WO1997042340A1

WO1997042340A1 - Ob receptor isoforms and nucleic acids encoding them

Info

Publication number: WO1997042340A1
Application number: PCT/US1997/007521
Authority: WO
Inventors: C. Thomas Caskey; John W. Hess; Patricia Hey; Michael S. Phillips
Original assignee: Merck & Co., Inc.
Priority date: 1996-05-06
Filing date: 1997-05-02
Publication date: 1997-11-13
Also published as: EP0900282A1; JP2000512486A; EP0900282A4; CA2253832A1

Abstract

The ob receptor has numerous isoforms resulting from alternative splicing; three novel isoforms, designated c', f, and g are disclosed. The nucleic acids encoding these isoforms are taught. Also part of the invention are vectors containing the nucleic acid encoding the receptors, host cells transformed with these genes, and assays which use the genes or protein isoforms.

Description

TITLE OF THE INVENTION

OB RECEPTOR ISOFORMS AND NUCLEIC ACIDS ENCODING THEM

FIELD OF THE INVENTION This invention relates to ob receptor protein isoforms, to DNA and RNA sequences encoding them, and to assays using the receptor isoform proteins.

BACKGROUND OF THE INVENTION Recently the identification of mutations in several genes involved in the onset of obesity in rodents have been identified. Of particular interest are mutations discovered in the peptide hormone, leptin, which is a component of a novel signal transduction pathway that regulates body weight (Zhang et al. 1994, Nature 372:425-432; Chen et al. 1996, Cell 84:491-495). Leptin was initially discovered by the positional cloning of the obesity gene, ob, in mice. Two different ob alleles have been identified: one mutation causes the premature termination of the leptin peptide resulting in a truncated protein, and the other mutation changes the transcriptional activity of the obesity {ob) gene, resulting in a reduced amount of circulating leptin.

There is a correlation between a decrease in the levels of biologically active leptin and the overt obese phenotype observed in oblob mice. Recombinant leptin has been shown to induce weight loss in the oblob mouse but not in the diabetic phenotype dbldb mouse (Campfield et al. 1995, Science 269: 546-549; Halaas et al. 1995, Science 269: 543-546; Pellymounter et al. 1995, Science 269:540-543; Rentsch et al. 1995, Biochem. Biophys. Res. Comm. 214:131 -136; and Weigle et al. 1995, /. Clin. Invest. 96:2065-2070). Although the synthesis of leptin occurs in the adipocyte, its ability to decrease food intake and increase metabolic rate appears to be mediated centrally by the hypothalamus. Injection of recombinant leptin into, the third ventricle of the brain elicits a similar response as peripheral administration of leptin. Furthermore, the recent cloning of the human receptor for the leptin, the ob-receptor (OB-R), reveals that it is transcribed in the hypothalamus (Tartaglia et al. 1995, Cell 83: 1263-1271 ; Stephens et al. 1995, Nature 311: 530-532). In addition, a mutation that results in premature termination of the long-form of the mouse OB-R, which is preferentially expressed in the hypothalamus, appears to be responsible for the obese phenotype of the dbldb mouse (Lee et al. 1996, Nature 379:632-635; Chua et al. 1996, Science 271 :994-996; and Chen et al. 1996, Cell 84:491 -495). The OB-R from wild type (lean) rats and from rats having the fatty mutation (both heterozygous and homozygous/α ) have been isolated and sequenced. (Patent Application Serial Nos. , Attorney Docket Nos. 19642PV and 19642PV2, filed

February 22, 1996 and March 22, 1996, which are hereby incoφorated by reference.)

Various isoforms of the OB-Rs have also been identified. These isoforms are due to alternative splicing. For example, in the mouse the a form has 5 amino acids following the Lysine at 889; the b form has 273 amino acids after Lysine 889; the c form has 3 amino acids after Lysine 889; and the d form contains 1 1 amino acids after Lysine 889.

