The present vβniϊøn relates generally to gastrin-binding protein, genetic sequences encoding same and antagonists thereto.

The hormone gastiin was first described as a stimulant of gastric acid secretion by Edkins (1905). A 17 amino acid form of gastrin was subsequently purified from pøiαiie antral mucosa and sequenced (Gregory and Tracy, 1964). Cloning of a porcine gastrin cDNA revealed that gastrin _j γ was derived from a 104 amino acid pro omKne by proteolytic processing and C-terminal amidation ( Yoo et al., 1982). More recently there has been renewed interest in gastrin with the recognition that it acts as an autocrine growth factor for colon carcinoma cells in vitro (Hoosein et aL. 1988). The related hormone cholecystokinin (CCK) simulates enzyme secretion from the pancreas and contraction of the gall-bladder (Mutt, 1980).

Receptors for hormones of the gastrin/CCK family can be divided into three classes which differ in hormone and antagonist affinity (Thumwood, Ji and Baldwin 1991). The CCK-A receptor on pancreatic acinar cells has highest affinity for sulphated CCKg, low affinity fαrgastrin-u, and is inhibited by the antagonist 1364,718 (Chang and Lotti, 1986) more strongly than by the antagonist L365,260 (Lotti and Chang, 1989). In contrast, ifae gastrin/CCK-B receptor on gastric parietal cells and in brain has a similar affinity for sulphated CCKg and gastrin^ (Soil et al.. 1984), and is inhibited by 1365,260 more strongly than by 1364,718. The third class, gastrin/CCK-C receptor on gastric (Weinstock and Baldwin, 1988) and colonic (Hoosein et al.. 1988) carcinoma cell lines, has low affinity for both gastrin _j γ and CCKg and is not inhibited by either 1364,718 or 1365,260 (Thumwood, Ji and Baldwin, 1991). AU classes are inhibited by the general gastrin/CCK antagonists proglumide and benzotript (Rovati et al.. 1967; Hahne et al.. 1981).

Despite the abundance of information on binding of gastrin and CCK to their receptors, until the present invention, little was known about the structure of the

gastrin CCK receptors themselves. In particular, nothing was known of the sequence of any gastrin/CCK receptor. A need, therefore, exists for a gastrin/CCK receptor to be purified, sequenced and cloned.

In accordance with the present invention, an approximately 78 + 5 kDa gastrin- binding protein as determined by SDS-PAGE has been purified and genetic sequences encoding same cloned. The present invention, therefore, provides an abundant source of gastrin-binding protein for which antagonists and antibodies can be obtained. Such antagonists and antibodies will be useful in the treatment of gastrin- binding protein-associated colon cancers, the regulation of acid production and in ulcer therapy. In further accordance with the present invention, the gastrin-binding protein is identified on the basis of agonist and antagonist binding studies (Figure 6) as a gastrin/CCK-C receptor. Although not wishing to limit the scope of the present invention to any one hypothesis or putative mode of action, the data set forth herein suggest that gastrin-binding protein could be a common subunit of all gastrin binding receptors.

Accordingly, one aspect of the present invention is directed to an isolated mammalian gastrin-binding protein.

Hereinafter "gastrin-binding protein" is referred to as "GBP". By "GBP" as used in the specification and claims is meant a molecule having an amino acid sequence substantially corresponding to that shown in Figure 2 and includes any or all parts thereof (including coding sequence and/or extraneous sequences thereto) and any single or multiple amino acid substitutions, deletions and/or additions to the amino acid sequence. The term GBP also extends to the molecule in its unglycosylated form and further extends to any single or multiple substitutions, deletions and/or additions to its naturally occurring or unnaturally occurring glycosylation pattern. Furthermore, even though porcine GBP is exemplified herein, the present invention clearly extends to all mammalian GBPs and in particular human and porcine GBP,

the former which shares approximately 90% identity with porcine GBP. Accordingly, reference herein to "GBP" includes the porcine GBP shown in Figure 2, to mammalian (e.g. human) equivalents thereof and to any mutants, derivatives or portions thereof whether such mutants, derivatives or portions are in the amino acid sequence, carbohydrate moiety or both. Furthermore, "substantially corresponding" also means a degree of homology between a nucleotide or amino acid sequence and that shown in Figure 2 of approximately at least 50%, preferably at least 60%, more preferable at least 70% and even more preferably at least 85-95%.

