METHOD FOR MAKING INSULIN PRECURSORS AND INSULIN PRECURSOR ANALOGUES HAVING IMPROVED FERMENTATION YIELD IN YEAST
BACKGROUND
Yeast organisms produce a number of proteins that have a function outside the ceil. Such proteins are referred to as secreted proteins. These secreted proteins are expressed initially inside the cell in a precursor or a pre-form containing a pre-peptide sequence ensuring effective direction (translocation) of the expressed product across the membrane of the endoplasmic reticulum (ER). The pre-peptide, normally named a signal peptide, is generally cleaved off from the desired product during translocation. Once entered in the secretory pathway, the protein is transported to the Golgi apparatus. From the Qolgi, the protein can follow different routes that lead to compartments such as the cell vacuole or the cell membrane, or it can be routed out of the cell to be secreted to the external medium (Pfeffer et al. (1987) Ann. Rev. Biochem. 56:829-852).
Insulin is a polypeptide honmone secreted by p-cells of the pancreas and consists of two polypeptide chains, A and B, which are linked by two inter-chain disuiphide bridges. Furthermore, the A-chain features one intra-chain disuiphide bridge.
The hoRmone is synthesized as a single-chain precursor proinsulin (preproinsulin) consisting of a prepeptide of 24 amino acid followed by proinsulin containing 86 amino acids, in the configuration: prepeptide - B - Arg Arg - C - Lys Arg -A, in which 0 is a connecting peptide of 31 amino acids. Arg-Arg and Lys-Arg are cleavage sites for cleavage of the connecting peptide from the A and B chains.
Three major methods have been used for the production of human insulin in microorganisms. Two involve Escherichia colt\ with either the expression of a large fusion protein in the cytoplasm (Frank et al. (1981) in Peptides: Proceedings of the 7TH American Peptide Chemistry Symposium (Rich & Gross, eds.), Pierce Chemical Co., Rockford, IL pp 729-739), or use a signal peptide to enable secretion into the peripiasmic space (Chan et al. (1981) PNAS 78:5401-5404). A third method utilizes Saccharomyces cerevisiae to secrete an insulin precursor into the medium (Thim et al, (1986) PNAS 83:6766-6770). The prior art dis-cLoses a limited number of insulin precursors which are expressed in either £ coli or Saccharomyces cerevisiae, vide US 5,962,267, WO 95/16708, EP 0055945, EP 0163529, EP 0347845 and EP 0741188.

SUMMARY OF THE INVENTION
The present invention features novel connecting peptides (mini C-peptides) which confer an increased production yield in insulin precursor molecules and insulin precursor analog molecules when expressed in a transformed microorganism, in particular in yeast. Such insulin precursors or insulin precursor analogs can then be converted into insulin or insulin analogs by one or more suitable, well known conversion steps.
The connecting peptides of the present invention contain at least one Gly and will generally be shorter, than the natural human C peptide which, including the flanldng dibasic cleavage sites, consists of 35 amino acids. Thus the novel connecting peptides will in general not be of more than 15 amino acid residues in length and preferably not more than 10 amino acid residues. Typically the novel connecting peptides will be of up to 9, 8, 7 or 5 amino acid residues and will preferably not be of more than 4 amino acid residues in length.
As in the natural human insulin molecule, the connecting peptide will contain a cleavage site at its C and N termini enabling in vitro cleavage of the connecting peptide from the A and B chains. Such cleavage sites may be any convenient cleavage sites known in the art, e,g. a Met cleavable by cyanogen bromide; a single basic amino acid residue or a pair of basic amino acid residues (Lys or Arg) cleavable by trypsin or trypsin like proteases; Acromo-bactor lyticus protease or by a carboxypeptidase protease. The cleavage site enabling cleavage of the connecting peptide from the A-chain is preferably a single basic amino acid residue Lys or Arg, preferably Lys.
Alternatively cleavage of the connecting peptide from the B chain may be enabled by cleavage at the natural Lys829 amino acid residue in the B chain giving rise to a desBSO insulin precursor or desB30 insulin precursor analog. The desired B30 amino acid residue may then be added by well known in vitro, enzymatic procedures.
In one embodiment the connecting peptide will not contain two adjacent basic amino acid residues (Lys,Arg), In this embodiment, cleavage from the A-chain may be accomplished at a single Lys or Arg located at the N-terminal end of the A-chaIn and the natural Lys in position B29 in the B-chain.
The connecting peptide may comprise more than one Gly but preferably not more than 5. The connecting peptide will preferably not comprise more than 3 Gly and most preferred it will only comprise a single Gly. The Gly may be immediately N-terminal to the cleavage site adjacent to the A chain.
Furthennore, the B27 (atom CG2) will typically have a proximity to the A1 (atom CA) of less than 5 A,
Accordingly, in one aspect the invention is related to insulin precursors or insulin precursor analogs comprising a connecting peptide (C-peptide) being cleavable from the A and

B chains said connecting peptide comprising at least one Gly, wherein the B27 (atom CG2) has a proximity to the A1 (atom CA) of less than 5A.
In another aspect, the present invention is related to insulin precursors or insulin precursor analogs comprising a connecting peptide (C-peptide) being cleavable from the A and B chains said connecting peptide comprising at feast one Gly and a cleavage site enabling cleavage of the peptide bond between the A-chain and the connecting peptide, wherein one Gly is immediately N-terminai to said cleavage site.
In another aspect, the present invention is related to insulin precursors or insulin precursor analogs comprising a connecting peptide (C-peptide) being cleavable from the A and B chains said connecting peptide comprising at least one Gly, wherein the connecting peptide is of up to 6 amino acid residues in length.
In a further aspect, the present invention is related to insulin precursors or insulin precursor analogs comprising a sequence of formula:
B(1-27) - X3 - X2 - X1 - Y - A(1-21). wherein
X1 comprises 1-5 amino acid residues in length comprising at least one Gly,
X2 is one of Pro, Lys, Ala, Arg or Pro-Thr at position 29 of the B chain,
X3 is one of Pro, Asp, Lys, or the at position 28 of the B chain, and Yis Lys or Arg.
In one embodiment X1 is 1-4,1-3 or 1-2 amino acid residues in length.
In a further aspect, the present invention is related to insulin precursors or insulin precursor analogs comprising a sequence of formula:
B(1-27) - X3 - X2 - X1 - Y - A(1-21), wherein X1 comprises a Gly immediately N-terminal to Y,
X2 is one of Pro, Lys, Ala, Arg or Pro-Thr at position 29 of the B chain,
X3 is one of Pro, Asp, Lys, or lie at position 28 of the B chain, and
Y is Lys or Arg.
In one embodiment, Xi is 1-15,1-10. 1-8, 1-5 or 1-3 amino acid residues in length.
In a further aspect, the present invention is related to insulin precursors or insulin precursor analogs comprising a sequence of formula:
B(1-27)-X3 -X2-Xi-Y-A(1-21). wherein X1 comprises at least one Gly,
X2 is one of Pro, Lys, Ala, Arg or Pro-Thr at position 29 of the B chain.
X3 is one of Pro, Asp, Lys, or the at position 28 of tine B chain, and
Y is Lys or Arg.
and wherein the B27 (atom CG2) has a proximity to the A1 (atom CA) of less than 5 A.

