|1.||DNA encoding a functional serotonin 5HTlc recep¬ tor.|
|2.||cDNA of claim 1.|
|3.||cDNA of claim 2 having the nucleic acid sequence shown in Figure 1.|
|4.||Isolated, functional serotonin 5HTlc receptor encoded by the JDNA of claim 1.|
|5.||Serotonin 5HTlc receptor of claim 4 having the amino acid sequence shown in Figure 1.|
|6.||A plasmid comprising the DNA of claim 1.|
|7.||The plasmid of claim 6 designated pSRlc and deposited under ATCC Accession No. 67636.|
|8.||A plasmid adapted for expression in a mammalian cell which comprises the cDNA of claim 2 and the regulatory elements necessary for expression of the cDNA in the mammalian cell.|
|9.||The plasmid of claim 8 designated pMV7347.|
|10.||A mammalian cell comprising the DNA of claim 1.|
|11.||A mammalian cell comprising the cDNA of claim 2.|
|12.||A mammalian cell comprising the plasmid of claim 8.|
|13.||A transfected NIH3T3 cell comprising the plasmid of claim 9.|
|14.||The transfected NIH3T3 cell designated SR3T3 and deposited under ATCC Accession No. CRL 9651.|
|15.||A method for determining whether a ligand is capa¬ ble in vivo of binding to the serotonin 5HTlc receptor which comprises contacting a mammalian cell of claim 12 with the ligand under conditions associated with in vivo binding of ligands to the serotonin 5HTlc receptor, detecting the presence of any of the ligand bound to the serotonin 5HTlc receptor and thereby determining whether the li¬ gand binds to the serotonin 5HTlc receptor.|
|16.||A method of claim 15, wherein the mammalian cell is SR3T3.|
|17.||A method of detecting the expression of the sero¬ tonin 5HTlc receptor on the surface of a cell which comprises obtaining total mRNA from the cell and contacting the mRNA so obtained with the cDNA of claim 2 under hybridizing conditions, detecting the presence of mRNA hybridized to the cDNA, and thereby detecting the expression of the serotonin 5HTlc receptor by the cell.|
|18.||A DNA probe useful for detecting nucleic acid encoding the serotonin 5HTlc receptor comprising a nucleic acid molecule of at least about 15 nucleotides having a sequence complementary to a sequence included within the sequence shown in Figure 1.|
|19.||An antibody directed to the serotonin 5HTlc recep¬ tor.|
|20.||A monoclonal antibody directed to an epitope of the serotonin 5HTlc receptor present on the sur¬ face of a cell and having an amino acid sequence included within the amino acid sequence shown in Figure l.|
|21.||A method of detecting the presence of the sero¬ tonin 5HTlc receptor on the surface of a cell which comprises contacting the cell with a monoclonal antibody of claim 20 under conditions permitting binding of the monoclonal antibody to the receptor, detecting the presence of the monoclonal antibody bound to the cell, and thereby the presence of the serotonin 5HTlc receptor on the surface of the cell.|
|22.||A method of screening drugs to identify drugs which specifically interact with, and bind to, the serotonin 5HTlc receptor on the surface of a cell which comprises contacting the mammalian cell of claim 12 with a plurality of drugs, determining those drugs which bind to the mammalian cell, and thereby identifying drugs which specifically in¬ teract with, and bind to, the serotonin 5HTlc receptor.|
ISOLATED 5HT1C RECEPTOR, MAMMALIAN CELLS
EXPRESSING SAME AND USES THEREOF
This application is a continuation-in-part of U.S. Serial No. 162,654, filed February 29, 1988, the con¬ tents of which are hereby incorporated by reference.
Background Of The Invention
Throughout this application various publications are referenced by Arabic numerals within parentheses. Full citations for these references may be found at the end of the specification immediately preceding the claims. The disclosures of these publications in their entire¬ ties are hereby incorporated by reference in this ap¬ plication in order to more fully describe the state of the art to which this invention pertains.
Serotonin, 5-hydroxytryptamine (5HT) , is a biogenic amine that functions as a neurotransmitter (12), a hormone (13) , and a mitogen (14) . Serotonin modulates many forms of synaptic transmission and is believed to exert a number of effects on the growth of neurons in early development. In the spinal cord, serotonin is involved in the inhibitory control of sensory input and in the facilitation of motor output (15, 16). In the cortex, transmission at serotonergic synapses contrib- utes to affective and perceptual states, and these synapses represent a major site of action of psycho- tropic drugs such as LSD (17) . Serotonergic neurons project to diffuse regions of the brain and exert their physiological effects by binding to cell surface recep-
tors. To date, six serotonin receptor subtypes (5HTla- Id, 2 and 3) (previously designated 5HT-1A-1D, 2 and 3) have been defined on the basis of their pharma¬ cological properties (18) .
Individual receptor subtypes reveal characteristic differences in their abilities to bind a number of ligands, but the structural basis for the distinct ligand-binding properties is not known. Physiologists and pharmacologists have attempted to specify particu¬ lar biological functions or anatomical locations for some receptor subtypes, but this has met with limited success.
Similarly, the biochemical mechanisms by which these receptors transduce signals across the cell surface have been difficult to ascertain without having well- defined cell populations which express exclusively one receptor sub-type. Serotonin receptor subtypes couple to different intracellular second messenger signaling systems, including the regulation o adenylate cyclase activity (5HTla and 5HTlb) (19, 20, 21) and phospho- lipase C activities (5HTlc and 5HT2) (22, 23). The activation of these second messenger pathways by sero¬ tonin modulates the excitable properties of both cen¬ tral and peripheral neurons (24, 25, 26, 27). Sero¬ tonin receptors are also thought to be linked to the direct modification of ion channel states, and are implicated in mechanisms associated with pain, migraine headaches, and motor control. Moreover, drugs which bind to the serotonin receptor may be useful in treat¬ ing depression. One difficulty which this involves is the prior difficulty in examining a specific interac¬ tion of a drug with the serotonin 5HTlc receptor alone.
The methods provided by this invention provide a simple and qualitative assay to assess this interaction and eliminate the lack of specificity associated with the prior art use of tissue preparations as a semi-defined source of serotonin receptors. The expression of functional receptors in Xenopus oocytes has provided a sensitive assay for detection of mRNA encoding serotonin receptors, in particular the 5HTlc receptor, that couples via inositol phospholipid signaling systems (5, ,6, 28). This invention differs from the closest prior art in that NIH 3T3-SR cells can be obtained in any quantity desired, and provides the investigator with a source of receptors which is consistent in its molecular characteristics. Furthermore, this cell line also provides the investigator with a cellular environment in which ligand-receptor interactions can be measured using a simple spectrofluorimetric assay.
Although it has recently been reported that a cDNA fragment encoding the serotonin 5HTlc receptor has been cloned, this fragment does not encode and is not useful in producing functional serotonin 5HTlc receptors. The cDNA clone encodes only the carboxyl-terminal portion of the 5HTlc receptor and was isolated by hybrid-deple- tion of choroid plexus mRNA coupled with oocyte expres¬ sion (29) . Applicants, in the present invention, have combined cloning in RNA expression vectors with an electrophysiological assay in oocytes to isolate and characterize the expression of a functional cDNA clone encoding the entire 5HTlσ receptor.
Summary of the Invention
The present invention provides DNA encoding a function¬ al serotonin 5HTlc receptor.
The invention also provides a plasmid comprising the DNA encoding a functional serotonin 5HTlc receptor which is designated pSR-lc and deposited under ATCC Ac¬ cession No. 67636.
Additionally, the present invention provides a plasmid adapted for expressiςn in a mammalian cell which com¬ prises the cDNA encoding a functional serotonin 5HTlc receptor and the regulatory elements necessary for expression of the cDNA in the mammalian cell.
The present invention further provides the transfected NIH3T3 cell designated SR3T3 and deposited under ATCC Accession No. CRL 9651.
In addition, the invention provides a DNA probe useful for detecting nucleic acid encoding the serotonin 5HTlc receptor comprising a nucleic acid molecule of at least about 15 nucleotides having a sequence complementary to a sequence included within the sequence shown in Figure 1.
The invention herein also concerns an antibody directed to the serotonin 5HTlc receptor.
This invention additionally concerns a monoclonal anti¬ body directed to an epitope of the serotonin 5HTlc receptor present on the surface of a cell and having an amino acid sequence included within the amino acid sequence shown in Figure 1.
