Proc. Nadl. Acad. Sci. USAVol. 91, pp. 3666-3669, April 1994Neurobiology

Selective 5-HT1Da serotonin receptor gene expression in trigeminalganglia: Implications for antimigraine drug development

(mlrale/seroton receptor subtypes)

G. WILLIAM REBECK*, KENNETH I. MAYNARDtI, BRADLEY T. HYMAN*, AND MICHAEL A. MOSKOWITZ*tf§*Neurology and tNeurosurgery Services, *Stroke Research Laboratory, Massachusetts General Hospital, Harvard Medical School, Boston, MA 02114

Communicated by Susan E. Leeman, January 3, 1994

ABSTRACT Pharmacological data suggest that the actionsof -antimiine drugs such as sumatriptn may be medated by5-HT1D-lke serotonin recepors on trigemnhovascular nerveendings. We sought molecular evidence for the expreion ofanmRNA species encoding the 5-HT1D receptor subtype in guineapig and human trigeminal g i, using the polymerase chainreaction with oigon tides uniquely homologous to thecoding sequences of the 5-HT1B/D family (human 5-HTmDa and5-HT1DP; rat 5-HTo). A single band of predicted size wasobserved In samples from guinea pig trgminal gangl; se-quence analysis revealed the presence of a single message,which was 85% and 71% identical to the human 5-HTl) and5-HT1Dp receptor DNA sequences, respectively. Similar anal-yses of postmortem human trigemal gangia revealed thepresence of 5-HTnm., but not S-HTmDp, receptor message.Inasmuch as one recent report found thatmRNAendig onlythe 5-HT1Dp receptor subtype was expressed by vascularsmooth muscle of the central nervous system, the presentfindings suggest the importance of developing selective5-HTlDa receptor ago s as a strategy to reduce the risk ofmyocardlal infarction and possibly stroke that complicates theacute treatment of migaine headache.

The 5-HT1D serotonin receptor subtype comprises a family ofat least two closely related receptors encoded by distinctgenes (1, 2) designated 5-HTDDa and 5-HT1DP (2) or 5-HT1Dand 5-HT1I (3). Pharmacological evidence supports the ex-istence of prejunctional serotonin (5-hydroxytryptamine,5-HT) receptors resembling 5-HTj and 5-HT1B on trigemi-novascular axons transmitting nociceptive information fromguinea pig and rat meninges, respectively (4, 5). Serotoninagonists and antimigraine compounds such as sumatriptanand ergot alkaloids bind to these receptor sites to block thedevelopment of meningeal neurogenic inflammation and tosuppress nociceptive activity within trigeminovascular affer-ents (6-10).

Receptors for the 5-HTlD-like subtype have also beenpharmacologically characterized on postjunctional vascularsmooth muscle, where they mediate vasoconstriction (11).Recently, Northern blot analyses provided evidence for theselective expression of the 5-HT1Dq receptor gene in pialblood vessels (12). Activation of contractile postjunctionalreceptors on human coronary and cerebral arteries has beenlinked to the development of coronary vasospasm (13, 14),myocardial infarction (15), and possibly stroke, complicatingthe treatment of migraine.Our own studies were undertaken to examine the possible

expression of a selective receptor subtype which differenti-ates the pre- and postjunctional 5-HT1D receptor family. Weused polymerase chain reaction (PCR)-based amplification toshow expression ofmRNA species homologous to members

ofthe 5-HT1D receptor family in guinea pig trigeminal ganglia.Using a similar strategy, we also demonstrated the presenceof the 5-HTma, but not the 5-HT1Dp, receptor messages inpostmortem human trigeminal ganglia. Thus, agonists bind-ing selectively to the 5-HTDa receptor subtype could lead tomore specific antimigaine therapies with less potential riskof complications relating to constriction of vascular smoothmuscle.