In this embodiment X1 is typically 1-15,1-10,1-9,1-8,1-5,1-4 or 1-3 amino acid residues in length.
in the above fomnula X1 will typically contain 1-5 Gly, preferably 1-3 and more pre-fen-ed only one Gly molecule.
The amino acid residues In X1 can be any codable amino acid residue and may be the same or different with the only proviso that at least one amino acid residue in Xi is Gly.
In one embodiment, X3 is Asp and X2 is Lys. This embodiment encompasses the insulin precursor analogs containing an Asp in position B28 of the B chain (termed hereinafter "Asp826P")- In another embodiment X2 is Lys and X3 is Pro. In a further embodiment the sequence X1- Y is selected from the group of: (a) Glu-Glu-Gly-Lys(SEQ ID N0:1, (b) Glu-Gly-Lys, (c) Ser-GIy-Lys, (d) Asn-Gly-Lys, (e) Thr-Gly-Lys, (f) Asp-Gly-Lys, (g) Met-Gly-Lys, (h) Ala-Gly-Lys, (i) His-Gly-Lys and (]) Gly-Lys.
In still further specific embodiments, X1 is 1-3 amino acid residues; X3 is Lys and X2 is Pro. In a further embodiment, X1 is 1-3 amino acid residues, X3 is Asp and X2 is Lys. In another embodiment X2 is Pro, X3 is Lys and X1 is 1-2 amino acid residues of which one is Trp or Phe.
In another embodiment X3 is Lys, X2 is Pro-Thr and X1 consists of up to 15 amino acid residues of which one is Gly. In this embodiment X1 will contain a cleavage site at the C-terminal end, e.g a mono basic or dibasic (Lys, Arg) cleavage site.
In a specific embodiment, the mini C-peptide comprises the sequence Glu-GIy-Lys, Asn-Gly-Lys, or Asp-Gly-Lys.
The present invention is also related to polynucleotide sequences which code for the claimed insulin precursors or insulin precursor analogs. In a further aspect the present invention is related to vectors containing such polynucleotide sequences and to host cells containing such polynucleotide sequences or vectors.
In another aspect, the invention relates to a process for producing the insulin precursors or insulin precursor analogs in a host cell, said method comprising (i) culturing a host cell comprising a polynucleotide sequence encoding the insulin precursors or insulin precursor analogs of the invention under suitable conditions for expression of said precursor or precursor analog; and (ii) isolating the precursor or precursor analog from the culture medium.
In still a further aspect, the invention relates to a process for producing insulin or insulin analogs in a host cell, said method comprising (i) culturing a host cell comprising a polynucleotide sequence encoding an insulin precursor or insulin precursor analogs of the invention; (ii) isolating the precursor or precursor analog from the culture medium and (iii)

converting the precursor or precursor analog into insulin or an insulin analog by in vitro enzymatic conversion.
In one embodiment of the present invention the host cell is a yeast host ceil and in a further embodiment the yeast host cell is selected from to the genus Saccharomyces. In a further embodiment the yeast host cell is selected from the species Saccharomyces cere-visiae.
In a related aspect, the invention features a mini C-peptide in an insulin precursor or insulin precursor analog wherein the amino acid residues of the C-peptide exhibit sufficient flexibility to allow several geometric arrangements of the C-peptide to accommodate an atomic distance between B27 CG2 and A1 CA less than 5 A,
BRIEF DESCRIPTION OF THE DRAWINGS
Fig. 1 represents the pAK721 S. cerevisiae expression pfasmid expressing the LA19 leader- EEAEAEAEPK(SEQ ID NO:2)-lP(AlaAlaLys) fusion protein.
Fig. 2 is the DNA sequence and inferred amino acid sequence of the encoded fusion protein (a-factor-leader-EEAEAEAPK(SEQ ID NO:3)- AspB26lP portion of pAK1150 used as PCR template (SEQ ID N0:4 and 5),
Fig.3 is the DNA sequence encoding a leader-AsplP fusion protein with a synthetic mini C-peptide (DGK or AspGlyLys) generated by randomized optimization (SEQ ID NO:6 and 7). The mini C-peptide (DGK) is indicated by underining.
Fig. 4 shows the solution structures of AspIP(AspGlyLys) as backbone lines of ensemble of 20 converged structures,
Fig.5 shows a ribbon presentation of Aspb26lP(AspGlyLys). The figure is produced using MOLSCRIPT (Kraulis (1991) J. Appl. Crystallog. 24:946-950). Amino acid residue annotation is derived as follows: B1-B29 (B chain) are numbered 1-29, residues C1-C3 (C chain) are numbered 30-32. and residues A1-A21 (A chain) are numbered 33-53.
Fig. 6 is the ID proton NMR spectrum for Asp^®IP(Asp Gly Lys) recorded at 27°C at 1.0 mM concentration in 10%/90% D2O/H2O with 10 mM phosphate buffer at pH 8.0.
Fig. 7 is DNA and inferred amino acid sequence of the expression cassette expressing the YAP3-TA39-GluGluGlyGluProLys(SEQ ID N0:8)-Asp^lP fusion protein with a synthetic mini C-peptide (DGK or AspGlyLys) (SEQ ID N0:9 and 10).
Fig. 8 is DNA and inferred amino acids sequences of the expression cassette expressing the YAP3-TA57-GluGluGlyGluProLys(SEQ ID NO:8)-Asp^2®IP ^^^^^ protein with a synthetic mini C-peptide (DGK or AspGlyLys) (SEQ ID NOS: 11 and 12).

DETAILED DESCRIPTION Abbreviations and nomenclature.
By ''connecting peptide" or "C-peptide" is meant the connection moiety "C" of the B-C-A polypeptide sequence of a single chain preproinsulin-like molecule. Specifically, in the natural insulin chain, the C-peptide connects position 30 of the B chain and position 1 of the A chain. A "mini C-peptide" or "connecting peptide" such as those described herein, connect B29 or B30 to A1, and differ in sequence and length from that of the natural C-peptide,
By "IP" is meant a single-chain insulin precursor in which a desBSO chain is linked to the A chain of insulin via a connecting peptide. The single-chain insulin precursor will contain correctly positioned disuiphide bridges (three) as in human insulin.
With "desBSO" or "B(1-29)" is meant a natural insulin B chain lacking the B30 amino acid residue, "A(1-21)" means the natural insulin A chain, "B(1-27)" means the natural B chain lacking the B28, 829, and B30 amino acid residues; "AspB26P" means a single-chain insulin precursor with aspartic acid at position 28 of the B-chain and no C-peptide (B29 is linked to A1). The mini C-peptide and its amino acid sequence is indicated in the three letter amino acid code in parenthesis following the IP; Thus "Asp®IP(MetTrpLys)" means a single-chain insulin precursor with aspartic acid at position 28 of the B-chain arid a mini C-peptide with the sequence Met-Trp-Lys connecting B29 to A1,
By "insulin precursor" is meant a single-chain polypeptide which by one or more subsequent chemical and/or enzymatic processes can be converted into human insulin.
By "insulin precursor analog" is meant an insulin precursor molecule having one or more mutations, substitutions, deletions and or additions of the A and/or B amino acid chains relative to the human insulin molecule. The insulin analogs are preferably such wherein one or more of the naturally occurring amino acid residues, preferably one, two, or three of them, have been substituted by another codable amino acid residue, in one embodiment, the instant invention comprises analog molecules having position 28 of the B chain altered relative to the natural human insulin molecule. In this embodiment, position 28 is modified from the natural Pro residue to one of Asp, Lys, or lie. In a preferred embodiment, the natural Pro residue at position B28 is modified to an Asp residue. In another embodiment Lys at position 629 is modified to Pro; Also, Asn at position A21 may be modified to Ala, Gin, Glu, Gly, His, lie, Leu, Met, Ser, Thr, Trp. Tyr or Val, in particular to Gly, Ala, Ser, or Thr and preferably to Gly. Furthermore, Asn at position 83 may be modified to Lys. Further examples of insulin precursor analogs are des(B30) human insulin, insulin analogs wherein Phe has been de- leted; insulin analogs wherein tiie A-chain and/or the B-chain have an N-terminal extension