Brief Description of the Figures
Figure 1. Nucleotide - sequence and deduced amino acid sequence of the rat 5HTlc receptor. The sequenc¬ ing protocol is shown diagramatically. Sequences were determined for individual restriction fragments after subcloning into M13 vectors or with synthetic oligonu¬ cleotides as internal primers. The coding region is indicated by the heavy bar. Restriction sites are shown as indicated (E = EcoRI, H = Hind III, S = Sal I, Sa = Sac I) . Numbers in the left-hand margin indicate nucleotide position. DNA sequence of cDNA clone pSR-lc was determined by the chain termination method of Sanger et al. (1) .
Figure 2. Table of codon usage in 5HTlc DNA.
Figure 3. Cloning strategy for isolation of a functional 5HTlc cDNA clone.
A. Choroid plexus total RNA was fractionat¬ ed by sucrose gradient sedimentation to enrich for serotonin receptor mRNA. 50 μg of choroid plexus total RNA was centrifuged through a 5%-25% (w/v) sucrose gradient as described in Experimental Details. Col- lected fractions were assayed for absorbance at 260 nm (o-o) and for serotonin receptor activity by voltage clamp analysis. Fraction 1 corresponds to the top of the gradient (smaller RNA's) , and fraction 25 to the bottom (larger RNA's) . Positive fractions 19 and 20 (stippled bar) were identified by voltage-clamp record¬ ing of injected oocytes.
B. RNA from these fractions was used to construct a cDNA library in the bacteriophage expres-
sion vector λ ZAP in which cDNA inserts (hatched rect¬ angle) are flanked by promoters for T3 and T7 RNA polymerases (black boxes) . DNA derived from pools of clones was digested with the restriction endonuclease Not I, cleaving the DNA at the position shown.
C. These truncated DNA templates were tran¬ scribed in vitro with T7 RNA polymerase in the presence of the cap precursor GpppG to produce functional RNA copies of the cDNA inserts.
D. Xenopus oocytes were injected with this RNA, cultured for three days, and assayed for sensitiv¬ ity to serotonin by voltage-clamp recording of 5HT- activated currents.
E. A pool of cDNA clones giving a positive response was identified and progressively subdivided into smaller pools (sib selection) until a single posi¬ tive clone was obtained.
Figure 4. Expression of functional serotonin re¬ ceptors in Xenopus oocytes after injection of choroid plexus poly A+ RNA.
A. Voltage clamp recording from a Xenopus oocyte injected with 0.5 ng poly A+ RNA (10 dilution) isolated from rat choroid plexus. The oocyte membrane
—6 potential was maintained at -50 mV, and serotonin (10 M) was applied by superfusion for the duration indicat¬ ed.
B. Current responses evoked by application of serotonin (10~ M) in oocytes injected with progres¬ sive dilutions of rat choroid plexus poly A+ RNA. The
RNA stock was at an initial concentration of 1 ng/nl. Each oocyte was injected with 50 nl of diluted RNA, incubated for 3 days, and voltage-clamped at -50 mV. Detectable serotonin-induced membrane currents were obtained reliably with dilutions of up to 1 in 10 poly A+ RNA. Plots indicate mean and standard error of serotonin-evoked currents obtained from 2-6 oocytes.
Figure 5. Characteristic responses to serotonin by
Xenopus oocytes injected with natural or synthetic RNA, TTyyppiiccaall rreessppoonnssee ttoo 1100~~ M serotonin of a choroid plex- us mRNA-injected oocyte,
Figure 6. Response to 10~ M serotonin of an oocyte injected with RNA prepared in vitro from a pool of 100,000 cDNA clones.
Figure 7. Response to 10~ M serotonin of an oocyte injected with RNA prepared in vitro from a sin¬ gle cDNA clone (Z347) .
Figure 8. Injection of RNA from pSR-lc cDNA clone confers serotonin sensitivity in Xenopus oocytes. Transcription of RNA was performed as described in Experimental Details, and dilutions of an RNA stock (1 mg/ml) were injected into Xenopus oocytes. Voltage clamp analysis was performed 24-48 hours after injec¬ tion and the mean serotonin-induced current plotted as a function of progressive dilution of receptor RNA. Serotonin-induced membrane current could be detected reliably at dilutions of 1 in 10 of transcribed RNA (10 ng) . For further details see Experimental Details.
Figure 9. Serotonin sensitivity of Xenopus oocytes injected with RNA transcribed in vitro from λ ZAP cDNA
A. Inward current evoked by serotonin (10
M) in oocytes injected with RNA transcribed from a pool
5 of 10 λ ZAP cDNA clones (approximately 100 ng) (top c trace) and from a single λ ZAP cDNA clone (pSR-lc) at 10 —4 dilution (5 pg) (bottom trace) . Both oocytes were held at -50 mV.
B„ Serotonin-evoked currents after serial 0 dilution of RNA transcribed in vitro from the λ ZAP cDNA clone pSR-lc.. The undiluted RNA was at a concen¬ tration of 1 ng/nl. Each oocyte was injected with 50 nl of diluted RNA. Cells were incubated for 3 days and voltage-clamped at -50 mV. Each point represents the *> mean and standard error of serotonin-evoked currents obtained from 3-7 oocytes.
Figure 10. A model for the transmembrane topology of the rat 5HTlc receptor. The receptor is shown as 0 containing seven hydrophobic transmembrane regions. By analogy with the structure of rhodopsin, the amino terminus is located on the extracellular face of the lipid bilayer and the carboxyl terminus is located on the cytoplasmic side. Asterisks indicate serine resi¬ dues that represent potential phosphorylation sites. The star indicates an asparagine residue in the amino terminal region that is a potential N-glycosylation site.
Figure 11. Alignment of the amino acid sequence of the rat 5HTlc receptor (top) , hamster 3 2 -adrenergic receptor (BAR, second row) (64) , bovine substance K receptor (SK, third row) (30) ; human a- adrenergic receptor (A2AR, fourth row) (65) ; rat uscarinic M3
receptor (M3, fifth row) (66); and bovine rhodopsin (RHOD, sixth row) (42) . The amino acid residues en¬ closed by solid lines represent residues that are iden¬ tical in at least 2 of the 5 reference sequences when compared to the 5HTlc receptor. Sequences present in the highly variable third cytoplasmic loop are not included, but the number of residues present in this loop is indicated in parentheses. The first 25 amino acids of the rat M3 sequence have not been included here. Roman numerals and brackets denote the seven
10 putative transmembrane domains.
Figure 12. Expression of 125I-LSD binding sites on
SR3T3 cells and the pharmacological profile of ligand specificity. Membranes were prepared from control
15 NIH3T3 cells and from SR3T3 cells expressing pSR-lc cDNA inserted into the retroviral expression vector pMV7 (2). 125 I-LSD (lnM) was incubated with 0.2 mg membrane protein aliquots suspended in Tris-HCl buff¬ er. No specific binding was observed with untrans-
20 fected control NIH3T3 cells, determined by addition of 10 —5 unlabeled serotonin. In contrast, SR3T3 membranes exhibited 60-70% specific 125I-LSD binding, when deter- mined by addition of unlabeled serotonin (10 —5M) or by comparison with untransfected 3T3 cells. The IC_ n s for nςf 125 inhibition of I-LSD binding by mianserin and spi- perone were also determined by addition of unlabeled drugs for 5 minutes prior to and during addition of
125 I-LSD. Each point represents the mean determina- tions. Similar results were obtained in two separate
Figure 13. Activation of 5HTlc receptors in trans- formed fibroblasts elevates intracellular Ca 2+ concen¬ tration. Untransformed cells (NIH3T3) or transformed
cells expressing pSR-lc cDNA (SR3T3) were loaded with 2+ the Ca -sensitive dye indo-1. Changes in the level of intracellular free-Ca 2+ following exposure to 5 μm serotonin (hatched peak) were monitored with a flow cytometer to measure the ratio of emissions at 400 and 2+
500 nm. In each case, the resting intracellular Ca concentration in the absence of serotonin is also shown (open peak) . For NIH3T3 cells, the two peaks are coin¬ cidento For SR3T3 cells, 95% of the population showed an elevation in intracellular Ca 2+ in the presence of serotonin. The mean resting Ca concentration in
NIH3T3 and SR3T3 cells was similar.