It would be desirable to be able to further experiment with various isoforms in order to better understand obesity, and to be able to clone and produce novel ob receptor isoforms to use in assays for the identification of ligands which may be useful in understanding obesity and for its prevention and treatment.

DETAILED DESCRIPTION OF THE INVENTION

This invention relates to novel ob receptor isoforms designated c\ f and g which are substantially free from associated membrane proteins. It also relates to substantially purified ob receptor isoform c', f and g proteins. These isoforms are present in various species, including rat, mouse and human. Another aspect of this invention is to nucleic acids which encode OB receptor isoforms c', f or g. The nucleic acid may be any nucleic acid which can encode a protein, such as genomic DNA, cDNA, or any of the various forms of RNA. Preferably, the nucleic acid is cDNA.

This invention also includes vectors containing a OB-R isoform c', f or g gene, host cells containing the vectors, and methods of making substantially pure OB-R isoform c', f or g protein comprising the steps of introducing a vector comprising a OB-R isoform c', f or g gene into a host cell, and cultivating the host cell under appropriate conditions such that OB-R isoform c', f or g is produced. The OB-R isoform c', f or g so produced may be harvested from the host cells in conventional ways.

Yet another aspect of this invention are assays which employ OB-R isoform c', f or g. In these assays, various molecules, suspected of being OB-R isoform c', f or g ligands are contacted with a OB-R isoform c', f or g, and their binding is detected. In this way agonists, antagonists, and ligand mimetics may be identified. A further aspect of this invention are the ligands so indentified.

BRIEF DESCRIPTION OF THE FIGURES

FIGURE 1 is the amino acid sequence of wild type rat OB-R.

FIGURE 2 is the cDNA sequence of wild type rat OB-R. FIGURE 3 is the cDNA sequence encoding rat isoform.

FIGURE 4 is the cDNA specific for Rat isoform c'.

As used througout the specification and claims, the following definitions apply:

"Substantially free from associated membrane proteins" means that the receptor protein is not in physical contact with any membrane proteins.

"Substantially purified OB-receptor isoform c', f or g" means that the protein isoform is at least 90% and preferably at least 95% pure. "Wild type" means that the gene or protein is substantially the same as that found in an animal which is not considered to have a mutation for that gene or protein.

"fa" means that the gene or protein is substantially the same as that found in a rat homologous for ύ\t fatty mutation.

"Substantially the same" when referring to a nucleic acid or amino acid sequence means either it is the same as the reference sequence, or if not exactly the same, contains changes which do not affect its biological activity or function.

It has been suprisingly found, in accordance with this invention that the OB-R exists in a large variety of isoforms, including three novel ones, form c', f and g. These isoforms apply to all species, but for convenience, throughout the specification and claims, numberings of amino acids and nucleotides will use the rat wild type sequences (FIGURES 1 and 2) as a reference. However, it is to be understood that this invention is not limited to rat wild type proteins and nucleic acids and specifically includes rat (wild type and fatty), mouse, and human OB-R isoform c', f and g proteins and nucleic acids.

OB-R isoform f differs from wild type protein in that after the Lysine at position 889 (referring to the rat sequence in

FIGURE 1), there are six amino acids, ending at an Asparagine residue at position 895. In the cDNA, the codons are then followed by a Stop codon. One cDNA for rat isoform f is shown in FIGURE

3; this invention specifically includes all various cDNAs encoding an isoform f protein. The superscripted numbers refer to protein position numbers. Lys*89 _Iso890 _Met891 _Pro892 _Gly893 _Arg«94 _Asn895

In the human isoform f, Lysine 891 corresponds to the rat Lysine 889, the same six amino acids follow Lysine 889.

In a particularly preferred embodiment of this invention, the OB-R isoform f is from rat origin. OB-R isoform g differs from the wild type in that it is much shorter that the wild type sequence. The following eighteen amino acids are found at the beginning of the protein with the superscript numbers indicating their position. The Arginine at position 18 is spliced to a large fragment of the wild type molecule, beginning at the Proline at position 166 (in both mouse and human). This isoform then extends for the remainder of the wild type molecule.