The present invention is particularly directed to isolated GBP which includes recombinant GBP as well as molecules prepared by the sequential addition of amino acids or groups of amino acids in a defined order to produce a GBP molecule or its equivalent. The preferred form of GBP according to the present invention is its recombinant form although other non-naturally occurring (e.g. synthetic forms) or naturally occurring variants are encompassed by the instant invention. Such forms of GBP may have the naturally occuring glycosylation pattern or an altered glycosylation pattern or may be unglycosylated.

Advantageously, the GBP of the present application is biologically pure which means that a preparation comprises at least 40%, preferably at least 60%, more preferably at least 80% and most preferably at least 90% GBP relative to other molecules as defined by weight, gastrin-binding activity and/or antibody-binding activity. The GBP is thus referred to as being "isolated". Reference to "biologically pure" or "in substantially pure form" in relation to other molecules, such as antagonists, has the same meaning as above in relation to synthetic GBP.

The molecular weight of the GBP of the present invention is approximately 78 ± 5 kDa as determined by SDS-PAGE. One skilled in the art will immediately recognise that this figure may vary depending on the method employed to calculate or determine the molecular weight.

The isolated GBP of the present invention is useful in the development of antagonists to the binding of gastrin to cell membrane-associated gastrin/CCK receptors and is useful, in soluble form, in inhibiting, reducing and/or otherwise competing with gastrin-receptor binding. Such interruption to the gastrin-receptor interaction is useful in controlling gastric acid production and in inhibiting growth of colon carcinomas and in ulcer therapy.

Accordingly, another aspect of the present invention contemplates a method for interrupting gastrin-GBP interaction in a mammal which method comprises administering to said mammal an effective amount of soluble GBP for a time and under conditions sufficient for said GBP to compete with cell membrane-associated GBP for binding with gastrin. Such competition for gastrin-receptor binding can be used to reduce the consequences of gastrin-receptor binding such as, but not limited to, acid secretion and in the control of colon carcinomas and/or other cancers which at least in part proliferate due to the presence of gastrin-receptor and/or receptor- receptor interaction and also in ulcer therapy. The GBP has the same meaning as above and includes that part of the molecule required to bind, or interfere with, the binding of, gastrin to GBP. The GBP may be based on porcine, human or other mammalian GBPs. The intended mammal to be treated is preferably a human. The effective amount will vary depending on the condition and mammal to be treated and the amount of gastrin-receptor interference required and may range from lμg/kg to about 100 mg/kg body weight although effective amounts outside this range may be used without departing from the scope of the present invention. For example, an effective amount of from about 5μg/kg to about 20mg/kg of body weight may also be used. Furthermore, the administration of the GBP may be alone or in combination or conjunction with other molecules such as antagonists, antacid compounds or anti-cancer agents. Additionally, the method of the present invention is not limited to the use of a GBP from a mammal to treat the same species of mammal (i.e. homologous use). GBP of one mammal may be equally effective, more effective or otherwise beneficial for use with a differenct species of mammal

(i.e. heterologous use). For example, porcine GBP may be useful in treating humans.

Administration may be by any convenient route such as by intravenous, intra- peritoneal, intramuscular or substaneous injection or by the infusion, oral, suppository or intranasal spray routes. Gene therapy may also be used as well as the expression in a gut organism, such as E.coli. of a clone encoding GBP and where the GBP is capable of secretion from said gut organism. The gut organism may grow in the gut or may be administered in such a sufficient quantity that it need not proliferate. In the latter case, the gut organism may be mutated such that it remains viable while being unable to grow.

Another aspect of the present invention provides an assay for antagonists to gastrin- GBP interaction using the GBP described herein. In one such assay, prokaryotic or eukaryotic cells engineered to express GBP are exposed to labelled gastrin or CCK such as ^* ^^I- Leu^ ^** *] gastrin _j or di-^^I-[norleucine^] gastrin _j γ with and without potential antagonists. The extent to which a potential antagonist can interfere with the binding of labelled gastrin provides a convenient assay for suitable antagonist. One skilled in the art will immediately recognize many variations to this assay, such as labelling antagonists rather than gastrin. .All such variations are encompassed by the present invention.

In an alternative assay, various derivatives and analogues of GBP can be used in assays measuring cell proliferation by pH] thymidine incorporation. An example of such an assay is described by Thumwood, Ji and Baldwin, (1991).