METHODSTissue Preparation. Three male Hartley guinea pigs (200-

250 g) were housed under diurnal lighting conditions andkilled by decapitation. The trigeminal ganglia (36-52 mg)were removed from the skull base within 5 min along with thefollowing regions dissected from brain: frontal cortex, cere-bellum, hippocampus, and striatum. Tissue samples werefrozen in liquid nitrogen and stored at -80oC.Human trigeminal ganglia were removed from the skull

base of eight individuals 5-18 hr after death. Ganglia werecarefully dissected from the surrounding tissue and frozen at-800C.RNA Isolation. Tissue samples were sonicated in GTC

buffer [4 M guanidinium thiocyanate/10 mM Tris, pH 7.5/1mM EDTA/0.5% (wt/vol) N-lauroylsarcosine/1% (vol/vol)2-mercaptoethanol], and total RNA was extracted as de-scribed (16). In brief, the cell lysate was extracted twice withphenol/chloroform and nucleic acids were precipitated in2-propanol. RNA was resuspended in GTC buffer, precipi-tated in ethanol, washed with 70%6 ethanol, and suspended in100 MI of water. Human RNA samples were digested withRNase-free DNase (Boehringer Mannheim), extracted twicewith phenol, precipitated, and suspended in water. Theconcentration of RNA was measured by UV absorbance at260 nm.cDNA Preparation. Five micrograms of total RNA was

denatured at 700C and annealed to 10 pmol of (di)17. cDNAwas generated from the oligo(dT)-primed RNA by treatmentwith Moloney murine leukemia virus reverse transcriptase(Boehringer Mannheim) at 370C for 1 hr and the reaction wasterminated by heating at 900C for 5 min. The cDNA poolswere diluted to 100 id.

Oligonucleotide Primers. Oligodeoxynucleotides were syn-thesized on an Applied Biosystems model 392 DNA/RNAsynthesizer and used without purification. Primers W72 andW73 were complementary to the human 5-HT1Da receptorgene (Fig. 1) (17) and highly homologous to the rat 5-HT1Band human 5-HT1D sequences, but not other 5-HT receptorgene subtypes, including the 5-HT1D pseudogene (18). Primerpairs specific for the human 5-HT1D0 and 5-HTD receptor

Abbreviation: 5-HT, 5-hydroxytryptamine (serotonin).§To whom reprint requests should be addressed at: MassachusettsGeneral Hospital, Wellman 423, Boston, MA 02114.IThe sequences reported in this paper have been deposited in theGenBank data base (accession nos. U07022).

3666

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertisement"in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Dow

nloa

ded

by g

uest

on

July

26,

202

1

Proc. Natl. Acad. Sci. USA 91 (1994) 3667

I *E*IW72

347 bp

- I a

W73 W144 W145

149 bp

1131

I--ElUW144 W146 1173

158 bp

FIG. 1. Location ofamplification primers in the human 5-HTwD, and 5-HTjD receptor genes. The coding regions ofthe 5-HT1D. and 5-HT1Dpreceptor genes, which are contained in single exons, are represented as open bars, with base pairs numbered 1-1131 (5-HT1DJ) and 1-1173(5-HT1Dp). Shaded regions represent transmembrane domains of the gene products. The primer pair W72/W73 amplifies 347 bp. Thesubtype-specific primer pairs W144/W145 and W144/W146 amplify 149 and 158 bp, respectively.

genes were also made (W144, 100%o sequence identity to5-HTlDa and 90% identity to 5-HT1DS; W145, 100%o identityto 5-HTlD, and 45% identity to 5-HT1DP; W146, 100%o identityto 5-HT1D9 and 45% identity to 5-HT1Da) (Fig. 1). 5-HTIDa-specific primers amplify 149 bp, whereas 5-HTlDP-specificprimers amplify 158 bp (see Fig. 4).