and insulin analogs wherein the A-chain and/or the B-chafn have a C-terminal extension. Thus one or two Arg may be added to position B1.
The term "immediately N-terminai to" is meant to Illustrate the situation where an amino acid residue or a peptide sequence is directly linked at its C-terminal end to the N-tenminal end of another amino acid residue or amino acid sequence by means of a peptide bond.
In the present context, the term "functional analog of insulin" and the lilce, is meant to indicate a polypeptide with a similar biological action as the native human insulin protein.
By a distance shorter than 5 A between two amino acid residues is meant the shortest inter-atomic distance less than 5 A between any atom in the first amino acid and any atom in the second amino acid. Atomic distances are measured from three-dimensional structures of the molecule determined either by NMR (WQthrich, K,, 1986, NMR of Proteins and Nucleic Acids, Wiley, New York) or by X-ray crystallography (Drenth, J., 1994, Principles of Protein X-ray crystallography, Springer Veriag Berlin). A distance from one amino acid to another is measured as the shortest inter-atomic distance between any atom in the first amino acid and any atom in the second amino acid if not stated differently.
The present invention features novel mini C-peptides connecting position 29 of the insulin B chain and position 1 of the insulin A chain which significantly increased production yields in a yeast host ceil. By the term 'significantly increased production," "increased fermentation yield," and the like, is meant an increase in secreted amount of the insulin precursor molecule or insulin precursor analog molecule present in the culture supernatant compared to the yield of an insulin precursor or insulin precursor analog with no aromatic amino acid residue in the mini C peptide. An "increased" fermentation yield is an absolute number larger than the control; preferably, the increase is 50% or more larger than the control (AspB® IP) level; even more preferably, the increase is 100% or more larger than control levels.
"POT is the Schizosaccharomyces pombe triose phosphate isomerase gene, and "TPI1" is the S. cerevisiae triose phosphate isomerase gene.
By a "leader" is meant an amino acid sequence consisting of a pre-peptide (the signal peptide) and a pro-peptide.
The term "signal peptide" is understood to mean a pre-peptide which is present as an N-teniiinal sequence on the precursor form of a protein. The function of the signal peptide is to allow the heterologous protein to facilitate translocation into the endoplasmic reticulum. The signal peptide is nom^ally cleaved off in the course of this process. The signal peptide 5 may be heterologous or homologous to the yeast organism producing the protein. A number of signal peptides which may be used with the DNA constnjct of tine invention including yeast

aspartic protease 3 (YAPS) signal peptide or any functional analog (Egel-Mitani et al. (1990) YEAST 6:127-137 and US 5.726,038) and the a-factor signal of the MFa1 gene (Thomer (1981) in The Molecular Biology of the Yeast Saccharomyces cerevisiae, Strathem et a!,, eds„ pp 143-180, Cold Spring Harbor Laboratory, NY and US 4,870,00.
The term "pro-peptlde" means a polypeptide sequence whose function is to allow the expressed polypeptide to be directed from the endoplasmic reticulum to the Golgi apparatus and further to a secretory vesicle for secretion into the culture medium (i.e. exportation of the polypeptide across the cell wall or at least through the cellular membrane into the periplasmic space of the yeast cell). The pro-peptide may be the yeast a-factor pro-peptide, vide US 4,546,082 and 4,870,008. Alternatively, the pro-peptide may be a synthetic pro-peptide, which is to say a pro-peptide not found in nature. Suitable synthetic pro-peptides are those disclosed in US 5.395,922; 5,795,746; 5,162,498 and WO 98/32867. The pro- peptide will preferably contain an endopeptidase processing site at the C-terminal end, such as a Lys-Arg sequence or any functional analog thereof.
The polynucleotide sequence of the invention may be prepared synthetically by established standard methods, e.g. the phosphoamidite method described by Beaucage et al. (1981) Tetrahedron Letters 22:1859-1869, or the method described by Matthes et al. (1984) EMBO. Journal 3:801-805. According to the phosphoamidite method, oligonucleotides are synthesized, for example, in an automatic DNA synthesizer, purified, duplexed and ligated to form the synthetic DNA construct. A currently preferred way of preparing the DNA construct is by polymerase chain reaction (PCR).
The polynucleotide sequence of the invention may also be of mixed genomic, cDNA, and synthetic origin. For example, a genomic or cDNA sequence encoding a leader peptide may be joined to a genomic or cDNA sequence encoding the A and B chains, after which the DNA sequence may be modified at a site by inserting synthetic oligonucleotides encoding the desired amino acid sequence for homologous recombination in accordance with well-known procedures or preferably generating the desired sequence by PCR using suitable oligonucleotides.
The invention encompasses a vector which is capable of replicating in the selected microorganism or cell line and which carries a polynucleotide sequence encoding the Insulin precursors or insulin precursor analogs of the invention. The recombinant vector may be an autonomously replicating vector, /,e., a vector which exists as an extra-chromosomal entity, the replication of which is independent of chromosomal replication, e.g., a plasmid, an extra-chromosomal element, a mini-chromosome, or an artificial chromosome. The vector may contain any means for assuring self-replication. Alternatively, the vector may be one which, when introduced into the host cell, is integrated into the genome and replicated together with

the chromosome(s) into which it has been integrated. Furthermore, a single vector or plasmid or two or more vectors or plasmids which together contain the total DNA to be introduced into the genome of the host cell, or a transposon may be used. The vectors may be linear or dosed circular plasmids and will preferably contain an element(s) that permits stable integration of the vector into the host cell's genome or autonomous replication of the vector in the cell independent of the genome.
In a preferred embodiment, the recombinant expression vector is capable of replicating in yeast organisms. Examples of sequences which enable the vector to replicate in yeast are the yeast plasmid 2 |im replication genes REP 1-3 and origin of replication.
The vectors of the present invention preferably contain one or more selectable markers which permit easy selection of transformed cells, A selectable marker is a gene the product-of which provides for biocide or viral resistance, resistance to heavy metals, prototrophy to auxotrophs, and the like. Examples of bacterial selectable markers are the dal genes from Bacillus subtilis or Bacillus licheniformis, or markers which confer antibiotic resistance such as ampicillin, kanamycin, chloramphenicol or tetracycline resistance. Selectable markers for use in a filamentous fungal host cell include amdS (acetamidase), argB (ornithine carbamoyl-transferase), pyrG (orotidine-5'-phosphate decarboxylase), sC (sulfate adenyltransferase) and trpC (anthranilate synthase. Suitable markers for yeast host cfejls are ADE2, HIS3, LEU2, LYS2, MET3, TRP1, and URA3. A preferred selectable marker for yeast is the Schi-zosaccharomyces pompeJP] gene (Russell (1985) Gene 40:125-130).
In the vector, the polynucleotide sequence is operably connected to a suitable promoter sequence. The promoter may be any nucleic acid sequence which shows transcriptional activity in the host cell of choice including mutant, truncated, and hybrid promoters, and may be obtained from genes encoding extracellular or intracellular polypeptides either homologous or heterologous to the host cell.
Examples of suitable promoters for directing the transcription in a bacterial host cell, are the promoters obtained from .the £ coH lac operon, Streptomyces coelicolor agarase gene (dagA), Bacillus subtilis levansucrase gene (sacB), Bacillus licheniformis alpha-amylase gene {amyL), Bacillus stearothermoptiilus maltogenic amylase gene {amyM), Bacillus amylo-liquefaciens alpha-amylase gene {amyQ), and Bacillus licheniformis penicillinase gene (penP). Examples of suitable promoters for directing the transcription in a filamentous fungal host cell are promoters obtained from the genes for Aspergillus oryzae TAKA amylase, Rhi-zomucor miehei aspartic proteinase, Aspergillus niger neutral alpha-amylase, and Aspergillus niger acid stable alpha-amylase. In a yeast host, useful promoters are the Sac-charomyces cerevisiae Mai, TPl, ADH or PGK promoters.

