CA2094512A1 - Fusion polypeptides - Google Patents

Abstract

A non-naturally-occurring fusion protein comprising an extension peptide portion covalently linked at its C-terminus to the N-terminus of a biologically active portion is disclosed. The extension peptide portion can be removed by DPP IV cleavage. A use of fusion proteins with DPP IV cleavable extension peptide portions in medecinal preparations is disclosed. A method of purifying desired proteins from a mixture containing a fusion protein is disclosed.

Description

2~ 5~2 WO 92/10~76 PCr/US91/0915 FUSION POLYPEPTIDES
FIELD OF THE INVENTION
The present ;nvention relates to non-naturally occurring fusion polypeptides containing N-terminal portions cleavable by dipeptidylpeptidase IV (I:)PP IV).
BACKGROUND OF THE INVENTION
The techniques of molecular biology, specifically recombinant DNA technology, allow for the production of relatively large quantities of desirable biologically active polypeptides.
Fur~ermore, ~e genetic information encoding the polypeptides may be modified ~o produce relatively large quantities of modified polypeptides. Modifications made to the polypeptides are often used to improve their activity or facilitate their production and/or preparation.
Accordingly, much effort has been made to determine what modifications are desirable in order to increase, enhance or otherwise alter the biological activity of desired polypeptides. In addition, there is a great deal of work being do~e to modify desired polypeptides to facilitate their production and purification.
Naturally produced polypeptides are often initially biosynthesized as larger precursors which are then trimmed by a series of proteolytic cleavages to produce the final products.
Accordingly, several proeeases exist which recognize and cleave specific amino acids and/or arnino acid sequences. l~ese proteases participate in a conversion of a precursor protein to the fimal polypeptide product.
Once such protease is dipeptidylpeptidase IV (DPP IV) (EC 3.4.14.~). DPP lV was first reported in Hopsu-Havu, V.K. and G.G. Glenner, Histo. Chemie 3:197-201 (1966) and has been shown to be present in many mamrnalian tissues. DPP IV is presently cormnercially available from Enzyme Systems Products (Dublin, California). DPP IV recognizes specific arnino acid sequences on the N-terminus of proteins. Specifically, DPP IV will cleave a dipeptide from the N-terminus when the second arnino acid from the N-terrninus is proline (Pro), hydroxyproline (Hyp), alanine (Ala), serine (Ser), and threonine CThr) and any amino acid is at the N-terminus residue position provided if proline or hydroxyproline is not the amino acid residue third from the N-terminus. DPP IV activity is more efficient when proline or alanine is the second amino acid from the N-terminus and is usually most efficient when that position is occupied by proline. The activity of DPP IV in the stepwise cleavage of "PRO"
parts of precursors of naturally occurring peptides is widely reported.
Modern technology has made possible the high level production of biologically active proteins. Important polypeptides can be synthesi~ed using peptide synthesizers or in host cells using recombinant DNA technology. Often, biologically active proteins are administered as drugs. Numerous examples 0xist in which active proteins are used as therapeutics, prophylactics or to enhance or repress traits. Since DPP IV and other proteases degrade proteins, these drugs , 2 ~ 9 ~
WO 92/10~76 , ~ PCI/US91/~915:

are susceptible to degradation. ThUst a problem of using biologically active polypeptides as drugs is that their sustained presence is diminished and they must therefore be administered more frequently.
The rapid developme,nts in recombinant DNA methodology which have allowed ~the S production of polypeptides, proteins, and their analogs in large quantities in a very short period of time have created a need to isolate in highly efficient and predictable manners these proteins from complex mixtures including the total arnount of protein produced by the host cells and ~ose in the grow~ medium. The purification of heterologous polypeptides produced by host cells can be very expensive and can cause denaturation of the protein product itself. An 10 overview of protein purification techniques is provided in the Background Art section of U.S.
Patent Number 4,782,137 issued Nov. 1, 1988 to Hopp et al., incorporated herein by reference.
To circumvent the limitations in the art and provide better methods, recombinant DNA
technology may be used to provide desired polypeptides in the form of non-naturally occuring 15 proteins which contain a linker peptide that may be used as a ligand or other target for purification means. For example, U.S. palent number 4,782,137 relates to synthesis of a non-naturally oscuring peptide containing an antigenic linker peptide. The non-naturally occuring protein can be passed through a column containing in~nobilized antibodies which bind to the antigenic linker, thus isolating the non-naturally occuring protein. U.S. Patent 4,569,794 20 relates to a process of purifying non-naturally occuring proteins which contain N-terminal e~ensions that have an affinity for immobilized metals. The non-naturally occuring proteins bind to immobilize~ metal ions in a column. One problem with these methads is that the linker peptide is often undesirable and removal of the linker can be diff~cult.
The present invention relates to non-naturally occurring fusion protei~s which comprise 25 a core protein portion and an N-terminal extension which is cleavable by DPP IV. According to the present invenlion, a non-naturally occuring protein is provided wherein the extension attached to ~e core protein is not an N-terminal extension that occurs in na~ure attached to the core peptide; hence, non-naturally occurring fusion protein. The present invention relates to prodrugs which are DPP IV cleavable non-naturally occuring proteins wherein the core protein 30 portion is a biologically active protein. The present invention relates to DPP IV cleavable non-oaturally occuring proteins use~ul in purification methods whereby the N-terminal extension provides a feature or property which facilitates purification of the non-naturally occuring protein.
The present invention provides non-naturally occuring proteins which have N-terminal 35 extensions that are cleavable by DPP IV such that exposure of the non-naturally occuring protein to DPP IV results in conversion of the non-naturally occuring protein to a desirable , ~Y45`1~
, ~
WO 92/1~76 PCI/US91/0~15 protein. When used as a prodrug, the non-naturally occuring protein is processed into a biologically active protein in vivo using DPP IV present in ~e target species. When used in a purification process, non-naturally occuring protein can be purified by using i~s specifically designed N-terminus as a ligand and then processed with DPP IY to remove the N-terrninal extension and liberate a desired pro~ein.
The present invention allows for the production of a desired protein as a non-naturally occuring protein that is later converted to the desired protein when exposed to DPP IV.
Prodrugs are converted to drugs over a course of time using the patients' endogenous DPP IV, ~hereby achieving sustained presence of the active drug in a patient and reducing the frequency of administration. Pure desired proteins can be isolated using the present invention by producing and purifying non-naturally occuring proteins and then processing the non-naturally occuring proteins in vitro with DPP IV to produce the desired protein.
SUMMARY OF THE INYENTION
The present invention relates to a non-naturally occurring fusion protein comprising an extension peptide portion covalently linked at its C-terminus to the N-~erminus of a core protein portion, said extension peptide portion being of the formula:
A-X-Y(x~-y)n wherein A is optional and when present is methionine;
n is 0-20;
X is selected from the group consisting of all naturally occurring amino acid residues;
X' is selected from the group consisting of all naturally occurring amino acid residues except proline and hydroxyproline;
Y is selected from the group consisting of proline, hydroxyproline, alanine, serine and ~5 threonine except when n is 0 then Y is selected from the group consisting of alanine, serine and threonine.
The present i~vention also relates to the use of such non-naturally occuring proteins in medicinal preparations and to a method of purifying desired proteins from a mixture containin~
such non-naturally occuring proteins and impurities comprising the steps of selectively contacting said non-naturally occuring protein with material which immobilizes said non-naturally occuring protein, removing said impurities, separating said non-naturally occuring proteins from said material, contacting said non-naturally occuring protein with DPP IV, and isolating said desired protein.
INFORMATION DISCLOSURE
U.S. Patent No. 4,569,794 issued February 11, 1986 to Smith et al relates to theprocess of purifying proteins and compounds useful in such processes. The invention describes .
~, . .:, ' , :

.
.

2~94~1`2 ~;
WO 92/la576 PCI`/lJS91/09152 , ' 4-a process of isolating f~sion proteins which have biologically active polypeptides at ~e C-tenninal end and an N-terminal extension lirlker that is a metal ion chelating linker. The fusion peptide has an affinity to immobilized metal ions. Impurities can be removed by passing a mixture containing the fusion protein through a colurnn containing immobilized metal ions.
s The fusion protein becomes associated with the metal ions and only the impurities are eluted.
Upon changing conditiorls the fusion peptide is liberated from the immobilized metal ions thus resulting in purified fusion protein.
U.S. Patent No. 4,782,137 issued November 1, 1988 to Hopp et al., discloses the synthesis of a fusion protein having a highly antigenic N-terminal portion and a desired I0 polypeptide at the C-terminal portion. According to Hopp et al., the fusion proteins are purified from crude superrlatant by passing crude supernatent through a column containing immobilized antibodies which recognize the antigenic portion of the fusion protein. The immobilized antibodies keep the protein in the column while the undesired components of the supernatent are eluted. The colusnn conditions can then be changed to cause the antigen-15 antibody complex to dissociate. The fusion protein is then eluted and collected.
U.S. Patent No. 4,734,399 issued March 29, 1988, to Felix, et al. relates to grow~honnone releasing factor analogs. Several analogs are disclosed which have end terminals of Tyr-Ala and His-Ala. However, these molecules are not fi~sion proteins but rather core proteins only. The N-terminal dipeptides of Felix, et al. are part of the bGRF analog core 20 molecule.
European patent application Publication Number 0 220 958, published May 6, 1987 relates to selective chemical removal of N-terminal residues. The invention rela~es to a process and compounds ussful in the process to remove N-tersninal residues from desired polypeptides.
The desired polypeptide exists as a fusion protein having the desired polypeptide link at the N-2S terminal to a linker having the formula X-Pro. Upon exposure of the fusion protein to specific buffer conditions a diketopiperazine of the X-Pro portion of the fusion protein is formed ànd cleaved, thereby producing the desired polypeptide from the fusion precursor. The fusion proteins of EPO 220,958 ('958) is not included in the present invention because according to the present invention, when the N-terminal extension is only a dipeptide, i.e., when A is absent, n is zero and X is a naturally-occuring amino acid, Y is either alanine, serine or thr~onine. Tbus, whenever the extension is a dipeptide, it is X-Ala, X-Ser or X-Thr. The '958 application teaches chemical, not enzymatic, cleavage of the dipeptide X-Pro. The dipeptide X-Ala, X-Ser and X-Thr are not susceptible to the type chemical cleavage taught by the '958 application that cuts the X-Pro extension from the core protein.
Australian Patent Application Document No. AU-A-12709/88 discloses fusion proteins which contain afflnity peptides useful in immobilized metal affinity chromatography (IMAC).

2~9~5~2 :~ Wo92/10576 Pcr/ussl/o9l~2 _5 Th0 affinity peptides disclosed contained at least two neighboring histidine r~sidues. The IMAC purifica~ion means disclosed requires a special synthetic chemistry for making nitrilo-triacetic acid (NTA) resins.
Tallon, M.A., et al., Biochem. 2b:7767-7774 (1987) relate to synthesis of extended S analogs of the tridecapeptide ~-factor from Saccharomyces cerevisiae. The synthesized analogs are extended cr-factors, which represent sequences of naturally occurring pro-c~-factor coded for in the MF~l structural gene.
Kriel, G. et al, Eur. J. Biochem. 111:49-58 ~1980) describes the stepwise cleavage of the N-tenninal portion of melittin precursor (Promelittin) by dipeptidylpeptidase IV.
10 Promelittin is the main constituent of honeybee venom. In the amino acid sequence of the N-terminal portion of the precursor, every second residue is either proline or alanine. When promellitin is e~posed to DPP IV isola~ed from pig kidney, the N-terminal region of the precursor is cleaved in a stepwise fashion producing the mature protein. Promelittin, unlike fusion proteins according to the present invention, is a naturally-occurririg protein.
1$ Julius, D., et al, Cell, Vol 32:839-852 (March 1983) relates to the role of membrane bound DPP IV in the processing of yeast cY-factor from a larger precursor polypeptide. The yeast a-factor, unlilce fusion proteins according to the present invention, is a naturally-occurring protein.
Mollay, C. et al, Eur. J. Biochem. 160:31-35 (1986) describes thè isolation of DPP IV
20 from the skin secretion of Xenopus laevis. The activity of DPP lV is discussed.
Mentlein, R., FEB, Vol. 234, No. 2, pp. 251-256 auly 1988) reviews proline residues in the maturation and degradation of peptide hormones and neuropeptides. It is reporte~ that in mammals, proline specific proteases such as DPP IV are not involved in the biosynthesis of regulatory peptides but may play an important role in the degradation of them. Thus, it is 25 concluded that while in vertebrates and lower vertebrates precursor proteins rely on DPP IV to convert precursors to mature forms, the processing of regulatory proteins in mammals generally uses DPP IV as a degradation protease.
Frohman, L. A. et al. J. Clin. Invest. 78:906-913 (1986) }eport that human growth hormone releasing factor ~GE~F) and its analogs are rapidly degenerated in vivo in humans 30 and in vifro by plasma DPP IV.
Frohman, L. A. et al. J. Clin. Invest. 83:1533-1540 (1989) report that human growth honnone releasing factor ~GRF) and its analogs are rapidly degenerated in ViYo in humans and in vitro by plasma DPP IV.
Kubiak, T.M., et al, Drug Metabolism and Disposition, Vol. 17, No.4,pp. 393-397 3~ (1989)refer to the in vitro metabolic degradation of bovine growth horrnone releasing factor (bGRF) analogs in bovine and porcine plasma and the correlation with plasrna DPP IV activity.

.

Wo 9~/10576 2 0 9 ~ ~ ~ 2 Pcr~U~9l/O9~

1 he bGRF analogs tested had an ~AIa residue at position 2- of the N-terminus. lt is reported that the metabolic degradation of bGRF in plasma is due to ~he presence of DPP IV in the plasma.
Hong, W., et al, Biochemistry, 28:8474-8479 (1989) report the expression of 5 enzymatically active DPP IV in Chinese hamster ovary cells after transfection.Kreil, G., TIBS 15:23-26 (January 1990) reviews of the stepwise cleavage of dipeptides by DPP's in the conversion of precursors to final products. The precursors, described by Kreil are nahlrally~ccurring proteins. The fusion proteins of the present invention are non-naturally-occurring filsion proteins.
Boman, et al., J. Biol. Chem. 264:5852-5860 (1989) demonstrated that a dipeptidyl peptidase isolated from cecropia pupae (with similar specificity to DPP IV) was able to remove natural N-terminal sequences of Ala-Pro-Glu-Pro from the N-terminal of synthetic copies of the natural precursors of cecropia R and B. The preprocecropin disclosed by Boman is a naturally-occurring protein.
Dalboge, H., et al, Bio/technology, 5: 161-164 (February 1987) disclose converting E.
coli produced precursor of human growth hormone (hGH) to authentic hGH in vitro. The N-terminal extension of the precursor is removed by dipeptidypeptidase I.
Dalboge, H., et al, I:EBS, Vol. 246 (1,2):89-93 (March 1989) discloses the cloning and e~pression of IL-l,B precursor and its conversion to IL-l,B by removal of the precursor's N-20 terminal extension using dipeptidypeptidase I.
Dalboge, H., et al, FEBS, Vol. 266 (1,2):1-3 aune 1990) refer to in vivo processing of N-terminal methionine in E. coli. It is reported that the remo~al of the N-terminal methionine from e~tended human growth hormone was dependent upon the amino acid adjacent to the methionine.
Hopp, T.P. et al., Bio/Technol. 6:1204-1210 (October 1988), disclose addition of an cight amino acid peptide to the N-terminus of a desired recombinant Iymphokine in order tO
provide an antigenic N-terminus which can be used in immunoaffinity purification. This publication corresponds to U.S. Patent No. 4,782,137 described above.
Smith, M. C., et al., J. Biol. Chem. Vol. 263, 15:7211-7215 (1988) disclose e~perimental results supporting the hypothesis that specific metal chelating peptides on the NH2 terminus of a small peptide can be used to purify that protein using metal ion affinity chromatography. This reference provides specific data regarding one of the examples in the above described U.S. Patent Number 4,569,794. Specifically, the use of the metal chelating peptide His~Trp linked ~o either luteinizing hormone-releasing hormone or proinsulin allows the chimeric peptide to be p~rified using IMAC whereas control molecules not containing the His-Trp linlcer c~nnot be recovered in the like manner.