Met* Phe² Gln3 Thi⁴ Pro^ Arg° He? Val⁸ Pro⁹ Gly 10

His 1 1 Lys ^{1 2} Asp ^{1 3} Leu¹⁴ He^{1 5} Ser*6 Lys ¹? Arg^{1 8} Pro^ό. . .

After Pro 166, the remainder of the protein may be the same as wild type, or, alternatively it could also contain another isoform variation, such as isoform a, b, c, d, e, or f.

A particularly preferred embodiment is the rat isoform g.

OB-R isoform c' is similar to the OB-R isoform c which was previously described [Lee et al., Nature 379: 632-635]. After Lysine at position 889, it only has three amino acids, Val⁸90 τhr891 Phe⁸92 _Stop. As can be seen, isoform c' differs from isoform c in that the final amino acid is phenylalanine rather than valine found in isoform c. Further, there are untranslated sequences in the DNA encoding isoform c' which do not appear to be present in isoform c. The cDNA encoding the rat isoform c' is given in FIGURE 4. In humans, the Val, Thr, Phe follow Lysine 891.

One aspect of this invention is the molecular cloning of these various isoforms of OB-R. The wild type and/α receptor proteins contain an extracellular, a transmembrane domain. In the rat, the extracellular domain extends from amino acids 1 -830; the transmembrane domain is from amino acids 839-860; and the cytoplasmic domain is from amino acids 860-1 162. Similar domains have bene identified for the mouse and human proteins. This invention also includes isoform c', f and g proteins which lack one or more of these domains. Such deleted proteins are useful in assays for identifying ligands and their binding activity.

In the rat wild type protein, amino acids 1-28 form a signal sequence; thus the mature proteins extend from amino acids 28-1162. The mature protein isoforms form yet another aspect of this invention. This differs somewhat from the signal sequence of 1 - 22 reported for mouse and human OB-R; the mature mouse and human isoforms form yet another aspect of this invention. The OB-R isoform c', f or g gene can be introduced into virtually any host cell using known vectors. Preferred host cells include E. coli as well as mammalian and yeast cell lines.

One of ordinary skill in the art is able to choose a known vector which is appropriate for a given host cell; generally plasmids or viral vectors are preferred. The OB-R isoform c', f or g gene may be present in the vector in its native form, or it may be under the control of a heterologous promoter, and if desired, one or more enhancers, or other sequences known to regulate transcription or translation. The host cell containing the OB-R isoform c\ f or g gene is cultured, and the OB-R isoform c', f or g gene is expressed. After a suitable period of time the OB-R c', f or g isoform protein may be harvested from the cell using conventional separation techniques.

A further aspect of this invention is the use of an OB-R c', f or g isoform in assays to identify OB-R c', f or g isoform ligands. A ligand binds to the OB-R isoform receptor, and in vivo may or may not result in an activation of the receptor. Ligands may be agonists of the receptor (i.e. stimulate its activity), antagonists (inhibit its activity) or they may bind with little or no effect upon the receptor activity.

In an assay for ligands, an OB-R isoform of this invention is exposed to a putative ligand, and the amount of binding is measured. The amount of binding may be measured in many ways; for example, a ligand or the OB-R isoform being investigated may be labeled with a conventional label (such as a radioactive or fluorescent label) and then put in contact with the OB-R isoform under binding conditions. After a suitable time, the unbound ligand is separated from the OB-R isoform and the amount of ligand which has bound can be measured. This can be performed with any of the OB-R isoforms of this invention; alternatively the amount of binding of the various isoforms can be compared. In a competitive assay, both the putative ligand and a known ligand are present, and the amount of binding of the putative ligand is compared to the amount of binding to a known ligand. Alternatively, the putative ligand's ability to displace previously bound known ligand (or vice- versa) may be measured. In yet other embodiments, the assay may be a heterogeneous one, where the OB-R isoform may be bound to a surface, and contacted with putative ligands. Dectection of binding may be by a variety of methods, including labelling, reaction with antibodies, and chomophores.