Yet another aspect of the present invention is directed to an antagonist of GBP. Conveniently, although not essentially, the antagonist is obtained through an assay procedure as described above. Such an antagonist will be useful in controlling acid production, growth of colon carcinomas and/or in ulcer therapy. The antagonist may

be a peptide such as a peptide fragment of the gastrin molecule or a chemical molecule. Although not traditionally considered as an antagonist, this term also extends to an antibody wherein the antibody is a polyclonal, or more preferably monoclonal antibody, prepared against specific parts of GBP. Preferably the antagonist will provide a high degree of specificity and may, for example, allow for inhibition of colon carcinoma growth or excessive gastric acid production while not adversely affecting natural colon growth or gastric acid production. Generally, the antagonists of the present form will be in substantially pure form as defined above.

Another aspect of the present invention contemplates pharmaceutical compositions comprising mammalian GBP or an antagonist thereto, alone or in combination with other active molecules such as antacid compounds or anti-cancer agents or agents useful in ulcer therapy. Examples of suitable active molecules include chemical compounds and cytokines. The pharmaceutical compositions of the present invention will be useful in causing interruption of gastrin-receptor binding and/or receptor- receptor interaction as hereinbefore described. The preparation of pharmaceutical compositions is well known in the art and reference can conveniently be made to Remington's Pharmaceutical Sciences. 17th edition, Mack Publishing Company, Easton, Pennsylvania, USA. The GBP useful in a pharmaceutical composition has the same meaning as above and is preferably of human or porcine origin and most preferably has the amino acid sequence or is encoded for by a nucleotide sequence substantially setforth in Figure 2.

Yet another aspect of the present invention provides genetic sequences encoding GBP. The genetic sequences may be deoxyribonucleotides or ribonucleotides and preferably comprise the coding sequence, or its complementary sequence, as shown in part or in full in Figure 2. The complementary sequence is useful, for example, as a probe to isolate or screen for similar genetic sequences from various sources. The present invention extends to any single or multiple nucleotide substitutions, deletions and/or additions to the sequence shown in Figure 2 as well as to any

sequence showing preferably at least 60%, preferably at least 70% and more preferably at least 80% homology to the nucleotide sequence described herein. The present invention, therefore, extends to any nucleotide sequence encoding GBP including its derivatives, homologues and functional equivalents.

The genetic sequences encoding GBP described herein may be single or double stranded, alone or in combination with a vector and preferably operably linked to a promoter such as in an expression vector. The vector may be capable of replication and/or expression in one or both of eukaryotic and/or prokaryotic cells.

The present invention also encompasses the genetic sequences for GBP operably linked to a promoter, with or without an associated vector molecule, in a eukayotic or prokaiyotic cell and to such eukaryotic or prokaryotic cells containing said genetic sequences.

The present invention is further described by reference to the following non-limiting Figures and Examples.

Figure 1 is a diagrammatic representation showing the alignment of clones encoding GBP.

Clones PCR 1/2, 8/5 and 8/4 were isolated by conventional PCR with the indicated oligonucleotides as primers (see Table 1 for sequences) and oligo-dT primed porcine gastric ucosal cDNA as template. Clone RACE was isolated by anchored PCR with oligonucleotides 5A (a nondegenerate version of 5, based on the sequence of clone PCR 8/4) and 13 as nested primers and oligonucleotide 15-primed porcine gastric mucosal cDNA as template. Oligonucleotide 7 was used to screen for clone RACE. Clone LIV 1 was isolated by screening a porcine cDNA library with an 800 bp fragment from clone PCR 8/4 encompassing the 5' end to the Qa. I site. A partial restriction map of the GBP cDNA is indicated, with the following restriction enzymes: B, Bgl II; C, CM; E, Eco RI; P, PstI; S, Smal.

Figure 2 is a schematic representation showing the nucleotide sequence and predicted amino acid sequence of the GBP cDNA.

Nucleotides are numbered on the right and the amino acids on the left, where +1 is the N-terminal residue of the mature protein. The putative polyadenylation sequence and the nucleotide binding site are underlined. The asparagines of the 3 potential N- glycosylation sites are asterisked. Peptides isolated from the 78 kDa GBP are boxed.

Figure 3 is a photographic representation showing the size and tissue distribution of the mRNA encoding the GBP.

Poly A+ mRNA was prepared from porcine liver (L), and from the mucosa of porcine gastric antrum (A), corpus (C), and forestomach (F), and separated by electrophoresis on 1% (w/v) agarose-formaldehyde gel. The RNA was transferred to a nitrocellulose filter and probed with a nick-translated 800 bp fragment of clone PCR 8/4 or mouse 7-actin cDNA. The positions of the origin (O) and of the 28s and 18s ribosomal RNA are indicated.