! ~ ~wo 92/10576 2 0 9 ~L 512 PCr/US91/0915~
_7 Hochuli, E. et al., J. Chromat. 411:177-184 (198~) disclosed a nitrilotriase~ic acid absorbent useful for metal chelate affinity chromatography. It is reported that the disclosed absorbent when charged with Ni2+ is useful in binding to peptides and proteins containing neighbori~g histidine residues.
Ljungquist, C. et al., Eur. J. Biochem. 186:563-569 (1989) disclose the use of the metal chelating peptide Ala-His-Gly-His-Arg-Pro in multiplicities of two, four and eight together with a colurnn con~aining immobilized Zn2+ ions. According to Ljungquist use of this metal chelating peptide with zinc columns provides unexpectedly good purifisation of the fusion proteins.
DETAILED DESCRIPTION OF THE INVENTION
As used herein, ~e terms "non-naturally occurring fusion protein", "non-naturally occurring fusion polypeptiden, nfusion polypeptides" and "fusion proteins" refer interchange-ably to proteins and polypeptides which do not normally occur in nature and which comprise a core protein portion and an ex~ension portion.
As used herein ~core protein", "core protein portion" and "polypeptide portion" refers to the portion of a fusion polypeptide which is located at the C-terminus end of the molecule and which, absent the extension portion, would be a desired polypeptide and/or a biologically active protein including naturally occuring biologically active proteins and polypeptides and analogs and mutants thereof.
As used herein "N-terminal extens;on" refers to the first up to about 45 amino acids starting at the N-terminus and which are not part of the core protein.
As used herein "prodrugs" refers to fusion proteins wherein the biologically desired portion is a biologically active protein useful as a drug.
As used herein "biologically active protein" and "biologically active polypeptides" refer 25 to interchangeable proteins and polypeptides which possess biological activity.
As used herein "desired protein" and "desirable protein" refer interchangeable to pr~teins and polypeptides which are sought in pure form.
As used herein "extension portion" refers to the portion of a fusion protein which is an N-terminal extension and which is not part of the biologically desired portion.
As used herein "DPP IV cleavable N-tenninal extension portion" refers to the extension portion of a fusion protein which has an amino acid seguence that can be removed by the stepwise cleavage by DPP IV.
Ln ~e Sequence Listing Section, some amino acid residues have been designated Xaa in Seg ID. The following descriptions apply:
ln Seq ID No. 3 Xaa29 represents C-terminally amidated Argininyl residue.
In Seq ID No. 4 Xaa29 represents C-terminally amidated Argininyl residue.

: ~ , . . . ~
~ . . . . .

~ ~ 9 ~
w~ 92/10576 - ~ CT/US9l/0915 ln Seq ID No. S Xaa~9~represents C-terminally amidated Argininyl residue.
In Seq ID No. 14 Xaa29 represents C-terminally amidated Argininyl residue.
In Seq ID No. 18 Xaa31 represents C-terminally amidated Argininyl residue.
In Seq ID No. 19 Xaa33 represents C-terminally amidated Argininyl residue.
In Seq ID No. 20 Xaa39 represents C-terminally arnidated Argininyl residue.
In Seq ID No. 21 Xaa45 represents C-terminally amidated Argininyl residue.
ln Seq ID No. 24 Xaa27 represents C-terminally amidated Argininyl residue.
ln Seq ID No. 25 Xaa31 represents C-terminally amidated Argininyl residue.
ln Seq ID No. 26 Xaa33 represents C-terminally amidated Argininyl residue.
In Seq ID No. 27 Xaa35 represents C-terminally amidated Argininyl residue.
In Seq ID No. 28 Xaa37 represents C-terminally amidated Argininyl residue.
In Seq ID No. 29 Xaa33 represents C-terminally amidated Argininyl residue.
In Seq ID No. 30 Xaa35 represents C-terminally amidated Argininyl residue.
ln Seq ID No. 31 Xaa37 represents C-terminally amidated Argininyl residue.
In Seq ID No. 32 Xaa39 represents C-terminally amidated Argininyl residue.
In Seq ID No. 33 Xaa45 represents C-tenninally amidated Argininyl residue.
In Seq ID No. 34 Xaa43 represents C-terminally amidated Argininyl residue.
In Seq ID No. 35 Xaa45 represents C-terminally amidated Argininyl residue.
In Seq ID No. 36 Xaa31 represents C-terminally amidated Argininyl residue.
In Seq ID No. 37 Xaa31 represents C-terminally amidated Argininyl residue.
In Seq ID No. 38 Xaa31 represents C-terminally amidated Argininyl residue.
In Seq ID No. 39 Xaa31 represents C-terminally amidated Argininyl residue.
In Seq ID No. 40 Xaa31 represents C-terminally amidated Argininyl residue.
In Seq ID No. 41 Xaa33 represents C-terminally amidated Argininyl residue.
In Seq lD No. 42 Xaa3l represents C-terminally amidated Argininyl residue.
In Seq ID No. 43 Xaa33 represents C-terminally amidated Argininyl residue.
The present invention relates to improved proteins and polypeptides. According to the present invention, biologically active polypeptides are first produced as fusion proteins which contain the two portions: a first portion which represents the core protein portion; and, a 30 second portion which is an N-terminal extension portion that is covalently linked at its carboxy terminus to the amino terminus of the first portion. The N-terminal extension portion of the filsion polypeptide possesses an amino acid sequence which renders it subject to cleavage by the dipeptidylpeptidase IV (DPP IV).
A fusion protein according to the present invention has the for nula:
3~ Extension portion - Core protein portion wherein "Extension portion" represents a DPP IV cleavable N-tenninal extension; " - "

- WO 9~/10576 2 0 9 ~ 5 ~ 2 PCI/US91/09152 _9 represents a cova~ent peptide bond; and, "core protein portion" represents any desired peptide which is liberated from the Extension portion by DPP IV processing.
The Extension portion of a fusion protein according to the present invention has an amino acid sequence according to the formula:
S A-X-Y(X-Y)n wherei~ A is optional, and when present is methionine;
n represents the number of sequentially linked X'-Y groups, that number representing ~rom 0 to 20 of such groups, preferably 0 to 10 groups.
X is selected from t~e group consisting of any naturally occuring amino acid;
Y is selected from the group consisting of proline, alanine, serine, and threonine, except when n - 0, then Y is selected from the group consisting of alanine, serine, and threonine;
X' is selected from the group consisting of any naturally occuring amino acid except proline or hydroxyproline;
According to the formula, when n = 1, there are two Y residues. Further, it is possible to have up to twenty one Y residues and twenty X' residues in a single embodiment.
Individual Y residues and X' residues respectively can be any residue of the group from which they are selected. That is, all of the individual Y residues do not have to be the same in a given embodiment. Similarly, in an embodiment with more than one X' residue, each individual X' residue present can be any amino acid residue except proline and hydroxyproline irrespective of what residue any other X' residue may be. Each individual Y and X' residue respectively must conform to the rules for that particular group and all that is necessary is that the various individual residues at the specihc posi~ions follow the rules as articulated above.
Fusion proteins in which (A) is present as methionine (Met) represent sequences useful for ~he production of biologically active proteins by recombinant DNA methods in E. coli. The Met sequence present in these precursors usually will be processed by the E. coli enzymatic system or some other means which can be perfor~ned by a person with ordinary skill in the art.
Protein synthesis in E. coli is, under normal circumstances, initiated at the translation initiation codon AUG coding for Met. As a consequence, the newly synthesized polypeptides have a methionine residue as their N-terminal amino acid. E. coli possesses an enzymatic activity with ~e capacity to effectively remove N-terminal Met when the Met N-terminal residue is atjacent to an amino acid with a relatively small side chain like Gly, Ala or Ser as well as Pro. Highly specific removal of the N-tenninal Met can be accomplished using cyanogen bromide mediated cleavage of Met. However, for that procedure to be successful, the N-terminal Met must be dle only Met in the entire protein sequence; otherwise the cleavage will take place after each Met in the sequence. Accordingly, for fusion proteins containing internal Met sequences, the - :

. ' Wo 92~10576 2 0 9 4 ~ ~ 2 ~ US91/U915~ ^

second amino acid ~rom the N-terminus must be Pro, Gly, Ala or Ser if the Met is to be removed by the E. coli enzymatic system.
In addition to fusion polypeptides, the present invention relates to: recombinant DNA
molecules which comprise DNA sequences that encode the fusion polypeptides; methods of S using the recombinant DNA molecules; methods of using the fusion polypeptides including methods of purifying desired polypeptides and methods of delivering drugs which comprise administering prodrugs that are converted from precursor to biologically active forms by stepwise proteolytic removal of the N-terminal extension in vivo.
Production of fusion polypeptides can be accomplished using standard peptide synthesis 10 or recombinant DNA techniques both well known to those having ordinary skill in the art.
Peptide synthesis is the preferred method of making polypeptides which comprise about 50 arnino acids or less. For larger molecules, production in host cells using recombinant DNA
technology is preferred.
Fusion polypeptides which contain N-terminal portions that are recognized and cleaved 15 by DPP IV are useful and advantageous over unmodified polypeptides comprising only the core protein portion. The present invention describes two areas of particular utility. The first use is to provide fusion polypeptides, termed "prodrugs", which comprise biologically active polypeptides that are useful as drugs covalently linked to DPP IV cleavable N-terminal extensions. Tbese proforms can be converted into biologically active forms upon cleavage by 20 DPP IV in the body of a human or other animal that has been administered the prodrug.
Accordingly, the present invention relates to fusion polypeptides useful as prodrugs, use of fusion polypeptides in a medicinal preparation and to a method of delivering biologically active polypeptides to a patient. A second use for fusion polypeptides according to the present invention is in protein purification processes in which the N-terminal extension is the 25 component of the polypeptide which renders it effective in a purification method and which N-terminal extension is then removable by cleavage using DPP IV. Accordingly, the present invention relates to fusion polypeptides useful in purification procedures and to a method of purifying desired polypeptides. These uses serve as examples to illustrate the utility of the present invendon and are not meant to limit the invention in any way.
For bo~ uses? the core protein portion of the fusion protein is liberated from the e~ctension portion by DPP IV activity. ln the case of fusion proteins used in purification methods, it is undesirable that the core proteins be substrates for DPP IV cleavage. That is, it is preferred that DPP lV not be able to cleave the core protein after the extension portion has been removed. It is often most desirable that when a core protein is a DPP IV substrate, it is 35 delivered as a prodrug. ln such cases, the prodrug can result in sustained presence of the core protein since some of the DPP lV found in vivo (e.g. in plasma, kidney tissue and liver tissue) -~ WO92/10576 2~94512 P~/US91/09152 will be used to process N-terrninal extensions and, therefore, delay core protein degradation.
That is, the ex~ension portion of the fusion p~otein can act as a substrate for DPP IV and competitive inhibitor, delaying the DPP IV action on the core protein thereby temporarily protecting the core protein.
As used herein, "prodrug" means a fusion protein which contains a DPP IV cleavable N-terminal extension covalently linked to a core protein portion that is a biologically active polypeptide use~ul as a drug. Prodrugs according to the present invention can be administered as an individual proform or in combination with other compounds. The preferred embodiment is a well defined individual forrn of a prodrug. In ei~her case, the proforms are processed by naturally occurring DPP IY normally found in the body.
The advantage of administering a p~odrug in a medicinal preparation is that it delays activity and/or provides for extended presence of the biologically active protein. Prodrugs can remain active longer than unmodified molecules. Prodrugs can exist in a non-active state until such time elapses that a sufficient portion of the extension portion is degraded and the molecule becomes active. Prodrugs, therefore, can act as a time delayed drug delivery system.
Furthermore, different N-terminal extensions are degraded at different rates, depending on their length and the specific residues present in their amino acid sequence. Combinations of different forms of prodrugs having a variety of N-terminal extensions can be provided which can provide a sustained, steady level of active drug in a patient over a course of time. Prodrugs, therefore, can act as a time delayed drug delivery system.
As described above, DPP IV cleaves off a dipeptide from the N-terminus of a polypeptide provided certain residues occupy certain positions. As used herein, "position one"
refers to the amino acid residue position at the N-terïninus. As used herein, "position two"
refers to the amino acid residue position which is immediately adjacent to position one and which is second from the N-terminus. As used herein, "posi~ion three" refers to the amino acid residue position which is immediately adjacent to position two and which is third from the N-terrninus. The cleavage which will remove the N-terminal dipeptide occurs between position two and position tbree provided amino acid three is not proline or hydroxyproline and amino acid two is one of five amino acids: proline (Pro), hydroxyproline (Hyp), alanine (Ala), serine (Ser), or threonine CI~r).
DPP IV cleaves the N-terminal residues at a different rate depending upon which of the four amino acid residues is present at position two. In most cases, DPP IV cleaves most efficiently when position two is occupied by Pro and it is next most efficient when position two is occupied by Ala. When position one is occupied by tyrosine, phenylalanine or histidine, DPP IV works at about ~e same rate when position two is occupied by Pro or Ala. DPP IV is ne~t most efficient when position two is occupied by Ser. It is least efficient when Thr : ~ .
. . .. ~ ~
. . - : ., ~ . .