Figures 14 and 15. RNA blot analysis of the tissue distribution of 5HTlc receptor mRNA. Poly A+ RNA (2- 20 μg) prepared from different brain regions (Figure 14) and peripheral tissues (Figure 15) (7, 34) was subject¬ ed to electrophoresis through an 0.8% agarose-formalde- hyde gel, blotted onto Gene Screen (New England Nucle- ar) and hybridized with " a 32P-labeled probe prepared from the 3 kb EcoRI cDNA insert from pSR-lc. Tissue regions are indicated above each lane. Filters were again hybridized with a 32P-labeled human α-actin probe to assess relative amounts of RNA loaded in each lane. The 5HTlc mRNA was judged to be approximately 5.2 kb, using 18S and 28S ribosomal RNA as standards.
Detailed Description of the Invention
This invention relates to the cloning and isolation of DNA encoding a functional 5HTlc receptor. Although the specific DNA molecule which has been cloned is of rat origin, those skilled in the art will readily appreci¬ ate that such cloned rat DNA may be used as a hybrid¬ ization probe to recover DNA encoding the serotonin 5HTlc receptor from other sources including specifical¬ ly other mammalian sources, and most particularly human sources.
In one embodiment this invention provides cDNA encoding a functional serotonin 5HTlc receptor, e.g., the cDNA having the nucleic acid sequence shown in Figure 1.
In another embodiment this invention provides for the first time a means for obtaining isolated, functional serotonin 5HTlc receptor by expressing DNA, such as cDNA encoding the receptor in a suitable host, such as a bacterial, yeast, or mammalian host cell using meth¬ ods well known in the art and recovering the functional serotonin 5HTlc receptor after it has been expressed in such a host, again using methods well known in the art.
In a further embodiment this invention provides cloned DNA encoding serotonin 5HTlc receptor, typically cloned into a plasmid such as pBR322 or into a bacteriophage, such as λ bacteriophage. One example of such a plasmid is the plasmid pSR-lc described in greater detail hereinafter and deposited with the American Type Culture Collection (ATCC) in Escherichia coli under the designation DJ200 and under ATCC Accession No. 67636. This deposit and the other deposit made in connection with this invention were made pursuant to, and in sat-
isfaction of the provisions of the Budapest Treaty On The International Recognition of The Deposit of Micro¬ organisms For The Purposes of Patent Procedure and were made with the ATCC, 12301 Parklawn Drive, Rockville, Maryland 20852.
In a still further embodiment, this invention provides plasmids adapted for expression in a mammalian cell which comprises DNA, e.g., cDNA, encoding functional serotonin 5HTlc receptor and the regulatory elements necessary to express such DNA in the mammalian cell. Once again, those skilled in the art will readily ap¬ preciate that numerous plasmids may be constructed utilizing existing plasmids and adapted as appropriate to contain the regulatory elements necessary to express the DNA in the mammalian cell. In this regard numerous mammalian cells may be used including, for example, the mouse fibroblast cell NIH 3T3, CHO cells, HeLa cells, etc. One example of such a plasmid is the plasmid designated pMV7-347 described more fully hereinafter which was constructed from the plasmid pSR-lc already mentioned and the plasmid pMV7 described in detail by Maddon, et al. (2) .
Such expression plasmids may be used to transfect ma - malian cells or DNA encoding the serotonin 5HTlc recep¬ tor may be otherwise introduced into mammalian cells, e.g., by microinjection, to obtain mammalian cells which comprise DNA, e.g., cDNA or a plasmid, encoding the serotonin 5HTlc receptor. In one presently pre- ferred embodiment this invention provides an NIH3T3 cell transfected with the plasmid pMV7-347 which has been designated SR3T3 and deposited under ATCC Acces¬ sion No. CRL 9651.
This invention further provides a method for determin¬ ing whether a ligand, such as a known or putative drug, is capable in vivo of binding to functional serotonin 5HTlc receptor. This method comprises (a) contacting a mammalian cell, such as the SR3T3 cell, expressing functional serotonin 5HTlc receptor on its surface, with the ligand under conditions which are known to prevail, and thus to be associated with, in vivo bind¬ ing of the ligands to the serotonin 5HTlc receptor, (b) detecting the presence of any of the ligand being test¬ ed bound to the serotonin 5HTlc receptor on the surface of the cell, and (c thereby determining whether the ligand binds to the serotonin 5HTlc receptor.
This invention still further provides a method of de¬ tecting the expression of the serotonin 5HTlc receptor by a cell, particularly on the surface of the cell, such as a human brain cell, which comprises obtaining total mRNA from the cell using well known methods and contacting the mRNA so obtained with the DNA, e.g., cDNA, encoding the serotonin 5HTlc receptor under hy¬ bridizing conditions, detecting the presence of mRNA hybridized to the cDNA, and thereby detecting the ex¬ pression of the serotonin 5HTlc receptor by the cell.
This invention also provides a DNA probe useful for detecting in a sample, such as a sample of human brain cells, nucleic acid encoding the serotonin 5HTlc recep¬ tor. Such a probe comprises a nucleic acid molecule of at least about 15 nucleotides having a sequence comple¬ mentary to a sequence included within the sequence shown in Figure 1. Such nucleic acid probe technology is well known to those skilled in the art who will readily appreciate that such probes may vary greatly in length and may be labeled with a detectable label, such
as a radioisotope or fluorescent dye, to facilitate detection of the probe.
In yet another embodiment this invention provides an antibody directed to the serotonin 5HTlc receptor. Such an antibody may be serum-derived or monoclonal and may be prepared using methods well known in the art. For example, cells such as SR3T3 cells may be used as immunogens to raise such an antibody. Alternatively, synthetic peptides may be prepared using commercially available machines and the amino acid sequence shown in Figure 1. As a still further alternative, DNA, such as a cDNA whose sequence is shown in Figure 1 or. a frag¬ ment thereof, may be cloned and expressed and the re¬ sulting polypeptide recovered and used as an immunogen. in one embodiment the antibody is a monoclonal antibody directed to an epitope of the serotonin 5HTlc receptor present on the surface of a cell and having an amino acid sequence included within the amino acid sequence shown in Figure 1.
Still further this invention provides a method of de¬ tecting the presence of the serotonin 5HTlc receptor on. the surface of- a cell which comprises contacting the cell with a monoclonal or serum-based antibody directed to an exposed epitope on the receptor under conditions permitting binding of the antibody to the receptor, and detecting the presence of the antibody bound to the cell, and thereby the presence of the serotonin 5HTlc receptor on the surface of the cell. Such a method is useful in determining whether a given cell, particular¬ ly a given brain cell, is defective relative to the expression of functional 5HTlc receptor on the surface of the cell.
Finally this invention provides a method of screening drugs to identify drugs which specifically interact with, and bind to, the serotonin 5HTlc receptor on the surface of a cell. This method comprises contacting a mammalian cell which is expressing functional serotonin 5HTlc receptor with a plurality of drugs, known or putative, determining those drugs which bind to the mammalian cell, and thereby identifying drugs which specifically interact with, and bind to, the serotonin 5HTlc receptor.
As a first step in determining the structure and func¬ tion of serotonin receptors, a cDNA clone encoding one of these subtypes, lc, has been isolated. This partic¬ ular species was chosen for two reasons: (i) it is highly expressed in a particular brain subregion; and (ii) it couples to an intracellular signaling pathway compatible with its ability to be functionally ex¬ pressed in Xenopus laevis oocytes.
Autoradiographic analysis of brain sections using a variety of radioligands ( 3H-mesulergme, 3H-serotonin
125 and I-LSD) led to the discovery of the 5HTlc recep¬ tor and identified the choroid plexus as a site of high receptor density. This vascularized, non-neuronal tis¬ sue sits in the ventricles of the brain, where it pro¬ duces much of the cerebrospinal fluid. High resolu¬ tion autoradiography has localized 5HTlc receptors to the epithelial cell layer of the choroid. Quantitative binding studies have shown that the specific activity of 125 I _ LSD b i n ding sites is at least 10-fold higher in the rat choroid plexus than in any other tissue exam¬ ined. It has also been shown that serotonin can stimu¬ late the hydrolysis of phosphatidylinositol in the choroid plexus and that this effect is mediated through
its interaction with the 5HTlc receptor.