In another assay, the OB-R isoforms of this invention may be used in a "trans" activation assay. Such assays are described in U.S. Application Serial No. , Attorney Docket No. 19686PV, which was filed on April 22, 1996 and which is hereby incoφorated by reference. In this assay, a cell which expresses an OB-R isoform of this invention (either naturally or through recombinant means) is transfected with a reporter gene construct comprising a minimal promoter, a leptin activation element and a reporter gene. Transcription of the reporter gene is dependant upon activation of the leptin activation element. Binding of a ligand to the receptor isoform activates the leptin activation element, which then allows transcription of the reporter gene.

The following non-limiting Examples are presented to better illustrate the invention. EXAMPLE 1

Preparation of mRNA and cDNA from rat tissues

Tissues were collected from lean and fa/fa Zucker rats and snap frozen in liquid nitrogen. The tissues collected included: hypothalamus, pituitary, lung, liver, kidney, heart, adrenal glands, smooth muscle, skeletal muscle, and adipose tissue. The tissues were homogenized with a Brinkmann Polytron homogenizer in the presence of guanadinium isothiocyanate. mRNA was prepared from hypothalamus, lung, and kidney according to the instructions provided with the messenger RNA isolation kit (Stratagene, La Jolla, CA). cDNA was prepared from approximately 2 μg of mRNA with the Superscript™ choice system (Gibco/BRL Gaithersburg, MD). The first strand cDNA synthesis was primed using 1 μg of oligo(dT)i 2-18 primer and 25 ng of random hexamers per reaction. Second strand cDNA sythesis was performed according to the manufacturer's instructions. The quality of the cDNA was assessed by labeling an aliqout (l/10^tn) of the second strand reaction with approximately 1 μCi of [α-³²P]dCTP (3000 Ci/mmol). The labeled products were separated on an agarose gel and detected by autoradiography.

EXAMPLE 2

Preparation of a hypothalamic cDNA library

Approximately 3.6 μg of phosphorylatedBitXI adapters (Invitrogen, San Diego, CA) were ligated to approximately 3 μg of cDNA prepared as described in Example 1. The ligation mix was then diluted and size-fractionated on a cDNA sizing column (Gibco/BRL Gaithersburg, MD). Drops from the column were collected and the eluted volume from the column was determined. An aliqout from each fraction was analyzed on an agarose gel. Fractions containing cDNA of greater than or equal to 1 kb were pooled and precipitated. The size -fractionated cDNA with the Bst XI adapters was ligated into the prokaryotic vector pcDNA II (Invitrogen, San Diego, CA). The vector (4 μg) was prepared for ligation by first cutting with the restriction endonuclease Bst XI, gel purifying the linearized vector, and then dephosphorylating the ends with calf intestinal phosphatase (Gibco/BRL, Gaithersburg, MD) according to the manufacturers instructions. The ligation contained approximately 10-20 ng of cDNA and approximately 100 ng of vector and was incubated overnight at 14°C. The ligation was transformed into 1 ml of XL-2 Blue Ultracompetent cells (Stratagene, La Jolla, CA) according to the manufacture's intructions. The transformed cells were spread on 133 mm

Colony/Plaque Screen filters (Dupont/NEN, Boston, MA), plated at a density of 30,000 to 60,000 colonies per plate on Luria Broth agar plates containing 100 μg/ml Ampicillin (Sigma, St. Louis, MO).