WO 92/10~76 2 0 ~ 1 ~ 11 2 ;

occupies the secorl~ position.
Using this information, a ~à~iety of N-terminal extensions can be designed which are processed a~ different rates. Thus, a medicament can be administered which comprises cither a specific prodrug or a combination of prodrug ~orrns. The prodrugs, bearing an assortment of 5 N-tenninal extensions, will each be processed at a rate which is dependent upon their amino acid sequences. This combination of prodrug forrns can be forrnulated to comprise a series of prodrugs that are processed into active polypeptides across a spectrum of time.
lhe length and amino acid sequence residue makeup are controlling factors in the rate of DPP IV cleavage. Extensions containing all or mostly all alternating Y=Pro will be 10 processed the fastest while those containing Y=Thr will be converted the slowest. In addition it is known ~at dipeptidyl units X-Pro where X is either Glu or Asp are cleaved much slower than their counterparts where X is a neutral or basic amino acid residue. Since extensions can comprise different residues (of the four) at each cleavage position in the extension, an extremely high number of varia~ions and permutations can exist.
Any biologically active polypeptide can be used as a polypeptide drug. PCT patent application number PCT/US90/02923, incorporated herein by reference, PCT patent application number PCTIUS91/08248, incorporated herein by reference, and U.S. Patent Application Serial Number 07/368,231, incorporated herein by reference and each disclose bovine grourth hormone releasing factor analogs which can be used in a medicinal preparation as a prodrug 20 according to the present invention. Any of the analogs taught in these applications can be used as a core peptide portion of a fusion protein according to the present inventian. Fusion proteins comprising such core protein portions linked to extension portions may be produced by those having ordinary skill in the art using well known methods.
Other examples of embodiments of the present invention include hormones, receptors, 25 enzyrnes, storage proteins and blood proteins. Specific examples include: Vasoactive Intestinal Peptide (VIP); human ,B-casomorphin; Gastric lnhibitory Peptide (GIP); Gastric Releasing Peptide (GRP); human Peptide HI; human Peptide YY; fragment 7-37 of glucagon-like peptide-l; glucagon-like peptide-2; substance P; Neuropeptide Y; human Pancreatic Polypeptide;
insulin-like grow~ factor-1 (IGF-1); human grou th hormone (hGH); boYine grow~ hormone 30 (bGH); porcine growth hormone (pGH); prolactin ~RL); human, bovine, porcine or ovine growth holmone releasing factor (GRF); interleukin-l,B (IL-1~); EGF; IGF-2; glucagon;
corticotropin releæsing factor ~CRF); dynorfin; somatostatin-14; endothelin; transforming growth factor ~ ~TGF-cY); transforming growth factor ,B (TGF-,B); interleukin~; interleukin~;
nerve growth factor (NGF); tumor necrosis factor ~NF); insulin; fibroblast growth factor 35 (FGF); interferon; CD4; and interleukin-2 (IL-2) or their synthetic or biosynthetic analogs.
These polypeptides can also be used to forrn the core protein portion of fusion proteins ` r wo 92/10576 2 0 9 9L 5 1 2 PCMJS91/09152 according to the lpresent invention. These polypeptides are mean~ only to serve as examples of embodiments and are not meant to limit the scope of the present invention.
Smaller fusion proteins according to the present invention can be synthesized, for e~ample, by solid-phase methodology utilizing an Applied Biosystems 43()A peptide synthesizer S (Applied Biosystems, Foster City, California) as described in detail in PCTIUS90/02923 and ~7/36~,231.
For larger molecules, production in host cells using recombinant DNA is preferred.
Tnere are several different methods available to one having ordinary skill in the art who wishes to use recombinant DNA technology to produce fusion proteins. Typically, genes encoding desired polypeptides are inserted in expression vectors which are then used to transform or trarLsfect suitable host cells. The inserted gene is then expressed in the host cell and the desired polypeptide is produced. To produce the fusion polypeptides of the present invention in a like manner, an additional DNA sequence is included in the gene insert. Specifically, DNA
encoding the N-terminal extension residues is operably linked to the 5' end of the gene encoding the desire~ polypeptide. This additional genetic material must be placed downstream from the promoter of the expression vector so that it is under the control of the promoter.
Additionally, it must be placed in proper reading frame with the gene so ~at the expressed protein product includes the N-terminal extension residues covalently linked to the desired polypeptide.
Therefore, to produce fusion proteins according to the present invention using recombinant DNA technology, oligonucleotides must be designed which encode the amino acid sequence of the desired N-terrninal extension and these oligonucleotides must be operably inserted upstrearn of the ~' end of the gene encoding the core protein portion, generating a chimeric gene. The techniques to make oligonucleotides and the techniques used to producing a 2~ chimeric gene are well knojwn to those having ordinary skill in the art.
In addition to the utility of fusion proteins as prodrugs, the present invention relates to the purification ant processing of biologically active recombinant polypeptides. lhe desired biologically active reoombinant polypeptides are most preferably produced in a scluble form or secr~ted from the host. According to the present invention, the extension portion of the fusion protein can be recognized by purification means. The fusion protein is purified from the material present in the secretion media or extraction solution it is contained in and then processed to remove the extension por~ion from the core protein portion, thus producing purified desired protein. Accordingly, desired proteins most suited for processing as fusion proteins according to the present invention are those biologically active polypeptides which are IIOt thernselves substrates for DPP IV cleavage.
In accordance with the present invention, a gene sequence encoding for a desired .

.~ ~
: ~ . . .
.

2~94~I2 r WO 92/10576 PCl/US91/09152 protein is isolated, synthesized or other.wise obtained and operably linked to a DNA sequence coding an extension portiori. The hybrid gene containing the gene for a d~sired protein operably linked to a DNA sequence encoding an extension portion is referre~ to as a chimeric gene.
Methods and materials for preparing chimeric genes and recombinant vectors, transforming or trans~ecting host cells using the sarne, replicating the vectors in host cells and expressing biologically active foreign polypeptides and proteins are described in Principles of Gene Manipulation, by Old and Primrose, 2nd edition, 1981 and Sambrook et al., Molecular Cloning, 2nd Edition, Cold Spring Harbor Laboratory Press, NY (1989), both incorporated herein by reference.
The present invention relates to rec~mbirlant chimeric genes which encode fusion- proteins, expression vectors containing the same, hosts transformed or transfected with these expression vectors, and process for obtaining these genes, expression vectors, and hosts transforrned or transfected with said vectors.
The present invention may be used to purify any prokaryotic or eukaroytic protein that can be expressed as the product of recombinant DNA technology in a transformed or transfected host cell. These recombinant protein products include hormones, receptors, er~nnes, storage proteins, blood proteins, mutant proteins produced by protein engineering techniques, or synthetic proteins. The desired polypeptides produced may include HIV RNase H, tPA, IL-1, IL-1 receptor, CD4, human nerve growth factor, sCD4-PE40, human respiratory syncytial virus (RSV) FG chimeric glycoprotein (See U.S. Patent Application Serial No.
071543,780, incorporated herein by reference), EGF, IGF-l, IGF-2, glucagon, corticotropin releasing factor (CRF), dynorfin, endothelin, transforming growth factor ~Y CIGF-a), Pseudomonus endotoxin 40 ~E40), transforming growth factor-,B ~GF-~), insulin and analogs thereof.
Examples of purification means include IMAC and immunoaffinity. Other purification means which employ the use of extension peptides that can be removed using DPP I~ are within the contemplated scope of the present invention.
In one embodiment of the present invention, fusion proteins comprising a biologlcally active polypeptide portion and an extension portion which is a metal chelating peptide are useful i~ arl immobilized metal affinity chromatography system.
ITDmobilized Metal Ion Affinity Chromatography (IMAC) for fractionating proteins was first disclosed by Porath, J. et al., Nature 258:598-599 (19~5). Porath disclosed derivatizing a resin with imi~odiacetic acid (IDA) and chelating metal ions to the IDA~erivatized resin.
3~ Porath disclosed proteins could be immobilized in a column which contained irmnobilized metal ions. It involves attaching a cornmonly used iminodiacetic acid (IDA) to a matrix followed by .

~09~12 wo 92tlO576 Pc~r/ussl/o9l5 chel$ing a metal ion to the IDA-containing resin. The proteins bind to the metal ion(s) through functional groups of amino acid residues capable of donating electrons. Potential electron donating amino acid residues are cysteine, histidines, and tryptopham Proteins interact with metal ions through one or more of these amino acids wi~h electron donating side chains.
Smith et al. discloses in U.S. Patent No. 4,569,794, incorporated by reference herein, that certain amino acids residues are responsible for the binding of the protein to the ~mmobilized me~al ions. However, the bound protein can be eluted by lowering the pH or using competitive counter ligands such as imidazole if histidine side chains are involved in the binding. Histidin~containing di- or tripeptides in proteins have been used to show that IMAC
is a selective purification technique. Accordingly, Smith et al. discloses using recombinant DNA techniques to produce a fusion protein comprising a metal chelating peptide covalently liked to a desired polypeptide. The metal chelating peptide is an extension portion that is effectively a handle to the desired polypeptide. This handle can be used in protein purification.
Use of IMAC technology with metal chelating peptides having alternating His residues is disclosed in U.S. Patent Application Serial No. 07/506,605, which is incorporated herein by reference. U.S. patent application Serial No. 07/~06,605 discloses specific metal chelating peptides which provide unexpectedly superior results in the IMAC purification of a fusion protein when the metal chelating peptide comprises three to six alternating His residues.
Followiog the teachings of U.S. patent application Serial No. 07/506,605 and U.S. Patent No.

4,569,794, it is possible to employ the cornrnonly used IDA resin in IMAC for the purification of fusion proteins having a metal chelating peptide por~ion with at least three alternating histidine residues which are constituents of DPP IV-recognized sequences. Construction of fusion proteins and their use in an IMAC system is taught by U.S. Patent No. 4,569,794.
Construction and use of a metal chelating peptide portion comprising alternating His residues is taught in U.S. Patent Application Serial No. 07/506,605. By providing a ~usion protein with DPP IV recogrlizable residues between alternating His residues the present invention provides a fusion protein which can be purified using IMAC technology and subsequently processed with DPP IV to yield a desired polypeptide.
According to this embodiment of the present invention, the extension portion is a metal chelating peptide which Gall be represented by the formula:
A-X-Y(X -~D
and further, wherein A is optional, and when present is methionine;
n represents the number of sequentially linked X'-Y groups, that number representing from O to 20 of such groups, preferably 0 to 10 groups.
X is selected from the group consisting of any naturally occuring amino acid;
Y is selected from the group consisting of proline, alanine, serine, and threonine, ~ .

209~2 WO 92/1~76 Pcr/uss1/o9l5 except when n = p, then Y is selected ~rom the group consisting of alanine, serine, and threonine;
X' is selected from ehe group consisting of any naturally occuring arnino acid excep~
prolin0 or hydroxyproline;
wherein at least two ~o three residues designated X' and, optionally, X are Histidine (His). Preferably, Y is Pro and n is at least 3. When treated with DPP IV, the N-terrninal e~tension is cleaved in a stepwise fashion, producing the biologically active polypeptide provided the biologically active polypeptide is not itself a DPP IV substrate.
One example of a fusion protein includes an extension portion having the formula His-y(His-y)n wherein n=3 so 8, and Ys are Pro, Hyp, Ala, Ser or Thr, Pro being the most preferred. Another exarnple of a fi~sion protein includes an Extension portion having the formula X-Y(EIis-Y)n wherein n is 3 to 8, X is any naturally occurring amino acid; and Ys are Pro, Hyp, Ala, Ser or Thr, Pro being the most preferred.
Since N-terminal Met is a consequence of protein synthesis in E.coli and it is known to be processed by the E. coli enzymatic system when the adjacent amino acid is Pro, Gly, Ala or Ser, the following extensions represent sequences useful for the IMAC puriflcation and 20 cleavage of biologically active peptides or proteins expressed intracellularly in E. coli by recombinant DNA techniques.
Met-X-Y(HiS-y)n wherein n = 3 to 8, X is Pro, Gly, Ser, or Ala; and Ys are Pro, Hyp, Ala, Ser, or Thr, Pro being the most preferred.
In another exunple, if the peptide or protein desired is to be secreted from a given host after transformation or transfection then the vectors could be designed so as to secrete the protein or polypeptide using an extension portion which ~acilitate transport, such as:
X~Y~(EIiS~Y)n wherein n = 3 to 8, X could be any naturally occurring amino acid compatible with the secretion system frorn a given host and Ys are Pro, Hyp, Ala, Ser, or Thr.
Ano~er protein purification system which uses fusion proteins and which is well suited for DPP IV processing technology is immunoaffinity purification. U.S. Patent No. 4,782,137 issued November 1, 1988 to Hopp et al., incorporated herein by reference, discloses the synthesis of a fusion protein having a highly antigenic N-terminal portion and a desired polypeptide at the C-terrninal portion. According to Hopp et al., the fusion proteins are purified from crude supernatent by passing crude supernatent through a column containing - WO 92/lû576 PCr/US91/0915 -17-~
irnmobilized antibodies which recognize the antigenic portion of the fusion protein. The immobilized antibodies keep the protein in the coluTnn while the undesired components of the supernatant are eluted. The column conditions can then be changed to cause the antigen-antibody complex to dissociate.
S According to the presnt invention, the highly aneigenic N-terminal portion of the fusion protein is an extension portion which contains DPP IV recognizable residues. After collec~ion as described in the Hopp patent, the fusion protein according to the present invention can be e~cposed to DPP IV, thereby removing the extension portion. One of ordinary skill in the art could practice the immunoaffinity purification system of Hopp with N-terrninal e~ctensions according to the present invention.
The embodiments and exarnples described herein serve to illustrate the nature of the present invention and are not meant to limit the scope of the invention. Contemplated equivalents include fusion proteins which have N-terminal extensions which can be processed by at least one other means such that removal of the extension is due to a combination of means. Contemplated equivalents also include fusion polypeptides comprising chemically modified a nino acid residues.
EXAMPLES
E~cample 1 SyTlthetic Prodrugs which are Fusion Prodrugs Having Core Proteins that are l)PP IV Substrates Fusion polypeptides that can be synehesized and administered as prodrugs have a DPP
IV degradable N-terminal extension covalently linked to the N-terminal of the biologically active polypeptide. The formula for these prodrugs can be expressed as the formula:
extension portion - core protein drug portion wherein "extension portion" represents a DPP IV cleavable N-terminal extension; " - "
represents a covalent peptide bond; and, "core protein portion" represents any desired peptide which is liberated from the extension portion by DPP IV processing. In this example the core protein of the fusion protein is a potential substrate for DPP IV following removal of the e~ctension portion.
Synthetic prodrugs can be produced using peptide synthesis techniques well known in 30 ~e ar~.
In one embodiment, the core protein portion is epidermal growth factor (EGF) and the extension portion is Gly-Pro-Phe-Ala:
Gly 4-Pro~3-Phe~2-Ala~l-lEGF].
In another embodiment, the core protein portion is glucagon and the extension portion 35 is Ala-Pro-Ph~Ala:
Ala~-Pro~3-Phe~2-Ala~1 -~GLUCAGON] .

', WO 92/1~576 2 0 9 4 ~ ~ ~ PCI`/U~91lO915~ ~
~ - 1 8-In another embodiment, ~e core protein ponion is IAla15 Leu27]-bGRF (1-29~NH2 (Seq ID 3) ~nd the extension portion is Tyr-Ala:
Tyr~2-Ala~l-{[Ala15 Leu27]-bGRF (1-29)NH2}.
Example 2 Synthetic P~odrugs Which are Fusion Proteins Having Core Proteins that are not DPP IV Substrates Fusion polypeptides that can be synthesized and administered as prodrugs have a DPP
IV degradable N-terminal extension covalently linked to the N-terminal of the biologically active polypeptide. The formula for these prodrugs can be expressed as the formula:
exterJsion portion - core protein drug portion wherein "extension portion" represer~ts a DPP IV cleavable N-terminal extension; " - "
represents a covalent peptide bond; and, "core protein portion" represents any desired peptide which is liberated from the extension portion by DPP IV processing.
Synthetic prodrugs can be produced using peptide synthesis techniques well known in ~e art.
In one embodiment, the core protein portion is a bGRF analog, ~Val2,Ser8~28,Leu27~-bGRF (1-33)0H (Seq ID 1), and the extension portion is Gly-Pro-Tyr-Ala:
Gly~ Pro~3 Tyr~2 Ala-1 {[Val2Ser8~28AIa15LeU27]-bGRF (1-33)0H}.
In another embodiment, the core protein portion is corticotropin releasing factor (CRF) and the e~tension portion is Gly-Pro-Phe-Ala:
Gly~4-Pro~3-Phe~2-Ala~l-[CRF].
In another embodiment, the core protein portion is dynorfin and the extension portion is Phe-Pro-Pbe-Ala:
Phe~ Pro~3 Phe~2 Ala~l [DYNORFIN]
In another embodiment, the core protein portion is somatostatin-28 and the extension portion is Gly-Pro-Phs-Pro:
C;ly~-Pro~3-Phe~2-Pro~l-[SOMATOSTATlN-28] .
In another embodiment, the core protein portion is endothelin and the extension portion is Ala-Pro-Phe-Ala:
Ala~-Pro~3-Phe~2-Ala~l-[ENDOTHELIN] .
In another embodiment, the core protein portion is a bGRF analog [Ile2Ser8 28Alal5 Leu27]-bGRF (l~O)OH (Seq ID 2) and the extension portion is Phe-Ala:
phe-2 Ala-1 ~[l1e2 Ser8,28 AlalS Leu27]-bGRF (I 4Q)OH}.
In another embodiment, the core protein portion is ple2AlalsLeu27]-bGRF tl-29) NH2 (Seq ID 4) and the extension portion is Tyr-Ser:
Tyr~2-Ser~l-{[lle2 Alal5 Leu27]-bGRF (1-29)NH2}.
E~ample 3 Sustained Presence of bGR~ Analog Leu27-bGRF (1-29)NH2 .