W72W73W144W144W145W146

5'-ACC-AGG-AAG-CTC-CAC-ACC-CC-3'5'-GCC-TTG-GCC-TGC-CGC-CAG-AA-3'5'-CCT-ATA-TGG-CCG-GAT-CTA-C-3'5'-CCT-ATA-TGG-CCG-GAT-CTA-C-3'5'-AGT-GCC-CCT-CAT-GGA-GGC-TG-3'5'-TGG-GCA-CGT-CGG-GAA-CCC-GC-3'

bp 1%-215 (a)bp 542-523 (a)bp 645-663 (a)bp 678-6% (p)bp 793-774 (a)bp 835-816 (P)

PCR Amplffication and Product Separation. DNA was

amplified by PCR in a programmable thermal controller,model PTC-100 (MJ Research, Cambridge, MA). Reactionmixtures (50 pLI) contained 10 A1 ofcDNA pool, 200 ,M eachdNTP, 200 nM primers, and 2.5 units of Taq DNA polymer-ase (Boehringer Mannheim) in 10 mM Tris, pH 8.3/1.5 mMMgCl2/50mM KC1/0.01% gelatin. PCR ofguinea pig sampleswas conducted for 30 cycles ofdenaturation at 94°C for 1 min,annealing at 58°C for 1 min, and extension at 72°C for 2 min.PCR of human samples was conducted for 35 cycles of thesame protocol or 40 cycles of denaturation at 94°C for 1 minand annealing/extension at 63°C for 1.5 min.PCR products were separated by gel electrophoresis using

a 2% or 3% agarose gel in 90 mM Tris/90 mM boric acid/2mM EDTA containing ethidium bromide (1 ug/ml; FisherScientific). Upon completion, the gel was viewed under aUVlight and photographed.The DNA amplified from guinea pig cDNA using the W72

and W73 primers was sequenced with thefmol PCR sequenc-ing protocol (Promega). The W72 and W73 primers wereend-labeled with [y-32P]ATP (New England Nuclear) andextended in the presence ofdideoxynucleotides. The reactionproducts were separated in a 6% acrylamide/urea gel, whichwas then dried and exposed to film.

RESULTS

Approximately 1 pg of total RNA per mg of wet weight wasextracted from each of the guinea pig tissues and used togenerate cDNA pools. Primers directed to conserved regionsofthe 5-HT1Da gene were used forPCR amplification (Fig. 1).Amplified DNA ofthe predicted size (-350 bp) was producedfrom samples of trigeminal ganglia and cerebella in each ofthree separate experiments (Fig. 2). PCR products did notappear if the reverse transcriptase step was omitted. Evi-dence for expression ofa 5-HTID receptor gene was not foundin the guinea pig striatum, hippocampus, and neocortex (Fig.

2), although previous studies had demonstrated 5-HT1D re-ceptor expression in human tissues (2, 3).The amplifiedDNA from the guinea pig trigeminal ganglion

cDNA was sequenced by a PCR cycle sequencing protocol.A single DNA sequence was found, which was homologousto the appropriate region ofthe human 5-HTD receptor genes(Fig. 3). The guinea pig nucleotide sequence was 85% and71% identical to the human 5-HTlDa and 5-HT1D receptorsequences, respectively; the deduced 116-amino acid se-quence was 92% and 74% identical, respectively.These data suggested that the message for the 5-HT1Da

receptor was expressed in guinea pig trigeminal ganglia.However, without the known sequences of the receptorsubtypes in guinea pig, it was not possible to definitivelydetermine whether mRNA species of both 5-HTlDa and5-HT1D subtypes were present. To better address the ques-tion of expression of 5-HTlD receptor subtypes, we took asimilar PCR-based approach using trigeminal ganglia fromhumans. Ganglia from two postmortem cases showed con-vincing evidence of 5-HTlDa receptor mRNA. Ganglia fromsix other cases were negative, possibly due to degradation ofthe mRNA after death, although there was no obviousrelationship between postmortem time and receptor messagedetection.

Primers specific for 5-HTlDa and 5-HT1Dp receptor sub-types were designed (Fig. 1) and used to determine whichsubtypes were expressed in the human trigeminal ganglion

A B C D E

748k625 -502 f369/_246123

FIG. 2. Agarose gel electrophoresis of PCR-amplified productsfrom mRNA of trigeminal ganglia and various brain regions. RNAwas isolated from guinea pig trigeminal ganglion (lane A), cerebellum(lane B), corpus striatum (lane C), hippocampus (lane D), and frontalneocortex (lane E). Five micrograms of RNA was reverse tran-scribed and 5% ofthe resultant cDNA was amplified by using primersW72 and W73. The amplified DNA products were separated in a 2%agarose gel and stained with ethidium bromide. The outer lanescontained size standards of a 123-bp ladder.