The polynucleotide construct of the invention will also typically be operably connected 0 a suitable terminator. In yeast a suitable terminator is the TPI terminator (Alber et al. 1982) J. Mol. AppI, Genet. 1:419-434).
The procedures used to llgate the polynucleotide sequence of the invention, the pro-noter and the terminator, respectively, and to insert them into suitable yeast vectors contain-ng the information necessary for yeast replication, are well known to persons skilled in the art. It will be understood that the vector may be constructed either by first preparing a DMA construct containing the entire DNA sequence of the invention, and subsequently inserting this fragment into a suitable expression vector, or by sequentially inserting DNA fragments containing genetic information for the individual elements (such as the signal, pro-peptide, mini C-peptide, A and B chains) followed by ligation.
The present invention also relates to recombinant host cells, comprising a colynucleotide sequence encoding the insulin precursors or the insulin precursor analogs of the invention. A vector comprising such polynucleotide sequence is introduced into the host sell so that the vector is maintained as a chromosomal integrant or as a self-replicating extra-chromosomal vector as described eariier. The term "host cell" encompasses any progeny of a parent cell that is not identical to the parent cell due to mutations that occur during replication. The choice of a host cell will to a large extent depend upon the gene encoding the polypeptide and its source. The host cell may be a unicellular microorganism,, e.g., a prokaryote, or a non-unicellular microorganism, e.g., a eukaryote. Useful unicellular cells are bacterial cells such as gram positive bacteria including, but not limited to, a Bacillus cell, Streptomyces cell, or gram negative bacteria such as £ coll and PseUdomonas sp. Eukaryote cells may be mammalian, insect, plant, or fungal cells. In a preferred embodiment, the host cell is a yeast cell. The yeast organism used in the process of the invention may be any suitable yeast organism which, on cultivation, produces large amounts of the insulin precursor and insulin precursor analogs of the invention. Examples of suitable yeast organisms are strains selected from the yeast species Saccharomyces cerevisiae, Saccharomyces Iduyveri, Schh zosaccharomyces pombe, Sacchoromyces uvarum, Kluyvercmyces lactls, Hansenula poly-morpha, Pichia pastoris, Pichia methanoilca, Pichia kluyveri, Yanvwia lipolytica, Candida sp,, Candida utilis, Candida cacaoi, Geothchum sp., and Geotrichum fennentans.
The transformation of the yeast cells may for instance be effected by protoplast fornia-tion followed by transformation in a manner known per se. The medium used to cultivate the cells may be any conventional medium suitable for growing yeast organisms. The secreted insulin precursor or insulin precursor analogs of the invention, a significant prpportion of which will be present in the medium in correctly processed form, may be recovered from the medium by conventional procedures including separating the yeast cells from the medium by

centiifugation, filtration or catching the insulin precursor or insulin precursor analog by an ion exchange matrix or by a reverse phase absorption matrix, precipitating the proteinaceous components of the supernatant or filtrate by means of a salt, e.g. ammonium sulphate, followed by purification by a variety of chromatographic procedures, e.g. ion exchange chromatography, affinity chromatography, or the like.
The insulin precursors and insulin precursor analogs of the invention may be expressed with an N-terminal amino acid residue extension, as described in U,S, Patent No, 5,395,922, and European Patent No. 765,395A, both of which patents are herein specifically incorporated by reference. The extension is found to be stably attached to the. insulin precursor or insulin precursor analogs of the invention during fermentation, protecting the N-terminal end of the insulin precursor or insulin precursor analog against the proteolytic activity of yeast proteases such as DPAP. The presence of an N-terminal extension on the insulin precursor or insulin precursor analog may also serve as a protection of the N-terminal amino group during chemical processing of the protein, i.e. it may serve as a substitute for a BOC (t-butyl-oxycarbonyl) or similar protecting group.
The N-terminai extension may be removed from tiie recovered insulin precursor or insulin precursor analog by means of a proteolytic enzyme which is specific for a basic amino acid (e.g.. Lys) so that the terminal extension is cleaved off at the Lys residue. Examples of such proteolytic enzymes are trypsin or Achromobacter lythus protease .
After secretion to the culture medium and recovery, the insulin precursor or insulin precursor analogs of tiie invention will be subjected to various in vitro procedures to remove the possible N-terminal extension sequence and the mini C-peptide to give insulin or the desired insulin analog. Such methods include enzymatic conversion by means of trypsin or an Achromobacter lyticus protease in the presence of an L-threonine ester followed by conversion of the threonine ester of the insulin or insulin analog into insulin or the insulin analog by basic or acid hydrolysis as described in US patent specification No, 4,343,898 or 4,916,212 or Research Disclosure, September 1994/487 the disclosures of which are incorporated by reference hereinto.
As described below, IPs with synthetic C-peptides were constructed featuring a Gly residue (Examples 1). A Saccharomyces cerevisiae expression plasmid containing a DNA sequences of formula 1 was constiucted by PCR and used to transform a S. cerevisiae host cell. The amount of insulin analog produced was measured as a percentage of the control level Asp^^lP lacking mini C-peptide (Table 1); The novel C-peptides of tine invention containing a Gly in the sequence Xi of the mini C-peptide increased yields by up to 4-fold levels. As described below in Example 2 for Asp^^®IP(Asp Gly Lys), the mini C-peptides of the invention result in a region of flexibility between B27 and A1 which all allow a proximity of

A1 to B27 measured as the atomic distance between A1 (atom CA) and B27 (atom CG2) (e.g., less than 5 A). Accordingly, the invention encompasses mini C-peptide constructs which induce the structural effects shown in Example 2 below.
The present invention is described in further detail in the following examples which are not in any way intended to limit the scope of the invention as claimed. The attached Figures are meant to be considered as integral parts of the specification and description of the invention. All references cited are herein specifically incorporated by reference for all that is described therein.
EXAMPLES General Procedures
All expressions plasmids are of the C-POT type, similar to those described in HP 171 142, which are characterized by containing the Schizosaccharomyces pombe triose phosphate isomerase gene (POT) for the purpose of plasmid selection and stabilization in S. cer-evisiae. The plasmids furthermore contain the S. cerevisiae triose phosphate isomerase promoter and terminator. These sequences are similar to the corresponding sequences in plasmid pKFN1003 (described in WO 90/100075) as are all sequences except the sequence of the EcoRl-Xbal fragment encoding the fusion of the leader and insulin precursor. In order to express different fusion proteins, the £coRI-Xdal fragment of pKFN1003 is simply replaced by an EcoRI-Xbal fragment encoding the leader insulin precursor or leader insulin precursor analog of interest. Such EcoRl-Xba\ fragments may be synthesized using synthetic oligonucleotides and PCR according to standard techniques.
Yeast transformants were prepared by transfomiation of the host strain: S. cerevisiae
strain MT663 {MATalMATa pep4-3/pep4-3 HIS4/his4 tpi::LEU2/tpi::LEU2 Cir^. The yeast strain MT663 was deposited in the Deutsche Sammlung von Mikroorganismen und Zellkul-turen in connection with filing WO 92/11378 and was given the deposit number DSM 6278.
MT663 was grown on YPGaL (1% Bacto yeast extract, 2% Bacto peptone, 2% galactose, 1% lactate) to an O.D. at 600 nm of 0.6. 100 ml of culture was harvested by centrifuga-tion, washed with 10 ml of water, recentrifuged and resuspended in 10 ml of a solution containing 1.2 M sorbitol, 25 mM Na2EDTA pH = 8.0 and 6.7 mg/ml dithiotreitoL The suspension was incubated at 30°C for 15 minutes, centrifuged and the cells resuspended in 10 ml of a solution containing 1.2 M sortDitol, 10 mM NazEDTA, 0.1 M sodium citrate, pH 0 5,8, and 2 mg Novozym®234. The suspension was incubated at 30*'C for 30 minutes, the cells collected by centrifugation, washed in 10 ml of 1,2 M sorbitol and 10 ml of CAS (1.2 M sorbitol, 10 mM CaCl2. 10 mM Tris HCI (Tris = Tris(hydroxymethyi)aminomethane) pH = 7.5) and resuspended in 2 ml of CAS. For transfomiation, 1 ml of CAS-suspended cells was mixed with