WO 92/1057S 2 0 9 ~ 512 PCr/US91/091~2 A bGRF analog, Leu27-bGRF (1-29)NH2, its sequence shown as Seq ID 5, can be administered as a medicament comprising the core protein shown in Seq lD S and a variety of N-terminally extended prodrugs.
Several versions of prodmgs can be made by well known methods using Seq ID 5 as 5 the core protein portion. Extension portions for these Seq ID 5-based prodrugs are lle-Ala, Gly-Pro-Ile-Pro, Seq ID 6, Seq ID 7, Tyr-Ala, Gly-Pro-Tyr-Ala, Seq ID 8, Seq ID 9, Seq ID
10, Seg ID 11, Seq ID 12, Seq ID 13, Tyr-Ala-Tyr-Ala and Val-Ala.
Example 4 Sustained Presence of bGRF Analog [l~r2AlalSLeu27]-bGRP (1-29)NH2 A bGRF analog, [Thr2Ala1SLeu27]-bGRF (1-29)NH2, its sequences shown as Seq lD
lO 14, can be administered as a rnedicament comprising the core protein portion shown in Seq ID
14 and a variety of N-terminally extended prodrugs. Three versions of the prodrug were made having extension portions of Tyr-Thr, Tyr-Ser, and Tyr-Thr-Tyr-Thr, respectively.
Example 5 Fusion Proteins which Contain HIV RNase H and N-Terminal ~xtensions A strategy to purify chimeric proteins from recombinant E. coli is described based on 15 metal chelating peptide domains containing alternate histidines, with affinity for an immobilized metal ion. Vectors are constructed to direct ~he synthesis of f~lsion proteins using HIY RNase H as the core protein. As shown below, these fusion proteins are designed to possess alternatin~g histidines for purification by immobilized metal ion affinity chromatography (IMAC) -and alternating prolines Ot alternating alanines ~or DPP IV cleavage to remove the metal 20 chelating peptide (mcp).
The preferred DPP IV cleavable N-terminal extensions according to the pr~sent invention are outlined as follows:
Fusion protein H~RH/mcp #1 comprises an N-terminal extension of Seq ID 1~ linkedto HIV RNase H:
2S Met~11-Pro~l0-Ala~9-His~8-Pro~7-His~-Pro~5-His~-Pro~3-His~2-Ala~1-[HlV
RNase H]
Fusion protein HIVRH/mcp #2 comprises an N-terminal extension of Seq ID 16 linked to HIV RNase H:
Met'l l-Ala~10-Pro~9-His~8-Ala~7-His~-Ala~5-His~-Ala~3-His~2-Ala~l-[HlV
30 RNase H]
Fusion protein HIVRH/mcp #3 comprises an N-terminal extension of Seq ID 17 linlced to HIV RNase H:
Met~l l-Gl jl0-Pro~9-His~8-Pro~7-His~-Pro~s-His~-Pro~3-His~2-Ala~l-[HIV
RNase Hl These fusion proteins are cloned and expressed in E. coli, and are purified using DEAE
chromatography and RP-HPLC. N-terminal sequencing is used to characterize the fusion , . . . . ............ . . .
.'' ' : ' :

6 2 0 9 4 ~ ~ 2 PCr/l IS91/09152 proteins. Application of the alternating histidine-containing fusion proteins to the purification of recombinant proteins by IMAC and subsequent removal of the N-terminal extension by DPP
IY confinn the utility of the present invention.
Cons~oction of Chimeric Genes Containing HIV RNase H Gene All recombi~ant DNAs are prepared by standard techniques. Oligonucleotides cor-r~sponding to ~e metal chelating peptide/cleavage sequence a e constructed, purified, annealed and ligated to a gene encoding HIV RNase H to fonn a c~imeric gene.
To prepare expression vectors encoding altern~te histidines/DPP IV recognized cleavage residues/HIV-RNaseH, a chimeric gene is inserted into the final expression vector. E~pression 10 vectors containing the chimeric gene constructs are used to transform E. coli by standard techniques. Expression of the genes in E. coli results in the production of the fusion proteins encoded by the chimeric genes. These fusion proteins contain HIV RNase H amino acids and an N-terminal extension which contains alternate histidines (metal chelating peptide) and alternate prolines or aianines.
15 Preparation of Crude E. coli Extracts and lsolation of Fusion Proteins for Sequencing Approximately 3 g of E. coli cell paste is suspended in 30 ml of 0.25 M potassium phospbate, pH 7.2 containing 1 mM dithiothreitol a)TT), EDTA, phenylme~hylsulfonyl fluoride (PMSF), aod benzamidine HCL, 10 mg/liter aprotinin, leupeptin, and bestatin. This suspension is passed through a French Press three times to break the cells. Cell Iysates are 20 centrifuged at 12,000 rpm for 1 hr. The supernatant is removed and solid ammonium sulfate added to 70% saturation. After stirring for 1 hr, the suspension is centrifuged at 12,000 rpm for 1 hr. The supernatant is discarded and the pellet is redissolved in 2.25 mls of 50 mM Tris pH 7.5 containing 1 mM Dl~, PMSF, and benzamidine. The solution is then dialyzedovernight in 20 mM Tris, 50 mM NaCI, 1 mM DTI, 10~ glycerol, and 0.1 mM EI:TA pH25 7.5 (~uffer A) at 4C. The dialysate is collected, diluted with one volume of Buffer A, and applied to a 10 ml column of washed DEAE cellulose equilibrated in Buffer A. The run through is collect~d batchwise and the column further washed with 50 mls of Buffer A. These solutions are collect~d, pooled, and concentrated by 70% ammonium sulfate precipitation and resuspended in 2 mls of Buffer A and dialyzed as described above. Concentrated RH is used 30 for characterization by N-terminal sequence anaiysis.
Puriflcation of l:usion Proteirls Using IMAC
The feasibility of using a metal chelating peptide for the purification OI recombinant proteins from crude extracts can be dernonstrated by using the following chimerics expressed in recombinant E. coli with HIV RNase H as the model pro~ein. Fusion proteins HIV RNase 35 H/mcp #1, HIV RNase H/mcp #2 and HIV RNase H/mcp #3 are each purified.
IMAC columns are prepared as follows. Chelating Sepharose Fast Flow from ~.

209~2 Pharmacia is wæ~ed thoroughly with Milli-Q water on a scintered ~lass filter. The gel is the resuspended in water to form a slurry. The slurry is poured carefully into a glass column (Pharmacia) to a volume of 6 m!s (1 x 7 cm). After the gel h~s settled, the colurnn is wash~d with S volumes of 50 mM EDTA (ethylenediaminetetraacetic acid) pH 8Ø Following this, the 5 column is washed with 5 volumes of 0.2 N NaOH and 5 Yolumes of Milli-Q water. Th~
colurnn then is charged with 5 volumes of 50 mM NiSo4 (or ~nC12 or CuSO4). Finally, the column is washed with S bed volumes of equilibration buffer. T~e equilibration buffer is made up of 20 mM Tris pH 8.0, containing 500 mM NaCI, 1 mM PMSF, 1 mM ben~amidine, 10mg/L leupeptin, and 10 mg/L aprotinin.
The column has been equilibrated with at least 5 volumes of equilibration buffer. 5-10 mls of crude recombinant E. coli extract are applied to the colurnn by gravity. After all the crude material has entered the column, the column is washed with 10 column volumes of equilibration buffer containing 1.0 M NaCI, instead of 500 mM NaCl, pH 8Ø
The column is then eluted with increasing concentrations of imidazole in the 15 equilibration buffer at pH 8Ø For the earlier experiments, a large number of elutions are performed for each experiment to determine the concentration at which the chimeric eluted.
Later this elution is simplified and usually just three imida ole concentrations ase used: 35 mM, 100 rnM, and 300 mM imidazole in the equilibration buffer, pH 8Ø Ten bed volumes of each imidazole buffer are used. Between elutions, the column is washed with 10 volumes of 20 equilibration buffer. Finally, the column is stripped with S bed volumes of 50 mM EDTA pH
8.0 to determine if any protein is still bound to the column. The flow rates for the columns are 1.0 rnl/min. 5 ml fractions are collected. The columns are run at room temperature.
Commercially available Pierce protein assay kits are used to determine the protein content o~ the samples.
2~ HIV RNase H activity is determined by the method described in Becerra, S. P. et al, FEBS 270(1,2):7~80 (September ~990), incorporated herein by reference.
Co~lversion of the N-terminal extended fusion proteins to mature proteins Cornmercially available DPP IV purified from human placenta (Enzyme Systems Products, Dublin, Ca.) with a specific activity of ~200 mU per mg protein is used. One U is 30 equivalent to hydrolysis of 1 umole of a synthetic substrate, Ala-Pro-7-amino~-2 trifluoromedlyl coumarin in one minute at pH 7.8. Enzymatic conversion is carried out by Incubation of the ~sion protein (about 1-100 mg) at a concentration of 1-10 mg/ml with DPP
IV at 25 degrees C for 30 minutes at an enzyme to substrate ratio of 1:100 (w/w). The desired polypeptide is recovered from the uncleaved fusion protein by IMAC. The authenticity is 3~ confirmed by N-terminal sequence analysis.
13xample 6. Processing of bGRF Analog prodrugs in bovine plasma in vitro.

. ~
' ' wo 92/1~76 2 0 9 ~ ~ ~ 2 Pcr/USsl/0915~ `

Table I sum~r3zes representative experiments to demonstrate generation of the core peptide, [Leu27]-bGRF(1 29)NH2 (Seq ID 5) from its three N-~erminally extended analogs:
Tyr 4-Ala~3-Tyr~2-Ala~l-{[Leu27]-bGRF(1-29)NH2} (Seq ID 19),11e~2-Ala~l-{~Leu27]-bGRF(I-29)NH2} (Seq 11) 18) and Tyr~2-Ala~l-{[Leu27~-bGRF(1-29)NH2} (Seq ID 25) upon incubation 5 with bovine plasma in vi~ro. The only metabolites detec~ed in the incubation mixtures were those which were products of DPP-IV-related cleavages.:
Tyr~-Ala~3-Tyr~2-Ala~l-{[Leu27~-bGRF(1-29)~H2} (Seq ID 19~ was sequentially ~onverted over time to Tyr~2-Ala l-~[Leu27]-bGRF(1-29)NH2} (Seq ID 25), [Leu27]-bGRF(l-29)NH2 (core pe~tide, Seq ID S) and fimally to [Leu27]-bGR~(3-29)NH2 (Seq. ID 24).
Tyr~2-Ala~l-~[Leu27]-bGRF(1-29)NH2} (Seq ID 2S) was converted to [Le*7]-bGRF(1-29)NH2 (core peptide, Seq ID 5) and then to [Leu27]-bGRF(3-29)NH2 (Seq ID 24). The core peptide itself, [Leu27]-bGRF(1-2g)NH2 (Seq ID 5) was converted to [Leu27]-bGR~(3-29)NH2 (Seq ID 24) by plasma DPP-IV.
Even though no other me~abolites were observed under the HPLC conditions used in15 the experiments, the peptide [Leu27]-bGRF(3-29)NH2 (Seq ID 24) disappeared over time which indicates that other~ non-DPP-lV related, proteolyses must have also been taking place but at significantly slower rates. Not only was the core bGRF peptide generated from the DPP-IV-cleavable bGRF prodrugs shown here but also the half-life of the core protein genera~ed from the fusion p~otein was significantly prolonged in vitro as compared with the half-life of the core 20 peptide (Seq ~D 5) incubated directly with bovine plasma ~able 1). Moreover, the time at which the core peptide released from the prodrug is present seems to correlate well with the prodrug extension length: Half-life of Seq ID 5 generated from the prodrug with four amino acid residues in the extension part (Seq ID 19) was longer than that one derived from proGRFs having only two amino acids in the extension like in Seq ID 18 or 2S.
25 E~ample 7 In uivo and in w~ro bioactivi~y of bGRF Analog prodrugs.
As shown in Table 2, plasma grow~ hormone ~GH) levels were elevated when Holstein steers were injected iv with analog Seq lD 18. At the dose of 0.2 nmollkg body weight, the induced growth homone levels in plasma were comparable to those generated upon iv ~njection wi~ the same dose of the core peptide (Seq ID 5). It is important to stress that the extended 30 peptide Seq ID 18 had only ca ~æ inherent potency of the core peptide Seq ID S when both were tested in the in vitro pituitary cell assay for GH release. Therefore, the comparable in viw activity of these two peptides seems to indicate that the core peptide could have been released from the extended peptide in vivo.
The sarne pa~tern of GH release in vivo was also observed for the treatment with bGRF
35 prodrugs having ~our arnino acids residues in the extension (Seq ID 19) as shown in Table 3.
lllere wæ no significant difference in the in vivo induced GH release upon treatment with ~t~9~i2 `~ WO 92tlO576 Pcr~u~91/09152 prodrugs having either 4 or 2 amino acids in the extension (peptides Seq lD 19 and Seq ID 18, respectively) despite the significant difference in the in vitro half-life of the core peptide gener$ed ~rom these two bGRF prodrugs in virro as indicated in Table 1. The in vivo growth honnone release was rapid and the same for the core peptide Seq ID S as well as for Seq ID 19 S and 18, with no difference in the time of the GH peak following the challenge with bGRF
analogs.
Our interpretation of ~iese r esults is that most likely the rapid GH release from prodrugs in vivo is due to t~e overall high tissue and organ DPP-IV levels which are ca. 100 fold higher than the plasma DPP-IV concentration. This will explain the differenc0 in the rate of prodrugs processing in viYo as compared with the cleavages observed in the in vitro e~tperirnents summarized in Table 1. It is also feasible that the half-life of the core peptide generated from its prodrug precursors was extended in vivo but not suf~lciently to show an altered (extended) growth hormone release. It is known that GH release in vivo is modulated by the stimulatory effect of bGRF and inhibitory action of somatostatin (somatotropin release inhibitory faceor, SRIF). In the meal-fed steer model of Moseley, et al. a. Endocr. 17, 253-259, 1988), animals are injected iv with GRF two hours prior to feeding for the reason that the pituitary is more responsive to a GRF challenge before feeding versus following feeding.
Factors associated with feeding such as release of gut/pancreatic SRIF may interfere with the ability of the pituitary to release GH. In other words, no GH will be released from the pituitary even in the presence of (iRF during the SRIF overtone. Normally, in the unchallenged meal-fed steer model, serum GH concentration declines to basal levels for 3 to 6 hrs after feeding (so called trough period) with another exogenous episodic GH pulse at S to 8 hrs following feeding. In response to GRF injection 2 hrs before feeding, the GH response is rapidly occurring within 5-20 min. and GH level remains elevated for 120 to 240 min. before returning to baseline. In the case of GRF prodrugs tested so far, only the first exogenous GH
peak was elevated, the second one did not show any treatment effect. It is possible that the half-life of the core GRF generated from prodrugs in our experiments was not extended long enough to allow for the core peptide to be present in the circulation in sufftcient concentrations to affect the second sxogenous burst of exogenous G}I, usually 4 6 hours after the first one.
Taken together, our results support the general prodrug concept disclosed here because:
(i) bGRF prodrugs having DPP-lV-cleavable N-terminal extensions were processed successfully to produce the core peptide(s) in bovine plasma in vitro via DPP-IY mediated cleavages; (ii) the in vitro half-life of the core peptide generated from the prodrugs was significantly longer and was a ~anction of the N-terminal extension length in the prodrug; (iii) the fact that the bGRF
prodrugs with very low inherent potency were as effective as the core peptide in the release of GH in vivo indicates that most likely the core peptide was generated in vivo as anticipated.