Figure 4 is a schematic representation showing an alignment of the amino acid sequences of the rat bifunctional enzyme and the porcine GBP. Optimal alignments between the rat bifunctional enzyme (BE) and the porcine GBP were generated with the program ALIGN, using the mutation data matrix, a matrix bias of +6 and a break penalty of 6 (Dayhoff, 1979). Identities are indicated by asterisks and gaps introduced to maximize similarity are indicated by hyphens. The sequences of the mature proteins were 33.3% similar.

Figure 5 is a graphical representation showing a hydropathy profile of the GBP. Positive values indicate hydrophobic regions according to Kyte and Doolittle (1982) determined with a span of 21 amino acids. Amino acid numbers are indicated along the abscissa, beginning at the N-terminus of the mature protein.

Figure 6 is a graphical representation showing binding -of gastrin to the GBP

expressed in CHO cells.

Binding of di-^^I-Enorleucine^J-gastrin _j -y to 10^ CHO cells transfected with pMEXneo containing the full-length GBP cDNA ( ■ ) or with the vector alone (▼) was measured in triplicate as described in the Materials and Methods section, in the presence of increasing concentrations of unlabelled [norleucine ¹⁵ ]-gastrin _j 7 (A) or the general gastrin/CCK antagonist proglumide (B). Bars represent one standard deviation.

EXAMPLE

1. Materials and Methods

Isolation of GBP

GBP was isolated from pig gastric mucosal membranes by wheat germ agglutinin or concanavalin A sepharose chromatography followed by DEAE sepharose chromatography. The GBP was then reduced and carboxymethylated and further purified by preparative gel electrophoresis (see Baldwin et al, 1986 and Baldwin et al, 1987). A single N-terminal sequence was determined on the intact GBP. Following digestion of the GBP with trypsin, several peptides were isolated by reverse phase HPLC and sequenced (see generally Baldwin et al, 1987 Supra and Table 1).

Isolation of cDNA Clones by Polvmerase Chain Reaction (PCR)

Poly (A) ⁺ mRNA was selected from total RNA prepared from the corporeal mucosa of porcine stomach (Chirgwin et al.. 1979) by chromatography on oligo-dT cellulose. First strand oligo-dT-primed cDNA was synthesized with an Amersham (Bucks, UK) kit according to the manufacturer's instructions. Clone PCR 1/2 was isolated by a PCR with multiply degenerate 31mer primers l(768-fold) and 2(192-fold) (Table 1) based on either end of the N-terminal amino acid sequence of a GBP isolated from

SUBSTITUTE SHEET

porcine gastric mucosal membranes. The amplification was carried out in a lOO l reaction volume containing 50mM KC1, lOmM Tris (pH 8.3), 1.5mM MgC^, 0.1% (w/v) gelatin, 0.1% (v/v) Triton X-100, 200μM of each dNTP, 50pmol of each primer and cDNA (lOOng). After denaturation at 95°C for 3 min 1.5U Taq polymerase (Stratagene, La Jolla, CA) was added, and the mixture was overlaid with lOOμl paraffin oil. The samples were then manually cycled 30 times between water baths at 95°C (1 min), 37°C (2 min) and 50°C (3 min). Further rounds of PCR were then performed between a unique 31mer primer 8, whose sequence was based on the sequence of clone PCR 1/2, and multiply degenerate primers 4(28mer, 128- fold) or 5(25mer, 384-fold) based on the sequences of tryptic peptides AT3 and AT6, respectively (Table 1). The PCR products were digested with the appropriate restriction enzymes and subcloned into pGEM3Z (Promega, Madison, WT .

Isolation of Clones bv cDNA Library Screening

A porcine liver library λgtlO (Clontech, Palo .Alto, CA) was probed with an 8000bp fragment from clone PCR 8/4 (encompassing the 5' end to the CJa I site) which had been labelled with [α-^^PldATP by nick translation. Filters were hybridized in 6x SSC, 5x Denhardt's solution, 0.1% (w/v) SDS, 50μg/ml salmon sperm DNA and 30% (v/v) formamide at 42°C and washed in 6x SSC, 0.1% (w/v) SDS at the same temperature. The single positive plaque (LIV 1) obtained was subjected to two further rounds of screening, and phage DNA was prepared by the DEAE method (White and Rosenzweig, 1989). The insert was excised with Eco RI, and the 3 resultant fragments (265, 744 and 1093 bp) subcloned into pUC.