approx, 0.1 mg of plasmid DMA and left at room temperature for 15 minutes. 1 ml of (20% polyethylene glycol 4000, 10 mM CaCl2, 10 mM Tris HCI, pH = 7.5) was added and the mixture left for a further 30 minutes at room temperature. The mixture was centrifuged and the pellet resuspended in 0.1 ml of SOS (1.2 M sorbitol, 33% v/v YPD. 6,7 mM CaCy and incubated at 30**C for 2 hours. The suspension was then centrifuged and the pellet resuspended in 0.5 ml of 1,2 M sorbitol. Then, 6 ml of top agar (the SC medium of Sherman et al. (1982) Methods in Yeast Genetics, Cold Spring Harbor Laboratory) containing 1.2 M sorbitol plus 2,5% agar) at 52*^0 was added and the suspension poured on top of plates containing the same agar-solidified, sorbitol containing medium.
S. cerevisiae stain MT663 transformed with expression plasmids were grown in YPD for 72 h at SO^'C. Quantitation of the insulin-precursor yield in the culture supematants was performed by reverse-phase HPLC analysis with human insulin as an external standard (Snel & Damgaard (1988) Proinsulin heterogenity in pigs. Horm. Metabol, Res. 20:476-488).
Example 1.
Construction of Synthetic C-peptides With a Glycine Residue.
Synthetic genes encoding fusion proteins consisting of Asp^^lP associated with a leader sequence consisting of a pre-peptide (signal peptide) and a pro-peptide, were constnicted us-ingPCR'under standard conditions (Santbrook et al, (1989) Molecular Cloning, Cold Spring Harbor Laboratory Press) and E.H.F. polymerase (Boehringer Mannheim GmbH, Sandhoefer Strasse 116, Mannheim, Germany). The resulting DMA fragments were isolated and digested with endonucleases and purified using the Gene Clean kit (BiolOI Inc., La Jolla, CA, USA). Standard methods were used for DNA ligation and transformation of E, coli cells were performed by the CaCb method (Sambrook et al. (1989) supra). Plasmids were purified from transformed £ coli cells using QIAGEN columns (QIAGEN, Hilden. Germany). Nucleotide sequences were determined using the ALF Pharmacia Biotech DNA sequencing system with purified double-stranded plasmid DNA as template. Oligonucleotide primers for PCR were obtained from DNA technology (Ariius, Denmark),
Secretory expression of the Asp^^lP in S. cerevisiae was performed using S. cerevisiae strain MT663 and the 2 |j.m based yeast expression vector CPOT (see Fig. 1) as described in Thim, L. et al, (1986) Proa Natl. Acad. ScLUSA 83:6766-6770. The yeast expression vector contains the Schizosaccharomyces pombe triose phosphate isomerase gene (POT) for plasmid selection and stabilization in S. cerevisiae. Furthennore, the S. cerevisiae triose phosphate isomerase gene {TPI1) promoter and terminator are used for transcription initiation and termination of the recombinant gene encoding the leader-Asp^^lP fusion protein. Secretion

of the Asp26lP was facilitated by the a-factor leader, although a variety of known yeast leader sequences may be used.
As shown in Fig. 1, the pAK721 S. cerevisiae expression plasmid expressing the LA19 leader-EEAEAEAEPK(SEQ ID N0:2)-1P fusion protein was constructed based on the S. cerevisiae-E. coli shuttle POT plasmid (U.S. patent 5,871,957). L-IP indicates the fusion protein expression cassette encoding the leader-IP fusion protein, TPl-PROMOTER is the S. cerevisiae TPI1 promoter and TPI-TERMINATOR is the S. cerevisiae TPH terminator; TPl-POMBE indicates the S. pombe POT gene used for selection in S. cerevisiae] ORIGIN indicates a S. cerevisiae origin of replication derived from the 2|xm plasmid; AMP-R indicates the p-!actamase gene conferring resistance toward ampicillin, facilitating selection in £ coii and
ORIGIN-PBR322 indicates an £ co//origin of replication.
DMA encoding a number of fusion proteins of leader sequences and Asp^^lP with different mini-C-peptides was generated by PCR using appropriate oligonucleotides as primers, as described below. Standard methods were used to subclone DMA fragments encoding the leader-AspB26lP fusion proteins into the CPOT expression vector in the following configuration: leader-Lys-Arg-spacer-AspB26®IP, where Lys Arg is a potential dibasic endoprotease processing site. To optimize processing of the fusion protein by the S. cerevisiae Kex2.endoprotease, DNA encoding a spacer peptide, e.g. EEAEAEAPK (SEQ ID NO:3), was inserted between the DNA.encoding the leader and the AspB26lP (Kjeldsen et al. (1996) Gene 170, 107-112.), However, the present of the spacer peptide is not mandatory. The mature AspB26lP was secreted as a single-chain N-terminally extended insulin precursor analogue with a synthetic mini C-peptide connecting LysB29 and Gly^\ After purification of the Asp^^lP and proteolytic removal of the N-terminal extension and the synthetic mini C-peptide, a threonine amino acid residue (Thr^^°) may be added to Lys^^^ by enzyme-mediated transpeptidation, to generate Asp^^° human insulin (Markussen, et al. (1987) in "Peptides 1986" (Theodoropoulos, D., Ed.), pp. 189-194, Walter de Gruyter & Co., Berlin,).
Development of synthetic mini C-peptides was performed by randomization of one or more codon(s) encoding the amino acids in the mini C-peptide. All synthetic mini C-peptides feature an enzymatic processing site (Lys) at the C-terminus which allows enzymatic removal of the synthetic mini C-peptide (U.S. Patent No. 4,916,212, herein specifically incorporated by reference). Randomization was performed using doped oligonucleotides which introduced codon(s) variations at one or more positions of the synthetic mini C-peptides. Typically one of the two primers (oligonucleotides) used for PCR was doped. An example of an oligonucleotides pair used for PCR generation of leader-AsplP with randomized synthetic mini C-peptides used to generated synthetic mini C-peptides with the general fomiula: Xaa-Gly-Lys (XGK) are as follows:

Primer A: 5TAAATCTATAACTACAAAAAACACATA-3' (SEQ ID NO:13) and
Primer B: 3'-CCAAAGAAGATGTGACTGTTCNNMCCCTTCCCATAGCAACTTGTTACAACAT-
GAAGATAGACAAGAAACATGGTTAACCTTTTGATGACATTGATCAGATCTTTGATTC-S'
(SEQ ID NO: 14), where N is A, C, G, or T and M is C or A.
PCR was typically performed as indicated below: 5 p.1 Primer A (20 pmol), 5 |al Primer B
(20 pmol), 10 ^il 10X PCR buffer, 8 pi dNTP mix, 0.75 ^il E.H.F. enzyme, 1 \x\ pAK1150 plasmid
as template (approximately 0.2 pg DMA) (SEQ ID N0:3), and 70.25 pi distilled water.
Typically between 10 and 15 cycles were performed, one cycle typically was 94*^ C for
45 sec; 55*" C for 1 min; 72*^ C for 1.5 min. The PCR mixture was subsequently loaded onto an
2 % agarose gel and electrophoresis was performed using standard techniques. The resulting DIslA fragment was cut out of the agarose gel and isolated by the Gene Clean kit.
Fig. 2 shows the sequence of pAK1150 DNA used as template for PCR and infened amino acids of the encoded fusion protein (a-factor"leader-EEAEAEAPK(SEQ ID N0:3)-Asp^^lP of pAK1150 (SEQ ID NO:4 and 5). The pAK1150 plasmid is similar to pAK721 shown in Fig. 1. The a-factor-leader's C-terminus was modified to introduce a Nco 1 re.striction endonuclease site, which changes the inferred amino add sequences lihked to LysArg from SerLeuAsp to SerMeta. Moreover, the encoded AspB26lP does not feature a mini C-peptide but LysB29 is directly connected to GlyA1
The purified PCR DNA fragment was dissolved in water and restriction endonucleases buffer and digested with suitable restriction endonucleases (e.g. Bgl ll and Xba ]) according to standard techniques. The Bglll-Xbal DNA fragments were subjected to agarose electrophoresis and purified using The Gene Clean Kit
The expression plasmid pAK1150 or a similar plasmid of the CPOT type (see Fig. 1) was digested with the restriction endonucleases Bgl II and Xba I and the vector fragment of 10765 nucleotide basepairs isolated using The Gene Clean Kit.
The two digested and isolated DNA fragments (the vector fragment and the PCR fragment) were ligated together using T4 DNA ligase and standard conditions. The ligation mix was subsequently transformed into a competent E. coli strain (R-, M+) followed by selection with ampicillin resistance. Plasmids from the resulting £ coli's were isolated using QIAGEN columns.
The plasmids were subsequently used for transformation of a suitable S. cerevisiae
strainMT663 (MATa/MATa pep4-3/pep4-3 HIS4/his4 tpi::LEU2Api::LEU2 Cir*). Individual transformed S. cerevisiae clones were grown in liquid culture, and the quantity of Asp^^lP

