Method of screening for compounds that bind P2x receptor

This invention relate to the P.sub.2X -purinoceptor, its preparation and uses.

The P.sub.2X -purinoceptor is a ligand-gated ion channel; that is, the receptor itself forms an ion channel which opens when extracellular adenosine 5'-triphosphate (ATP) binds to the receptor. There are five other classes of neurotransmitter receptors (nicotinic acetylcholine, glutamate, glycine, GABA.sub.A and 5-HT.sub.3); these form a structurally related superfamily of ligand-gated ion channels (Barnard, Trends Biochem. Sci. 17, 368-374, (1992)). The P.sub.2X -receptor now identifies a new family of this type of receptor. The unique structure of this receptor, the widespread distribution of this receptor throughout the body, and the numerous physiological roles this receptor may play, make it an important protein that can be used to identify new, therapeutically effective, compounds for the treatment of a number of pathological states.

In 1929 the eminent physiologist Szent-Gyorgyi described powerful cardiovascular actions of extracellular purine nucleosides (e.g. adenosine) and nucleotides (e.g. ATP) (Drury & Szent-Gyorgyi, J. Physiol. 68 213-237 (1929)), but it was not until 1972 that pharmacological evidence was provided to suggest the existence of distinct receptors for extracellular ATP (ie. that recognise ATP but not adenosine) (Burnstock, Pharmacological Reviews 21 509-581 (1972)). The seminal and subsequent work on this area by Burnstock and colleagues was largely unaccepted throughout the 1970s and early 1980s until the development of a range of relatively selective ligands and techniques for directly measuring ATP release overwhelmingly substantiated Burnstock's hypothesis (Barnard et al., Trends Pharmacol. Sci. 15 67-70 (1994)). In the past four or five years, unequivocal evidence for the role of ATP as a neurotransmitter has been provided for sympathetic control of blood flow to the intestine and smooth muscle tone (contractility) in genitourinary tissue such as vas deferens, bladder and ureter (Barnard et al. (loc. cit.) and Evans & Surprenant, Brit. J. Pharmacol. 106 242-249 (1992)). Substantial indirect evidence also exists for the role of ATP as a neurotransmitter in a number of distinct neurones in the spinal cord, autonomic ganglia and certain nuclei in the central nervous system (Bean, Trends Pharmacol. Sci. 15 67-70 (1992), Evans et al., Nature 357, 503-505 (1992) and Edwards et al., Nature 359 144-147 (1992)).

Purinoceptors are classified as P.sub.1 (adenosine as ligand) and P.sub.2 (ATP as ligand). The P.sub.2 receptors are subclassified into two broad types--those that are 7-transmembrane receptors that couple to G-proteins (P.sub.2Y, P.sub.2U, P.sub.2T, and perhaps P.sub.2Z) and those that form a directly gated ion channel (P.sub.2X). Pharmacological and/or physiological evidence for subtypes of each of these types of receptors exists. The most recent nomenclature for these receptors is shown below.

P.sub.2X P.sub.2Y P.sub.2Z Type Ligand-gated channel G-protein coupled Non-selective pore Subtype P.sub.2X, P.sub.2X2, P.sub.2X3 P.sub.2Y, P.sub.2Y2, P.sub.2Y3

Various P.sub.2 receptors have previously been cloned. P.sub.2Y1 was cloned by the Barnard/Burnstock group (Webb et al., FEBS Lett. 324 219-225 (1993)) based on homology with other 7-TM G-protein coupled receptors. This group used PCR technology and primers based on conserved domains of the second and sixth transmembrane regions to screen a mammalian brain cDNA library and, with final success, an embryonic chick whole-brain cDNA library.

P.sub.2Y2 /P.sub.2U was cloned by the Julius laboratory (Lustig et al., Proc. Nat'l. Acad. Sci. USA 90 5113-5117 (1993)) by expression cloning in the oocyte from cDNA obtained from a NG108-15 neuroblastoma cell line.

P.sub.2Y3 /P.sub.2T was also obtained by the Barnard/Burnstock group using the same probe and embnryonic brain cDNA library used to obtain the P.sub.2Y1 receptor (Barnard et al., Trends Pharmacol. Sci. 15 67-70 (1994)).

However, as yet, cloning of the P.sub.2X receptor has remained an elusive goal. The prior cloning exercises undertaken for the other P.sub.2 receptors do not provide an adequate lead to enable the P.sub.2X receptor to be cloned. First, all the above purinoceptors are G-protein activation of one or more second messenger systems. There are over 200 currently identified proteins which belong to this 7-TM/G-protein coupled family. Agonists at these receptors activate cascades of intracelluar transduction pathways, often involving several enzymes; the response of the cell is inherently slow (several seconds to minutes) and changes in excitability are subtle if they occur. In contrast, the P.sub.2X receptor is a fundamentally different type of purinoceptor that incorporates an ion channel. Activation of P.sub.2X receptors is rapid (milliseconds), has predominately local effects, and brings about immediate depolarisation and excitation.

Secondly, the tissue distribution of the P.sub.2X receptor is distinctly different from other purinoceptors, and the physiological roles differ from other purinoceptors.

One of the principal established ways to clone a receptor is based on sequence relatedness of the nucleotides that encode the amino acids of the receptor protein; it depends on there being a fairly high level of homology between a known sequence and that of the unknown receptor. This method was used to clone the P.sub.2Y1 from (above). Several laboratories, including that of the applicants, invested significant effort in obtaining the P.sub.2X receptor using PCR techniques and primers based on conserved regions of various ligand-gated ion channels (ie. nicotinic ACh, GABA, glutamate, 5-HT.sub.3). This approach failed. With hindsight, this failure can be rationalised, as it can now, but only now, be seen that the structure of the P.sub.2X receptor bears no homology with any of these ligand-gated ion channels. For the same reason, approaches based on fragment hybridisation would not succeed.

However, by adopting a different approach, it has now been found possible to clone the P.sub.2X receptor, and it is on this achievement that the present invention is in part based.

According to a principal aspect of the present invention, there is provided a recombinant or isolated DNA molecule encoding a P.sub.2X receptor, wherein the receptor:

(a) has the amino sequence shown in FIG. 1, FIG. 2, FIG. 3 or FIG. 4; or

(b) is substantially homologous to the sequence shown in FIG. 1, FIG. 2, FIG. 3 or FIG. 4;

or a fragment of such a DNA molecule, which fragment includes at least 15 nucleotides taken from nucleotides 1 to 813 shown in FIG. 1, the full nucleotide sequences shown in FIG. 2 and 3, or from nucleotides 1 to 1744 shown in FIG. 4.

The sequence shown in FIG. 1 is a cDNA sequence that encodes a rat vas deferens P.sub.2X receptor. This sequence is 1837 bases in length and encodes a protein of 399 amino acids. As was determined after the receptor was cloned, approximately one half of the protein-encoding sequence, from nucleotides 814 onwards, had been discovered previously but the function of the previously cloned sequence was not known except that it appeared to be implicated in apoptotic cell death (Owens et al., Mol. Cell. Biol. 11 4177-4188 (1991); the Owens et al. sequence lacks a translation initiation site and could not be made into protein. (In FIG. 1, the upstream portion of the reported sequence of Owens et al., namely PQLAHGCYPCPPHR, which is not shared with the P.sub.2X receptor, is shown for comparative purposes and does not form part of the invention.)