Lacking any biochemical or immunological tools with which to purify or easily detect serotonin receptors, a functional assay has been employed as the basis for identifying a cDNA clone encoding this receptor. Miledi and co-workers were the first to show the utili¬ ty of Xenopus laevis oocytes as an expression system for studying neurotransmitter receptors and ion chan¬ nels from mammalian brain (4, 5). The microinjection of brain poly A+ RNA into these cells leads to the appearance of functional receptors at the cell surface, and renders the oocyte responsive to a number of neuro- transmitters, including serotonin. Using two electrode voltage clamp techniques, this response can be detected elect ophysiologically as an inward membrane current carried by chloride ions. Detailed analysis of the mechanism involved has shown that the application of serotonin leads to the hydrolysis of phosphatidylino- sitol-4,5-biphosphate and the production of inositol triphosphate (IP-) . IP- then evokes the release of calcium from intracellular stores, leading to the open¬ ing of endogenous calcium-dependent chloride channels located in the oocyte plasma membrane. More recently, Lubbert et al. have shown that following injection of rat brain mRNA into oocytes the predominant serotonin receptor subtype detected is 5HTlc. They have also demonstrated that the rat choroid plexus is enriched for 5HTlc receptor message, and that electrophoretic fractionation of plexus RNA yields a partially purified preparation in the 5 to 6 kb size range which encodes an active receptor. Using a hybrid-depletion proce¬ dure, in combination with the oocyte assay, this group has isolated a partial 5HTlc. receptor cDNA clone from a mouse choroid plexus library (6) .
Specifically, this invention relates to the first iso¬ lation of a functional cDNA clone encoding the rat 5HTlc receptor, e.g., the rat 5HTlc receptor, using the oocyte expression assay to screen in vitro transcripts generated from a choroid plexus cDNA expression li¬ brary. The serotonin receptor is shown to belong to the family of rhodopsin-like signal transducers which are distinguished by their seven-transmembrane configu¬ ration and their functional linkage to G-proteins. A mammalian cell line expressing functional 5HTlc recep¬ tors at the cell surface has been constructed, as de¬ termined by pharmacologic and physiologic methods, thus establishing the first well-defined, cultured cell line with which to study a particular serotonin receptor sub-type. It has been found that the 5HTlc receptor is present throughout the central nervous system, im¬ plying that it may be involved in a variety of neuro- nal functions.
The invention will be better understood by reference to the Experimental Details which follow, but it will be readily appreciated by those skilled in the art that these examples are only illustrative of the invention as described more fully in the claims which follow thereafter.
The cloning of most neurotransmitter receptors has required the purification of receptor, the determina¬ tion of partial protein sequence and the synthesis of oligonucleotide probes with which to obtain cDNA or genomic clones. However, the serotonin receptors have not been purified and antibodies to the receptors have not been generated. Applicants have therefore designed a cDNA expression system that permits identification of functional cDNA clones encoding serotonin receptors in the absence of* protein sequence information. A similar approach has been used to isolate a cDNA encod¬ ing the bovine neuropeptide substance K receptor (30) .
Applicants' cloning strategy (Figure 3) is based on quantitative considerations of the following findings.
5 High levels of serotonin 5HTlc receptors (10 per cell) are expressed in the choroid plexus (3, 31), a non- neuronal cell type in the central nervous system asso¬ ciated with the production of cerebrospinal fluid. Xenopus oocytes injected with choroid plexus mRNA ex¬ hibit a serotonin-evoked inward current. Application of serotonin appears to liberate inositol phosphates that raise intracellular Ca 2+ levels, leading to the opening of Ca2+-dependent chloride channels (5, 6, 28) .
Patch clamp recording has demonstrated that the conduc¬ tance of a single serotonin-activated chloride channel is about 3 pS (28). Thus, the opening of 10 chloride channels will result in the generation of a readily- detectable current of 100 nA. Assuming that the trans¬ lational efficiency of the oocyte approximates that of a plexus cell, injection of RNA from ten plexus cells, or as little as 2 pg of poly A+ RNA, should lead to the expression of 10 serotonin receptors on the oocyte
surface. Parenthetically, the amplification of sig¬ naling usually associated with G protein coupled recep¬ tors (32, 33) suggests that considerably fewer than 10 receptors need be occupied to open 10 chloride channels. To determine experimentally the sensitivity of the oocyte expression assay, oocytes were injected with serial dilutions of poly A+ RNA isolated (7, 34) from the rat choroid plexus. Injection of as little as 5 pg of poly A+ RNA (or 5 fg of pure receptor mRNA, assuming 0.1% abundance) was sufficient to generate an inward current of about 100 nA (Figure 4) . Since it is possible to inject up to 50 ng of RNA into an oocyte, it should be possible to detect a cDNA clone encoding the 5HTlc receptor, even if present at an exceedingly low frequency in a cDNA expression library.
A cDNA library was therefore constructed from choroid plexus mRNA in λ ZAP, a vector permitting the in vitro transcription of cDNA inserts. RNA transcribed from populations of cDNA clones were screened for their ability to induce serotonin-evoked currents in Xenopus oocytes. A single clone encoding a functional sero¬ tonin 5HTlc receptor was purified from a large popula¬ tion of clones by procedures of sib selection (Figure 3).
Female, oocyte-positive Xenopus laevis frogs were ob¬ tained from Xenopus I and collagenase from Worthington.
A ZAP and T3 RNA polymerase were purchased from Strata- gene. T7 RNA polymerase and diguanosine triphosphate (GpppG) was purchased from Pharmacia and AMV reverse transcriptase was purchased from Life Sciences, Inc. All other reagents used for cDNA synthesis or in vitro
transcription were supplied by Pharmacia, New England Biolabs or Boeringer-Mannheim. Ultra-pure sucrose (RNAse free, for density gradients) was obtained from Schwarz/Mann, Inc.
Serotonin was purchased from Sigma, 125I-LSD from
Microinjection of Xenopus oocytes
Oocytes were surgically removed from frogs and separat¬ ed manually into small groups of 10 or 20. Cells were enzymatically dissociated by treatment in Barths medium containing 2% Ficoll (w/v) and 2 mg/ml collagenase, for 2 hours at room temperature with constant agitation. Oocytes were stored in Barths containing 2% Ficoll, penicillin and streptomycin, at 18 C. Cells were typi¬ cally injected with between 50 and 100 nl of sample. Microinjection pipettes were pulled from 25 μl capillar¬ ies (Drummond) using a vertical microelectrode puller. A Narishige manipulator, fitted with a simple Clay- Adams micropipette was used as the injection device. Oocytes were injected while under Barths with Ficoll and incubated at 18 C for 1-3 days before assaying.
Oocytes injected 1 to 3 days prior to recording were placed in a continuous-flow chamber perfused with frog Ringer solution. All drugs were introduced through this system by switching the perfusion line inlet. The membrane potential of the oocyte was voltage-clamped at -50 mV with two microelectrodes (2-5 megohm resis¬ tance; filled with 3M KC1) using an Axoclap 2A with virtual ground and a remote switchable headstage. Membrane currents were recorded through the virtual ground and filtered at 10 Hz. Responses were either
digitized at 5 Hz and stored on a PDP 11/73 microcom¬ puter or recorded on a Gould chart recorder.
RNA isolation and Northern blot analysis RNA was extracted from tissues by homogenization in guanidiniu and precipitation with LiCl, as described by Cathala et al. (7). Polyadenylated RNA was prepared by one cycle of binding to oligo(dT)-cellulose (Pharmacia, Type 7) .
Northern analysis of RNA samples was carried out ac¬ cording to Kru lauf et al. (8) .
Sucrose gradient fractionation of choroid plexus RNA Approximately 250 μg of total RNA, extracted from choroid plexus tissue obtained from 12 adult rats, was dissolved in 200 μl of 0.1% SDS, 5mM Tris-HCl (pH 7.3), 0.5mM EDTA and loaded onto a 12-ml linear sucrose gra¬ dient (35) [5-25% (w/v) sucrose containing 0.1M NaCl, lO M Tris-HCl (pH 7.3), 0.5% SDS, ImM EDTA]. The sam¬ ple was centrifuged in a Beckman SW 41Ti ultracentri- fuge rotor at- 21,000 rpm for 15.5 hours at 22*C. 0.45 ml fractions were collected manually from the top and the RNA recovered by ethanol precipitation. Each sam¬ ple was dissolved in 20 μl of water, 1 μl was removed for oocyte injection and the remainder was stored at - 80'C.
cDNA library construction
Positive fractions from the sucrose gradient were com¬ bined and the RNA concentrated by ethanol precipita¬ tion. The first strand of cDNA was synthesized using oligo-(dT).. ,, formation (Pharmacia) as a primer (36) and actinomycin D to reduce hairpin formation. The second strand was synthesized according to the method
of Okayama and Berg (9) . The double-stranded cDNA was methylated with EcoRI methylase and the ends flushed with T4 DNA polymerase. Phosphorylated EcoRI linkers (Pharmacia) were ligated to the cDNA and excess linkers were cleaved off with EcoRI (37) . cDNA was separated from free linkers by chromatography on Ultragel AcA 34 (LKB) (38) . cDNAs were inserted into the EcoRI site of λ ZAP (Stratagene) according to the instructions of the manufacturer. Recombinant bacteriophage was propa¬ gated using the bacterial strain supplied by Stratagene (BB4) and Bluescript plasmids were rescued from ZAP clones by M13 excision according to the instructions of the manufacturer.
