appended claims.

Ii O -

Claims

CLAIMS:

1 A DNA sequence encoding a polypeptide characterlsed as ( 1} capable of binding to an EGF receptor, {( 2) having substantially the same loop structure as a naturally occurring growth factor capable of binding to an E Gt receptor ("natural ligand") and ( 3) having at least about 30 percent homology with a natural ligand wherein said polypeptide comprises three domains, each of which is an amino acid sequence having at least about 30 % homology to fragments of the same or different natural ligands, and having the same relative position as said fragments of said natural ligands, said domains being defined as:

a first domain of at least about 11 N-terminal amino acids beginning at about amino acid Wa'6 to aa 1 and terminating at about amino acid aall to aas 15; a second domain of at least about 11 central fragment amino acids beginning at about amin'o acid aall to a&Is and terminating at about amino acid aa 25 to aa 29; a third domain of at least about 14 amino C- terminal amino acids beginning at about amino acid aa 25 to aÀ 29 and terminating at about aa 40 to a 553; and optionally said DNA sequence durther encodes a fourth domain of at least about 6 extreme N-terminal amino acids beginning at about aa-7 to a'l and terminating at about amino acid aÀ'45 to aa 7, with optionally an amino acid sequence capable of being selectively cleaved insetted at the carboxy terminus of said fourth domain, jolned to the amino terminus of said first domain with the proviso that where said polypeptide has an amino acid sequence identical to a natural ligand, said DNA sequence encoding said fourth domain is present and encodes other than the extreme N- terminal amino acids of the natural ligand.

l I - 2 The DNA sequence according td Claim 1, wherein said natural ligand is VGP, 'TGF-lpha, MGF, ZGP, 8 P O F or amphiregulin.

3 'The DNA sequence according to Claim 1, wherein said DNA sequence further encodes a leader sequence, with optionally an amino acid sequence capable of being selectively cleaved inserted at the carboxy terminus ot said leader sequence, joined to the amino terminus of said first domain, or when said fourth domain is present, to the amino terminus of said fourth domain.

4 The DNA sequence according to Claim 3, wherein said leader sequence is an N-terminal fragment of a lambda N-protein, lamba era protein, or amphiregulin.

The DNA sequence according to Claim 3, wherein said leader sequence is a signal sequence or a mutant thereof.

S The DNA sequence according to Claim 5, wherein said signal sequence is an alkaline phosphatase signal sequence, a simian TOF-betal signal sequence, or a lactis killer toxin signal sequence.

7 The DNA sequence according to any one of Claims 1-6, wherein at least one of said domains is at least about 30 % homologous to a fragment of VGF, TGOF- alpha, SC? or SFPOF.

8 The DNA sequence according to any one of Claims 1, 4 and 6, wherein said DNA sequence encodes V Gr.

l J ?;N I I - 9 The DNA sequence according to any one o- Claims 1, 4 and 6, wherein said DNA sequence encodes TGP.

The DNA sequence according to any one of Claims 1, 4 and 6, wherein said DNA sequence encodes NIP 11 The DNA sequence according to Claim 1, wherein said DNA sequence encodes SFGF.

12 The DNA sequence according to Claim 1, wherein said sequence comprises at least one alteration from the sequence encoding a natural liqand resulting in the following changes in amino acid sequencet BGD to HG Tj GMYC to GYA Ci RCS to VC Bl CLHGDC to CLHGTC: and GMYCRC to GYACVC.