Neurobiology: Rebeck et al.

Dow

nloa

ded

by g

uest

on

July

26,

202

1

Proc. Natl. Acad. Sci. USA 91 (1994)

1 50 1005HTlDalpha ACCAGGAAGC TCCACACCCC TGCCAACTAC CTGATTGGCT CCCTGGCCAC CACCGACCTC TTGGTTTCCA TCTTGGTAAT GCCCATCAGC ATCGCCTATA CCATCACCCA

111111111 11 11 11 1111111 III 11111 11111111 11 1111 11 III H IM 11111 11 III 11111guinea pig ... ...... TGCCAACTAT CTCATCGAAT CCTTGGCCAC CACTGACCTT TTGGTTTCTA TCCTGGTCAT GCCAATCAGC ATCGCATACA CCACCACCCG

11 11111 111111 II 1I 11 1 I11111 11 II I 1111111 11 1III HIMII 1111 1 1111 15HTID beta ACCCGGAAAC TGCACACCCC GGCTAACTAC CTGATCGCCT CTCTGCGAGT CACCGACCTG CTTGTGTCCA TCCTGGTGAT GCCCATCAGC ACCATGTACA CTGTCACCGG

150 200CACCTGGAAC TTTGGCCAAA TCTTGTGTGA CATCTGGCTG TCCTCTGACA TCACGTGCTG CACAGCCTCC ATCCTGCATC TCTGTGTCAT TGCTCTGGAC AGGTACTGGG CAATCACAGA1111111111 1111111111 11 1111 11 1111111 11 11111 1111 1111 11111 III 11111 1111111111 1111111111 1111111111 1111111111 11111 11CACCTGGAAC rTTGGCCAAA TCCTGTGCGA CATCTGGGTG TCCTCCGACA TCACCTGCTG CACGGCCTCC ATCCTCATC TCTGTGTCAT TGCTCTGGAC AGGTACTGGG CCATCACGGA

11111 11111 11 11 11111 11 11 11 1111 1111 11 11 11111111IIIHIII 111111111111111111111CCGCTGGACA CTGGGCCAGG TGGTCTGTGA CTTCTGGCTG TCGTCGGACA TCACTTGTTG CACTGCCTCC ATCCTGCACC TCTGTGTCAT CGCCCTGGAC CGCTACTGGG CCATCACGGA

250 300 347TGCCCTGGAA TACAGTAAAC GCAGGACGGC TGGCCACGCG GCCACCATGA TCGCCATTGT CTGGGCCATC TCCATCTGCA TCTCCATCCC CCCGCTCTTC TGGCGGCAGG CCAAGGC11111111 11111 11 11 11 11111 11 I 1111 11 11111 1111 11111111 1111111111 111111

CGCCCTGGAG TACAGCAAGC GTCGCACAGC GGGCCATGCT GGGGCCATGA TCGCGGCGGT CTGGGTCATC TCCATCTGTA TCTCCATCCC TCCGCTC..................

111 11 111 111 I 11I I 111 IIII H IM 1111 11 II 111 I 111 11CGCCGTGGAG TACTCAGCTA AAAGGACTCC CAAGAGGGCG GCGGTCATGA TCGCGCTGGT GTGGGTCTTC TCCATCTCA TCTCGCTGCC GCCCTrCTTC TGGCGTCAGG CTAAGGC

FIG. 3. Comparison of guinea pig and human 5-HT1D receptor gene sequences. DNA amplified from the guinea pig trigeminal ganglion byusing primers W72/W73 was sequenced and compared with the two human 5-HTD receptor subtypes 5-HTlD and 5-HT1D8. The 20 bp on the5' and 3' ends of guinea pig sequence denoted by dots represent the W72 and W73 primers.

samples. Amplified DNA of the expected size was detectedin both specimens when PCR was performed with 5-HT1Da-specific primers, but not with 5-HTlD-specific primers (Fig.4, lanes A and B). Both sets ofprimers successfully amplifiedDNA of expected sizes from genomic DNA (lanes C and D)and did not amplify DNA fromRNA samples not treated withreverse transcriptase (lanes E and F). As a further positivecontrol, human striatal tissue, which is positive for both5-HT1Da and 5-HT1Dp receptor messages (2, 3), was used inthe PCR. DNA was successfully amplified from striatumcDNA using both 5-HTlDa and 5-HT1D9 receptor-specificprimers (Fig. 5).