Preferably the FIG. 1 sequence fragments are taken from nucleotides 1-810. Often the FIG. 4 sequence fragments are taken from nucleotides 1-777.

The sequence shown in FIG. 2 is a cDNA sequence that encodes a rat superior cervical ganglion P.sub.2X receptor.

The sequence shown in FIG. 3 is a cDNA sequence that encodes a rat dorsal root ganglion P.sub.2X receptor.

The sequence shown in FIG. 4 is the cDNA sequence that encodes a human P.sub.2X receptor. The cDNA was isolated from the human urinary bladder using a rat P.sub.2X probe. It is 2643 bases long and encodes a 399 amino acid protein having an amino acid sequence which is highly homologous with the amino acid sequence of the rat P.sub.2X receptor isolated from rat vas deferens and with the rat P.sub.2X receptors isolated from a rat superior cervical ganglion and from a rat dorsal root ganglion. Recently we have become aware of an expressed sequence tag corresponding to residues 1745-1933 (Proc. Natl. Acad.Sci. USA 91,10645-10649 (October 1994).

Sequences which are substantially homologous to the FIG. 1, FIG. 2, FIG. 3 or FIG. 4 amino acid sequence include those which encode proteins having at least 40%, 50%, 60%, 70%, 80%, 90%, 95% or 99% homology in increasing order of preference. A protein having at least 99% homology with the amino acid sequence of FIG. 1, FIG. 2, FIG. 3 or FIG. 4 will have no more than four amino acid variations from such a sequence. Preferred substantially homologous sequences include P.sub.2X sequences from other species. Thus for the rat P.sub.2X receptor sequences a preferred substantially homologous sequence is a human P.sub.2X sequence. One method of determining sequence homology is disclosed in W R Pearson and D J Lipman, Proc. Natl Acad Sci USA 85:2444-2448 (1998).

Fragments may of course be larger than 15 nucleotides. Fragments encoding substantially the whole of the P.sub.2X rat receptors or human receptor may be expected to share the biological activity of the receptor, or at least some of its biological activities. Shorter fragments may be useful for encoding one or more selected domains of the receptor, or simply as probes for detecting or identifying other useful DNA sequences, including those encoding substantially homologous proteins. Fragments of at least 20, 30 or 50 nucleotides may be more frequently of use than shorter ones.

DNA molecules of the invention are useful for a number of purposes. First, and not least, the P.sub.2X cDNA shown in FIG. 1, in FIG. 2, in FIG. 3 and in FIG. 4 enables the relevant proteins to be expressed in living cells. This would not be possible with fragments of the cDNA. However not only are fragments of DNA within the scope of the invention, for the various purposes mentioned above, but also genomic and other sequences of DNA (including synthetic DNA and "minigenes", which include at least one, but not all, of the introns naturally present in the gene) are included within its scope. cDNA sequences encoding the rat receptor proteins or human P.sub.2X receptor protein may be preferred in some circumstances because such sequences are smaller than either genomic or minigene DNA and therefore more amenable to cloning manipulations. The P.sub.2X receptor protein can be stably expressible in chinese hamster ovary (CHO) cells, as will be described below.

Still on the subject of expression, while it would be possible to express genomic DNA in eukaryotic cells, it is much more difficult to manipulate the DNA for insertion into host cells due to the larger size that commonly results from introns. The size is particularly important for the expression of RNA; very long cRNAs--the size of whole genes--are difficult to make in sufficient quantity. On the other hand, expression from RNA is much preferred at least for the investigation of ion channel proteins, because the Xenopus oocyte is sufficiently large to be studied easily by electro-physiological methods.

Secondly, the cDNA sequences encode proteins that, in their predicated folding within the membrane, differ from other known proteins. This is advantageous because, based on historical precedent, this will lead to the discovery of a large family of related proteins and these may have functional roles unrelated to signalling mediated by ATP.

Thirdly, knowledge of the protein sequences encoded by rat and human P.sub.2X cDNA allows the development of molecular models that predict the detailed disposition within the membrane. It further allows the correctness of such models to be determined by expression of mutagenised proteins. These two approaches are advantageous because they may premit the molecular design of complementary therapeutic agents that activate or block the receptor.

Fourthly, the P.sub.2X cDNA sequences allow the distribution of the RNA that encodes this receptor, as well as the receptor protein itself, to be mapped in human tissues. RNA distribution can be determined by in situ hybridisation. Such hybridisation studies are disclosed in the present examples. Knowledge of a deduced amino acid sequence from cDNA allows synthetic peptides to be made that can be used to generate antibodies that selectively recognise a P.sub.2X receptor. Thus a P.sub.2X protein can be mapped by immunohistochemistry. This may suggest novel therapeutic applications for drugs that activate or block the P.sub.2X receptor, that can not be predicted on the basis of less sensitive current methods for localising the receptor (radioactive ligand binding).

Fifthly, rat P.sub.2X cDNA is advantageous because it can allow the isolation of a closely related cDNA from human tissue.

Sixthly, the isolation of the human P.sub.2X cDNA clone will enable a human genomic clone to be obtained. It is probable that mutations of this gene will be discovered that lead to human genetic disease. The analysis of such mutations may lead to appropriate treatments of diseases or disorders caused by such mutations.

In one aspect of the present invention rat vas deferens P.sub.2X receptor was cloned by a method which does not require prior inference about structure. Tissues were chosen that were believed to be rich in the RNA for the receptor of interest. A number of tissue sources were tried but they did not provide RNA that led to ATP responses in oocytes. Eventually, vas deferens was chosen. From extracted polyadenylated RNA, a cDNA library or bank that corresponds as far as possible to the DNAs in the tissue was constructed. It was not assured, either before work began or until it was satisfactorily completed, that a satisfactory cDNA library in which the rat P.sub.2X gene was represented could be constructed; nevertheless, this was achieved in plasmid pBKCMV.

An individual clone within the library that contains the rat vas deferens P.sub.2X cDNA of interest was detected by progressive fractionation of the library; at each step the fraction was tested to determine whether RNA made from it can direct the formation of the protein of interest. More specifically, RNA was transcribed in vitro from the cDNAs in the library (approximately 2 million) and the RNA ("cRNA") mixture was injected into immature Xenopus oocytes. cRNA is very susceptible to inadvertent enzymatic degradation, so all procedures were carried out under sterile conditions. The cDNA pools were made by the miniprep procedure and therefore contained large amounts of E. coli RNA; this difficulty was overcome by precipitating any RNA before the cRNA was transcribed.

Detection of the protein can in principle be done by radioactive ligand binding or by a functional response. The activation of G proteins in the Xenopus oocyte and the subsequent cellular response was used to obtain the P.sub.2Y2 /P.sub.2U receptor. In the present work, a decision was made to use the opening of the integral ion channel of the P.sub.2X as the response. Individual oocytes were screened two days after injection to determine whether they had made P.sub.2X receptor protein in their membrane. This was done by recording the current flowing across the oocyte membrane when ATP (30 .mu.M) was applied to the outside of the oocyte; if the P.sub.2X receptor has been produced, a small transient current would be expected. However, testing for expression of the receptor was not straightforward, as some batches of oocytes exhibit responses to ATP because they naturally express other kinds of ATP receptor. This difficulty was overcome as follows: when an oocyte responded to ATP with the expected current this was further tested by blockade with a P.sub.2X receptor antagonist (suramin). The cDNA fraction that gave let to the positive response in such an oocyte was further divided, and each fraction was again tested. Such progressive fractionation led to isolation of a single clone. The insert in the plasmid was sequenced; the sequence is shown in FIG. 1. This sequence was used to design PCR primers which were used in the cloning of cDNA encoding a P.sub.2X receptor from a rat superior cervical ganglion (see FIG. 2). A similar procedure was then used in the cloning of cDNA encoding a P.sub.2X receptor from a rat dorsal root ganglion (see FIG. 3).