WO 92/10576 2 0 9 4 !~ ~ 2 PCI /US91/09152 ~

E~ample 8 Prepara~ion of Gly~-Pro~3-Ile~2-Pro~l {[Leu27~ bGRF(1-29)NE~2}, trifluoroacetate salt.
The synthesis of the GRF analog peptide Seq ID 26, having the formula:
Gly-Pro-Ile-Pro-Tyr-Ala-Asp-Ala-lle-Phe-Thr-Asn-Ser-Tyr-Arg-Lys-Val-Leu-Gly-Gln-Leu-Ser-S Ala-Arg-Lys-Leu-Leu-Gln-Asp-lle-Leu-Asn-Arg-NH2 (as the CF3COOH salt) is conducted in a stepwise manner as in procedure A which is des~ribed in published PCT patent application PCT/US90/02923 incorporated herein by reference. Amino.a;cid analysis, theoretical values in parantheses: Asp 4.16 (4); Thr 1.07 tl); Ser 1.81 (2); Qlu`2.07 (2), Pro 1.98 (2); Gly 1.99 (2); Ala 2.99 (3); Val 1.13 (1), lle 2.84 (3), Leu 5.08 (5); Tyr 1.93 (2); Phe 0.96 (1); Lys 2.04 (2); Ar~ 2.97 (3).
Example 9 Preparation of Tyr~-Ala~5-Gly 4-Pro~3-1le~2-Pro~l {[Leu27] bGRF(1-29)NH2}, trifluoroacetate salt.
The synthesis of the GRF analog peptide Seq ID 27 which comprises Seq ID 6 as the extension po~tion and which has the formula:
#3H-Tyr-Ala-Gly-Pro-Ile-Pro-Tyr-Ala-Asp-Ala-Ile-Phe-Thr-Asn-Ser-Tyr-Arg-Lys-Val-Leu-Gly-Gln-Leu-Ser-Ala-Arg-Lys-Leu-Leu-Gln-Asp-Ile-Leu-Asn-Arg-NH2 (as the CF3COOH salt) is ~onducted in a stepwise manner as in procedure A which is described in published PCT patent applicatiorl PCT/US90/02923 incorporated herein by reference. Amino acid analysis, theoretical values in parantheses: Asp 4.04 (4); Thr 1.03 (1); Ser 1.74 (2); Glu 2.aS (2), Pro 1.99 (2); Gly 2.00 (2~; Ala 4.01 (4); Val 1.28 (1), Ile 2.84 (3), Leu 5.09 (5); Tyr 2.94 (3);
Phe 0.97 (1); Lys 2.07 (2); Arg 3.00 (3).
Example 10 Preparation of Lys~8-Pro~~-Tyr~-Ala~5-Gly~-Pro~3-Ile~2-Pro~l ~Leu27]
bGRF(1-29)NEI2}, trifluoroacetate salt.
The synthesis of the GRF analog peptide Seq ID 28 which comprises Seq ID 7 as the extension portion and which has the formula:
Lys-Pro-Tyr-Ala-Gly-Pro-lle-Pro-Tyr-Ala-Asp-Ala-lle-Phe-Thr-Asn-Ser-Tyr-Arg-Lys-Val-Leu-Gly-Gln-Leu-Ser-Ala-Arg-Lys-Leu-Leu-Gln-Asp-lle-Leu-Asn-Arg-NH2 (as the CF3COOH salt) is conducted in a stepwise manner as in procedure A which is described in published PCT
patent application PCT/US90/02923 incorporated herein by reference. Amino acid analysis, theoretical values in parantheses: Asp 4.04 (4); Thr 0.95 (1); Ser 1.78 (2); Glu 2.04 (2), Pro 2.91 (3); Gly 1.98 (2); Ala 3.91 (4); Val 0.96 (1), lle 2.86 (3), Leu 5.08 ~5); Tyr 3.06 (3);
Phe 0.97 (1); Lys 3.06 (3); Arg 3.08 (3).
~ample 11 Preparation of Gly~-Pro~3-Tyr~2-Ala~l ~[Leu2~] bGRF(1-29)NH2}, trifluoroacetate salt.
The synthesis of the GR~ analog peptide Seq ID 29 having the formula:
#6 Gly-Pro-Tyr-Ala-Tyr-Ala-Asp-Ala-lle-Phe-Thr-Asn-Ser-Tyr-Arg-Lys-Val-Leu-Gly-Gln-Leu-` Wo 92tllJ576 2 ~ 9 ~ 5 1 2 P(~lrtUSsl/091~2 Ser-Ala-Arg-Lys-Leu-Leu-Gln-Asp-Ile-Leu-Asn-Arg-NH2 (as the CF3COOH salt) is conducted in a stepwise manner as in procedure A which is described in published PCT patent applicacion PCI/US90/02g23 incorporated herein by reference. Amino acid analysis, theoretical values in paranth~ses: Asp 4.01 (~); Thr 0.9S (1); Ser 1.80 (2~; Glu 2.02 (2); Pro 0.97 (l); Gly 1.98 (2); Ala 3.91 (4); Val 0.99 (1), lle 1.89 (2), Leu 5.08 (5); Tyr 3.05 (3); Phe 0.98 (1); Lys 2.03 (2); Arg 3.06 (3).
E~ample 12 Preparation of Tyr~-Ala~~-Gly 4-Pro~3-Tyr2-Ala~1 {[Leu27] bGRF(1-29)NH2}, trifluoroacetate salt.
The synthesis of the GRF analog peptide Seq ID 30 which comprises Seq ID 8 as the extension portion and which has the formula:
#7 Tyr-Ala-Gly-Pro-Tyr-Ala-Tyr-Ala-Asp-Ala-Ile-Phe-Thr-Asn-Ser-Tyr-Arg-Lys-Val-Leu-Gly-Gln-Leu-Ser-Ala-Arg-Lys-Leu-Leu-Gln-Asp-lle-Leu-Asn-Arg-NH2 (as the CF3COOH salt) is conducted in a stepwise manner as in procedure A which is described in published PCT patent application PCT/US90102923 incorporated herein by reference. Amino acid analysis, theoretical values in parantheses: Asp 4.07 (4); Thr 0.96 (1); Ser 1.79 (2); Glu 2.02 (2); Pro 0.99 (1); Gly 1.95 (2); Ala 4.80 (5); Val 0.96 (1), lle 1.87 (2), Leu 5.09 (~); Tyr 4.11 (4);
Phe 0.97 (1); Lys 2.06 (2); Arg 3.08 (3).
E~ample 13 Preparation of Lys~8-Pro~7-Tyr~-Ala~S-Gly~-Pro~3-Tyr~2-Ala~l {[Leu27] bGRF(1-29)NH2}, trifluoroacetate salt.
The synthesis of the GRF analog peptide Seq ID 31 which comprises Seq ID 9 as the extension portion and which has the formula:
#8 Lys-Pr~-Tyr-Ala-Gly-Pro-Tyr-Ala-Tyr-Ala-Asp-Ala-lle-Phe-Thr-Asn-Ser-Tyr-Arg-Lys-Val-Leu-Gly-Gln-Leu-Ser-Ala-Arg-Lys-Leu-Leu-Gln-Asp-lle-Leu-Asn-Arg-NH2 (as tbe CF3COOH
salt) is conducted in a stepwise manner as in procedure A which is described in published PCT
patent application PCT/US90/02923 incorporated herein by reference. Amino acid analysis, theoretical values in parantheses: Asp 4.06 (4); Tbr 0.95 (1); Ser 1.78 (2); Glu 2.ûl (2); Pro 1.95 (2); Gly 1.96 (2); Ala 4.81 (5); Val 0.95 (1), lle 1.87 (2), Leu 5.09 (5); Tyr 4.12 (4);
Phe 0.96 (1); Lys 3.08 (3); Arg 3.10 (3).
Example 14 Preparation of Phe~10-Ala~9-Lys~8-Pro~7-Tyr~-Ala~5-Gly~-Pro~3-Tyr~2-Ala~
~1Leu27] bGRF(1-29)NH2}, trifluoroacetate salt.
The synthesis of tl e GRF analog peptide Seq lD 32 which comprises Seq ID 10 as tbe e~tte~sion portion and which has tbe formula:
#9 Ph~Ala-Lys-Pro-Tyr-Ala-Gly-Pro-Tyr-Ala-Tyr-Ala-Asp-Ala-lle-Pbe-Thr-Asn-Ser-Tyr-Arg-Lys-Yal Leu-Gly-Gln-Leu-Ser-Ala-Arg-Lys-Leu-Leu-Gln-Asp-lle-Leu-Asn-Arg-NH2 (as the CF3COOH salt) is coslducted in a stepwise manner as in procedure A which is described in published PCT patent application PCT/US90/02923 incorporated herein by reference Amino 209~512 7~ Pcr/ussl/o9ls2`

acid analysis, ~heoretical values in paran~heses; Asp 4.16 (4); Thr 1.01 (1); Ser 1.89 (2); Glu 2.08 (2); Pro 1.91 (2); (ily l.g3 (2); Ala 5.72 (6); Val 0.97 (1), lle 1.90 (2), Leu 5.09 (5);
Tyr 4.09 (4); Phe 1.99 (2); Lys 3.08 (3); Ar~ 3.04 (3).
Example 15 Preparation of Gly~l2-Pro~ll-Phe~l0-Ala~9-Lys~8-Pro~7 Tyr~ Ala~5 Cil 4 P ~3 S Tyr~2-Ala~l ~[Leu~7] bGRF(1-29)NH2}, trifluoroacetate salt.
The synthesis of the GRF analog peptide Seq ID 33 which comprises Seq ID 11 as ~he extension portion and which has the formula~
#lû Gly-Pro-Phe-Ala-Lys-Pro-Tyr-Ala-Gly-Pro-Tyr-~-Tyr-Ala-Asp-Ala-lle-Phe-Thr-Asn-Ser-Tyr-Arg-Lys-Val-Leu-Gly-Gln-Leu-Ser-Ala-Arg-Lys-Leu-Leu-Gln-Asp-lle-Leu-Asn-Arg-NH2 (as the CP3COOH salt) is conducted in a stepwise manner as in procedure A which is described in published PCT patent application PCTlUS90/02923 incorporated herein by reference.
Arnino acid analysis, theoretical values in parantheses: Asp 4.08 (4); Thr 0.96 (1); Ser 1.79 (2); Glu 2.07 (2); Pro 2.88 (3); Gly 2.94 (3); Ala 5.74 (6); Val 0.96 (1), lle 1.88 (2), Leu 5.13 (5); Tyr 4.11 (4); Phe 1.99 (2); Lys 3.10 (3); Arg 3.09 (3).
Example 16 Preparation of Val~14-Pro~l3-Gly-12-Pro~1l-Phe~l0-Ala~9-Lys~g-Pro~7-Tyr~-Ala 5-Gly~-Pro~3-Tyr~2-Ala~l {[Leu27] bGRF(1-29)NH2}, trifluoroacetate salt.
The synthesis of the GRF analog peptide Seq ID 34 which comprises Seq ID 12 as the extension portion and which has ehe formula:
#11 Val-Pro-Gly-Pro-Phe-Ala-Lys-Pro-Tyr-Ala-Gly-Pro-Tyr-Ala-Tyr-Ala-Asp-Ala-lle-Phe-Thr-20 Asn-Ser-Tyr-Arg-Lys-Val-Leu-Gly-&ln-Leu-Ser-Ala-Arg-Lys-Leu-Leu-Gln-Asp-Ile-Leu-Asn-,4rg-NH2 (as ~e CF3COOH salt) is conducted in a stepwise manner as in procedure A which is described in published PCT patent application PCT/US90/02923 incorporated herein by reference. Amino acid analysis, theoretical values in parantheses: Asp 4.W (4); Thr 0.96 (1);
Ser 1.81 (2); Glu 2.04 (2); Pro 3.80 (4); Gly 2.99 (3); Ala 5.92 (6); Val 1.98 (2), Ile 1.89 (2), 25 Leu 5.10 (5); Tyr 4.09 (4); Phe 1.99 (2); Lys 3.06 (3); Arg 3.08 (3).
Example 17 Preparation of Arg~16-Pro l5-Val~14-Pro~l3-Gly~l2-Pro~11 Phe~l0 Ala 9 Lys 8 Pro~7-Tyr~-Ala~S-Gly'4-Pro'3-Tyr'2-Ala~l {[Leu27] bGR~:(1-29)NH2}, trifluoroacetate salt.
The synthesis of the GRF analog pepeide Seq ID 35 which comprises Seg lD 13 as the 30 extension portion and which has the formula:
#12 Arg-Pro-Val-Pro-Gly-Pro-Phe-Ala-Lys-Pro-Tyr-Ala-Gly-Pro-Tyr-Ala-Tyr-Ala-Asp-Ala-Ile-Phe-Thr-Asn-Ser-Tyr-Arg-Lys-Val-Leu-Gly-Gln-Leu-Ser-Ala-Arg-Lys-Leu-Leu-Gln-Asp-lle-Leu-Asn-Arg-NH2 (as the CF3COOH salt) is conducted in a stepwise manner as in procedure A
which is described in published PCT patent application PCT/US90/02923 incorporated herein 35 by reference. Amino acid analysis, theoretical values in parantheses: Asp 4.10 (4); Thr 0.98 (l); Ser 1.84 (2); Glu 2.03 (2); Pro 4.82 (5); Gly 2.97 (3); Ala 5.91 (6); ~al 1.99 (2), lle 1.88 :; WO 92/10576 PCI/US91/09152 (2), Leu 5.09 (5); Tyr 4.10 (4); Phe 1.98 (2); Lys 3.03 (3); Arg 4.09 (4).
Fxample 18 Preparation of Val~2-Ala~l g[Leu27] bGRF(1-29)NH2}, ~rifluoroacetate salt.
The synthesis of the GRF analog peptide Seq ID 36 having the formula:
#14 Val-Ala-Tyr-Ala-Asp-Ala-lle-Phe-Thr-Asn-Ser-Tyr-Arg-Lys-Val-Leu-Gly-Gln-Leu-Ser-Ala-S Arg-Lys-Leu-Leu-Gln-Asp-Ile-Leu-Asn-Arg-NH2 (as the CF3COOH salt) is conducted in a stepwise manner as in procedure A which is described in published PCT patent application PCT/US90/02923 incorporated herein by reference. Amino acid analysis, ~heoretical va]ues in parantheses: Asp 3.98 (4); Thr- 0.89 (1~; Se~ 1.76 (2); Glu 2.02 (2); Gly 1.05 (1); Ala 3.87 (4);
Val 1.85 (2), lle 1.77 (2), Leu 5.17 (5); Tyr 2.04 (2); Phe 0.97 (1); Lys 2.07 (2); Arg 3.06 10 (3).
Exarnple 19 Preparation of Tyr~2-Thr~l {~Alal; Leu27] bGRF(1-29)NH2~, trifluoroacetate salt.
The synthesis of the GRF analog peptide Seq ID 37 having the formula:
#15 Tyr^Thr-Tyr-Ala-Asp-Ala-lle-Phe-l'hr-Asn-Ser-Tyr-Arg-Lys-Val-Leu-Ala-Gln-Leu-Ser-Ala-15 Arg-Lys-Leu-Leu-Gln-Asp-Ile-Leu-Asn-Arg-NH2 (as ~e CF3COOH salt) is conducted in a stepwise manner as in procedure A which is described in published PCT patent application PCTtUS90/02923 incorporated herein by reference. Amino acid analysis, theoretical values in parantheses: Asp 4.06 (4); Thr 1.86 (2); Ser 1.77 (2); Glu 2.07 (2), Ala 3.98 (4); Val 1.08 (1), lle 1.89 (2), Leu 5.14 (5); Tyr 2.94 (3); Phe 0.96 (1); Lys 1.99 (2); Arg 3.04 (3).
20 Example 20 Preparation of Tyr~2-Thr~l {[lle2 Ala15 Leu27] bGRF(1-29)NH2}, trifluoroacetate salt.
The synthesis of the ~:iRF analog peptide Seq ID 38 having the formula: -#16 Tyr-Thr-Tyr-lle-Asp-Ala-lle-Phe-Thr-Asn-Ser-Tyr-Arg-Lys-Val-Leu-Ala-Gln-Leu-Ser-Ala-Arg-Lys-Leu-Leu-Gln-Asp-Ile-Leu-Asn-Arg-NH2 (as the CF3COOH salt) is conducted in a 25 stepwise manner as in procedure A which is described in published PCT patent application PCT/US90/02923 incorporated herein by reference. Amino acid analysis, theoretical values in parantheses: Asp 4.07 (4); Thr 1.87 (2); Ser 1.75 (2); Glu 2.07 (2~, Ala 2.94 (3); Val 1.09 (1~, Ile 2.87 (3), Leu 5.12 (5); Tyr 2.92 (3); Phe 0.96 (1); Lys 2.00 (2); Arg 3.05 (3).
Example 21 Preparation of Tyr~2-Thr~l ~[Thr2 Alal5 Leu27] bGRF(1-29)NH2}, trifluoroacetate salt.
The synthesis of the GRF analog peptide Seq ID 39 having the fo~nula:
#17 Tyr-Thr-Tyr-Thr-Asp-Ala-lle-Phe-Thr-Asn-Ser-Tyr-Arg-Lys-Val-Leu-Ala-Gln-Leu-Ser-Ala-Arg-Lys-Leu-Leu-Gln-Asp-lle-Leu-Asn-Arg-NH2 (as the CF3COOH salt) is conducted in a stepwise manner as in procedure A which is described in published PCT patent application 3S PCT/US90/02923 incorporated herein by reference. Amino acid analysis, theoretical values in parantheses: Asp 4.05 (4); Thr 2.68 (3); Ser 1.77 (2); Glu 2.07 (2), Ala 2.90 (3); Val 1.08 (1), WO 92/lû576 2 0 9 4 ~ ~ 2 PCr/~S91/Og~2 ~