Isolation of Clones bv RACE-PCR

The RACE-PCR procedure (Frohman et al.. 1988) was used to amplify cDNA corresponding to the 5' end of the mRNA. First strand cDNA specifically primed with oligonucleotide 15 was synthesized from poly(A) ⁺ liver mRNA using a

SUBSTITUTE SHEET

Promega cDNA synthesis kit, according to the manufacturer's instructions. The cDNA was tailed with dATP and terminal deoxynucleotidyl transferase (IBI, New Haven, CT) and used for successive rounds of RACE-PCR with oligos 5A or 13 as nested primers and a poly T primer. Amplification was carried out on 1/10 of the diluted tailed cDNA reaction as described above except that 1.5U Taq polymerase (Promega, Madison, WI) was used and the samples placed in an automated heating/cooling block (Hybaid, Middlesex, UK) programmed for 5 cycles at 94°C (40s), 37°C (30s), 71 °C (lmin), followed by 30 cycles at 94°C (40s), 53°C (30s), 71 °C (lmin). An aliquot of the second round PCR reaction resolved on a 1.5% (w/v) agarose gel and visualized by ethidium bromide staining. Southern blot analysis (Southern, 1975) using oligonucleotide 7 labelled with [γ-^P] ATP and polynucleotide kinase (New England Biolabs, Beverly, MA) confirmed that a band of 239 bp was a 5' extension of the transcript. The remainder of the second round PCR reaction was extracted with chloroform to remove the paraffin oil and DNA was precipitated by the addition of 1/10 volume 3M sodium acetate (pH 5.0) and 2.5 volumes ethanol and digested with Bam HI and ECQ RT. The 239 bp fragment was sliced from an ethidium-stained 1.5% (w/v) agarose gel, purified using a Geneclean kit (BIO 101, La Jolla, CA) and subcloned into pGEM3Z.

Nucleotide Sequencing

cDNA fragments were subcloned from pGEM3Z into both M13mpl7 and mpl9 vectors (New England Biolabs, Beverly, MA) according to standard methods (Maniatis et al.. 1982). Complete nucleotide sequences of both DNA strands were obtained by the dideoxynucleotide chain termination method (Sanger et al.. 1977) with single-strand M13 DNA as template, using Sequenase DNA polymerase (United States Biochemical Co., Cleveland, OH), as described by the manufacturer.

SUBSTITUTE SHEET

Northern Analysis

Total poly(A) ⁺ mRNA was isolated from porcine liver, and from the mucosa of porcine gastric antrum, corpus and forestomach (Chirgwin et al.. 1979). Samples of 2μg were run on 1% (w/v) agarose formaldehyde gels, transferred to nitrocellulose (Genescreen, DuPont, Boston, MA.) and probed with a nick-translated 800 bp fragment of clone PCR 8/4 or mouse γ-actine cDNA as recommended in the manufacturer's instructions.

Constructions of Full-length GBP cDNA

A full-length cone of the cDNA encoding the GBP was constructed as follows. The 5' fragment (840bp) was generated by PCR (94°, 1 min; 60°,1 min; 72°,2 min; 30 cycles) between primers 30 and 31 (Fig.1) with liver cDNA specifically primed with oligonucleotide 15 as template. The 3' fragment (2200bp) was generated by PCR (94°, 1 min; 60°, 1 min; 72°,3 min; 30 cycles) between λgtll forward and reverse primers (Promega, Madison, WT) with LIV I phage DNA as template. Both fragments were blunt cloned into -Sjnal-cut pUC19. The 5' fragment fEco RI-Bgl IT) and the 3" fragment (Bgl π-Kpn I) were then joined via the internal Bgl II site (Fig. 1) and ligated into Ego RI/Kpn I-cut pGEM-3Z. The full length clone was released by digestion at the Kpn I sites in the two partial copies of the pUC19 polylinker and ligated into the Kpn I site of the mammalian expression vector pMEXneo (a derivative of the vector pDM16 (Martin-Zanca et al.. 1986; Oskam et al., 1988)).

CHO Cell Transfection

CHO cells were transfected by electroporation (Andreason and Evans, 1988) with either pMEXneo containing the full-length GBP cDNA or with pMEXneo alone. Transected CHO cells were selected in Alpha minimal Eagle's medium containing

SUBSTITUTE SHECT

10% (v/v) FCS, 2mM glutamine and lmg/ml G418 on the day after transfection. Clones were then expanded in media containing 500μg/ml G418, and maintained in complete media containing lOOμg/ml G418.