DNA in accordance with the invention will usually be in recombinant or isolated form and may be in the form of a vector, such as a plasmid, phagemid, cosmid or virus, and in some embodiments contains elements to direct expression of the protein, for example in a heterologous host. Non-expressible vectors are useful as cloning vectors.

Although DNA in accordance with the invention may be prepared synthetically, it is preferred that it be prepared by recombinant DNA technology. Ultimately, both techniques depend on the linkage of successive nucleotides and/or the ligation of oligo- and/or poly-nucleotides.

The invention enables, for the first time, P.sub.2X receptor to be prepared by recombinant DNA technology and hence free from protein with which it is naturally associated or contaminated (such as the P.sub.2U or, particularly, P.sub.2Y receptor, or other ATP receptors or binding proteins), and this in itself forms another aspect of the invention. The protein will generally be associated with a lipid bilayer, such as a cell, organelle or artificial membrane. P.sub.2X receptor prepared by expression of DNA in accordance with the first aspect may be glycosylated, but does not have to be. Generally speaking, receptor proteins and ion channels that are glycosylated will also function after carbohydrate removal or when expressed in cells that do not glycosylate the protein. However, there are often important quantitative differences in the function between the glycosylated and non-glycosylated protein. In the case of the rat was deferens P.sub.2X receptor, we believe that the native protein is glycosylated because it has a molecular weight of 62 kd when purified from the rat vas deferens, as compared to the molecular weight of 45 kd for the cloned protein. Similar results were obtained for the human P.sub.2X receptor (see later).

There are also several asparagine residue in the extracellular domain that are likely sites of sugar attachment.

Knowledge of the amino acid sequence of a P.sub.2X receptor enables the protein or peptide fragments of it to be prepared by chemical synthesis, if required. However, preparation by expression from DNA, or at least translation from RNA, will usually be preferred.

Particularly useful peptide fragments within the scope of the invention include epitopes (which may contain at least 5, 6, 7, 10, 15 or 20 amino acid residues) of the P.sub.2X receptor which are immunologically non-cross reactive with the RP-2 polypeptide disclosed in Owens et al., loc. cit.

A P.sub.2X receptor, and fragments of it, can be used to prepare specified polyclonal and monoclonal antibodies, which themselves form part of the invention. Polyclonal and monoclonal antibodies may be prepared by methods well established in the art. Hybridoma and other cells expressing monoclonal antibodies are also within the invention.

RNA encoding a P.sub.2X receptor, transcribable from DNA in accordance with the invention and substantially free form other RNAs, also forms part of the invention, and may be useful for a number of purposes including hybridisation studies, in vitro translation and translation in appropriate in vivo systems such as Xenopus oocytes.

The invention also relates to host cells transformed or transfected with a vector as described above. Host cells may be prokaryotic or eukaryotic and include mammalian cells (such as COS, CHO cells and human embryonic kidney cells (HEK 293 cells)), insect cells, yeasts (such as Saccharomyces cerevisiae) and bacteria (such as Escherichia coli). Host cells may only give transient expression of the receptor, as in the case of COS cells, but for preference the host cells are stably transfected with the vector. Host cells which appropriately glycosylate the receptor are preferred. A CHO cell line or any other cell line that stably expresses a P.sub.2X receptor can be used for electrophysiological, calcium-influx, calcium-imaging and ligand-binding studies. Host cells which do not express the receptor may still be useful as cloning hosts.

A P.sub.2X receptor prepared by recombinant DNA technology in accordance with the invention has a number of uses, either in situ in a membrane of the expression host or in in vitro systems. In particular the receptor can be used as a screen for compounds useful in a variety of human (or other animal) diseases and conditions, as will now be briefly described. Such compounds include those present in combinatorial libraries, and extracts containing unknown compounds (e.g. plant extracts).

Epilepsy Epilepsy results from overexcitation of distinct neurones in specific regions of the brain, in particular in the hippocampus. Functional ATP P.sub.2X receptors are known to be present in some hippocampal neurones. If the P.sub.2X receptors are expressed on inhibitory interneurons, then receptor agonists would be therapeutically useful. If the receptor is expressed on principal (pyramidal or granule) cells, then receptor antagonists will be useful. If will now be possible to determine which classes of neuron express the receptor.

Cognition Hippocampal neurones respond to ATP by activation of a P.sub.2X receptor; these areas are of primary importance to cognition. It is now possible to determine the cellular localisation of the P.sub.2X receptor with in the hippocampus; depending on this localisation, either agonists or antagonists might be effective to enhance memory.

Emesis The acute trigger for emesis is rapid contraction of smooth muscle of the upper gastrointestinal tract. Activation of ATP P.sub.2X receptors present on smooth muscle of the GI tract, in particular the stomach and trachea, results in strong, rapid muscle contractions. P.sub.2X -antagonists selective for visceral smooth muscle could be useful for emesis. Furthermore, P.sub.2X receptors are known to be expressed in the nucleus of the tractus solitarious (Ueno et al., J. Neurophysiol. 68 778-785 (1992)) and may be involved in transmission from primary visceral afferents; this could be blocked by selective P.sub.2X antagonists.

Pain First, P.sub.2X receptors are expressed in dorsal horn neurones of the spinal cord. Activation of these neurones by ATP causes fast depolarizing, excitatory responses (Jahr & Jessell, Nature 304 730-733 (1983)); if a component of the transmission from nociceptive fibres is mediated by ATP then this could be blocked by a P.sub.2X antagonist. Secondly, ATP is one of the most noxious substance known when applied intradermally. This is because it activates directly the peripheral terminals of small diameter nociceptive fibres; it is known that the cell bodies in the dorsal root ganglion express P.sub.2X receptors. A P.sub.2X antagonist would be a peripherally active analgesic, and is likely to be effective in migraine.

Asthma Bronchial smooth muscles contract in response to activation of P.sub.2X receptors. This may occur in response to ATP released from sympathetic nerves, or from local immune cells. P.sub.2X antagonists may help to prevent stimulus-evoked spasms of bronchial smooth muscle and thereby reduce the frequency and/or severity of asthmatic attacks.

Peripheral vascular disease It is becoming clear that ATP and not noradrenaline is the primary vasoconstrictor neurotransmitter in small resistance arteries--those that comprise over 70% of total peripheral resistance. This has been shown for many vessels (Westfall et al., Ann. N.Y. Acad. Sci. 603 300-310 (1991)). A selective antagonist could be used for local collateral vasodilation.

Hypertension Hypertension that is associated with increased sympathetic tone could be treated with P.sub.2X receptor antagonists, because ATP is a major excitatory transmitter to many resistance vessels in several species including man (Westfall et al., loc. cit. and Martin et al., Br. J. Pharmacol. 102 645-650 (1991)).

Diseases of the immune system A molecule identical to part of the P.sub.2X receptor has been cloned from thymocytes that have been induced to die (Owens et al., loc. cit.).