lle 1.89 (2), Leu 5.20 (5); Tyr 2.87 (3); Phe 0.93 (1); Lys 2.01 (2); Arg 3.07 (3).
Fxample 22 Preparation of Tyr~2-Ser~l ~[Thr2 Ala15 Leu27] bGRF(1-29)NH2}, trifluoroacetate salt.
The synthesis of the GRF analog peptide Seq ID 40 having tbe fonnula:
5 #18 Tyr-Ser-Tyr-Thr-Asp-Ala-1Je-Phe-Thr-Asn-Ser-Tyr-Arg-Lys-Val-Leu-Ala-Gln-Leu-Ser-Ala-Arg-Lys-Leu-Leu-Gln-Asp-11e-Leu-Asn-Arg-NH2 (as the CF3COOH salt) is conducted in a stepwise manner as in procedure A which is described, jh~published PCT patent application PCT/US90/02923 incorporated herein by reference,~ ,A~rnino acid analysis, theoretical values in parantheses: Asp 4.~1 (4); Thr 1.82 (2); Ser 2.64 (3); Glu 2.05 (2), Ala 2.90 (3); Val 1.04 (1), 10 lle 1.87 (2), Leu 5.16 (5); Tyr 2.92 (3); Phe 0.94 (1); Lys 2.01 (2); Arg 3.04 (3).
Example 23 Preparation of Tyr 4-Thr~3-Tyr~2-Thr~1 g[Thr2 Ala15 Leu27] bGRF(1-29)NH2}, trifluoroacetate salt.
The synthesis of the C;RF analog peptide Seq ID 41 having the formula:
#19 Tyr-Thr-Tyr-Thr-Tyr-Thr-Asp-Ala-Ile-Phe-Thr-Asn-Ser-Tyr-Arg-Lys-Val-Leu-Ala-Gln-15 Leu-Ser-Ala-Arg-Lys-Leu-Leu-Gln-Asp-11e-Leu-Asn-Arg-NH2 (as the CF3COOH salt) is conducted in a stepwise manner as in procedure A which is described in published PCT patent application PCI'tUS90/02923 incorporated herein by reference. Amino acid analysis, theoretical values in parantheses: Asp 4.06 (4); Thr 3.66 (4); Ser 1.85 (2); Glu 2.05 (2), Ala 2.93 (3); Val 1.09 (1),11e 1.91 (2), Leu 5.15 (5); Tyr 3.91 (4); Phe 0.95 (1); Lys 2.00 (2);
20 Arg 3.04 (3).
E~ample 24 Preparation of Tyr~2-Ala~l {[Leu27] bGR~(1-29)NH2}, trifluoroacetate salt.
The synthesis of the GRF analog peptide Seq ID 42 having the formula:
Tyr-Ala-Tyr-Ala-Asp-Ala-lle-Phe-l'hr-Asn-Ser-Tyr-Arg-Lys-~lal-Leu-Gly-Gln-Leu-Ser-Ala-Arg-Lys-Leu-Leu-Gln-Asp-lle-Leu-Asn-Arg-NH2 (as the CF3COOH salt) is conducted in a stepwise 25 manner as In procedure A which is described in published PCT patent application PCT/IJS90/02923 incorporated herein by reference. Amino acid analysis, theoretical values in parantheses: Asp 4.01 (4); Thr 0.97 (1); Ser 1.88 (2); Glu 2.00 (2); Gly 1.02 (1); Ala 3.88 (4);
Val 0.97 (1), Ile 1.86 (2), Leu 5.03 (~); Tyr 3.03 (3); Phe 0.96 t1); Lys 2.18 (2); Arg 3.03 (3).
30 Example 25 Preparation of Tyr~-Ala~3-Tyr~2-Ala~l ~[Leu27] bGRF(1-29)NH2}, ~ifluoroacetate salt.
'rhe synthe~sis of the GRF analog peptide Seq ID 43 having the forrnula:
#13 Tyr-Ala-Tyr-Ala-Tyr-Ala-Asp-Ala-lle-Phe-Thr-Asn-Ser-Tyr-Arg-Lys-Val-Leu-Gly-Gln-Leu-Ser Ala-Arg-Lys-Leu-Leu-Gln-Asp-lle-Leu-Asn-Arg-NH2 (as the CF3COOH salt) is conducted 35 in a stepwise manner as in procedure A which is described in published PCT patent application PCT/US90102923 incorporated herein by reference. Amino acid analysis, theoretical values in . ~ . . .

` ?` 2 0 9 4 5 1 2 PCl/lJS91/091i2 parantheses: Asp 3.97 (4); Thr 0.90 (1); Ser 1.74 (2); Glu 1.98 (2); Gly 1.04 (1); Ala 4.85 (5);
Val 0.91 (1), lle 1.77 (2), Leu 5.13 (5); Tyr 4.14 (4); Phe 0.99 (1); Lys 2.07 (2); Arg 3.05 (3).

WO 92/10576 PClr/US91/09152 Table 1. ~n yitro potency and in vitro plasma stabili~y of selected GRF analogs.

, _. . ~
Peptide Sequence Seq. In Vilro In V~o ¦ ID No. Potency: P~ 2 [Leu27~-bGRF(1-29)NH2 ., ~.. 5 1.00 (E4p81) ~ t _ .
¦ llle~2-Pro~1 { [Leu27]-bGRF( 1 -29)NH2} . 18 0 .045 43P2 1 ) ~ Tyr~2 Ala l~[Leu27]-bGRF(1-29)NH2} tS 0.13 38 5 ¦ Tyr~4-Ala~3-Tyr~2-Ala~1{[Leu27]-bGRF(1-29)NH2} ¦ 19 0.052 2971#

* Peptides were tested in an in vi~ro bovine anterior pituitary cell culture as described by Friedman et al. (Int. J. Peptide and Protein Res. 37:14-20 [1991]).
~* Peptides were incubated at 30 ~M in bovine plasma in Yitro at 37DC as described in Kubiaketal. (DrugMet. Disp. 17;393-397 [1989]). Yaluespresentedhererelateto the half-life of [Leu ]-bGRF(1-29)NH~ (Seq. ID S) incubated directly in plasma or Leu27]-bGRF(1-29)NH2 (Seq. ID 5) generated from extended peptides Seg. lD No.
18, 25 or 19, respectively. Two experiments with two different plasma pools were run, Exp 1 and 2 as indicated in parentheses.
# Peptide Seq. ID 19 was tested against lLeu27~-bGRF(1-29)NH2 (Seq. ID 5) using a different bovine plasma pool. The half-life of Seq. ID S in this plasma specimen was 50.2 min.

.... . ... . . . .
.: .
., .

~ LIY4~
i;wo 92/10~76 PCI /US91/0915 ~3l-Table 2. Serum GH Response to IV Injections of Various Doses of [Leu27]-bGRF(l-29)NH2 (Seq ID S) and lle~2-Pro~l~[l,eu~7]-bGRF(1-29)NH~} (Seq ID 18) in Mc;~l-lFed Holstein Steers.~

_ Peak Heighl Time to Area 0-8 h (ng/ml) (Pmta.r~ ) (Unit) ~umber Dose of A' B' A' B' A' ~3a-Treatment nmol/kg Animals RnegspoDd- .
Sal~n~ ~-- ob 32 4b 32.4 89 89D 4 3 4,3b Seq ID 18 0.02 8/10' 71.2b' 76.9~'23' 18c 4.6~' 5.0~' ¦ Seq ID 5 0.20 9/lff 119.8' 130.9' ~ 14' 6.8d 7 0a ¦ Seq ID 18 0.20 10/lff 101.4'a 101.4'26' 26' 1 Seq l~s 18 20.0 10/10' 137.8a137.8' 23' 23' 10.1' ¦ SEM .04 9.1 9.4 8 8 .3 3 ¦ p Value .OOtsl .Cs07 .007 .04 .03 I .0001 .000 _ . __ _ _ A~imals were iDjected IV with peptides at the doses indicated 2 hrs be~ore fecding and procedures were as described by Moseley et al. J. EDdocrinology 117:253-259 (1988).
Analysis A includes all stecrs and Analysis B includes onsy steers respoDding to GRF
injection and control stecrs.
b,C~d,t Values with dilferent superscripts in a colurnn are significantly different (P~.OS).
;

. . , ,, , . :

20~4~1`2 WO 92/1 0576 PCI /US91/091;2 Table 3. Serum GH Response to.lV Irtiections of Various Doses of [Leu27~-bGRFtl-29)NH2 (Seq ID S) and~yr~-Ala~3-Tyr~2-Ala~l{[Leu27]-bGRF(1-29)NH2}
(Seq ID 19) in M~ Fed Holstein Steers.~

. Peak Height Time to Area 0-10 h (ng/ml) Peak (n~in) (Ur~it) Number Treat- Dose of A' B~ A B A B
meDt nrnol/kg Respond-Saline 30.9b 30.9b 45b ~ 3,9b 3.96 Seq ID 5 0.02 9/12 91.9 101.3 ~ 20 4.7D.~ 4--,7b.C-Seq ID 19 0.02 10/i2 88.5 92.4 30b 22 4.8 4.4b~' Seq ID 5 0.20 11/12 793~ 79-lb`' 21b 18 5~4 53 Seq ID 19 0.20 7/12' 97.4' 120.1' 53b 8' 5-4C 5.4c~d Seq ID 1g 2.0 9/12 74.5C 70.8bC 24b 18 6.9d 6.8d SEM .lO 15.9 17.3 14 8 .4 .4 _ . _ p Value .0001 .Oo .06 .28 .11 .0008 .007 EM5 . . . .1117 3042 3623 2260 763 228] 2483 Steers were injeaed IV with peptides at the doses indicated 2 hrs before feeding and procedures were as described by Moseley et al. ~. Endocrinology 117:253-259 (1988).
Analysis A includes all steers and Analysis B includes oD]y steers responding to GRF
injection and co~trol sleers.
b,C~d Values w~th different superscripts in a column are significanlly differeD~ (P~.05).