DISCUSSIONThese studies provide direct evidence for the expression of aprejunctional 5-HT1D receptor in trigeminal ganglia. UsingPCR with primers directed against sequences conserved in5-HTwD and 5-HT1Dl receptors from several species (W72/W73), we identified the presence ofanmRNA with homologyto the human 5-HT receptors in guinea pig trigeminalganglia. The partial sequence reported here aligns with the5-HT1Da subtype as defined by the recent cloning and se-quencing of the guinea pig forms of several 5-HT1 receptorgenes (T. Branchek, personal communication). More com-plete studies to define the tissue distribution of the 5-HTmDreceptors in guinea pig can now be attempted by using PCRprimers specifically designed to both the 5-HTID. and5-HT1Dp receptor sequences of guinea pig. No conclusions

A B C D E F

369

246

123-

FiG. 4. Agarose gel electrophoresis of PCR-amplifled DNA fromhuman trigeminal ganglion. 5-HTID, and 5-HTmp gene sequenceswere amplified from human trigeminal ganglion cDNA and humangenomic DNA samples.. Lanes A and B, human cDNA, lanes C andD, humangenomic DNA; lanes E and F, humanRNA not transcribedinto cDNA; each with 5-HTlDr- and 5-HTw~p-specific primers,respectively. Products were separated by electrophoresis through a3% agarose gel containing ethidium bromide.

about the presence or absence of a 5-HTI1D receptor mRNAin guinea-pig trigeminal ganglia can be made without thesefuture studies.

Evidence for the expression of 5-HT1Da, but not 5-HTID,receptor mRNA was found in human trigeminal ganglia. Theantimigraine drug sumatriptan binds with high affinity to5-HT1Da and 5-HT1Dp recognition sites (1), suggesting that itsactions may well be mediated by one or both receptorsubtypes, depending upon the extent to which the two genesare expressed. However, the importance of other 5-HTreceptor subtypes which bind sumatriptan cannot be ex-cluded. For example, recently an intronless gene encoding a5-HT1F (5-HT1E) receptor subtype to which sumatriptanbinds with high affinity was isolated from a human genomiclibrary (19-21). Sumatriptan also binds to 5-HT1A and 5-HT1Breceptors, albeit with somewhat lower affinities (22, 23).The pharmacological properties of the trigeminovascular

5-HT receptor resembles most closely the 5-HT1D (guineapig) and 5-HT1B (rat) receptors among the defined 5-HT1subtypes (4, 24), with notable exceptions (5, 24). 5-HT1B and5-HT1D receptor subtypes are often located prejunctionallyor presynaptically and are negatively coupled to neurotrans-mitter release (25-28). Within the rat (and presumably thehuman) trigeminovascular system, prejunctional receptors towhich sumatriptan binds inhibit neuropeptide release (24),thereby blocking the development of neurogenic inflamma-

A B C D E F

FIG. 5. Agarose gel electrophoresis of PCR-amplifled DNA fromhuman striatum cDNA samples. Human striatal RNA was used togenerate cDNA for PCR amplification (lanes A and C). Sampleswithout reverse transcriptase (lanes B and D) were included tomonitor for genomic DNA contamination. Primers specific for5-HT1Du (lanes A and B) and 5-HTwD (lanes C and D) receptor geneswere used to amplify DNA from striatum cDNA samples, but notfrom samples generated without reverse transcriptase. The 5-HTDpgene sequence was also amplified from human genomic DNA (laneE), but not water (lane F). The rightmost lane contained a 100-bpladder.