The selective expression in these conditions implies that a molecule closely related to the P.sub.2X receptor play a role in the apoptosis that is an integral part of the selection of immunocompetent cells. The molecule described by Owens et al. (RP-2) was incomplete and could not have been translated into protein. The cloning of the P.sub.2X receptor will now allow the isolation of full length RP-2 clones, their heterologous expression and the determination of their functional roles.

Irritable bowel syndrome ATP is an important transmitter to the smooth muscles of the intestinal tract, particularly in the colon. It is also a transmitter between neurons in the enteric nervous system, by activating P.sub.2X receptors (Galligan, Gastroenterology, in press). Antagonists at P.sub.2X receptors may therefore have utility in the management of this condition.

Premature ejaculation This could be prevented by preventing stimulus-evoked contraction of vas deferens smooth muscle. P.sub.2X receptors are highly expressed in this tissue; antagonists at this site would prevent vas deferens contractility during sympathetic excitation.

Cystitis P.sub.2X receptors may be implicated in increased bladder sensitivity in patients with cystitis. Thus antagonists of such P.sub.2X receptors may be useful in treating cystitis.

Useful agonists and antagonists identified as described above also form an aspect of the invention.

The cloning of the hP.sub.2X receptor is an important aspect of the present invention. hP.sub.2X is the first human member of a multigene family of ionotropic purinoceptors. Its strong similarity with P.sub.2X, isolated from rat vas deferens and with P.sub.2X isolated from rat superior cervical ganglion or from rat dorsal root ganglion, suggests that it is a human homolog of the rat proteins. The present inventors have found that differences between these two sequences are nearly all conservative substitutions of hydrophilic residues. Surprisingly, hP.sub.2X has only 41% identity with the other reported P.sub.2X receptor, that from rat PC12 cells (Brake et al, New structural motif for ligand-gated ion channels defined by an ionotropic ATP receptor Nature 371: 519-523 (1994)). The PC12 derived receptor was proposed to have a similar membrane topography and shares the conserved spacing of cysteine residues, indicated for the two smooth muscle sequences in FIG. 5.

The computed molecular weight of the hP.sub.2X polypeptide (45 kd) agrees with that of the in vitro translation product when made in absence of pancreatic microsomal membranes. A larger product, 60 kd, produced in presence of microsomes suggests glycosylation and supports the idea of a central extracellular domain. The predicted hP.sub.2X protein thus has the general features of other cloned members of this family (Valera et al, a new class of ligand-gated ion channel defined by P.sub.2X receptor for extracellular ATP Nature 371: 516-519 (1994); Brake--supra): a large, cysteine-rich extracellular central domain flanked by two transmembrane spans and short internal N- and C-termini.

The distribution of the hP.sub.2X mRNA was examined by northern blot analysis. Hybridisation of a principal 2.6 kb species was seen in all RNA samples tested, with the exception of brain. A smaller, 1.8 kb band, observed in spleen, and lung mRNAs could be due to a shorter 3' untranslated portion of the mRNA, as occurs for P.sub.2X mRNA from the rat vas deferens. The hybridisation observed in thymus, lung, spleen and liver RNA may reflect the content of smooth muscle in those organs. However, hP.sub.2X is likely to have roles in other cell types, as demonstrated by its presence in adrenal gland, and the hemopoertic cell line HL60. The strong induction of hP.sub.2X mRNA by HL60 differentiation may reflect a parallel observation in rat in which the smooth muscle form of P.sub.2X mRNA can be induced in immature thymocytes by dexamethasone (RP2 mRNA; Owens et al, Identification of mRNAs associated with programmed cell death in immature thymocytes J J Mole Cell Biol 11; 4177-4188 (1991)).

The present invention has enabled the first comprehensive pharmacological characterization of a cloned P.sub.2X -purinoceptor to be made. The time course of the responses to ATP and the sensitivity to .alpha., .beta., -methylene ATP are similar to those reported for the native hP.sub.2X in urinary bladder (Inue & Brading, Human, pig and guinea-pig bladder smooth muscle cells generate similar inward currents in response to purinoceptor activation Br J Pharmacol 103: 1840-1841 (1991)). Thus the functional properties of some native P.sub.2X purinoceptors can be obtained by the expression of a single molecular species. The agonist induced current recorded from ooctyes expressing the hP.sub.2X clone gives a direct measure of the activation of P.sub.2X -purinoceptors in a system with low levels of endogenous ectonucleotidase activity. The agonist profile 2MeSATP.gtoreq.ATP>.alpha.m .beta., -meATP for hP.sub.2X is similar to that of the cloned rat vas deferens P.sub.2X -purinoceptor. The high potency of .alpha., .beta., -meATP in whole tissue studies (.alpha.,.beta., -meATP>>2MeSATP.gtoreq.ATP) probably reflects, its resistance to ectonucleotidases.

The concentration-effect curves for ATP, 2MeSATP and 2-chloro-ATP were superimposable, indicating that these particular substitutions at the 2' position on the adenine ring do not affect agonist binding to the P.sub.2X -purinoceptor. The agonist activity of AP.sub.5 A is likely to be because diadenosine phosphates (AP.sub.5 A, and AP.sub.6 A) released from the platelets can act as vasoactive agents through activation of P.sub.2X -purinoceptors.

Preferred features of each aspect of the invention are as for each other aspect, mutatis mutandis.

The invention will now be illustrated by the following examples. The examples refer to the accompanying drawings, in which:

FIGS. 1A-1C show DNA and amino acid sequences of the rat vas deferens P.sub.2X receptor as determined in Example 2. (SEQ ID NO 4).

FIGS. 2A-2C show DNA and amino acid sequences of a rat superior cervical ganglion P.sub.2X receptor, as determined in Example 11. (SEQ ID NO 5).

FIGS. 3A-3C show DNA and amino acid sequences of a rat dorsal root ganglion P.sub.2X receptor, as determined in Example 12. (SEQ ID NO 6).

FIGS. 4A-4D show DNA and amino acid sequences of a human P.sub.2X receptor as determined in Example 6. (SEQ ID NO 7)

FIG. 5 shows the alignment of the predicted amino acid sequence of hP.sub.2X with the rat vas deferens P.sub.2X, and in vitro translation of hP.sub.2X protein.

TM1 and TM2 filled boxes indicate the hydrophobic regions and boxed amino acids indicate the differences between the two sequences,

o indicates conserved cysteine residues.

* Indicates potential sites of N-glycosylation.

FIG. 6 shows an SDS-PAGE analysis of .sup.35 S-methionine labelled hP.sub.2X protein. Lanes 1 and 2 show in vitro coupled transcription/translation of pBKCMV-hP.sub.2X cDNA in the absence and presence of microsomal membranes, respectively.

FIGS. 7 AND 8 show Northern analyses of the hP.sub.2X cDNA, wherein:

A) FIG. 7 shows Northern blot with 8 .mu.g of total RNA from differentiated HL60 cells.

0 indicates HL60 cells without treatment;

PMA2 and PMA3 indicate respectively cells treated 2 days, and 3 days with PMA;

DMSO indicates cells treated 6 days with DMSO;

dcAMP indicates cells treated 5 days with dibutyryl cAMP;

UB indicates 100 ng of polyA.sup.+ RNA from human urinary bladder; and

B) FIG. 8 shows distribution of hP.sub.2X in human tissues. Lanes contained 1 .mu.g polyA.sup.+ RNA except for the urinary bladder which contained 0.2 .mu.g of polyA.sup.+ RNA.