2~9~
. ..................................... .
W O 92/10~76 PCT/US91/09152 , SEQUENCE LISTING
(1) G~NER~L INFO~M~TION:
(i) APPLICANT: Kubiak, Teresa M.
Sharma, Satish R.
(Ll) TITLE OF INVENTION: Fu~ion Polypeptide~
~iii) NU~BER OF SEQUENCES: 42 (iv) CORRESPONDENCE ADDRESS:
(A) ADDRESSEE: Upjohn Company - Corp. Patents ~ Trademark~
~B) STREET: 301 Henrie~ta Street ~C) CITY: ~alamazoo (D) STATE: Michigan (E) COUNTRY: USA
(F) ZIP: 49001 . ~v) COMPUTER READABLE FORM:
~A) MEDIUM TYPE: diskette ~3M 3.5, DS double side 1.0 M~) (B) COMPU~ER: IBM PC compatible ~C) OPERATING SYSTEM: PC-DOS/MS-DOS
~D) SOFTWARE: WordPerfect 5.1 20(vi) CU~RENT APPLICATION DATA:
(A) APPLICATION NUMBER:
(B) EILING DATE:
(C) CLASSIPICATION:
(vii) PRIOR APPLICATION DATA:
(A) APPLICATION NUMBER: US07/626,727 (B) FILING DATE: 13/12/90 (vii) PRIOR APPLICATION DATA:
(A) APPLICATION NUMBER: US07/614,170 ~B) FILING DATE: lq/11/90 30(vii) PRIOR APPLICATION DATA:
~A) APPLICATION NUMBER: PCT/US90/02923 (B) FILING DATE: 30/05/90 (vii) PRIOR APPLICATION DATA:
(A) APPLICATION NVMRER: US07/368,231 ~B) FILING DATE: 16/06/89 (vii) PRIOR APPLICATION DATA:
~A) APPLICATION NUMBER: U507/506,605 (B) FILING DATE: 09/04/90 t~iii) ATTORNEY/AGENT INFORMATION:
(A) NAME: DeLuca, MarX
(B) REGISTRATION NUMBER: 33229 (C) REFERENCE/DOCKET NUMBER: 4595 (ix) TELECOMMUNICATION INFORMATION:
(A) TELEPHONE: 616 385 5210 ~B) TELEFAX: 616 385 6897 , ; ' WO 92/1û57h 2 0 ~ 4 5 ~ 2 Pcr/lJsg~ /09152 (2) INFORMA~ION FOR SEQ ID NO:1:
- ~i) SEQVENCE CHAR~CTERISTIC:
(A) L~3NGTH: 33 (B) TYPE: amino acid ~D) TOPOLOGY: lin~ar (xi) S~QUENCE DESCRIPTION: SEQ ID NO:l:

Tyr Val Asp Ala Ile Phe Thr Ser Ser Tyr Arg Lys Val Leu Ala Gln 1 5 10 ~ 15 ,~
L~u S~r Ala Arg Lya Leu Leu Gln Asp Ile Leu Ser Arg Gln Gln Gly Glu (3~ INFORMATION FOR SEQ ID NO:2:
~i) SEQUENCE C~ARACTERIS~IC:
(A) LENGTH: 40 (B) TYPE: amin~ acid (D) TOPOLOGY: linear ~xi) SEQUENCE DESCRIPTION: SEQ ID NO:2:
Tyr Il0 Asp Ala Ile Phe Thr Ser Ser Tyr Arg Ly6 Val Leu Ala Gln Leu Ser Ala Arg Lye Leu Leu Gln ABP Ile Leu Ser Arg Gln Gln Gly Gl~ Arg A~n Gln Glu Gln Gly Ala (4) INFORMATION FOR SEQ ID NO:3:
(i) SEQUENCE CHARACTERISTIC:
tA) LENGT~: 29 ~a) TYPE: amino acid : (D) TOPOLOGY: linear ~lx) FEATURE:
~0 (A) NAME/KEY: C-terminally amidated ArgLnLnyl re~idue ~B) LOCATION: Xaa29 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:3:

Tyr Ala Aep Ala Ile PhQ Thr Asn Ser Tyr Arg Lys Val Leu Ala Gln 4~ 1 5 10 15 WO 92/10576 pcr/us91 /o9l 5 Leu Ser Ala Ary ~ys Leu Leu Gln Asp I le Leu Rsn Xaa ( 5 ) INFO~3TION FOR SEQ ID NO: 4:
~i) SEQU~NCE CaAR~CTERISTIC:
( A ) LENGT~: 2 9 (B) TYPE^ amino acid (D) TOPOLOGY: linear ~ix) FEATURE:
(A) NAMEIKEY: C-termin~lly amidated Argininyl residue (B) LOCATION: Xaa29 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:4:

Tyr Ile Asp Ala Ile Phe Thr Asn Ser Tyr Arg Lys Vai Leu Ala Gln 1 5 10 l5 Leu Ser Ala Arg Ly~ Leu L~u Gln Asp Ile Leu Asn Xaa (6) INFORMATION FOR SEQ ID NO:5:
~i) SEQUENCE CHARACTERISTIC:
(A) LENGTH: 29 (B) TYPE: amino acid ~D) TOPOLOGY: linear (ix) FEATURE:
(A) NAME/XEY: C-terminally amidated Arginlnyl residue ~B) LOCATION: Xaa29 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:5:

Tyr Ala A~p Ala Ile Phe Thr A~n Ser Tyr Arg Lys Val Leu Gly Gln l 5 l0 15 Leu Ser Ala Ar~ Ly~ Leu Leu Gln A~p Ile Leu A~n Xaa (7) TNFORMATION FOR SEQ ID NO:6:
(i) SEQUENCE C~ARACTERISTIC:-(A) LENGTH: 6 ~B) TYPE: amino acid ~D) TOPOLOGY: linear ~xL) SEQUENCE DESCRIPTION: SEQ ID NO:6:
` 45 Tyr Ala Gly Pro Ile Pro . . .

209~L51-~
WO 92/10576 PCI/US~l/O91i' ; ~

(8) INFORMATION FOR SEQ ID NO:7:
5~i) SEQUENCE CHARACTERISTIC:
(A) LENGTH: 8 ..
~B) TYPE: amino acid (D) TOPOLOGY: linear (xi~ SEQUENCE D~SCRIP~ION: SEQ ID NO:7:
Lys Pro Tyr Ala Gly Pro lle Pro (9) INFORMATION FOR SEQ ID NO:8:
(i) SEQUENCE CHARACTERISTIC:
~A) LENGTH: 6 ~B) TYPE: amino acid (D) TOPOLOGY: linear ~xi) SEQVENCE DESCRIPTION: SEQ ID NO:8:

Tyr Al a Gl y Pro Tyr Al a (10~ INFORMATION FOR SEQ ID NO:9:
(i) SEQUENCE CHARACTERISTIC:
(A) LENaTH: 8 (B) TYPE: amino acLd (D) TOPOLOGY: linear ~xi) SEQUENCE DESCRIPTION: SEQ ID NO:9:
Ly Pro Tyr Ala Gly Pro Tyr Ala ~11) INFORMATION FOR SEQ lD NO:10:
(i) SEQUENCE CHARACTERISTIC: ' (A) LENGTH: 10 (B) TYPE: amino acid (D) TOPOLOGY: linear (xi) SEQUENCE DESCRIPTION: SEQ ID NO:10:

Phe Ala Ly~ Pro Tyr Ala Gly Pro Tyr Ala 1 5 . lo 2~i9~12 "WO 92tlO57~ PCI/US~1/0915 ( 12 ) INFOR~iArION FOR SEQ ID NO~
(i) SEQUENCE CHARACTERISTIC:
(~) L2NGTH: 12 ~B) TYPE: aminO aCid (D) TOPOLOGY: 1inear (Xi) SEQUENCE DESCRIPTION: SEQ ID NO:11:
G1Y PrO Phe A1a LYB PrO TYr A1a G1Y PrO TYr A1a 1~

(13) INFORMATION FOR SEQ ID NO:12:
(i) $2QUENCE CaARACTERISTIC:
(A) LENGTH: 14 (B) TYPE: aminO acid (D) TOPOLOGY: 1inear ~Xi) SEQVENCE DESCRIPTION: SEQ ID NO:12:
Va1 PrO G1Y PrO Phe A1a Ly~ Pro Tyr Ala Gly PrO TYr A1 (14) INFORMATION FOR SEQ 1D NO:13:
~) SEQUENCE CHARACTERISTIC:
(A) LENGTH: 16 ~B) TYPE: aminO aCid (D) TOPOLOGY: 1in8ar (Xi) SEQVENCE DESCRIPTION: SEQ ID NO:13:
Arg PrO Va1 PrO G1Y PrO Phe A1a LY~ PrO TYr A1a G1Y PrO TYr A1a 1 s lo 15 ~15) INFORMATION FOR SEQ ID NO:14:
(i) SEQUENCE CaARACTERISTIC:
(A) LENGTH: 29 ~B) TYPE: aminO aCid ~D) TOPOLOGY: 1inear (iX) FEATURE:
~A) NAME/KEY: C-termina11Y amidated ArgininY1 r~3idUe (B) LOCATION: Xaa29 (Xi) SEQUENCE DESCRIPTION: SEQ ID NO:14:
TYr Thr AgP A1a I1e Phe Thr ASn Ser TYr Arg LY9 Va1 LeU A1a G1n 4~ .
L~Y Sar A1a Arg LYS LeU LeU G1n A5P I1e LeU ASn Xaa .

.

W O 92/10;76 2 ~ 9 ~ ~ 1 2 P~T/US91/09152 (16) INFORMATION FOR SEQ ID NO:15:
(i) SEQUENCE C~AhACTERISTIC:
(A) LENGT'B~
(B) ~YPE: ~mino acid :
(D) TOPOLOGY: line~r ~xi) S~QUENCE DESCRIPTION. SEQ ID NO:15:
Het Pro Ala Hi8 Pro Hi~ Pro Hi~ Pro His Ala (17) INFORMATION FOR SEQ ID NO:16:
(i) 5EQUENCE CHARACTERISTIC:
tA) 1ENGTH: 11 (B) ~YPE: amino acid ~D) TOPOLOGY: linear ~xi) SEQUENCE DESCXIPTION: SEQ ID NO:16:

Met Ala Pro Hi~ Ala HLs Ala Hi~ Ala Hi~ Ala (18) INEORNATION FCR SEQ ID NO:17:
(i) SEQUENCE CHARACTERIST1C:
~A) LENGTH: 11 ~B) TYPE: amino acid ~D) TOPOEOGY: linear (xi) SEQUENCE DESCRIPTION: SEQ ID NO:17:

Met Gly Pro ~i8 Pro H~ Pro Hi~ Pro HLs Ala (19) INFORMATION FoR SEQ ID NO:18:
(i) SEQUEN OE CHARACTERISTIC:
~A) ~ENGTH: 31 (~1 TYPE: amino a~id (D) TOPO~OGY: linear ~ix) FEATURE:
- ~A) NAME/KEY: C-terminally amidated Arqininyl re~idue ~B) LOCATION: Xaa31 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:18:

~WO 92/10576 2 0 9 4 ~12 PCr/US91/09152 -39~
~le Pro Tyr Ala A~p Ala Ile Phe Thr Asn Ser Tyr Arg Lys Val Leu 1 5 . 10 1~
Gly Gln L~u ser Ala Arg LYR Leu Leu Gln A~p Ile Leu Asn Xaa (20) INFOR~TION FOR SEQ ID NO:l9:
. (i) SEQUENCE C~ARACTERISTIC:
(A) LENGTH: 33 ~B) TYPE: amino acid ~D~ ~OPOLOG~: linear (ix) FEATURE:
(A) NAME/~EY: C-ter~inally amida~ed Arqininyl residue lS (B) LOCATION: Xaa33 (xi) SEQUENCE DESCRIPTION: SEQ ID N0:19:
Tyr Ala Tyr Ala Tyr Ala Asp Ala Ile Phe Thr Ser Ser Tyr Arg Ly~
1 5 10 ~5 Val Leu Ala Gln Leu Ser Ala Arg Lys Leu Leu Gln Asp Ile Leu ser Xaa (21) INFORMATION FOR SEQ ID NO:20:
(i) SEQUENCE C~ARACTERISTIC:
~ ENGTH: 39 ~B) TYPE: amino acid ~D) TOPOLOGY: linear (ix) EEATURE:
~A) NAME/KEY: C-torminally amidatqd Argininyl re~ldue ~) LOCATION: Xaa39 ~xl) SEQUENCE DESC~IPTION: SEQ ID NO:20:

Fhe Ala Ly~ Pro Tyr Ala Gly Pro Tyr Ala Tyr Ala Asp Ala Ile Phe Thr Aan Ser Tyr Arg Lys Val Leu Ala Gln Leu Ser Ala Arg Lys Leu Leu Gln A~p Ile Leu Asn Xaa ..:
.: . : :

WO 92/10576 2 0 9 4 512 PCI/IJS91/0915~ `
~0-(22) INFORMATION FOR SEQ ID NO:21 ~i) SEQU~NCE CHARACTERISTIC:
~A) LENGTH: 45 ~ .
(B) TYPE: amino acid (D) TOPCLOGY:~ iinear (ix) FEATURE:
(A) NAMEIXEY: C-t~rminally amidatPd Argininyl rssidue (B) LOCATION: Xaa45 ~xi) SEQUENCE DESCRIPTION: SEQ ID NO:21:
~0 Arg Pro Val Pro Gly Pro Phe Ala Ly~ Pro Tyr Ala Gly Pro Tyr Ala Tyr Ala Asp Ala Ile Phe Thr A~n Ser Tyr Arg Lys Val Leu Gly Gln Leu Ser Ala Arg LYB Leu Leu Gln Asp Ile Leu Asn Xaa (23) INFORMATION FOR SEQ ID NO:22:
(i) SEQUENCE CHARACTERISTIC:
(A) LENGTH: 10 (B) TYPE: amino acld (D) TOPOEOGY: linear (xi) SEQUENCE DESCRIPTION: SEQ ID NO:22:

Phe Ala Lys Pro Tyr Ala Gly Pro Tyr Ala (24) INFORMATION FOR SEQ ID NO:23:
(i) SEQUENCE CHARACTERISTIC:
~A) ~ENGTH: 16 3~ (B) TYPE: amino acid (D) TOPO~OG~: l inear ` (xi) S~QUENCE DESCRIPT~ON: SEQ ID NO:2~:

Arg Pro Val Pro Gly Pro Phe Ala Lyg Pro Tyr Ala Gly Pro Tyr Ala (25) INFORMATION FOR SEQ ID NO:24:
.(i) SEQUENCE CHARACTERISTIC:
4~ ~A) LENGTH: 27 (~) TYPE: amino acid :.: ~:' ,, ' . . ' . . ., ' . :

;'` ~W O 92/10~76 2 ~ 9 ~ 5 1 ~ PCT/US91/091~
~1 -ID) $0P010GY: linear (ix) FEATURE:
(A) NA~E/REY: C-terminally amidated Arqininyl re~iduP
~B) LOCATION: Xaa27 S (xi) SEQUENCE DESCRIPTION: SEQ ID NO:24:

A~p Ala Ile Phe ~hr Asn Ser Tyr Arg Lys Val Leu Gly Gln L~u Ser l 5 lO lS

0 Ala ~rg Ly~ Leu Leu Gln A~p Ile Leu Asn Xaa 2s ~26) INFORMATION FOR SEQ ID NO:25:
~i) SEQUENCE CHARACTERISTIC:
(A) LENGTH: 31 (B) TYPE: amino acid (D) TOPOLOGY: linear (ix) FEATURE:
~A) NAME/KEY: C-terminally amidated Argininyl residue (B) LOCATION: Xaa3l ~xi) SEQUENCE DESCRIPTION: SEQ ID NO:25:

Tyr Ala Tyr Ala A~p Ala Ile Phe Thr Ser Ser Tyr Arg Lyq Val Leu - 25 1 5 lo 15 Gly Gln ~eu Ser Ala Arg Ly~ Leu Leu Gln Asp Ile Leu A~n Xaa (27) INFORMAT~ON FOR SEQ ID NO:26:
(i) SEQUENCE CHARACTERISTIC:
(A) LENGTH: 33 (B) TYPE: amino acid (D) TOPOLOGY: linear (ix) FEATURE:
(A) NAME~KEY: C-terminally amidated Argininyl re~idue (B) LOCATION: Xaa33 ~xi) SEQUENCE DESCRIPTION: SEQ ID NO:26:
Gly Pro Ile Pro Tyr Ala a~p Ala Ile Phe Thr A~n ser Tyr Arg Ly~

Val Leu Gly Gln Leu Ser Ala Arg Lys Leu Leu Gln A~p Ile Leu A~n .
, ' ~ ' ' ~.