Preparation of A DNA for in vitro transcription The cDNA library was amplified in pools of 20,000 clones. ' Recombinant phage were eluted from each plate (150mm) in 15 ml SM and kept separate. The majority of each phage stock (14 ml) was used to prepare DNA by a standard protocol (40) . Each DNA sample was treated with DNase-free RNase A (50 μg/ml, 37 β C, 30 min), fol¬ lowed by digestion with proteinase K (100 μg/ml, 37°C, 30 min) . After a series of phenol and chloroform-iso- amyl alcohol extractions, the DNA was concentrated by ethanol precipitation. When, in the course of sib se¬ lection, the pool size was reduced to 2000 clones, or smaller, phage stocks had to be re-amplified in order to obtain sufficient amounts of DNA for transcription.
For the initial screen, DNA derived from five pools of
5 20,000 clones was combined so as to represent 10 clones per sample.
Prior to transcription, approximately 20 μg of each DNA sample was digested to completion with the restriction endonuclease ' Not I, cleaving each recombinant molecule
downstream from the T7 promoter and after- the cDNA insert. This was followed by proteinase K digestion
(100 μg/ml, 37*C, 30 min.), organic (phenol-chloroform- isoamyl alcohol) extraction and ethanol precipitation.
DNA was dissolved in 5 μl TE.
In vitro transcription
Transcriptions were carried out in a 50 or 100 μl reac¬ tion volume at 40 ' C and contained 20 μg DNA template, 40mM Tris-HCl (pH 7.9), 7mM MgCl 2 , lOmM DTT, 2mM sper- midine, lOmM NaCl, 25 μg/ml BSA, 2000 units/ml human placental RNase inhibitor, 0.5mM ATP, CTP and UTP, 0.2mM GTP, ImM GpppG and 70 units T7 RNA polymerase. Incubation was for 1-2 hours. The reaction was termi¬ nated by extraction with phenol-chloroform mixture, followed with a chloroform extraction. Nucleic acid was recovered by ethanol precipitation and the pellet was dissolved in a small volume of water, usually 5 μl for microinjection into oocytes. A small amount of alpha- 32-p-UTP was included in the reaction cocktail to monitor the extent of RNA synthesis, which was deter¬ mined by TCA-precipitation of a small aliquot of the reaction products. Typically, between 3 and 10 μg of RNA was synthesized in these reactions, making the RNA/cDNA ratio roughly 5.
The DNA sequence of pSR-lc cDNA was determined by the chain termination method of Sanger et al. (1).
Transfection of NIH3T3 cells with pMV7-347 DNA was done by the standard method of calcium phosphate precipita¬ tion (10) . Transformants were selected in DME supple¬ mented with 10% calf serum (HyClone) and 1 mg/ml neomy- cin (G418; Difco) .
Flow cytometry analysis of serotonin-induced changes in intracellular Ca 2+ concentration Cells were removed from plates by treatment with Mg 2+ and Ca 2+-free Hepes buffered Hanks saline containing
0.5 mM EDTA.
Cells were loaded with Indo-1 in normal Hanks buffer supplemented with 0.1% BSA, 0.1% glucose and 10 μm Indo- l ester at 37*C for 5 minutes. Cells were washed in
5 Hanks and resuspended at 2.5 x 10 per ml, and analyzed using an Epics 753 flow cytometer. The intracellular
Indo-1 was excited with an Innova 90-5 argon ion lasar emitting 100 mW at 363 nm wavelength. The fluorescence emission was split with a 448 dichroic filter and de¬ tected with photomultipliers with 400.5 nm bandpass and 500 nm bandpass filters. Cells and serotonin (10 μM) were held in separate syringes and mixed in a "Y M connector just prior to entering the flow chamber. Further details of this method are given in Schieren and MacDermott (11) .
125 -I-LSD binding assays
Preparation of crude membranes and binding assays were carried out essentially as described by Yagaloff, et
125 . ' al. (3). I-LSD (2200 Ci/mmol; InM final concentra¬ tion) was added to approximately 0.2 mg of membrane protein aliquot in 1 ml 50mM Tris-HCl(pH 7.6), 10μM pargyline, 5mM EDTA, ImM ascorbate. 0.2 mg of crude
membrane protein derives from approximately 2 x 10 cells. Non-specific binding of 125I-LSD (not displace- able by 10 -7M mianseπn) to membranes prepared from untransfor ed NIH3T3 cells represented approximately
45% of that obtained with SR3T3 membranes. Mianserin displaceable binding was in the order of 50,000 cpm
125 I-LSD bound/mg protein. Samples were incubated, in the absence or presence of unlabeled competing drugs, at 37*C for 15 minutes. Samples were collected by vacuum filtration onto Whatman GF/B filter discs, washed three times with 10 mis each of ice-cold Tris buffer and counted in an LKB gamma counter.
pSR-lc DNA was linearized with NotI or Xhol endonu¬ clease and transcribed with T7 or T3 RNA polymerase to generate sense and anti-sense RNA probes, respective¬ ly. Transcription reactions (20 μl) contained 40 mM Tris-HCl (pH 7.5), 6 mM MgCl_, 2 mM spermidine, 10 mM NaCl, 10 mM DTT, 30 units placental RNase inhibitor, 0.5 mM each of ATP, CTP, and GTP, 250 μCi 35 S-UTP (1000 Ci/mmol; Amersham), and 2 μg linearized plasmid DNA and were incubated at 40'C, for 60 minutes. Template DNA was hydrolyzed with RQl DNase (Promega) according to the manufacturer. RNA probes were purified by two ammonium acetate-ethanol precipitations to give a final specific activity of 109 cpm/μg.
The fixation of adult rat brains and preparation of tissue sections was done according to published proce¬ dures (61) with the following exceptions: Hybridiza¬ tion buffer consisted of 50% formamide, 0.6 M NaCl, 10 mM Tris-HCl (pH 7.5), 0.02% Ficoll, 0.02% polyvinyl pyrolidone, 0.1% BSA, 1 mM EDTA, 10 μg/ml salmon sperm
DNA, 50 μg/ml yeast total RNA, 50 μg/ml yeast tRNA and 1 x 10 7 cpm/ml 35S-labeled probe. The final high-stnn-
gency wash contained 0.1X SSC, 0.05% sodium pyrophos- phate and 14 mM / 3-mercaptoethanol. Slides were dipped in Kodak NTB2 nuclear track emulsion (Eastman Kodak) , exposed at 4*C, and developed in Kodak D-19 developer.
Isolation of a functional serotonin receptor cDNA clone Oocytes were microinjected with serial dilutions of poly A+ RNA from the rat choroid plexus in order to estimate the sensitivity of the voltage clamp assay in measuring serotonin receptor activity (Figure 4) . The limit of detection is approximately 5 picograms, or the amount of mRNA equivalent to that which can be obtained from 20 mammalian cells. The ability to detect such low levels of receptor mRNA is a consequence of two factors, namely the signal amplification provided by the second messenger system, and the sensitivity of the voltage clamp method in measuring changes in membrane conductance.