FIGS. 9, 10 and 11 show the response of oocytes expressing hP.sub.2X to purinoceptor agonists, wherein:

A) FIG. 9 shows traces which show inward currents evoked by ATP, 2 me SATP and .alpha.,.beta., me ATP (0.1, 1, and 100 .mu.M). Records for each agonist are from separate oocytes;

B) FIG. 10 shows concentration response relationships of full P.sub.2X -purinoceptor agonists. Data are expressed relative to the peak response to 100 .mu.M ATP; and

C) FIG. 11 shows concentration response of partial P.sub.2X -purinoceptor agonists. Data are fitted with a Hill slope of 1 (n=4-8).

FIGS. 12 and 13 show the effects of P2-purinoceptor antagonists of hP.sub.2X mediated responses, wherein;

A) FIG. 12 shows concentration response curves for ATP in the presence of the P2-purinoceptor agonist suramin (1, 10 and 100 .mu.M) (n=4 for each point); and

B) FIG. 13 shows concentration dependence of suramin, DIDS PPADS and P5P in inhibiting the response to 10 .mu.M ATP (n=4 for each point).

FIGS. 14A-14D show the results of the functional characterisation of rat superior ganglion P.sub.2X receptors (as encoded by clone 3, described in Example 10). These experiments provided electrical recordings from transfected HEK293 cells.

Top left: Superimposed currents evoked by ATP (30 .mu.M) during the time are indicated by the bar. Holding potential was changed from -70 to 20 mV.

Top right: Peak current as a function of membrane potential.

Bottom left: Superimposed currents evoked by ATP, from 1 to 300 .mu.M.

Bottom right: Concentration-response curves for ATP and .alpha..beta.methylene-ATP (points are mean.+-.s.e. mean for 5-8 experiments).

FIGS. 15A-15C show the inhibition of currents caused by various substances acting on the clone 3 form of the P.sub.2X receptor (as described in Example 11), compared with PC12 and human bladder forms in HEK293 cells.

Top: inhibition by suramin.

Middle: inhibition by PPADS.

Bottom: inhibition by pyridoxal 5-phosphate.

EXAMPLES

(i) Rat Vas Deferens P.sub.2X Receptor

Example 1

Cloning of the Rat vas deferens P.sub.2X Receptor

Total RNA was isolated by the guanidinium isothiocyanate method (Sambrook et al., "Molecular Cloning: A Laboratory Manual" Cold Spring Harbor Laboratory Press, second edition (1989)) from vas deferens of 4 weeks old Sprague-Dawley male rats, and the poly A+ RNA was subsequently purified by oligo (dT)-cellulose. First strand cDNA primed with the sequence 5'-GAGAGAGAGAGCGGCCGCTTTTTTTTTTTTTTT-3' (SEQ ID NO 1) was synthesised with SUPERSCRIPT.TM. (BRL, Gaithersburg, Md., USA). After conversion of the cDNA to double stranded (Gubler & Hoffman, Gene 25 263-269 (1983)) EcoRI linkers were ligated to the cDNA, and the product was digested with NotI. The EcoRI-NotI cDNA of 1.3 to 9 kb was isolated by gel electropheresis, and a unidirectional library was constructed by ligation of the cDNA to pBKCMV (Stratagene, San Diego, Calif., USA) digested with the same enzymes. The library was electroporated into E. coli DH10B cells divided in 24 pools of 8.times.10.sup.4 clones. The plasmid DNA from the pools was prepared by minialkaline lysis followed by LiCl precipitation (Sambrook et al., loc. cit). NotI-linearised cDNA was transcribed in vitro with T3 RNA polymerase in the presence of the cap analogue m7GpppG (Sambrook et al., loc. cit). The in vitro transcribed RNA (cRNA) was concentrated to 4 mg/ml.

Example 2

Sequencing of the Rat vas deferens P.sub.2X Receptor cDNA

The cDNA insert was sequenced the exonuclease method (Henikoff Meth. Enzymol. 155 156-164 (1987)). The sequence is shown in FIG. 1.

Example 3

Functional characterisation of the Rat vas deferens P.sub.2X Receptor cDNA in Oocytes

50 nl (200 ng) of RNA was injected into defolliculated Xenopus oocytes. After incubation for 2-6 days at 18.degree. C., the oocytes were assayed for ATP-evoked currents by a two-electrode voltage clamp (GENECLAMP.TM.); one electrode is to hold the voltage constant (at -100 mV), and the other is to measure the currents. A cDNA pool which showed ATP induced currents was subdivided to obtain a single clone (P.sub.2X). Electrophysiological measurements were done at -100 mV, in a perfusion medium containing 96 mM NaCl, 2 mM KCl, 1.8 mM CaCl.sub.2, 1 mM MgCl.sub.2, 5 mM Hepes pH 7.6, and 5 mM sodium pyruvate. For dose-response curves and suramin inhibition, oocytes were injected with 100 ng P.sub.2X cRNA, and all recordings were performed at -60 mV, with Ba.sup.2+ substituted for external Ca.sup.2+ to prevent activation of endogenous Ca.sup.2+ -activated Cl.sup.- currents. Microelectrodes (0.5-2 M.OMEGA.) were filled with 3M KCl.

Example 4

Functional characterisation of the Rat vas deferens P.sub.2X Receptor cDNA in HEK 293 Cells

HEK 293 cells were transfected by the lipofectin method (Felgner et al., Proc. Nat'l. Acad. Sci. USA 84 7413-7417 (1987)) with P.sub.2X -plasmid. DNA concentration used was 1 mg/2 ml medium placed into a 35 mm petri dish containing four 11 mm diameter coverslips on which HEK cells were placed at 10,000 cells per coverslip. Cells were exposed to lipofectin/DNA for 6 h and recordings made 16-36 h later; 40-60% of cells from which recordings were made exhibited P.sub.2X responses. Currents were recorded from HEK 293 cells using whole-cell recording methods and the AXOPATCH.TM. 200 amplifier (Axon Instruments); patch pipettes (5 M.OMEGA.) contained (mM) Cs or K aspartate 140, NaCl 5, EGTA 11, HEPES 5. The external solution was (mM) NaCl 150, KCl 2, CaCl.sub.2 2, MgCl.sub.2 1, HEPES 5 and glucose 11; the pH and osmolarity of both solutions were maintained at 7.3 and 305 mosmol/l respectively. All recordings performed at room temperature. Data acquisition and analysis were performed using PCLAMP.TM. and AXOGRAPH.TM. software (Axon Instruments). Solutions for experiments examining calcium permeability of ATP currents in HEK cells contained (mM): internal solution NaCl 150, HEPES 5, CaCl.sub.2 0.5 and EGTA 5 (free calcium concentration about 5 nM); external sodium solution NaCl 150, glucose 11, histidine 5, CaCl.sub.2 2; external calcium solution CaCl.sub.2 115, glucose 11 and histidine 5. The pH and osmolarity of the solutions were 7.4 and 295 mosmol/l respectively. For single channel measurements, a GENECLAMP.TM. 500 amplifier and outside-out recording methods were used (Adelman et al., Neuron 9 209-216 (1992)). Wax-coated patch pipettes (5-10 M.OMEGA.) contained (mM) K-gluconate 115, HEPES 5, BAPTA 5 and MgCl.sub.2 0.5, external solution was 96 mM NaCl, 2 mM KCl, 1.8 mM CaCl.sub.2, 1 mM MgCl.sub.2, 5 mM Hepes pH 7.6, and 5 mM sodium pyruvate. ATP was applied by U-tube typically for 1 s; data was sampled at 5 kHz in 2 s segments beginning 300 ms prior to onset of agonist (ATP) application and filtered at 1 kHz.