secreted to the culture supematants were determined by RP-HPLC. The DNA sequence encoding the synthetic mini C-peptide of the expression plasmids from S. cerevisiae clones secreting increased quantity of the AspB26lP were then determined. Subsequently, the identified synthetic mini C-peptide sequence might be subjected to another round of randomization optimization.
An example on a DNA sequence encoding a leader-Asp^^®lP(AspGlyLys) fusion protein featuring a synthetic mini C-peptide (AspGlyLys) resulting from the randomized optimization process described are shown in Fig. 3 (SEQ ID NO:6 and 7).
Table 1 shows the insulin analogue precursors generated by the above method and production yield expressed as a percent of the control. Fermentation was at 30°C for 72 h in 5
ml YPD. Yield of the insulin precursor analogs was determined by RP-HPLC of the culture supernatant, and is expressed relative to the yield of insulin precursor of a control strain.. In the table, "a" indicates an a-factor leader in which the C-temninus up to the LysArg has
been modified from SLD (SerLeuAsp) to SMA (SerMetAla) and "ex4" is an N-terminal extension peptide with the amino acid sequence EEAEAEAPK (SEQ ID N0:3).


Example 2. Structure Determination of AspP(AspG{yLys) in Aqueous Solution by NMR Spectroscopy.
NMR spectroscopy. Samples for NMR were prepared by dissolving the lyophillzed protein powder in 10/90 D2O/H2O with a 10 mM phosphate buffer and adjusting the pH as desired by addition of small volumes of 1 M DCl or NaOD. All pH meter readings are without correction for isotope effects. Samples of AspB26lP(AspGlyLys) for NMR were prepared at concentrations ranging from 25M to 1mM at pH 8.0. Two-dimensional 1H-1H NMR spectra of ImM samples,
DQF-COSY (Piantini et al. (1982) J, Am. Chem. Soc. 104:6800-6801. Ranee et al. (1983) Biochem. Biophys. Res. Commun, 117:479-485), TOCSY (Braunschweiier et al. (1983) J. Magn. Reson. 53:521-528, Bax et al. (1985) J. Magn. Reson. 65:355-360) and NOESY (Jee-ner et al. (1979) J. Chem, Phys. 71:4546-4553) were recorded at 600 MHz on a Varian Unity Inova NMR spectrometer equipped witti a ^H/^^C/^^N tiiple resonance probe with a self-shielded triple-axis gradient coil using standard pulse sequences from the Varian user library. The operating temperature was set to 27°C. For each phase sensitive two-dimensional NMR specbiim
512 ti increments were acquired each with 2048 or 4096 real data points according to the TPPI-States method (Marion et al, (1989) J. Magn. Reson. 85:393-399). Spectral widths of 6983 Hz in both dimensions were used, with the earner placed exactly on the water resonance which was attenuated by using either saturation between scans for 1,5 seconds or selective excitation by a gradient-tailored excitation pulse sequence (WATERGATE, Piotto et al. (1992) J, Biomol. NMR 2:661-665). DQFCOSY spectra were recorded using a gradient enhanced version applying magic-angle gradients (Mattiello et al, (1996) J. Am, Chem. Soc. 118:3253-3261), For TOCSY spectra mixing times between 30 and 80 ms were used and for NOESY mixing times between 50 and 200 ms.
The processing of the two-dimensional NMR spectra was performed using the software package Xwinnmr (version 2,5, NMR processing software from Bruker Analytische Messtech-nik GmbH, D-76275 Ettlingen, Germany). Each dimension was processed witii shifted sine-bell apodization and zero-filling perfomried once in each dimension. Baseline connections were applied if necessary using Xwinnmr standard procedures. The spectral assignment, cross peak integration, sequence specific assignment, stereo specific assignment, and all other bookkeeping were performed using the program PRONTO (PRONTO Software Development and Distribution. Copenhagen Denmark) (Kjasr et al. (1991) AM TO ASI Series (Hoch, J, C, Redfield C, & Poulsen, F. M., Eds.) Plenum, New York). Chemical shifts are measured in ppm and the water resonance set to 4.75 ppm.
Structure calculations. Distance restraints for the subsequent structure calculation were obtained from integrated NOESY cross peaks classified as either weak, medium orsti-ong cor-

esponding to upper distance restraints of 5.5, 3.3, and 2.7 A, respectively. For distance re-straints involving methyl groups, an additional 0.5 A was added to the upper limit (Wagner et al, 1985) J. Moi. Biol. 196:611-639). Structure calculations were performed using the hybrid method combining distance geometry (Crippen et al. (1988) Distance Geometry and Molecular Conformation, Research Studies Press, Taunton, Somerset, England; Kuszewski et al. (1992) J.Biomol NMR 2:33-56) and simulated annealing based on the ideas of Nilges et al. (1988) FEBS Lett. 229:317-324 using X-PLOR 3.0 (Brunger (1992) X-PLOR Version 3.1: A System for X-ray Crystallograpliy and NMR ,Yale University Press, New Haven) according to the examples given by the X-PLOR manual (dg_sub_embed.inp, dgsa.inp, refine.inp). Residue numbers are derived from standard insulin residue numbering, residues in the B-chain are numbered B1-29, residues in the C-peptide (e. g. AspGlyLys) are numbered C1-C3 and residues in the A-chain are numbered A1-A21.
Spectral assignment of the NMR spectra followed for most resonances the standard sequential assignment procedure described by Wiithrich (1986 NMR of Proteins and Nucleic Acids, Wiley, New York). The standard assignment procedure fails when the amid proton of a particular amino acid residue exchanges to rapidly with protons in the water. At pH 8.0 this occurs for several amino acid residues, however, comparison with earlier mutant insulin NMR spectral assignments and identification of neighboring (in space) amino acid residues through NOEs allow an almost total spectral assignment. Analysis of the NOESY spectra showed that several amino add residues had a NOE network to the surrounding residues similar to what has previously been determined for other insulin molecules, i.e., human insulin HisB1g mutant (Ludvigsen et al. (1994) Biochemistry 33:7998-8006) and these similar connections are found for residues B1-B10, B13-B14, B17-B24 and A4-A21. Additionally the dihedral angle restraints for the above listed residues were adopted from those used previously (Ludvigsen et al. (1994) supra).
Several amino acids in particular B27-B29, C1-C3, A1-A3 have cross peaks patterns which are consistent with peptide chains that are less well ordered than commonly well-defined secondary structural elements. Thus additional NOEs were converted into distance restraints without any further classification than upper limits of 5.5 A or 6.0 A if a methyl group were included. An ensemble of 20 converged structures (Fig. 4) was calculated and the relevant parameters listed in Table 2 for the converged structures. Each NOE here Identical to a distance restraint is only counted once even though it might occur several times in the NOESY spectrum. Ramachandran plot quality assessment is standard quality parameters to evaluate local geometry quality. In general the described quality parameters are comparable to 2.5 A resolution of X-ray based protein stmctures (Laskowski et al. (1996) J, Bio-mol. NMR 8:477-486).