Given the degree of sensitivity which can be achieved with the physiologic assay, it is possible to screen a choroid plexus cDNA library for a functional serotonin receptor clone by synthesizing capped RNAs in vitro and injecting these transcripts into oocytes. In view of the frequency predicted for obtaining a full-length 5-6 kb cDNA required for the expression of an active sero¬ tonin receptor, it is necessary to first enrich the message population to be used in the construction of the library. This was accomplished by the method of sucrose density sedimentation using total RNA isolated from the choroid plexus (9, 35, 36, 37, 38). Fractions collected from such a gradient were assayed for sero¬ tonin receptor mRNA by voltage clamp analysis of micro¬ injected oocytes, revealing a positive peak in the high molecular weight range (see Figure 3A) . Consis¬ tent with reported data (6, 39) , injection of RNA in the size range 5-7 kb resulted in a serotonin-evoked inward current in oocytes. The RNA thus obtained was
used to construct a cDNA library in the cloning vector λ ZAP (9, 35, 36, 37, 38). This vector was chosen because bacteriophage T7 and T3 promoters flank the cDNA insertion site, enabling one to synthesize RNA transcripts in vitro from either direction. A size- enriched library representing 1.2 million clones was generated and DNA was prepared from pools of 20,000 recombinant phage. In order to expeditiously screen the library, DNA from five pools (i.e. 100,000 clones each) was combined, transcribed with T7 RNA polymerase (40) and the reaction products microinjected into oocytes. After a 3 day incubation period, the oocytes were subjected to voltage clamp analysis (this cloning scheme is summarized in Figure 3) . Figure 5 shows the typical response to ' io " M serotonin of a choroid plexus πtRNA-injected oocyte. Figure 6 shows an initial posi- tive response to 10 —6M serotonin of an oocyte injected with RNA prepared in vitro from the pool of 100,000 cDNA clones; although relatively small (20-50 nA) , this signal was believed to be genuine because there is no obvious source of false-positives or background noise. Further fractionation reduced the pool size step-wise by factors of ten, yielding a single cDNA clone (Z347) after four rounds of re-screening. At each stage, at least one pool of RNA generated serotonin-responsive currents. The magnitude of the serotonin-evoked cur¬ rent normalized per microgram of injected RNA in¬ creased as the sib selection proceeded. A response obtained from Z347 is shown in Figure 7.
A plasmid containing the cDNA insert still flanked by T7 and T3 promoters (p347) was rescued from Z347 by a λ excision procedure. Figure 8 shows the results of a dilution study in which various amounts of capped, synthetic RNA transcribed from pSR-lc DNA were injected
into oocytes and the responses to serotonin recorded. The limit of detection falls at or near 0.5 pg of in¬ jected RNA, an amount sufficient to generate a sero¬ tonin-evoked current of about 200 nA (Figure 9B) .
Assuming an average RNA size of 3kb, this is roughly
5 equivalent to 3 x 10 molecules per oocyte.
Pharmacologically, the rat choroid plexus 5HT1C recep¬ tor has been shown to bind the serotonin antagonists mianserin with high affinity (K.=15 nM) and spiperone with substantially lower affinity (K^δ μM) (23). This rank order of drug potency has been established by three independent methods: (i) competitive binding studies with 125I-LSD to choroid plexus membranes;
(ii) inhibition of serotonin-stimulated phos-phatidy- linositol turnover in choroid plexus tissue; and (iii) inhibition of oocyte responses to serotonin following injection of rat choroid plexus mRNA. For oocytes injected with pSR-lc-derived RNA, 1 μM mianserin completely abolishes responses to 10 nM 5-HT, whereas 1 μM spiperone is barely effective as an antagonist (not shown) . The relative potency of these antag¬ onists is therefore similar when assayed against 5HTlc receptors present in the choroid plexus or in oocytes injected with pSR-lc-derived RNA. Applicants conclude that pSR-lc encodes the lc subtype of serotonin receptor from the choroid plexus.
The nucleotide sequence and deduced amino acid sequence of serotonin receptor lc
The nucleotide sequence and deduced amino acid sequence of the 5HTlc receptor are shown in Figure 1. In addi¬ tion. Figure 2 shows the percent of codon usage in the 5HTlc receptor DNA. One long open reading frame ex¬ tending from an ATG codon at position 688 to a stop
codon at residue 2068 can encode a protein sequence 460 amino acids in length, having a relative molecular size of 51,899 kD. Stop codons are found in all three frames upstream of this ATG, and downstream of the termination site. Interestingly, the coding sequence is preceded by a long 5' untranslated region with mini¬ mum length of 700 bp. The functional 5HTlc receptor cDNA cloned by applicants is only 3 kb whereas the mRNA containing this sequence is more than 5 kb. It is likely therefore that the cDNA clone does not extend to the 3' terminus of the mRNA, but terminates artificial¬ ly in the 3'-untranslated region. In addition, nucleotides surrounding this putative initiation codon satisfy the consensus rules for a protein translation start site.
The 5HTlc receptor shares numerous sequence and struc¬ tural properties with the family of receptor molecules that has been predicted to span the lipid bilayer seven times (Figure 10) . This family includes rhodopsin and related opsins (41) , the α and β adrenergic receptors (42) , the muscarinic cholinergic receptors (43) , the substance K neuropeptide receptor (30) , the yeast mat¬ ing factor receptors (46, 47, 48) , and the oncogene c- mas (49) . Each of these receptors is thought to trans- duce extracellular signals by interaction with guanine nucleotide-binding (G) proteins (42, 50, 51).
In addition to the similarity in transmembrane topolo¬ gy, the serotonin receptor shares blocks of homology with other receptors of its kind, particularly in areas which are known to be highly conserved among members of this family (see Figure 1) . For example, in the second transmembrane domain, there is a stretch of amino acids whose sequence is nearly invariant, and reads: -N-Y-
F(I)-X-X-S(N)-L-A-X-A-D-L. In the case of the 5HTlc receptor sequence, this region is only slightly diver¬ gent from the consensus, reading: -N-Y-F-X-X-S-L-A-I-A- D-M-. Other invariant landmarks include: two cysteine residues, possibly disulfide-linked, within the oligo- saccharide side chain in the region preceding the first transmembrane domain (N 3g ) .
On the basis of structural homologies with rhodopsin, it is likely that the amino terminus of the 5HTlc re¬ ceptor is located on the extracellular side of the plasma membrane, with the carboxyl-terminus located intracellularly. In this scheme, three extracellular loops alternate with three intracellular loops to link the seven transmembrane domains.
Although the amino-terminal 65 amino acids of the 5HTlc receptor are thought to reside on the extracellular face of the membrane, the protein lacks a definable signal sequence. A potential N-linked glycosylation site is present at position 39. The receptor contains seven hydrophobic stretches (20 to 30 amino acids) that represent potential transmembrane domains. These do¬ mains constitute the regions of maximal sequence simi¬ larity with the other transmitter receptors of this class. The 5HTlc receptor shares 25% sequence identity with the 2 -adrenergic receptor, and 20% identity with the muscarinic and substance K receptors (Figure 11) .
A conserved aspartate residue is present within the second putative transmembrane domain and, by analogy with the jS 2 -adrenergic receptor, this residue may par¬ ticipate in ligand binding. Mutations in this con¬ served aspartate in the 2 ~adrenergic receptor reduce agonist binding (52, 53). In addition, amino acid
residues in this domain can be cross-linked to affinity ligands of the 0 2 ~adrenergic receptor (50, 51) . These data, taken together with the observation that all ligands capable of associating with this family of receptors have amino groups, suggest that aspartate participates in stabilizing the ligand in the plane of the membrane.
The third cytoplasmic loop, connecting transmembrane domains 5 and 6 and thought to interact with the dif¬ ferent G proteins (54, 55, 56), is of widely varying length in different .receptors. No sequence similari¬ ties within this domain are apparent among receptors that couple to common signaling systems. In the 5HT1C receptor, this loop consists of 77 amino acids, many of which are basic, a feature characteristic of other receptors in this family such as adrenergic and musca- rinic receptors.
In the case of the serotonin receptor, lysine and ar- ginine constitute 22% (17/77) of the total number of residues in this domain.