13 A DNA sequence encoding a polypeptide of interest and leader sequence of about 8 to 45 N- terminal amino acids, including modifications thereof of, from a bacteriophage lambda N-protein or Cro protein, amphiregulin, T 4 gene 32 or a bacterial alkaline phosphatase signal sequence, joined to the amino terminus of said polypeptide of interest, optionally via an amino acid sequence capable of being selectively cleaved, said polypeptide of interest comrising three domains, the sequence of each domain corresponding at least substantially to the sequence of a fragment of a naturally occurring growth factor capable of binding to an ZGF receptor ("natural ligand") and-having the same relative position as said fragments of said natural ligand, said domains being at least about 11 N-terminal amino acids beginning at about amino acid aa'6 to aa 1 and terminating at about amino acid aa 11 to a 15; a second domain of at least about 11 central I 13 - fragment amino acids beginning at about amino acid aa 11 to aa 1 S and terminating at about amino acid aa 25 to Àa 291 a third domain of at least about 14 amino C- terminal amino acids beginning t about amino acid aa 25 to aa 29 and terminating at about aa 40 to a 47 14 The DNA sequence according to Claim 13, wherein said leader sequence is about 32 or 33 N- terminal amino acids of said N-gene.

The DNA sequence according to Claim 14, wherein said polypeptide of interest is TGF-alpha, VGF, VGFA, VGPO Fa or ZGF.

16 The DNA sequence according to Claim 13, wherein slaid leader sequence is an alkaline phosphatase signal sequence or a modified alkaline phosphatase signal sequence.

17 The DNA sequence according to Claim 16, wherein said polypeptide ot intrest is VG Fa, n VGQ Fa, or OF, 1 B The DNA sequence according wherein said polypeptide of interest has amino acid sequence:

X MVVS FNOCPD TQF T C R F LVQ PACVC G I R C Q 1 V V L V O Y O R.

19 The DNA sequence according wherein said polypeptide of interest has amino acid sequence:

1 D-I P A I R L C G P 3 G D G Y TCRF LVQ BK X P A C V C 8 G I R C Q 1 V V L V D Y Q R.

to Claim 14, the following C P R GFG H G Y T to Claim 14, the following CL G G Y T I I i k I 14 - The DNA sequence according to Claim 14, wherein said polypeptide has the following amino acid sequences VV 8 HF N D C PD Q F CZ A R D I D G Y A C V C S RC RVV V DY QR.

21 The DNA sequence according wherein slaid leader sequence is about 21 amino acids of said Cro protein.

22 The DNA sequence according wherein said polypaptide of interest has amino acid sequences XVV H 9T NDCPD H Q P T C R F L V Q EK A C V C 8 0 1 R C Q V V L V D Y QR.

CF GT GYTG to Claim 13, N-terminal to Claim 19, the following C P G R G Y T 23 The DNA sequence according to Claim 13, wherein said leader sequence is about 8 amino N- terminal acids of T 4 gene 32.

24, The DNA sequence according to Claim 23, wherein said polypeptide of interest has the following amino acid sequences VVSH R N D C P D S TQ F C F GT C R LV Q X KP A C V C S GYTG I R C Q H V V L V D Y Q R.

An expression cassette which comprises in the direction of transcription, a transcriptional and translational initiation regulatory region functional in a host celli a DNA sequence encoding a polypeptide characterised as ( 1) capable of binding to an EGFOP receptor, ( 2) having substantially the same loop structire as a naturally occurring growth factor Al l capable of binding to an z GF receptor ("natural ligand") and ( 3) having at least about 30 percent homology with a natural ligand, wherein said polypeptide comprises three domains, each of which is an amino acid sequence having at least about 30 % homology to fragments of the same or different natural ligands, and having the same relative position as said fragments of said natural ligands, said domains being defined as:

& first domain of at least about 11 N-terminal amino acids beginning at about amino acid aa'6 to aa 1 and terminating at about amino acid aall to a 15 e S a second domain of at least about li central fragment amino acids beginning at about amino acid aall to aa 15 and terminating at about amino acid aa 25 to aa 291 a third domain of at least about 14 amino C- terminal amino acids beginning at about amino acid aa 25 to as 29 and terminating at about aa 40 to aa 53; and optionally said DNA sequence further encodes a fourth domain of at least about 6 extreme N-terminal amino acids beginning at about aa 7 to aa-l and terminating at about amino acid aa'45 to aa'7, with optionally an amino acid sequence capable of being selectively cleaved inserted at the carboxy terminus of said fourth domain, joined to the amino terminus of said first domain with the proviso that where said polypeptide has an amino acid sequence identical to a natural ligand, a DNA sequence encoding said fourth domain is present and encodes other than the extreme N- terminal amino acids of the natural ligand and a transcriptional and translational termination regulatory region functional in said host cell.