EXAMPLE 3

Screening a hvpothalamic cDNA library

Colonies on filters were replica plated onto a second filter set. The master filter was stored at 4°C for subsequent isolation of regions containing colonies that gave a positive hybridization signal. The replica filters were grown for several hours at 37°C until colonies were visible and then processed for in situ hybridization of colonies according to established procedures (Maniatis, et al. Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Publications, Cold Spring Harbor, NY, which is hereby incoφorated by reference). A Stratalinker (Stratagene, La Jolla, CA) was used to crosslink the DNA to the filter. The filters were washed at 55°C for 2 hours in 2x SSC and 0.5% SDS to remove bacterial debris. Eight to ten filters were then placed in a heat sealable bag (Kapak, Minneapolis, MN) containing 15-20 ml of l x hybridization solution (Gibco/BRL, Gaithersburg, MD) containing 50% formamide and incubated for 1 hour at 42°C. The filters were hybridized overnight with greater than 1 ,000,000 cpm/ml of the radiolabeled probe described below in lx hybridization buffer (Gibco/BRL, Gaithersburg, MD) containing 50% formamide at 42°C. The probe, a 2.2 kb fragment encoding the extracellular portion of the Ob-R was labeled by random priming with [alpha ³²P]dCTP (3000 Ci/mmole, Amersham, Arlington Heights, IL) using redi-prime (Amersham, Arlington Heights, IL). The probe was purified from unincoφorated nucleotides using a Probequant G-50 spin column (Pharmacia Biotech, Piscataway, NJ). Filters were washed two times with O.lx SSC 0.1 % SDS at 60°C for 30 min and then subjected to autoradiography. Individual regions containing hybridization positive colonies were lined up with the autoradiogram of the hybridized filter. These were excised from the master filter, and placed into 0.5 ml Luria broth plus 20% glycerol. Each positive was replated at a density of approximate 50-200 colonies per 100 by 15 mm plate and screened by hybridization as previously described. Individual positive colonies were picked and plasmid DNA was prepared from an overnight culture using a Wizard kit (Promega, Madison, WI).

EXAMPLE 4

Amplification of Lean Rat OB-receptor cDNA using PCR

To provide for a probe to screen the hypothalamic cDNA library, the rat OB receptor was initially obtained by PCR using degenerate primers based on the mouse and human OB- receptor amino acid sequences. A set of oligonucleotide primers, were designed to regions with low codon degeneracy. The pairing of the forward primers ROBR 2 (5'-CAY TGG GAR TTY CTI TAY GT-3') and ROBR 3 (5^*-GAR TGY TGG ATG AAY GG-3') corresponding to mouse amino acid sequences HWEFLYV and ECWMKG, with reverse primers ROBR 6 (5 '-ATC CAC ATI GTR TAI CC-3'), ROBR 7 (5'-CTC CAR TTR CTC CAR TAI CC-3'), ROBR 8 (5'-ACY TTR CTC ATI GGC CA-3 ) and ROBR 9 (5'- CCA YTT CAT ICC RTC RTC-3') representing mouse amino acids, GYTMWI, VYWSNWS, WPMSKV, and DDGMKW provided good yields of the appropriately sized products. The fragments of interest were amplified as long polymerase chain reaction (PCR) products by a modifying the method of Barnes (1994, Proc. Natl. Acad. Sci. 91 :2216-2220, which is hereby incoφorated by reference). In order to obtain the required long PCR fragments, Taq Extender (Stratagene, La Jolla CA) and the Expand Long Template PCR System (Boehringer Mannheim, Indianapolis, IN) were used in combination. The standard PCR reaction mix, in a final volume of 20 μl, contained 5 ng of template (lean rat cDNA), 100 ng of primers, 500 μM dNTPs, 1 X Buffer 3 from the Expand kit, 0.1 μl each of Taq Polymerase and Taq Expander. Reactants were assembled in thin walled reaction tubes. The amplification protocol was: 1 cycle of 92°C for 30 sec, followed by 32 cycles at 92°C for 30 sec, 45°C for 1 min. and 68°C for 3 min. using a Perkin-Elmer (Norwalk, CT) 9600 Thermal Cycler.