Description of the calculated staicture.
A representative structure resembling the average of the ensemble is displayed in Fig 5. Asp^^lP(AspGlyLys) is structurally similar to the native insulin structure for regions comprising residues B1-B10, B14-B23, A4-A21. The differences are mostly pronounced for regions in the vicinity of the connecting peptide in positions B26-B29, C1-C3, A1-A3 and less pronounced for residues B11-B13. The structure of Asp®IP(AspGlyLys) near the C-peptide is strikingly different from the native like structure in solution (Ludvigsen (1994) supra) and AspB26®IP(AIaAlaLys) stmcture in the crystal phase (Whittingham et al. (1998) Biochemistry 37:11516-11523). The connecting peptide of AspB26lP(AspGlyLys) is pooriy detemnined in terms of accuracy, but a few structural restraints obtained from the NOESY spectra (NOEs

between ThrB27 and GlyC2 and between ThrB27 and GlyA1) clearly indicate important structural arrangements of the C-peptide. The relative intense NOEs between ThrB27 (methyl group HG2) and GlyC2 (atom HA) and between Thr827 (methyl group HG2) and GlyA1 (atom HA) in a flexible region shows that these proton pairs are close in space (<5 A). The tight arrangement of Thr^^, Gly^^ and Gly ^^ defined as the atomic distance between B27 (CG2) and A1 (CA) is less than 5A, not seen previously in any single chain insulin molecule shows that the C-peptide accommodates this structural arrangement and in fact the C-peptide can do this in several ways which appears to be a prerequisite for the C-peptide. However, it is clear that the presence of Glycine in the connecting peptide allows more flexibility in the connecting peptide and subsequently less structural constraints are imposed on the neighboring amino acids in their quest to accommodate an optimal packing with the remainder of the insulin molecule. Secondly the arrangement of AspB26 LysB26, AspB26 and LysC3 charged side-chains creates a highly polar surface compared to other connecting peptides.
Under the conditions used for NMR both the spectra AspB26lP(AspGlyLys) are influenced by some degree of self-association but the exchange between monomer and dimer is on the timescale of NMR only observed here as an average between the two states. Below concentrations of 0.2 mM the degree of self-association does not change as seen by NMR at even lower concentrations (at least until 25 |iM). Table 3 shows chemical shifts of Asp^^^lP(AspGlyLys) at 27° Celcius obtained at 600 MHz, pH 8 in 10%/90% D2O/H2O with 10 mM phosphate buffer. Chemical shifts are referenced by setting the residual water signal to 4.75 ppm. N/A means no assignment AspB26IP(AspGlyLys) assignments (1-29 = B1-B29; 30-32 = C1-C3 and 33-53 = A1-A21). Table 4 provides the atomic coordinates of Asp^^IP(AspGlyLys) in PDB format. The structure selected to represent the ensemble (Fig. 5 and Table 4 atomic coordinates) has 84.8% residues in "favored" regions and 15.2% in "addi-tionally allowed" regions of the ramachandran plot as described in Tabl=

Example 3
The insulin analogue precursor Asp^^®lP(AspGlyLys) was produced culturing yeast strain
MT663 transfonned with an expression plasmid expressing either a YAP3-TA39-EEGEPK(SEQ ID N0:8)-Asp^^^lP(DGK) fusion protein or a YAP3-TA57-EEGEPK(SEQ ID N0:8)-Asp^^^lP(DGK) fusion protein, TA39 is a pro-sequence QPIDDTESNTTSVNLMADDT-ESRFATNrrU\GGLDWNLISMAKR(SEQ ID N0:15). The sequence EEGEPK(SEQ ID NO: 8) is an N-terminal extension to the B-chain of the insulin analogue. TA57 is a pro-sequence QPIDDTESQTTSVNLMADDTESAFATQTNSGGLDWGLISMAKR (SEQ ID NO: 16), cDNA encoding the leader sequences YAP3-TA39 and YAP3-TA57 and cDNA encoding the AspB26lP(DGK) and the N-terminal extension were cloned into an expression vector of the C-POT type using standard techniques (Sambrook J, Fritsch EF and Maniatis T, Molecular cloning, Cold spring Harbour laboratory press, 1989). The DNA and inferred amino acids sequences are shown in Fig 7(SEQ ID NO;9 and 10) and Fig 8 (SEQ ID NO:11 and 12).
Table 5 shows the yields. Fermentation was conducted at 30°C for 72 h in 5 mi YPD. IP yield was determined by RP-HPLC of the culture supernatant and is expressed relative to the IP yield of the strain yJB155,

New Patent Claims
1. An insulin precursor or insulin precursor analog comprising a connecting peptide (C-peptide) being cleavable from the A and B chains said connecting peptide comprising at least one Gly and wherein the B27 (atom CG2) has a proximity to the A1 (atom CA) of less than 5 A.
2. An insulin precursor or insulin precursor analog according to claim 1, wherein the
connecting peptide is of up to 10 amino acid residues in length.
3, An insulin precursor or insulin precursor analog according to claim 1, wherein the
connecting peptide is of up to 9 amino acid residues in length.
4. An insulin precursor or insulin precursor analog according to claim 1, wherein the :onnecting peptide is of up to 5, preferably up to 3 amino acid residues in length.
5. An insulin precursor or insulin precursor analog according to claim 1, wherein the :onnecting peptide comprises up to 5 Gly .
3. An insulin precursor or insulin precursor analog according to claim 1, wherein the connecting peptide comprises up to 3 Gly, preferably only one Gly .
7. An insulin precursor or insulin precursor analog according to claim 1, wherein the connecting peptide comprises a C-terminal Lys or Arg immediately N-terminal to the A chain.
3. An insulin precursor or insulin precursor analog according to claim 7, wherein one Gly amino acid residue is immediately N-terminal to the Lys or Arg .
3. An insulin precursor or insulin precursor analog according to claim 1 comprising a sequence of formula I:
8(1-27) - X3 - X2 - X1 - Y - A(1-21),
A/herein
X1 comprises at least one Gly ,
X2 is one of Pro, Lys, Ala, Arg or Pro-Thr at position 29 of the 8 chain,
X3 is one of Pro, Asp, Lys, or lie at position 28 of the 8 chain, and

Y is Lys or Arg.,
10. An insulin precursor or an insulin precursor analog according to claim 9, wherein Xi is 1-15 amino add residues in length.
11. An insulin precursor or an insulin precursor analog according to claim 9, wherein Xi is 1-10, amino acid residues in length.
12. An insulin precursor or an insulin precursor analog according to claim 9. wherein Xi is 1-8, preferably acid residues in length.
13. An insulin precursor or an insulin precursor analog according to claim 9, wherein Xi is 1-5 amino acid residues in length.
14. An insulin precursor or an insulin precursor analog according to claim 9, wherein Xi is 1-3 amino acid residues in length.
15. An insulin precursor or an insulin precursor analog according to claim 9, wherein Xi contains up to 5 Gly .
16. An insulin precursor or an insulin precursor analog according to claim 9, wherein Xi contains up to three, preferably only one Gly.
17. An insulin precursor or an insulin precursor analog according to claim 9, wherein Xa
is Asp and X2 is Lys.
18. An insulin precursor or an insulin precursor analog according to claim 9, wherein Xi -Y is selected from the group of:(a) Glu-Glu-Gly-Lys(SEQ ID N0:1), (b) Glu-Gly-Lys, (c) Ser-Gly-Lys, (d) Asn-Gly-Lys, (e) Thr-Gly-Lys, (f) Asp-Gly-Lys, (g) Me-Gly-Lys, (h) Ala-Gly-Lys. (i) His-Gly-Lys and Q) Gly-Lys.
19. An insulin precursor or an insulin precursor analog according to claim 9, wherein one Gly is immediately N-terminal to Y.
20. An insulin precursor or insulin precursor analog comprising a connecting peptide (C-peptide) being cleavable from the A and B chains said connecting peptide comprising at least one Gly and a cleavage site enabling cleavage of the peptide bond between the A-chain and the connecting peptide, wherein one Gly is immediately N-terminal to said cleavage site.