Example 5

Transfection of the Rat vas deferens P.sub.2X Receptor cDNA into CHO and HEK293 Cells

CHO cells were stably transfected by a method used for other ion channels (Claudio, Meth. Enzymol. 207 391-408 (1992)). Transfection was confirmed by a) electro-physiological recording and b) radioligand binding. ATP and other agonists (up to 30 .mu.M) caused rapidly desensitising inward currents in 14 of 14 CHO cells stably transfected, and had no effect in 45 of 45 non-transfected cells. [.sup.3 H].alpha..beta.methyleneATP binding was more than 600 cpm per million transfected cells with less than 80 cpm nonspecific binding.

Stable transfection of HEK293 cells was also achieved. This was confirmed by electrophysiological recording.

(ii) Human P.sub.2X Receptor

The materials and methods used in the human P.sub.2X receptor examples are set out below:

In Vitro translation In vitro coupled transcription/translation were performed using Promega's TNT Coupled reticulocyte lysate Systems with or without 2 .mu.l of canine pancreatic microsomal membranes (Promega). .mu.g Circular pBKCMV-hP.sub.2X (0.5 ug) was transcribed with the T3 RNA polymerase as described in the system manual in a 25 .mu.l reaction for 2 h are 30.degree. C. Synthesized proteins (5 .mu.l) were analysed by SDS-PAGE and autoradiography.

Differentiation of HL60 cells HL60 cells (human promyelocytes ATCC CCL240) were passaged twice weekly in RPMI-1640 supplemented with 25 mM HEPES, 2 mM Glutamax II, and 10% heat-inactivated fetal calf serum (GIBCO BRL). For each experiment 33.times.10.sup.6 cells were resuspended at 2.5.times.10.sup.5 cells/ml in medium containing either phorbol mystate acetate (100 nM), 1.1% DMSO, or dibutyryl cAMP (200 .mu.M) (SIGMA) for the indicated times.

Northern blot analysis PolyA.sup.+ RNAs were obtained from Clontech Laboratories Inc. (Palo Alto) except for the urinary bladder and HL60 mRNA which were prepared as described (Valera et al (1994)--supra). Samples were quantified by measuring the O.D. at 260 nm, and by staining the membrane with methylene blue. The RNA were fractionated on a 1% agarose--6% formaldehyde gel and electroblotted to a non-charged nylon membrane (BDH). Prehybridisation at 68.degree. C. was performed for 6 hours in hybridisation buffer (50% formamide, 5X SSC, 2% blocking buffer (Boehringer Mannheim ), 0.1% laurolylsarcosine, 0.02% SDS). Hybridisation was overnight at 68.degree. C. in fresh hybridisation buffer with a digoxigenin-UTP labelled riboprobe (100 ng/ml) corresponding to the entire hP.sub.2X sequence. The membrane was washed at 68.degree. C.; twice in 2X SSC+0.1% SDS, and twice in 0.1X SSC+0.1% SDS. Chemiluminescent detection of hybridisation was carried at room temperature as follows: the membrane was rinsed 5 min in buffer B1 (0.1 M maleic acid, 0.15 M NaCl, pH 7.5), saturated for 1 hour in 1% blocking buffer (B2), incubated 30 min with anti-digoxigenin-antibody alkaline phosphatase conjugated (750 u/ml, Boehringer Mannheim) diluted 1:15000 in B2, washed in B1+0.3% tween 20 (1X 5 min, 1X 15 min, 1X 1 h), equilibrated for 5 min in buffer B3 (0.1 M Tris HCl pH 9.5, 0.1 M NaCl, 50 mM MgCl.sub.2), incubated 45-60 sec in lumigen PPD (Boehringer Mannheim) diluted 1:100 in B3. The humid membrane was sealed in a plastic bag, incubated 15 min at 37.degree. C., and exposed 15 to 20 min to Hyperfilm-ECL (Amersham).

P.sub.2X expression into oocytes Human urinary bladder P.sub.2X cDNA, subcloned into the pBKCMV expression vector, was linearized with Notl, and transcribed in vitro with T3 polymerase in the presence of cap analogue m7G (5')ppp(5')G. Defolliculated Xenopus oocytes (Bertrand et al, Electrophysiology of neuronal nicotinic acetylcholine receptors expressed in Xenopus oocytes following nuclear injection of genes or cDNAs Meth Neurosci 4: 174-193 (1991)) were injected with 50 ng of human P.sub.2X in vitro transcribed RNA, and incubated at 18.degree. C. for 2-6 days in the ND96 solution (mM); NaCl96, KCl2, MgCl.sub.2 1, CaCl.sub.2 2, sodium pyruvate 5, HEPES 5, ph 7.6-7.5, penicillin (10 U/ml), and streptomycin (10 .mu.g/ml).

Electrophysiology Oocytes were placed in a 1 ml chamber and superfused at 2-3 ml/min with ND96 solution with 0.1 mM BaCl.sub.2 replacing the 2 mM CaCl.sub.2 to prevent activation of endogenous calcium-activated chloride currents (Barish, A transient calcium-dependent chloride current in the immature Xenopus oocytes J Physiol 342: 309-325 (1983)). Currents were measured using a two-electrode voltage-clamp amplifier (Geneclamp Axon Instruments) at a holding potential of -60 mV. Microelectrodes were filled with 3 M KCl (0.5-2 M.OMEGA.). Data were collected using PClamp software (Axon Instruments). ATP and other purinoceptor agonists were applied by a U-tube perfusion system (Fenwick et al, A patch clamp study of bovine chromaffin cells and their sensitivity to acetylcholine J Physiol 331: 577-597 (1982)) placed close (200-500 .mu.m) to the oocyte. Initial studies showed that reproducible responses (<10% variation in peak amplitude) could be obtained when ATP (at concentrations up to 1 mM) was applied to hP.sub.2X injected oocytes for 5 s every 10 mins. Concentration response relationships to ATP and its analogs were determined by measuring the peak amplitude of responses to a 5 s application of agonist applied at 10 min intervals. Responses to agonists were normalized in each oocyte to the peak response evoked by 100 .mu.M ATP; 100 .mu.M ATP was usually applied at the beginning and at the end of an experiment to determine if there was any rundown of the response. No inward current was recorded in uninjected oocytes in response to application of purinoceptor agonists at the maximal concentration used (n=3 for each agonist). Antagonists were applied both in the superfusate and together with ATP in the U-tube solution. Antagonists were superfused for 5-10 min prior to the application of ATP.

Data analysis Concentration response curves for purinoceptor agonists were fitted with a Hill slope of 1. Equi-effective concentrations i.e. concentration of agonist, giving 50% of the response to 100 .mu.M ATP, (EEC.sub.50) were determined from individual concentration response curves. For antagonists the concentration required to give 50% inhibition (IC50) of the response to 10 .mu.M ATP (approximately 90% of peak response to ATP) were determined. Data are presented throughout as mean.+-.SEM for a given number of oocytes.