This strategy produced a series of PCR products with the largest being approximately 2.2 Kbp amplified from primers ROBR 2 and ROBR 9. These products were subcloned for DNA sequence analysis as described below. The insert was excised from the cloning vector with the restriction endonuclease Ecυ RI, and fragments were separated from the vector by agarose gel electrophoresis. The fragments were eluted from the gel using a Prep-A-Gene kit (BioRad, Richmond CA) according to the manufacturer's instructions and radiolabeled as described above.

EXAMPLE 5

Subcloning of PCR products

PCR products of the appropriate size were prepared for subcloning by separation on an agarose gel, excising the band, and extracting the DNA using Prep-A-Gene (BioRad, Richmond, CA). PCR products were ligated into pCR™II (Invitrogen, San Diego, CA) according to the instructions provided by the manufacturer. The ligation was transformed into INVaF' cells and plated on Luria- Bertani plates containing 100 μg/ml ampicillin and X-Gal (32 μl of 50 mg/ml X-Gal (Promega, Madison, WI). White colonies were picked and grown overnight in Luria-Bertani broth plus 100 μg/ml ampicillin. Plasmid DNAs were prepared using the Wizard miniprep kit (Promega, Madison, WI). Inserts were analyzed by digesting the plasmid DNA with EcoRI and separating the restriction endonulease digestion products on an agarose gel,

Plasmid DNA was prepared for DNA sequencing by ethanol precipitation of Wizard miniprep plasmid DNA and resuspending in water to achieve a final DNA concentration of 100 μg/ml. DNA sequence analysis was performed using the ABI

PRISM™ dye terminator cycle sequencing ready reaction kit with AmpliTaq DNA polymerase, FS. The initial DNA sequence analysis was performed with Ml 3 forward and reverse primers, subsequently primers based on the rat OB-R sequence were utilized. Following amplification in a Perkin-Elmer 9600, the extension products were purified and analyzed on an ABI PRISM 377 automated sequencer (Perkin Elmer, Norwalk, CT). DNA sequence data was analyzed with the Sequencher program.

Claims

97 23- 13 -WHAT IS CLAIMED IS:

1. 0/>-receptor (OB-R) isoform c', f or g, sustantially free from associated proteins.

2. An OB-R isoform according to Claim 1 which is substantially pure.

3. An OB-R isoform according to Claim 1 which is a c' isoform.

4. An OB-R isoform according to Claim 1 which is an f isoform.

5. An OB-R isoform according to Claim 1 which is a g isoform.

6. An OB-R isoform according to Claim 1 which is from a rat.

7. An OB-R isoform according to Claim 6 which is from a wild-type rat.

8. An OB-R isoform according to Claim 6 which is from a fatty rat.

9. An OB-R isoform according to Claim 3 which is human.

10. An OB-R isoform according to Claim 4 which is human.

1 1. An OB-R isoform according to Claim 5 which is human.

12. An OB-R isoform according to Claim 3 which is from a mouse.

13. An OB-R isoform according to Claim 4 which is from a mouse.

14. An OB-R isoform according to Claim 5 which is from a mouse.

15. A nucleic acid encoding an OB-R of Claim 1.

16. A nucleic acid according to Claim 15 which is a cDNA.

17. A vector comprising a nucleic acid which encodes an OB-R of Claim 1.

18. A vector according to Claim 17 which is a plasmid.

19. A host cell containing a vector according to Claim 17.

20. A host cell according to Claim 19 which is E. coli, a mammalian cell, or a yeast cell.

21. An assay to determine if a putative ligand binds to an OB-R isoform c', f or g comprising: contacting the putative ligand with an OB-R isoform c', f or g, and determining if binding has occurred.

22. An assay according to Claim 17 wherein the ligand is labeled.

23. An assay to determine if a putative ligand binds to an OB-R isoform c', f or g which is a trans-activation assay.