21. An insulin precursor or insulin precursor analog according to claim 20, wherein the connecting peptide is of up to 15 amino acid residues in length.
22. An insulin precursor or insulin precursor analog according to claim 20, wherein the connecting peptide is of up to 10 amino acid residues in length.
23. An insulin precursor or insulin precursor analog according to claim 20, wherein the connecting peptide is of up to 9 amino acid residues in length.
24. An insulin precursor or insulin precursor analog according to claim 20, wherein the connecting peptide is of up to 5 amino add residues in length.
25. An insulin precursor or insulin precursor analog according to claim 20, wherein the connecting peptide is of up to 3 amino add residues in length.
26. An insulin precursor or insulin precursor analog according to claim 20, wherein the B27 (atom CG2) has a proximity to the A1 (atom CA) of less than 5 A.
27. An insulin precursor or insulin precursor analog according to claim 20, wherein the connecting peptide comprises up to 5 Gly.
28. An insulin precursor or insulin precursor analog according to claim 20, wherein the connecting peptide comprises up to 3 Gly, preferably only one Gly.
29. An insulin precursor or insulin precursor analog according to claim 20, wherein the cleavage site enabling cleavage of the peptide bond between the A-chain and the connecting peptide is Lys or Arg .
30. An insulin precursor or insulin precursor analog according to claim 20 comprising a sequence of formula I:
B(1-27)-X3 -X2-Xi-Y-A(1-21),
wherein
X1 comprises a Gly immediately N-terminal to Y,
X2 is one of Pro, Lys, Ala, Arg or Pro-Thr at position 29 of the B chain,
X3 is one of Pro, Asp, Lys, or lie at position 28 of the B chain, and
Y is Lys or Arg.

31. An insulin precursor or an insulin precursor analog according to claim 30, wherein Xi is 1-15 amino acid residues in length.
32. An insulin precursor or an insulin precursor analog according to claim 30, wherein Xi is 1-10 amino acid residues in length.
33. An insulin precursor or an insulin precursor analog according to claim 30, wherein Xi is 1-8 amino acid residues in length.
34. An insulin precursor or an insulin precursor analog according to claim 30, wherein Xi is 1-5 amino acid residues in length.
35. An insulin precursor or an insulin precursor analog according to claim 30, wherein Xi is 1-3 amino acid residues in length.
36. An insulin precursor or an insulin precursor analog according to claim 30, wherein Xi contains up to 5 Gly .
37. An insulin precursor or an insulin precursor analog according to claim 30, wherein Xi contains up to three, preferably only one Gly .
38. An insulin precursor or an insulin precursor analog according to claim 30, wherein X3 is Asp and X2 is Lys.
39. An insulin precursor or an insulin precursor analog according to claim 30, wherein X, -Y is selected from the group of:(a) Glu-Glu-Gly-Lys(SEQ ID N0:1) (b) G!u-Giy-Lys, (c) Ser-Gly-Lys, (d) Asn-Gly-Lys, (e) Thr-Gly-Lys, (f) Asp-Gly-Lys, (g) Me-Gly-Lys, (h) Ala-Gly-Lys, (i) His-Gly-Lys and G) Gly-Lys.
40. An insulin precursor or insulin precursor analog according to claim 30. wherein the B27 (atom CG2) has a proximity to the A1 (atom CA) of less than 5 A,
41. An insulin precursor or insulin precursor analog comprising a connecting peptide (C-peptide) being deavable from the A and B chains said connecting peptide comprising at least one Gly, wherein the connecting peptide is of up to 5 amino acid residues in length.
42- An insulin precursor or insulin precursor analog according to claim 41, wherein the connecting peptide is of up to 3 or up to 2 amino acid residues in length.

43. An insulin precursor or insulin precursor analog according to claim 41, wherein the connecting peptide comprises up to 3 Gly, preferably only one Oly •
44. An insulin precursor or insulin precurspr analog accordinrj to claim 41, wherein the connecting peptide comprises a Lys or Arg immediately N-terminal to the A chain.
45. An insulin precursor or insulin precursor analog according to claim 41, wherein one Gly immediately N-terminal to the Lys or Arg.
46. An insulin precursor or insulin precursor analog according to claim 41, wherein the Ml (atom CG2) has a proximity to the A1 (atom CA) of less than 5 A.
47. An insulin precursor ox insulin precursor analog according to claim 41 comprising a sequence of formula I:
B(1-27)-X3 -X2-Xi-Y-A(1-21), wherein
X1 is 1- 5 amino acid residues in length and comprises at least one Gly , X2 is one of Pro, Lys, Ala, Arg or Pro-Thr at position 29 of the B chain, X3 is one of Pro, Asp, Lys, or lie at position 28 of the B cnain, and
Yis Lys or Arg.
48. An insulin precursor or an insulin precursor analog according to claim 47, wherein Xi is 1-4 amino acid residues in length.
49. An insulin precursor ox an insulin precursor analog accordi-g to claim 47, wherein Xi is 1-3 amino add residues in length.
50. An insulin precursor or an insulin precursor analog accordi' g to claim 47, wherein Xi contains up to three, preferably one Gly .
51. An insulin precursor or an insulin precursor analog accordir g to claim 47, wherein X3 , is Asp and X2 is Lys.
52. An insulin precursor or an insulin precursor analog according to claim 47, wherein Xi - Y is selected from the group of:

(a) G!u-Glu-Gly Lys(SEQ ID NO: ), (b) Glu-Gly-Lys, (c) Ser-Gly-Lys, (d) Asn-Gly-Lys, (e) Thr-Gly-Lys, (f) Asp-Gly-Lys, (g) Me-Gly-Lys, (h) Ala-Gly-Lys, (i) His-Gly-Lys and (j) Gly-Lys.
53. An insulin precursor or an insulin precursor analog according to claim 47, wherein one Gly is immediately N-terminal to Y.
54. An insulin precursor or insulin precursor analog according to claim 47, wherein the B27 (atom CG2) has a proximity to the A1 (atom CA) of less than 5 A.
55. A polynucleotide sequence encoding an insulin precursor or an insulin precursor analog according to any of claims 1-55.
56. An expression vector comprising a polynucleotide sequence according to claim 55.

57. A host cell transformed with the vector claim 56.
58. A process for making an insulin precursor or an insulin precursor analog said method comprising (i) culturing a host cell comprising a polynucloetide sequence encoding an insulin precursor or an insulin precursor analog according to any of claims 1-54 under suitable culture conditions for expression of said precursor or precursor analog; and (ii) isolating the expressed precursor or precursor analog.
59. A process according to claim 58, wherein the host cell is a yeast host cell
60. A process for making insulin or an insulin analog, said method comprising (i) culturing a host cell comprising a polynucleotide sequence encoding the insulin precursor or an insulin precursor analog according to any of claims 1-54 under suitable culture conditions for expression of said precursor or precursor analog; (ii) isolating the precursor or precursor analog from the culture medium and (iii) converting the precursor or precursor analog into insulin or an insulin analog by in vitro chemical or enzymatic conversion.
61. A process according to claim 60, wherein the host cell is a yeast host cell line.

62. An insulin precursor or insulin precursor analog substantially as
herein described with reference to the accompanying drawings.
63. A process for making an insulin precursor or an insulin precursor
analog substantially as herein described with reference to the
accompanying drawings.

*