Gastrin Binding Assay

Binding experiments were performed essentially as described (Weinstock and Baldwin, 1988). Briefly, transfected CHO cells were grown to confluence, removed from culture dishes using 3mM EDTA in phosphate buffered saline (PBS) and incubated at 37°C for 5 minutes. Cells were pelleted (150 g, 5 min), washed in PBS once, and adjusted to a concentration of 10^/ml in PBS containing 0.2% (w/v) ovalbumin. 10" cells (lOO l volume) were added to triplicate tubes containing lO^cpm di-^I-p-Jle ^** ^] gastrin _j γ (Seet et al.. 1988) and the appropriate amount of cold [Nle^] gastrin _j 7 or proglumide (total volume 200μl). The tubes were rotated end-over-end at room temperature for 1 hour. Cells were then washed by centrifugation (10,000g, 1 min) through 0.3ml of an oil mixture containing dibutyl phthalate/dinonyl phthalate (3/1, V/V). The upper aqueous phase was removed, and 0.3ml of PBS was added before spinning again. Both aqueous and oil phases were discarded and the pellet was counted in a Packard gamma counter (Packard, Downers Grove, TL.).

2. Synthetic GBP

Isolation and Nucleotide Sequence of a GBP cDNA

A porcine cDNA clone (PCR 1/2) corresponding to the N-terminal amino acid sequence of the GBP was isolated by applying PCR to porcine stomach cDNA using multiply degenerate oligonucleotides 1 and 2 as primers (Fig. 1, Table 1). To obtain clones PCR 8/4 and PCR 8/5, PCR was performed between a unique primer 8 based on the sequence of clone PCR 1/2, and multiply degenerate primers 4 or 5, which

SUBSTITUTE SHEET

woe designed from the sequences of tryptic peptides AT3 and AT6, respectively. An 800 bp fragment from clone PCR 8/4 was then used to screen a porcine liver λgtlO cDNA library. An overlapping clone (LIV 1) of 2102 bp was isolated, which contained an in-frame termination codon and poly(A) tail. Since no clones encoding the 5' untranslated region or initiation codon were isolated from the library, the RACE-PCR procedure, described in Example 1, was used to obtain clone RACE, which encoded the upstream regions of the transcript. The composite nucleotide sequence from clones LIV 1, PCR 1/2, PCR 8/4, PCR 8/5, and RACE (Fig. 2) was 2744 bp long, in good agreement with the size (3 kb) estimated from Northern blots of stomach and liver mRNA (Fig. 3). The composite sequence was comprised of a 56 bp 5' untranslated region, a 2289 bp open reading frame, and a 399 bp 3' untranslated region. Although there was no in-frame termination codon upstream of the ATG codon at position 1 (Fig. 2), this ATG was the likely initiation codon as it was in favourable context according to Kozak (1987), with an A at position -3 and a G at position +4. In the 3' untranslated region there was a putative AACAAA polyadenylation signal 21 bp upstream of the poly(A) tract, instead of the common AATAAA (Proudfoot and Brownlee, 1976). The poly (A) tract at the 3' terminus of the clone consisted of a stretch of 30 adenosine residues which were interrupted by a single cytosine residue at position 2727.

Size of the GBP mRNA

Northern analysis indicated that the transcript encoding the GBP was expressed in liver and in all three sections of the stomach, with expression in the corpus being stronger than in antrum or forestomach. A single transcript of approximately 3 kb was present (Fig. 3).

SUBSTITUTE SHEET

Amino Acid Sequence of the GBP

The amino acid sequence translated from the composite nucleotide sequence of the GBP cDNA was in complete agreement with the amino terminal sequence and 8 internal peptide sequences derived from the GBP (Fig. 2). .Although no such similarity had been detected with any of the individual peptides, the cDNA-deduced amino acid sequence showed a significant degree of similarity (33.3%) to the amino acid sequence of a previously reported rat peroxisomal bifunctional enzyme, enoyl- CoA hydratase/3-hydroxyacyl-CoA dehydrogenase (Osumi et al.. 1985) (Fig. 4). The N- and C-terminal regions of the GBP also showed approximately 30% homology to rat mitochondiral enoyl CoA hydratase (Minami-Ishii et al.. 1989) and to pig mitochondrial 3-hydroxyacyl-CoA dehydrogenase, respectively (Kelly et al.. 1987) (Table 2).

With regard to the distantly (30%) homologous proteins listed in Table 2, there is no evidence that these proteins function as gastrin/CCK receptors. None of them has been shown to bind gastrin or CCK and in fact all of them are found in intracellular organelles and so could not function as gastrin/CCK receptors.