W O 92/10576 2 ~ 9 4 ~ ~ 2 PCT/US91/0~152 ~2-Xaa (28) INFOR~TION FOR SEQ ID NO:2'1~
5~i) SEQUENCE CHARACTERI~TIC:
(A) L~NGTH: 3~
~B) TYPE: amino acid D ~ TOPOLOGY: 1 i ne~r ~ix) FEATURE:
(A) NANE/KEY: C-terminally amidated Argininyl residue (~) LOCATION: Xaa35 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:27:

Tyr Ala Gly Pro Ile Pro Tyr Ala Asp Ala Ile Phe Thr Aqn ser Tyr l 5 l0 15 ~g Lys Val Lau Gly Gln Leu Ser Ala Arg Lys Leu Leu Gln A~p Ile 20 25 ~0 Leu A3n Xaa (29) I~FORMATION FOR SEQ ID NO:28:
ti) SEQUENCE CHARACTERISTIC:
(A) L~NGTH: 37 (B) TYPE: amino acid ~D) TOPOLOGY: linear ~ ix ) FEATUR~:
~A) NAME/XEY: C-terminally amidated Asgininyl re~idue (B) LOCATION 2 Xaa37 (xl) SEQ B NCE DESCRIPTION: SEQ ID NO:28:

Lyn Pro Tyr Ala Gly Pro Ile Pro Tyr Ala Asp Ala Ile Phe ~hr Asn l 5 l0 15 Ser Tyr Arg Lys Val Leu Gly Gln L~u Ser Ala Arg Lys Leu Leu Gln 20 2~ 30 , 40 A~p Ile Leu A~n Xaa (30) INFORNaTION FOR SEQ ID NO:29:
45 ~i) S QUENCE CHaRACTERISTIC:
~A) 1ENGTH: 33 2~9~5~2 wo ~2/10~7b PCI`/US91/091i2 ~3 -~B ~ TYPE: amino acid (D~ TOPOLOGY: linear (ix) FEATURE:
tA) NAME~REY: C-terminally amidated Arqininyl residue (B) LOCATION: Xaa33 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:29:

Gly Pro Tyr Ala Tyr Ala A~p Ala Ile Phe Thr Asn Ser Tyr Arg Lys Val Leu Gly Gln Leu Ser Ala Arg Ly~ Leu Leu Gln Asp Ile Leu A~n 20 2~ 30 Xaa (31) INFORMATION FOR SEQ ID No:30:
~i) 5EQUENCE CHARACTERISTIC:
~A) LENGTH: 35 (B) TYPE: amino acid (D) TOPOLOGY: linear ~ix~ FEATURE:
(A) NAME~REY: C-terminally amidated Argininyl re~idue (B) LOCATION: Xaa35 ~xi) SEQUENCE DESCRIPTION: SEQ ID NO:30:

~yr Ala Gly Pro Tyr Ala Tyr Ala Asp Ala Ile Phe Thr Asn Ser Tyr Arg Ly~ Val Leu Gly Gln Leu Ser Ala Arg Ly~ Leu LQU Gln A~p Ile Leu Asn Xaa (3~) INFORMATION FOR SEQ ID NO:31:
~i) SEQUENCE CHaRACTERISTIC:
(A) LENGTH: 37 (B) TYPE: amino acid (D) TOPOLOGY: linear (i~) FEATURE:
(A) NANE/XEY: C-terminally amidated Argininyl residue ~B) LOCATION: Xaa37 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:31:
Ly~ Pro Tyr Ala Gly Pro Tyr Ala Tyr Ala Asp Ala Ile Phe Thr A~n ~: .
:' ~ ' ' ' -WO 92/1~76 2 ~ 9 ~ ~ 1 2 PCI/US~/09152' `"' 1 5 ~0 15 Ser Tyr Arg Lys Val Leu Gly Gln~Leu Ser Ala Arg Ly3 Leu ~eu Gln 20 ....... ,; 25 30 S -,., Af3p Ile Leu Ann Xaa ~33) INFORM~ION FO~ SEQ ID NO:32:
(i) SEQUENCE CHaRACTERISTIC:
~A) LENGT~: 39 ~B) TYPE: amino acid ~D) TOPOLOGY: linear (ix) EEATU~E:
~A) NAMæ/REY: C-terminally amidated Argininyl residue (B) LCCATION: Xaa39 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:32:

Phe Ala Lya Pro Tyr Ala Gly Pro Tyr Ala Tyr Ala Asp Ala Ile Phe Thr Aan Ser Tyr arg Ly~ Val Leu Gly Gln Leu Ser Ala Arg Ly~ Leu Leu Gln A~p Ile Leu A~n Xaa 30 ~34) INFORM~TION FOR SEQ ID NO:33:
~$) 9EQVENCE CHARACTERISTIC:
~A) LENGTH: 41 ~B) TYPE: amino acid (D) TOPOLOGY: linear ~ iX ) EEATURE:
~A) ~MEIKEY: C-terminally amidated Argininyl r~$due ~B) LOCATION: Xaa41 ~xi) SEQUENCE DESCRIPTION: SEQ ID No:33:

Gly Pro Phe Ala Ly~ Pro Tyr Ala Gly Pro Tyr Ala Tyr Ala A~p Ala Ile Ph~ Thr A~n Ser Tyr Arg LYR Val Leu Gly Gln Leu Ser Ala Arg Ly~ Leu Leu Gln Aap Ile Leu A~n Xaa - ~. ~ . .. . . - - .,........ :

2 ~ 2 : ' . ' WO 92/10576 P~/US91/0915 ~5-(35) INFORM~TION FOR SEQ ID NO:34:
S (i) SEQUENCE CHARACTERISTIC:
tA) LENGTH: 43 (~) TYPE: amino acid (D) TOPOLOGY: linear (ix) FEA~URE:
~A) N~ME/~EY: C-terminally amidated Argininyl re~idue ~B) LOCATION: Xaa43 ~xi) SEQUENCE DESCRIPTION: SEQ ID NO:34:

Val Pro Gly Pro Phe Ala Ly3 Pro Tyr Ala Gly Pro Tyr Ala Tyr Ala l 5 l0 15 Asp Ala Ile Phe Thr Asn Ser Tyr Arg Lys Val Leu Gly Gln Leu Ser Ala Arg Lya Leu Leu Gln Asp Ile Leu Asn Xaa (36) INFORMATlON FOR S2Q ID NO:35:
(i) SEQUENCE CHARACTERISTIC:
(A~ LENGTH: 45 ~B) TYPE: amino acid (D) TOPOLOGY: linear (ix) FEATURE:
~A) NAME/XEY: C-terminally amidated Argininyl reRldue ~B) LOCATION: Xaa45 ~xi) SEQUENCE DESCRIPTION: SEQ ID NO:35:

Arg Pro Val Pro Gly Pro Phe Ala Lys Pro Tyr Ala Gly Pro Tyr Ala 1 5 l0 15 Tyr Ala Asp Ala Ile Phe Thr Asn Ser Tyr Arg Lyq Val Leu Gly G1D
: 20 25 30 :
Leu Ser Ala Asg Ly~ 1QU Leu Gln A~p Ile Leu A~n Xaa 35 40 . 45 ~37) INFORMATTON FOR SEQ ID NO:36:
~i) SEQUENCE CHARACTERISTIC:
~A) 1ENGTH: 3l -:

.. : . , : ., ::,:, :, .:. : :
., WO 92/10~76 2 0 9 4 ~ i 2 PCI /VS91/091S' ~6-~B) TYPE: amino acid (D) TOPOLOGY: linear ~ix) F~ATURE:
(A) NAMEtKEY: C-terminally amidated Argininyl re~idue (B) LOCATION: Xaa31 `.
~xi) SEQUENCE DESCRIPTION: sEQ Ii NO:36:
;,:
Val ~la Tyr Ala A~p Ala Ile Phe ~hr~Asn Ser Tyr Arg Lys Val Leu Gly Gln Leu Ser Ala Arq Lys Leu Leu Gln Asp Ile Leu Asn Xaa ~38) INFORMATION FOR SEQ ID NO:37:
(i) S~QUENCE CHARACTERISTIC:
- (A) LENGTH: 3l (B) TYPE: amino acid (D) TOPOLOGY: linear (ix) FEATURE:
tA) NA~EIKEY: C-terminally amidated Argininyl re~idue (B) LOCATION: Xaa31 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:37:

Tyr Thr Tyr Ala A~p Ala Ile Phs Thr Asn Ser Tyr Arg Ly~ Val Leu Ala Gln Leu Ser Ala Arg Ly~ Leu Leu Gln Asp Ile Leu A~n Xaa (39) INFORMATION FOR SEQ ID NO;38;
(i) SEQUENCE CHARACTERISTIC:
~A) LENGTH; 31 (B~ TYPE: amino acid ~D) TOPOLOGY: linear (i~) FEATU~E:
(A) NAME/KEY: C-terminally amidated Argininyl r~idue -~
~B) LOCATION: Xaa31 ~xi) SEQUENCE DESCRIPTION: SEQ ID NO:38:
Tyr Thr Tyr Ile A~p Ala Ile Phe Thr Asn Ser Tyr Arg Lys Val Leu Ala Gln Leu Ser Ala Arg Lys Leu Leu ~ln Asp Ile Leu A~n Xaa , :, ~

~U9~51~
..WO 92/10~76 PCr/US91/091~2 ~7-(40) INFORMATION FOR SEQ ID NO:39:
(i) SEQUENCE CHARACTERISTIC:
(A) LENGTH: 31 S (B) TYPE: amino acid ~D) TOPOLOGY: li~ar tix) FEATURE:
(A) NA~E/~EY: C-terminally amidatPd Argininyl xesidue (B) LOCATION: X~a31 (xi) SEQuENcE D~SCRIPTION: SEQ ID NO:39:

Tyr Thr Tyr Thr Asp Ala lle Phe Thr Asn Ser Tyr Arg Eys Val Leu 1 ; 10 15 Ala Gln Leu Ser Ala Arg Lys Leu Leu Gln Asp Ile Leu A~n Xaa (41) INFORMATION FOR SEQ ID NO:40:
(i) SEQUENCE CHARACTERISTIC:
(A) LENGTH: 31 ~B) TYPE: amino acid ~D) TOPOLOGY: linear (ix) FEATURE:
(A) NAME/KEY: C-terminally amidated Argininyl residue ~B) LOCATION: Xaa31 ~xi~ SEQUENCE DESCRIPTION: SEQ ID NO:40:

Tyr Ser Tyr Thr Asp Ala Ile Phe Thr Asn Ser Tyr Arg Lys Val Leu Ala Gln Eeu Ser Ala Arg Lys Leu Leu Gln Asp Ile Leu Asn Xaa (42) INFORM~TION FOR SEQ ID NO:41:
(i) SEQUENCE CHARACTE~ISTIC:
~A) LENGTH: 33 ~B) TYPE: amino acid 44 ~D~ TOPOLOGY: linear ~ix) FEATURE:
~A) NAME/XEY: C-terminally amidated Argininyl residue SB) LOCATION: Xaa33 ~xi) SEQUENCE DESCRIPTION: SEQ ID NO:41:
Tyr Thr Tyr Thr Tyr Thr A~p Ala Ile Phe Thr Aqn Ser Tyr Arg Lys : . ., :

. ~ , :

W 0 92/10576 2 0 9 A ~12 PCT/US91/09l~
~8-.

Val Leu Ala Gln Leu Ser Ala Arg Lys Le~ Le~ Gln Asp Ile Leu A~n S :.
Xaa ~

(43) INFORMATION FOR SEQ ID NO:42:
(i) SEQUENCE CHARACTERISTIC:
(A) LENGTB: 31 (B) TYP2: amino acid (D) TOPOLOGY: linear (ix) FEATURE:
(A) NAME/KEY: C-terminally amidated Argininyl residue (B~ LOCATION: Xaa3l (xi) SEQUENCE DESCRIPTION: SEQ ID No:42:

Tyr Ala ~yr Ala ABP Ala Ile Phe Thr Ser Ser Tyr Arg Lys Val Leu 1 5 lO 15 Gly Gln Leu Ser Ala Arg Lys Leu Leu Gln Asp Ile Leu A~n Xaa (44) INFORUATION FOR SEQ ID NO:43:
(i) SEQUENCE C~ARACTERISTIC:
(A~ LENGTH: 33 ~B) TYPE: amino ac~ d (D) TOPOLOGY: linear (ix) FEATURE:
~A~ NAME/KEY: C-terminally amidated Argininyl re~idue (8) LOCATION: Xaa33 (xl) SEQUENCE DESCRIPTION: SEQ ID NO:43:
3~ .
Tyr Ala Tyr Ala Tyr Ala A~p Ala Ile Phe Thr Ser Ser Tyr Arg LYB
l 5 l0 l~

; Val Leu Ala Gln Leu Ser Ala Arg Lys Leu Leu Gln A~p Ile Leu Ser Xaa

Claims (16)

1. A non-naturally-occurring fusion protein comprising an extension peptide portion covalently linked at its C-terminus to the N-terminus of a core protein portion, said extension peptide portion being of the formula:
A-X-Y(X'-Y)n wherein A is optional and when present is methionine;
n is 0-20;
X is selected from the group consisting of all naturally occurring amino acid residues;
X' is selected from the group consisting of all naturally occurring amino acid residues except proline and hydroxyproline;
Y is selected from the group consisting of proline, hydoxyproline, alanine, serine and threonine except when n is zero and A is absent then Y is selected from the group consisting of alanine, serine and threonine.

2. A non-naturally-occurring fusion protein according to claim 1 wherein A is present and X is selected from the group consisting of Pro, Gly, Ala and Ser.

3. A non-naturally-occurring fusion protein according to Claim 1 wherein n is 0-10.

4. A non-naturally-occurring fusion protein according to claim 1 wherein said biologically active polypeptide is selected from the group consisting of: bGRF analogs, EGF; IGF-2, glucagon; corticotropin releasing factor; dynorfin, somatostatin-14; endothelin; transforming growth factor or; Vasoactive Intestinal Peptide; human .beta.-casomorphin; Gastric Inhibitory Peptide; Gastric Releasing Peptide; human Peptide HI; human Peptide YY; glucagon-like peptide-1 fragment 7-37; glucagon-like peptide-2; substance P; Neuropeptide Y; human Pancreatic Polypeptide; insulin-like growth factor-1; human growth hormone; bovine growth hormone; porcine growth hormone; prolactin; human growth hormone releasing factor; bovine growth hormone releasing factor; porcine growth hormone releasing factor; ovine growth hormone releasing factor; interleukin -1.beta.; and interleukin-2.

5. A non-naturally-occurring fusion protein according to claim 1 wherein said extension peptide portion is selected from the group consisting of .

6. A non-naturally-occurring fusion protein according to claim 2 wherein n is 3-5.

7. A non-naturally-occurring fusion protein according to claim 6 wherein all X' residues are histidines.

8 A non-naturally-occurring fusion protein according to claim 7 wherein X is a histidine.

9. A non-naturally-occurring fusion protein according to claim 7 wherein three Y residues are proline.

10. Use of a non-naturally-occurring fusion protein according to claim 1 to prepare a medicament.

11. A use according to claim 10, wherein said medicament comprises additional fusion proteins having identical biologically active portions and different extension portions.

12. A use according so claim 10 wherein said biologically active portion of said non-naturally-occurring fusion protein is a bGRF analog.

13. A method of purifying desired proteins from a mixture containing a non-naturally-occurring fusion protein according to claim 1 and impurities comprising the steps of:
selectively contacting said fusion protein with material which immobilized said fusion protein;
removing said impurities;
separating said fusion proteins from said material;
combining said fusion protein with DPP IV; and isolating said desired protein.

14. A method according to claim 13 wherein said material is fixed in a column.

15. A method according to claim 13 wherein said material is an antibody which binds to said extension portion.

16. A method according to claim 13 wherein said material is immobilized metal ions and said extension portion comprises at least 3 consecutive X' residues that are histidines.