Both the carboxyl-terminal cytoplasmic domain and the third cytoplasmic loop contain sequences that may serve as substrates for phosphorylation by cyclic AMP-depen- dent protein kinase or protein kinase C. In addition, there are four serine residues in the carboxyl-terminal 12 amino acids which, by analogy with rhodopsin and the adrenergic receptors, may represent additional phos¬ phorylation sites. These potential sites of phos¬ phorylation may play a role in regulating the activity of the receptor molecule (57). Serine 312 in the vari¬ able loop lies within a consensus protein kinase C phosphorylation site, as does serine 382 , located in the
Receptor expression in transfected mammalian cells The deduced protein sequence of pSR-lc indicates that applicants have cloned a new member of the gene family encoding G-protein-coupled neurotransmitter receptors. To establish further that this clone encodes a sero¬ tonin receptor and to extend the ability to manipulate it, applicants have demonstrated that the introduction of this cDNA into mammalian fibroblasts renders these cells responsive to serotonin. The entire 3 kb EcoRI cDNA fragment from pSR-lc was subcloned into the expression vector pMV7 (58) . This vector contains a murine leukemia virus long terminal repeat which serves as a promoter for expression of the serotonin cDNA as well as an independent expression cassette encoding neomycin phosphotransferase. Transformed NIH3T3 cells resistant to neomycin were isolated and a single clone expressing significant levels of receptor mRNA (SR3T3) was identified by RNA blot analysis.
Radioligand binding experiments were performed to de¬ termine whether SR3T3 cells display serotonin recep¬ tors on their surface. It has been previously shown that crude membranes prepared from the rat choroid plexus bind 125I-LSD specifically and with high affini¬ ty. Membranes prepared from SR3T3 cells exhibited high affinity binding sites for ligands that interact with the 5HTlc receptor on choroid plexus cells (3, 59). SR3T3 cells expressed 10 3 to 104 high affinity binding sites per cell for I-labeled lysergic acid diethyl- amide ( 125I-LSD), whereas no specific high affinity sites were detected on the parental NIH3T3 cell line.
Moreover, the relative ability of specific antagonists to inhibit 125 I-LSD binding to SR3T3 cells paralleled
their potency on choroid plexus cells (Figure 12) (23) .
Mianserin (IC__ = 20 nM) was about two orders of magni- tude more effective m di .splaci.ng 125I-LSD binding than was spiperone (IC 5Q = 2 μM) . Binding of 125I-LSD was also displaced by serotonin with an IC 5Q value of 20 nM. The expression of specific high affinity binding sites for 5HTlc-selective ligands on transformed 3T3 cells provides independent confirmation that pSR-lc encodes a serotonin lc receptor. Applicants conclude that the receptors expressed by SR3T3 cells are pharma¬ cologically identical to those found in the choroid plexus.
Next, applicants determined whether serotonin can pro¬ voke the mobilization of intracellular calcium in transfected cells. This would provide a rapid and quantitative assay for ligand binding and transmembrane signaling.
To ascertain whether the binding of serotonin activates intracellular signaling pathways in transformed fibro- blasts, SR3T3 cells were loaded with the Ca 2+-sensitive dye indo-1 and analyzed in a fluorescence-activated cell sorter (11) . Indo-1 undergoes a characteristic and quantitative shift in its ratio of fluorescence emission at 400 and 500 nm wavelengths (excitation = 360 nm) as a function of Ca 2+ concentration and servreess as a quantitative measure of intracellular-free Ca 2'+ (60) . SR3T3 and control NIH3T3 cells loaded with indo- 1 were exposed to serotonin immediately before fluores¬ cence analysis to reduce the possibility of desensiti- zation that may result from prolonged exposure of cells to agonist. In this way, all cells are examined at a fixed time following ligand-receptor interaction. Ninety-five percent of the SR3T3 cells loaded with
indo-1 showed a marked increase in intracellular Ca2+ when exposed to serotonin, whereas control, untrans- fected NIH3T3 cells did not respond to serotonin (Fig¬ ure 13) (Table I). SR3T3 cells sorted in the absence of serotonin also remain at low, resting calcium lev¬ els. These experiments indicate that the introduction of pSR-lc cDNA in mammalian fibroblasts leads to the expression of functional serotonin receptors. Thus the 5HTlc receptor is capable of triggering the same trans- duction machinery regardless of the cell type in which it is expressed.
MATCH RANGE : 13 41
SUBTRACTION RANGE: 0 255
PERCENT NEGATIVE 5.7559
STD DEVIATION 33.4417
KURTOSIS 0.0248 STATISTICS RANGE 0... 255
PERCENT POSITIVES 94.244
STD DEVIATION 27.5321
KURTOSIS 0.0003 STATISTICS RANGE 41... 255
1 DJ -25 15/ 1/88 20:48 1P256 3T3 PNV7K1 5 BL RATIO /RATIO,FALS
2 DJ -26 15/ 1/88 20:57
1P256 3T3 PNV7K1 5 5HT RATIO /RATIO,FALS
Expression of 5HTlc serotonin receptor in the nervous system
RNA blot analysis and in situ hybridization (61) were performed to examine the expression of 5HTlc receptor mRNA in different brain regions and in peripheral tis- sues. Using a 35S-labeled anti-sense RNA probe for in situ hybridization, a heavy grain density associated with epithelial cells of the choroid plexus in the third, fourth and lateral ventricles was revealed. Strikingly, hybridization was not restricted to cells of the choroid plexus but appeared in numerous neuronal cell groups throughout the central nervous system. Labeled neurons were observed in the lateral habenula, whereas neurons in medial habenula were not labeled. Higher magnification of this region revealed the pres¬ ence of silver grains in neuronal perikarya. The 5HTlc mRNA was also observed in neurons in cortical struc¬ tures and in a variety of subcortical brain regions (62).
Filters to which poly A+ RNA from these tissues had been transferred were hybridized with a nick-translated probe prepared from the entire 3 kb EcoRI cDNA insert. A mouse alpha-actin probe was also used as an external calibration standard.
Although it is possible that the in situ analysis de¬ tects mRNA species other than that of the 5HTlc recep¬ tor, the distribution of 5HTlc receptor mRNA determined by in situ hybridization was supported by RNA blot analysis (Figures 14 and 15). An intense 5.2 kb RNA was observed in choroid plexus, and in a variety of other regions of the brain, including the basal gan¬ glia, hypothalamus, hippocampus, pons medulla, and spinal cord. Longer exposure reveals the presence of
a small amount of receptor mRNA in the olfactory bulb. Receptor RNA was not detected in the cerebellum or in liver, kidney, intestine, heart, and lung. Titration studies with purified pSR-lc RNA indicate that the 5HTlc receptor RNA comprises about 0.02% of the choroid plexus message population. The relative abundance of this RNA in other regions of the brain is at least ten times lower. In situ hybridization, however, indicat¬ ed that some neurons express receptor mRNA levels com¬ parable to those in choroid plexus cells. These find- ings demonstrate that the 5HTlc receptor is not re¬ stricted to epithelial cells of the choroid plexus, but is expressed in numerous discrete nueronal cell groups in many regions of the rat brain and spinal cord. This distribution of 5HTlc receptors suggests that this receptor subtype may mediate many of the central ac¬ tions of serotonin.
Diversity of receptor subtypes
Applicants have cloned and characterized a functional cDNA encoding the 5HTlc subclass of serotonin receptor. RNA transcribed from this cDNA clone (pSR-lc) confers serotonin-sensitivity to Xenopus oocytes. The expres¬ sion of pSR-lc in mouse fibroblasts results in the appearance of high affinity serotonin-binding sites on the cell surface. Exposure of these transformants to serotonin increases intracellular Ca 2+ levels. More¬ over, abundant expression of 5HTlc receptor mRNA is observed in subsets of neurons throughout the brain, suggesting that the 5HTlc receptor plays an important role in central neurotransmission.
The procedure used to isolate the functional 5HTlc receptor cDNA has combined cloning in RNA expression vectors with a sensitive electrophysiological assay for
serotonin receptor function in Xenopus oocytes. Simi¬ lar procedures have been used to isolate the cDNA en¬ coding the receptor for the neuropeptide substance K (30) , and the Na -glucose cotransporter (63) . The injection of as little as 5 pg of choroid plexus poly A+ RNA (1 fg of receptor mRNA) into oocytes was ade¬ quate to generate serotonin-evoked membrane currents. The sensitivity of the oocyte expression assay presum¬ ably results from the signal amplification associated with the coupling of these receptors to second messen¬ ger systems. In the oocyte, this assay is at present limited to receptors that activate second messenger systems, in particular phospholipase C, which elevates intracellular calcium levels. This cloning strategy may therefore be generally applicable to the isolation of genes encoding neurotransmitter and growth factor receptors that initiate a similar signal amplification, even if their mRNA's are present in exceedingly small amounts in the total RNA population..