26 A host cell comprising an expression cassette which includes'in the direction of transcription, a transcriptional and translational i 16 initiation regulatory region functional in a host cell; a DN& sequence encoding a polypeptide characterized as { 1) capable of binding to an EGF receptor, ( 2) having substantially the sams loop structure as a naturally occurring growth factor capable of binding to an EGF receptor ("natural ligand") and ( 3) having at least about 30 percent homology with a natural ligand, wherein said polypaptide comprises three domains, each of which I 1 an amino acid sequence having at least about 30 % homology to fragments of the same or different natural ligands, and having the same relative position as said fragments of said natural ligands, said domains being defined ass a flrst domain of at least about 11 N-terminal amino acids beginning at about amino acid sa-6 to aa 1 and terminating at about amino acid aa 11 to aa 15; a second domain of at least about 11 central fragment amino acids beginning at about amino acid aa 11 to aa 15 and terminating at about amino acid aa 25 to as 291 a third domain of at least about 14 amino C- terminal amino acids beginning at about amino acid aa 25 to aa 29 and terminating at about aa 40 to a& 53 i and optionally said DNA sequence further encodes a fourth domain of at least about 6 extreme N-terminal amino acids beginning at about aa-7 to aa 1 and terminating at about amino acid aa-45 to aa 7, with optionally an amino acid sequence capable of being selectively cleaved inserted at the carboxy terminus of said fourth domain, joined to the amino terminus of said first domain with the proviso that where said polypeptide has an amino acid sequence identical to a natural ligan, a DNA sequence encoding said fourth domain is present and encodes other than the extreme N-terminal amino acids of the natural ligand and a transcriptional and translational termination regulatory region functional in said host cell.

-_O r T; -.

_ 117; 27 A method for producing a polypeptide.

characterl Sed as { 1) capable of binding to an SGP receptor, ( 2) having substantially the same loop structure as a naturally occurring growth factor capable of binding to an El receptor ("natural ligand") and ( 3) having at least about 30 percent homology with a natural SG receptor binder, wherein said polypeptide comprises three domains, each of which is an amino acid sequence having at least about 30 t homology to fragments of the same or different natural ligands, and having the same relative position as siad fragments of said natural ligands, said domains being defined as a first domain of at least about 11 N-terminal amino acids beginnlng at about amino acid aa'6 to aa I and terminating at about amino acid aa 11 to aai S; a second domain of at least about 11 central fragment amino acids beginning at about amino acid aall to aa 15 and terminating at about amino acid aa 25 to a& 29 a third domain of at least about 14 amino C- terminal amino acids beginning at about amino acid aa 25 to aa 29 and terminating at about aa 40 to aa 53; and optionally a fourth domain of at least about 6 extreame N-terminal amino acids beginning at about aa 07 to aa 1 and terminating at about amino acid aa 45 to aa'7, with optionally an amino acid sequence capable of being selectively cleaved inserted at the carboxy terminus of said fourth domain, joined to the amino terminus-of said first domain with the proviso that where said polypaptide has an amino acid sequence identical to a natural ligand, said fourth domain is present and is other than the extreme N-terminal amino acids of the natural ligand; and a transcrilptional and translational termination regulatory region functional in said host call, said method comprising:

I 11 - growing a host cell containing an expression cassette which includes, in the direction of transcription, a transcriptional and translational initiation regulatory region functional in said host cell, a DNA sequence encoding said polypeptide, and'a transcriptional and translational termination regulatory region, in an appropriate culture medium whereby said polypeptide is expressed; and .-:; isolating said polypeptide.