Drugs Adenosine, adenosine 5'-monophosphate sodium salt (AMP), adenosine 5'-diphosphate sodium salt (ADP), adenosine 5'-triphosphate magnesium salt (ATP), adenosine 5'-O-(-3-thiophosphate) tetralithium salt (ATP-.gamma.-S), uridine 5'-triphosphate sodium salt (UTP), .alpha.,.beta.-methylene ATP lithium salt (.alpha.,.beta., -meATP), .beta.,.gamma.-methylene-D-ATP sodium salt (D-.beta.,.gamma.-meATP), 2'-3'-O-(4-benzoylbenzol)ATP tetraethylamonium salt (BzATP), 4,4'-diisothiocyanatostilbene 2,2'-disulphonic acid, disodium salt (DIDS) were obtained from Sigma. 2-MethylthioATP tetra sodium salt (2MeSATP) , 2-chloro-ATP tetra sodium salt, and .beta.-.gamma.-methylene-l-ATP (l-.beta.-.gamma.-meATP) were obtained from RB1. Pyridoxal 5-phosphate monohydrate (Aldrich), p1, p5-di[adenosine-5']pentaphosphate trilithium salt (AP5A) (Boehringer Mannheim), pyridoxal phosphate 6-azophenyl 2',4'-disulphonic acid (PPADS, gift of G. Lambrecht, University of Frankfurt) and suramin (Bayer) were tested. Drugs were prepared from frozen aliquots of stock solutions and diluted to give the required final concentration.

Example 6

Sequence and characteristics of hP.sub.2X from urinary bladder

Isolation of human P.sub.2X cDNA Human urinary bladder tissue was obtained from a cystectomy for a bladder tumor. The patient showed no symptoms of bladder instability or urodynamic abnormalities. Only those portions, surrounding the tumor, which appeared macroscopically normal (Palea et al--supra) were used. Total RNA was isolated by guanidinium isothiocyanate and poly A.sup.+ RNA was purified as described (Valera et al (1994)--supra). Preparation of a cDNA library in .lambda.gt10, random primer labelling of a rat smooth muscle P.sub.2X probe (Valera et al (1994)--supra), low stringency hybridisation screening and lambda phage DNA isolation were all done by standard protocols (Sambrook et al, Molecular Cloning, A Laboratory Manual, 2nd edn., Cold Spring Harbor Laboratory Press, New York (1989)). Several independent phage isolates were examined and the cDNA insert from one was chosen for subcloning into Eco RI-Not I digested pBKCMV. This 2677 bp hP.sub.2X cDNA was sequenced as described (Valera et al (1994)--supra).

The 2677 bp cDNA, hP.sub.2X, contained a single long open reading frame which corresponds to a protein of 399 amino acids (FIG. 4). This amino acid sequence is highly homologous with that of the P.sub.2X receptor, isolated from rat vas deferens (89% identity). There are two regions of hydrophobicity near either end of the protein which are sufficiently long to traverse the membrane but there is no hydrophobic N-terminal leader sequence. All five potential sites for glycosylation and all ten cysteine residues in the central section of the protein are conserved. In vitro translation of hP.sub.2X RNA in the presence of microsomes produced a 60 kD product, whereas translation in the absence of microsomes produced the 45 kD peptide (FIG. 6). 45 kD is the computed molecular weight, suggesting that the additional 15 kD results from glycosylation.

Some human urinary bladder P.sub.2X cDNA was used to transfect HEK293 cells. Stable transfection was confirmed by electrophysiological recording.

Example 7

Distribution of human urinary bladder P.sub.2X mRNA

The distribution of the human urinary bladder P.sub.2X mRNA was examined by northern analysis. A single 2.6 kb mRNA species was observed in bladder, placenta, liver and adrenal gland (FIG. 8). In thymus, spleen, and lung samples, the 2.6 kb band plus additional higher molecular weight RNAs of 3.6 and 4.2 kb were seen. A smaller additional RNA species of 1.8 kb was observed in spleen and lung. No hybridisation was detected with brain mRNA.

Example 8

Induction of hP.sub.2X mRNA in HL60 cells

A portion of the 3'-untranslated region had been previously deposited in the database (HSGS01701) as an expressed sequence tag for the differentiation of the human promyelocytic cell line, HL60 (Okubo unpublished). We examined the induction of hP.sub.2X mRNA in HL60 cells by Northern blot analysis (FIG. 7). HL60 cells can be differentiated into distinct lineages, depending on the inductant (Koeffler, Induction of Differentiation of Human Acute Myelogenous Leukemia Cells: Therapeutic Implications Blood 62: 709-721 (1983)). Induction of macrophage-like characteristics with phorbol diesters or granulocytic differentiation with DMSO or dibutryl cAMP, each produced an increase in P.sub.2X mRNA (FIG. 7, lane 6), HL60 RNA (lane 1-5) showed hybridisation of two bands (1.8 and 2.6 kb) and both of these were inducible. This contrasts with the bladder, where Northern analysis showed only a single RNA species (2.6 kb) (FIG. 7, lane 6).

Example 9

Pharmacological characterization of hP.sub.2X

Application of ATP (30 nM-1 mM) to oocytes injected with hP.sub.2X receptor RNA evoked inward currents (FIGS. 9, 10 and 11). Responses to low concentrations of ATP (30-300 nM) developed over 3-5 s. Higher concentrations of ATP (1 .mu.M) evoked responses which peaked within 1-1.5 s and then declined during the continued application of ATP (40-60% of the peak amplitude after 5 s). The current returned to control values on washout of ATP. The peak amplitude of the inward current evoked by ATP was concentration-dependent (FIGS. 9, 10 and 11) and could be fitted by a curve with a Hill slope of 1 with a EC.sub.50 of 0.82 .mu.M. When ATP (100 .mu.M) was applied for 5 s every 10 min, reproducible inward currents were recorded. This is in contrast to the responses of the P.sub.2X receptor clone from rat vas deferens where a second application of ATP (>1 .mu.M) applied 10 mins after the first, evoked an inward current that was .about.50% of the initial peak amplitude.

Concentration-response curves were constructed for a number of other P2 purinoceptor agonists (FIGS. 9, 10 and 11). 2meSATP, 2-chloro-ATP, .alpha.,.beta.-meATP and ADP were full agonists. BzATP, AP.sub.5 A and ATP-.gamma.-S produced maximal responses of about 65% of the maximal ATP response. The maximal responses to d and l-.beta.,.gamma.-meATP were not determined. Adenosine, AMP and UTP (100 .mu.M) evoked small inward currents (2.3.+-.1.5, 6.08.+-.2, and 3.7.+-.1.8% of the response to 100 .mu.M ATP respectively). The EEC.sub.50 values and relative potencies of purinoceptor analogs are summarised in Table 1 below.

TABLE 1 relative agonist EEC50 (.mu.M) potency ATP 0.82 1 2MeSATP 0.6 .+-. 0.1 1.36 2chloroATP 0.76 .+-. 0.1 1.08 AP5A 2 .+-. 0.2 0.41 .alpha.,.beta.-meATP 3.6 .+-. 1.6 0.23 BzATP 4.2 .+-. 2.2 0.20 ATP-.gamma.-S 10.6 .+-. 3.8 0.077 d,.beta.,.gamma.-meATP 24.1 .+-. 1.6 0.034 ADP 34.3 .+-. 16 0.024 EEC50: Equi-effective concentrations producing an inward current equivalent to 50% of the peak response to 100 .mu.M ATP. EECSO taken from individual fitted concentration response curves with a Hill slope of 1. EEC50 for ATP from mean data from all experiments. (n = 3-4).