Pharmacological studies, and more recently gene clon¬ ing, have established " that multiple receptor subtypes exist for most, if not all, neurotransmitters. The existence of multiple receptor subtypes provides one mechanism by which a single neurotransmitter can elicit distinct cellular responses. The variation in cellular response can be achieved by the association of individ¬ ual receptor subtypes with different G proteins and different signaling systems. Further flexibility is provided by the ability of distinct receptors for the same ligand to activate or inhibit the same second messenger system. For example, among the adrenergic receptors, β and 2 receptors activate adenylate cy- clase, α 2 receptors inhibit adenylate cyσlase and α-_ receptors activate phospholipase C pathways (52, 53).
A similar array of cellular responses can be elicited by serotonin in cells bearing different receptor sub¬ types. Both 5HTlc and 5HT2 receptors stimulate the phospholipase C-mediated production of inositol phos¬ phates, whereas 5HTla and 5HTlb receptors may regulate adenylate cyclase activity (19, 20, 21) or couple to G proteins that directly activate ion channels (24, 25, 26, 27) . The diverse neural actions of serotonin are thought to be mediated by activation of these distinct receptor subtypes. For example, the hallucinatory and perceptual disturbances associated with administration of LSD and other psychedelic serotonin analogs (67) are probably elicited by activation of cortical 5HT2 recep¬ tors (68) . In contrast, the inflammatory and pain- producing effects of serotonin are mediated via 5HT3 receptors on peripheral sensory endings (69) . The ability to express 5HT receptors in new cellular envi¬ ronments devoid of other receptor subtypes should per¬ mit the characterization of transduction systems asso- ciated with these specific recsptors. Applicants an¬ ticipate that the cloning of additional receptor sub¬ types will help to elucidate the mechanisms of action of serotonin in the nervous system.
List of References Cited
1. Sanger, S., et al., Proc. Natl. Acad. Sci. USA 74: 5463-5467 (1987) .
2. Maddon, P.J., et al.. Cell £7: 333-348 (1986).
3. Yagaloff and Hartig, J. Neurosciences 5_: 3178-3183 (1985) .
4. Sumikawa, S., et al., Proc. Royal Soc. London, Series B 223: 255-260 (1984).
5. Gunderson, C., et al., Proc. Royal Soc. London, Series B 219: 103-109 (1983).
6. Lubbert, H. , et al., J. Neurosciences 7_: 1159-1165 (1987).
7. Cathala, G., et al., DNA 2 : 329-335 (1983).
8. Krumlauf, et al.. Development £ : 603-617 (1987).
9. Okayama, H. and Berg, P., Mol. Cell. Biol. : 161- 170 (1982).
10. Wigler, et al.. Cell 1£: 777-785 (1979).
11. Schieren and MacDermott, J. Neurosci. Methods, in press. .
12. Takaki, M. , et al., J. Neurosciences 5 : 1769 (1985).
13. Kravitz, E.A., et al., J. Exp. Biol. 8_9: 159 (1980) .
14. Nemecek, G.M., et al., Proc. Natl. Acad. Sci. USA 8_3: 674 (1986) .
15. Fields, H.L., in Advances in Pain Research and Therapy, L. Kruger and J.C. Liebeskind, Eds. (Raven Press, New York, 1984), Vol. 6, pp. 241- 252.
16. White, S.R. and Neuman, R.S., Brai Res. 188: 119 (1980) .
17. Jacobs, B.L., in Hallucinogens: Neurochemical, Behavioral, and Clinical Perspectives, B.L. Jacobs, Ed. (Raven Press, New York, 1984) , pp. 183-202.
18. Peroutka, S.J., Ann. Rev. Neurosci. _11: 45 (1988).
19. DeVivo, M. and Maayani, S., J. Pharmacol. Exp. Ther. 18-* 248 (1986) .
20. Hoyer, D. and Schoekkter, P., Eur. J. Pharmacol. 147: 145 (1988).
21. Weiss, S., et al., Eur. J. Pharmacol. 120: 227 (1986) .
22. de Courcelles, D., et al., J. Biol. Chem. 260: 7603 (1985).
23. Conn, P.J. , et al., Proc. Natl. Acad. Sci. USA JT3: 4086 (1986) .
24. Aghajanian, G.K., in Serotonin Neurotrans ission and Behavior, B.L. Jacobs and A. Gelperin, Eds. (MIT Press, Cambridge, Massachusetts, 1981) , pp. 156-185.
25. Andrade, R. , et al., Science 234: 1261 (1986).
26. Andrade, R. and Nicoll, R.A., J. Physiol. 394: 99 (1987) .
27. Siegelbaum, S.A., et al.. Nature 299: 413 (1982).
28. Takahashi, T. , et al., Proc. Natl. Acad. Sci. USA 84: 5063 (1987) .
29. Lubbert, H. , et al. , Proc. Natl. Acad. Sci. USA 8.4: 4332 (1987).
30. Masu, Y., et al.. Nature 329: 836 (1987).
31. Pazos, A. and Palacios, J.M. , Brain Res. 346: 205 (1985).
32. Stryer, L. , in Cold Spring Harbor Symp. Quant. Biol., Molecular Neurobiology (Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 1983), Vol. 48, pp. 841-852.
33. Casey, P.J. and Gilman, A.G., J. Biol. Chem. 263: 2577 (1988).
34. Aviv, H. and Leder, P., Proc. Natl. Acad. Sci. USA 6£: 1408 (1972).
35. Mayuhas, 0. and Perry, R.P., Cell .16 , : 139 (1979).
36. Kraus, J.P., et al., Proc. Natl. Acad. Sci. USA 83: 2047 (1986) .
37. Huynh, T.V., et al., in DNA Cloning, I.D.M. Glov¬ er, Ed. (RL Press, Oxford, 1985), Vol. 1, p. 49.
38. Watson, CJ. and Jackson, J.F., in DNA Cloning, I.D.M. Glover, Ed. (RL Press, Oxford, 1985), Vol. 1, p. 79.
39. Sumikawa, K. , et al., Proc. Natl. Acad. sci. USA 8_1: 7994 (1984).
40. Maniatis, T. , et al., in Molecular Cloning (Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 1982) .
41. Nathans, J. and Hogness, D.S., Cell 3_4_: 807 (1983).
42. Dohlman, H.G., et al.. Biochemistry 2_6: 2657 (1987).
43. Bonner, T.I., et al.. Science 237: 527 (1987).
44. Kubo, T., et al.. Nature 32_3: 411 (1986).
45. Peralta, E., et al.. Science 236: 600 (1987).
46. Burkholder, A.C. and Hartwell, L.H., Nucl. Acids Res. , 13: 8463 (1985) .
47. Hagan, D.C., et al., Proc. Natl. Acad. Sci. USA 3.: 1418 (1986) .
48. Nakayama, N. , et al., EMBO J. ±: 2643 (1985).
49. Young, D., et al., Cell 45.: 711 (1986).
50. Dohl an, H.G., et al.. Biochemistry 1813 (1988) .
51. O'Dowd, B.F., et al., Ann. Rev. Neurosci., in press.
52. Chung, F.-Z., et al., J. Biol. Chem. 263: 4052 (1988) .
53. Strader, CD., et al., Proc. Natl. Acad. Sci. USA 84: 4384 (1987) .
54. Dixon, R., et al., EMBO J. 6 : 3269 (1987).
55. Strader, CD., et al., J. Biol. Chem. 262: 16439 (1987a) .
56. Dixon, R., et al.. Nature 326: 73 (1987a).
57. Sibley, D.R. , et al.. Cell 4j3: 913 (1987).
58. Kirschmeir, P., et al., DNA, in press.
59. Kadan, M.J. , et al., J. Neurochem. 3_: 601 (1984).
60. Grynkiewicz, G. , et al., J. Biol. Chem. 260: 3440 (1985).
61. Fremeau, R.T., et al.. Science 234: 1265 (1986).
62. Molineaux, S., et al., in preparation.
63. Hediger, M.A. , et al., Proc. Natl. Acad. Sci. USA 84.: 2634 (1987).
64. Dixon, R. , et al.. Nature 321: 75 (1986).
65. Kobilka, B. , et al.. Science 238: 650 (1987).
66. Bonner, T.I., et al.. Science 237: 527 (1987).
67. Leary, T. , et al., in The Psychedelic Experience (University Books, New Hyde Park, New York, 1964) .
68. Glennon, R.A. , et al., Life Sci. 3_5: 2505 (1984).
69. Richardson, B.P., et al.. Nature 316: 126 (1985).