28 l & plasmid comprising the functional domains of p SH Mll/NDO/IGF p R Mll/PAD/GE?; pa MB 1/PAK/EGF; p BM 1 l/N/TGFI p BM Il/NDP/VGFA p B 411/NDP/VG Fai p S Mll/PAD/n VG Fai p 9 Xll/PAX/n VG Fal p RSV/VGFI p Ac/SPGF; p Ac/VG Fp pt Ac/TTV 1 Tac Pak/ZGFI pl T Ct/ZG;F p TC Pt/n VG Fa; p Bl Xl/N/Tl Vl p B Mll/NDP/'m Vl p Bll/NDP/VTVI or p M 1 16/NDP/TVV.

29 A polypeptide characterised as ( 1) capable of binding to an KG? receptor, ( 2) having substantially the same loop structure as a naturally occurring growth factor capable of binding to an EP- receptor ("natural ligand") and ( 3) having at least about 30 percent homology with a natural ZG? receptor binder, wherein said polypeptide comprises three domains, each of which is an amino acid sequence having at least about 30 % homology to fragments of the same or dilfferent natural ligands, and having the same relative position as said fragments of said natural ligands, said domains being defined as:

a first domain of at least about 11 N-terminal amino acids beginning at about amino acid aa 6 to aa 1 and terminating at about amino acid aa 11 to aa 15;l S a second domain of at least about 11 central fragment amino acids beginning at about amino acid aall to aals and terminating at about amino acid aa 25 to aa 29; a third domain of at least about 14 amino Cterminal amino acids beginning at about amino acid aa 25 to aa 29 and terminating at about Àa 4 " to aa 53 p and optionally a fourth domain of at least about 6 extreme N-terminal amino acids beginning at about aa'7 to aa'l and terminating at about amino acid aa'45 to aa'7, with optionally an amino acid sequence capable of being selectively cleaved inserted at the carboxy terminus of said fourth domain, joined to the amino terminus of said first domain with the proviso that where said polypeptide has an amino acid sequence identical to a natural ligand, said fourth domain is present and has a sequence other than that of the extreme N-terminal amino acids from the natural ligand.

A polypeptide comprising the following amino acid sequence or a fragment thereof of at least 37 amino acids comprising at least about amino acids 1 to 37 i aa -24 -20 -15 -10 -5 -1 0 D S O N A I T T 8 P S T N A T T D I P A I R 10 S 15 20 L C G P E G D O Y C L H X 1 G D C I H X 2 A R D 30 35 40 0 G X 3 M Y C R C S a O Y T O S R C Q H V V 50 53 L V D Y Q R S X N P NT wherein X 1 is a bond or one amino acids X 2 is a bond or from 1 to 2 amino acids; x 3 is a-bond or I to 4 amino acids; and Xl, X 2 and 13 may be the same or different.

31 The polypeptide according to Claim 30, wherein at least one of said aa 14 is threonine; said aa 24 is tyrosins; said aa 25 is alanine; and said aa 27 is valine.

_ Ml _ - 32 The polypeptide according to Claim 30, wherein said amino acid sequence comprises aa" 6 to aa 47.

33 The polypeptide according to Claim 32, wherein said aa 14 is threonine, said aa 24 is tyrosine and said aa 25 is alanine.

34 The polypeptlde according to Claim 33, wherein said aa" 6 to aa 13 of said amine acid sequence are replaced with the followings aa -6 -1 1 5 10 13 VVS NF N D C P D 8 H TQFC F HG A polypeptide according to Claim 33, wherein said &a 14 to aa 27 of said amino acid sequence are replaced with the followings as 14 20 25 27 T CRFLVQ ZDKP ACV 36 The polypeptide according to Claim 35, wherein said aa 22 is replaced with glutamic acid.

37 A DNA sequence encoding a polypeptide having a sequence according to any of Claims 30-36.

38 A method as claimed in claim 27, wherein a procedure is followed which is substantially as indicated in the foregoing Examples section.

39 A polypeptide produced by a method as claimed in claim 27 or 38.

Published 1989 at The P&tent Office, State House, 66 '71 Hig Holborn, London W Cl R 4 TP Further copies maybe obtainedfrom The Patent O ice.

Sales Branch, St Mary Cray, Orpington, Kent BR 5 3RD Printed by Multiplex techniques ltd, St Mary Cray, Kent, Con 1/87