Example 10

Antagonist studies

The P2-purinoceptor antagonist suramin (1-100 .mu.M) shifted the concentration-response curve for ATP to the right. At 1 .mu.M suramin the shift was almost parallel. The dissociation equilibrium constant (K.sub.B) estimated from K.sub.B =1/(DR-1) where DR is the dose ratio was 130 nM. With higher concentrations of suramin the inhibition did not appear to be competitive. Under the present experimental conditions this K.sub.B estimate is higher than those reported previously for suramin (pA2 5.9, Trezise et al, Br J Pharmacol 112: 282-288 (1994)) pK.sub.B 5.2, von Kugelgen et al, Interaction of adenine nucleotides, UTP and suramin in mouse vas deferens: suramin-sensitive and suramin-insensitive components in the contractile effect of ATP Naunyn Schmiedeberg's Arch Pharmacol 342: 198-205 (1990)). The antagonism by suramin was fully reversed after 10 mins wash and indicates that the non-competitive antagonism at high concentrations is not due to irreversible binding of the antagonist to the receptor.

The putative P.sub.2X purinoceptor antagonists PPADS, DIDS and pyridoxal 5 phosphate (Ziganshin et al, Selective antagonism by PPADS at P.sub.2X purinoceptors in rabbit isolated blood vessels Br J Pharmacol 111: 923-929 (1994), Bultmann & Starke, Blockade by 4,4'-diisothiocyanatostilben-2,2'-disulphonate (DIDS) of P.sub.2X purinoceptors in rat vas deferens Br J Pharmacol 112: 690-694 (1994), Trezise et al, Eur J Pharmacol 259: 295-300 (1994)) inhibited inward currents evoked by 10 .mu.M ATP (approximately EC.sub.90 concentration) in a concentration dependent manner (FIGS. 12 and 13). Suramin PPADS and DIDS were equally effective in inhibiting ATP evoked currents (IC.sub.50 .about.1 .mu.M). The IC 50 for P5P was .about.20 .mu.M. PPADS and P5P antagonism was readily reversible on washout. In contrast, inhibitory effects of DIDS (100 .mu.M) were very slow to reverse on washout.

(iii) Rat Superior Cervical Ganglion P.sub.2X Receptor

Example 11

Isolation and functional expression of a cDNA encoding a P.sub.2X receptor from rat superior cervical ganglion (referred to herein as clone 3)

A 440 bp fragment was amplified by polymerase chainreaction (PCR) from rat testis cDNA, using degenerate primers based on conserved nucleotide sequences within the rat vas deferens P.sub.2X receptor cDNA and on the sequence of PC12 cDNA (Ehrlich H A (ed) PCR Technology MacMillan, Basingstoke (1989)). The primers used are given below:

Sense (SEQ ID NO 2) 5'T G T/C G A A/G A/G T I T T/C I G G/C I T G G T G T/C C C 3' Antisense (SEQ ID NO 3) 5'G C A/G A A T/C C T A/G A A A/G T T A/G T/A A I C C 3'

(wherein I=Inosine and "T/C" indicates that either T or C is present at the position indicated (this applies mutatis mutandis to the other alternatives given).

The cloned PCR fragment was labelled and used as a hybridization probe for screening a rat testis cDNA bank in .lambda.ZAP. One recombinant phage was positive, and its insert was excised and transferred to a plasmid (#432). This cDNA was 1500 bp with a single EcoR1 site (at position 1000, still in the open reading frame). The 5' end of the cDNA was too short to encode the entire N terminus.

Internal primers specific to the new sequence were made and the tissue distribution was tested by PCR. The candidate was present in mRNA prepared from phaeochromocytoma (PC12) cells, intestine and superior cervical ganglion (scg). The hybridization probe was therefore used to screen a rat scg cDNA bank in .lambda.gt10. From 30 initial positives, 20 pure phage DNA stocks were prepared; 19 were various portions of the candidate sequence, and the insert from one was transferred to plasmid (p457) and sequenced. The insert appeared to be a full length cDNA; it has a single open reading frame of 388 amino acids (FIG. 2). The insert from p457 was subcloned into pcDNA3 (p464) and used to transfect human embryonic kidney (HEK293) cells.

The functional characterisation of the clone illustrated in FIG. 2 (referred to herein as clone 3) was carried out by electrical recordings from transfected HEK293 cells and from oocytes injected with the in vitro transcribed RNA, as described in Example 4 for the rat vas deferens P.sub.2X receptor. Table A summarizes the main properties of clone 3 as compared to those of rat vas/human bladder cDNA clone, and the PC12 cDNA clone (provided by David Julius and Tony Brake of the University of California at San Francisco).

TABLE A Functional Properties of 3 cloned P.sub.2X Receptors bladder clone 3 PC12 kinetics desensitization very strong very little very little rundown profound very little very little ionic permeability monovalent no no no differences differences differences divalent (Ca.sup.++) high high high permeability permeability permeability Ca.sup.++ block none intermediate very strong agonist profile ATP 0.7 .mu.M 11 .mu.M 8 .mu.M .alpha.,.beta.-meATP 3 .mu.M >>100 mM >>100 .mu.M antagonist profile suramin 1 .mu.M <40% block 6 .mu.M PPADS 1 .mu.M <30% block 1 .mu.M P-5-P 6 .mu.M <40% block 6 .mu.M DIDS 1 .mu.M >100 .mu.M

The main functional properties of clone 3 are as follows. (a) The currents evoked by ATP show little or no decline during applications of several seconds; that is, there is little desensitisation (FIG. 14). (b) The relative permeabilities of the ionic pore to sodium, potassium, cesium, tetraethylammonium and to calcium are not different to those observed for the rat vas deferens/human bladder or the PC12 forms of the receptor. (c) Extracellular calcium (30 mM) inhibits the inward current through the P.sub.2X receptor channel of the PC12 form whereas it does not block current through the rat vas deferens/human bladder form; clone 3 is intermediate in sensitivity. (d) The effectiveness of agonists that are structurally related to ATP is the same as that found for the PC12 form; most notably, .alpha..beta.methylene ATP has little or no agonist action (FIG. 14). (e) Currents activated by ATP at the clone 3 receptor were much less sensitive to antagonism by suramin., pyridoxal 5'-phosphate and pyridoxal-6-azophenyl-2',4'-disulphonic acid (PPADS) than were similar current mediated by the other two forms (rat vas deferens/human bladder; PC12) (FIG. 15).

(iv) Rat Dorsal Root Ganglion P.sub.2X Receptor

Example 12

Isolation of a cDNA encoding a P.sub.2X receptor from a rat dorsal root ganglion

By using PCR with the same primers as used in Example 11 above, but using different cDNA sources, further P.sub.2X family members can be found.

Using this method, rat dorsal root ganglion P.sub.2X receptor cDNA was isolated. FIG. 1B shows the cDNA sequence of this clone (referred to herein as clone 6), together with the putative amino acid sequence. The portions underlined in this figure correspond to the PCR primers initially used.

A similar procedure to that described in Example 11 was then used to isolate the full length cDNA.