Structural Features of the GBP

The calculated molecular weights of the precursor and mature protein encoded by the GBP cDNA were 83,010 and 79,028, respectively.

The GBP sequence contained three potential N-linked glycosylation sites [Asn-X- Ser/Thr], at residues 106, 392 and 439. Residue 106 was probably glycosylated since no PTH-amino acid was detected in the corresponding cycle during the sequencing of peptide AT6; in contrast residues 392 and 439 are not glycosylated, since PTH-Asn signals were seen in the corresponding positions of peptides AT3 and ATI, respectively. Glycosylation can also be demonstrated indirectly by binding to

SUBSTITUTE SHECT

concanavalin A-Sepharose and wheat germ agglutinin-Sepharose, or directly with an N-glycan detection Mt or by a reduction of 3,000 in the apparent M _j . after treatment with N-glycanase. Hydropathy analysis of the GBP (Fig. 5) indicated the presence of several potential transmembrane domains. The domains centred on residues 110 and 335 did not appear to be functional since one contained a glycosylation site at residue 106, and the other contained at residues 332-337 the nucleotide binding site [Gly-X-Gly-X-X-Gly, X = any amino acid] which in the related bifunctional enzyme is associated with the dehydrogenase activity.

The sequence of the mature GBP was preceded by a 36 amino acid signal sequence, which resembled a mitochondrial leader peptide sequence in having Arg at -10, Phe at -8 and Ser at -5 relative to the mature N-terminus (Hendrick et al.. 1989). However, the GBP contains carbohydrate (see above), and to the present inventors knowledge, no mitochondrial protein has been shown to be glycosylated (Hart et al.. 1989).

GBP Expression

A full length clone encoding the GBP was constructed in pGEM-3Z, transferred into the expression vector pMEXneo, and transected into CHO cells. Expression of the GBP resulted in a 100% increase in binding (Fig. 6). Unlabelled gastrin- _j , and the general gastrin/cholecystokinin receptor antagonist proglumide (Rovati et al.. 1967; Hahne et al.. 1981), both competed for the -^^I-gastrin binding site, with IC50 values of 0.53μM and 1.2mM, respectively. No competition was observed with the gastrin receptor antagonist L365,260 (Lotti and Chang, 1989), or with the cholecystokinin receptor antagonist L364,718 (Chang and Lotti, 1986), at concentrations up to lμM. A gastrin binding site with similar characteristics was also present in lower abundance on CHO cells transected with vector only (Fig. 6).

Those skilled in the art will appreciate that the invention described herein is susceptible to variations and modifications other than those specifically described. It is to be understood that the invention includes all such variations and modifications. The invention also includes all of the steps, features, compositions and compounds referred to or indicated in this specification, individually or collectively, and any and all combinations of any two or more of said steps or features.

The single N terminal sequence from the intact GBP is shown over two lines above (the intact sequence is ...VAVTRI....). Note that tryptic peptide AT2 overlapped with the N-terminal sequence. Multiply degenerate oligonucleotides were synthesized on a DNA synthesizer Model 380A (Applied Biosystems, Foster City, CA) using phosphoramidate chemistry. Additional underlined nucleotides incorporating the following restriction sites (sense primers 1 and 8, Hind UJ; antisense primer 2, Eco RI; antisense primers 4 and 5, Clal) were added at the 5'termini of oligonucleotides to permit cloning of PCR product. The unique oligonucleotide 8 was synthesized on the basis of the nucleotide sequence of clone PCR 1/2.

TABLE 2. Comparison of the GBP and Fatty Acid Oxidation Enzymes

Percentage Homology

Alignment Porcine Rat Rat Porcine Score (S.D.) GBP ECH/HAD ECH HAD

Porcine 33.3 33.3 29.9 GBP

Rat 23.4 34.2 33.5 ECH HAD

Rat 14.2 11.8 ECH

Porcine 15.6 16.1 HAD

Optimal alignments of the mature proteins without signal sequences were generated with the program ALIGN, using the mutation data matrix, a matrix bias of +6 and a break penalty of 6 (Dayhoff, 1979). An alignment score of 3 standard derivations (S.D.) above the meaning is usually taken as an indication of relatedness. GBP, gastrin binding protein; ECH, enolyl CoA hydratase; HAD, 3-hydroxyacyl CoA dehydrogenase.

SUBSTITUTE SHEET

REFERENCES

Andreason,G.L. and Evans,G.A. (1988). BioTechni ues 6,, 650.

Baldwin,G.S., Chandler,R., Scanlon,D.B. and Weinstock,J. (1986). J.Biol.Chem 261. 12252-12257.

Baldwin,G.S., et al (1987) Protein Seq. Data Anal. 1: 7-12.

Chang,R.S.L. and Lotti, V.J. (1986). Proc.Natl.Acad.Sci USA S2, 4923-4926.

Chirgwin, J.M., Przybyla,A.E., McDonald,R.J., and Rutter,W.J. (1979) Biochem. IS, 5294-5299.

Dayhoff, M.O. (1979(. In Atlas of Protein Sequence and Structure ed. Dayhoff, M.O.

Duong,L.T., Hadac,E.M., Miller,L.J., and Vlasuk,G.P. (1989) J.Biol.Chem. 264. 17990-17996.

Edkins,J.S. (1905) Proc.Rov.Soc.B. 76, 376.

Frohman,M.A., Dush,M.K., and Martin,G.R. (1988) Proc.Natl.Acad.Sci. USA.

55, 8998-9002.

Gregory ,R. A. and Tracy ,H.J. (1964) Gut 5_, 103-117.

Hahne,W.F., Jensen,R.T., Lemp,G.F. and Gardner .D. (1981) Proc.Natl.Acd.Sci. USA 78, 6304-6308.

Hart,G.W., Haltiwanger,R.S., Holt,G.D., and Kelly ,W.G. (1989). Ann.Rev.Biochem.. 5.8, 841-874.

Hendrick,J.P., Hodges,P.E., and Rosenberg,L.E. (1989) Proc.Natl.Acad.Sci. USA. 6, 4056-4060.

Hoosein,N.M., Kiener,P.A., Curry,R.C, Rovati,L.C, McGibra,D.K. and Brattain,M.G. (1988) Cancer Res. 48_, 7179-7183.

Kelly ,D.P., Kim,J.-J., Billadeloo,J.J., Hainline,B.E., Chu,T.W., and Strauss,A.W. (1987) Proc.Natl.Acad.Sci. USA. Si, 4068-4072.

Kyte and Doolittle (1982). J. Mol. Biol. 157: 105-132.

Kozak,M. (1987) Nucl.Acids Res. 15_, 8125-8148.

Lotti, VJ. and Chang,R.S.L. (1989) Eur.J.Pharmacol. 1 , 273-280.

Maniatis,T., Fritsch,E.F., and Sambrook,J. (1982) in Molecular Cloning (A laboratory manual), p. 391, Cold Spring Harbor Laboratory, NY.

Martin-Zanca,D., Hughes,S.H. and Barbacid,M. (1986) Nature 3J9_, 743-748.

Minami-Ishii,N., Taketani,S., Osumi,T. and Hashimoto,T. (1989) Eur.J.Biochem. 185. 73-78.

Mutt,V. (1980) in Gastrointestinal Hormones, ed. Jerzy Glass,G.B. (Raven Press, New York), pp 169-221.

Oskam,R., Coulier,F., Ernst,M., Martin-Zanca,D. and Barbacid,M. (1988) Proc.Natl.Acad.Sci. USA S5, 2964-2968.

Osumi,T., Ishaϋ,N, Hijikata,M. et al J.Biol.Chem (1985 260. 8905-8910.

Proudfoot,N.J. and Brownlee,G.G. (1976) Nature. 263. 211-214.

Rovati,A.L., Casula,P.L. and Dar,G. (1967) Minerva Medica 5_S, 3651.

Sanger,F., Nicklen,S. and Coulsen,A.R. (1977) Proc.Natl.Acad.Sci. USA. 74, 5463-5467.

Seet,L., FabrLL., Nice,E.C. and Baldwin,G.S. (1988) Biomedical Chrom. 2, 159- 163.

Soll,A.H., Amirian,D.A., Thomas,L.P., Park,J., Elashoff .D., Beaven,M.A. and Yamada,T. (1984) Am.J.Physiol. 247, G715-G723.

Southem,E.M. (175) J.Mol.Bio. 2S, 503-517.

Thumwood,C.M., Ji, H. and Baldwin,G.S. (1991) Exp.Cell Res. 192. 189-192.

Weinstock . and Baldwin,G.S. (1988) Cancer Res. 48, 932-937.

White,B.A. and Rosenzweig,S. (1989) BioTechniques 7, 694-695.

Yoo,O.J., Powell,C.T. and Agarwal,K.L. (1982) Proc.Natl.Acad.Sci. USA 79, 1049-1053.