3668 Neurobiology: Rebeck et al.

Dow

nloa

ded

by g

uest

on

July

26,

202

1

Proc. Natl. Acad. Sci. USA 91 (1994) 3669

tion. Consistent with this formulation, expression of a geneencoding the 5-HT1B receptor was found in rat trigeminalganglion cells by in situ hybridization (29). The rat 5-HT1Breceptor gene exhibits remarkable sequence homology to thehuman 5-HT1Dp receptor gene. Hence, the lack of 5-HT1D8receptor message in human trigeminal ganglia was unex-pected because of evolutionary considerations and becauseof results suggesting the existence of prejunctional 5-HT1Btrigeminovascular receptors in the rat that mediate the effectsof sumatriptan (4). However, the expression of differentreceptor isotypes mediating identical responses in cells fromdifferent species has been observed previously (30, 31) andestablishes precedent for the unanticipated results obtainedin human trigeminal ganglia. The mechanisms of action ofsumatriptan in rat may be clarified by studies of a newlydescribed 5-HTlD-like receptor (32, 33) which, on the basis ofsequence homology, tissue distribution, and pharmacology,is likely to be the species homolog of the human 5-HT1Dareceptor.

In summary, our present findings illustrate that the 5-HTprejunctional receptor mRNA expressed by human trigemi-novascular neurons is likely to be the 5-HT1Da and not the5-HT1D9 receptor subtype. Our results also suggest that thereceptor subtype mediating constriction of cerebral and cor-onary arteries (11-15) is distinct from the 5-HT receptorsubtype mediating blockade of neural transmission. Conse-quently, it may be possible to design therapeutically useful,subtype-specific drugs which do not constrict coronary orpial vessels (11, 34, 35). At the same time, these new drugswould establish the relative importance of neurogenic versusvascular mechanisms in migraine.

We wish to acknowledge J. J. Wenniger and J. Reiter for technicalassistance and the Department ofNeuropathology, Dr. J. Golden, D.Dewing, and J. Gamba for assistance in obtaining postmortem tissue.M.A.M. and K.I.M. were supported by National Institutes of HealthGrant NS 21558 and G.W.R. by National Institutes of HealthTraining Grant NS07009. B.T.H. acknowledges the support of theBrookdale Foundation.

1. Weinshank, R. L., Zgombick, J. M., Macchi, M. J., Branchek,T. A. & Hartig, P. R. (1992) Proc. NatI. Acad. Sci. USA 89,3630-3634.

2. Hartig, P. R., Branchek, T. A. & Weinshank, R. L. (1992)Trends Neurosci. 13, 152-159.

3. Jin, H., Oksenberg, D., Ashkenazi, A., Peroutka, S. J., Dun-can, A. M. V., Rozmahel, R., Yang, Y., Mengod, G., Palacios,J. M. & Dowd, B. F. (1992) J. Biol. Chem. 267, 5735-5738..

4. Matsubara, T., Moskowitz, M. A. & Byun, B. (1991) Br. J.Pharmacol. 104, 3-4.

5. Moskowitz, M. A. (1992) Trends Pharmacol. Sci. 13, 307-311.6. Saito, K., Levine, L. & Moskowitz, M. A. (1988) Ann. Neurol.

24, 732-737.7. Buzzi, M. G. & Moskowitz, M. A. (1990) Br. J. Pharmacol. 99,

202-206.8. Buzzi, M. G., Dimitriadou, V., Theoharides, T. C. & Mosko-

witz, M. A. (1992) Brain Res. 583, 137-147.

9. Moskowitz, M. A., Nozaki, K. & Kraig, R. P. (1992) J. Neu-rosci. 13, 1167-1177.

10. Nozaki, K., Moskowitz, M. A. & Boccalini, P. (1992) Br. J.Pharmacol. 106, 409-415.

11. Hamel, E. & Bouchard, D. (1991) Br. J. Pharmacol. 102,227-235.

12. Hamel, E., Fan, E., Linville, D., Ting, V., Villemure, J.-G. &Chia, L.-S. (1993) Mol. Pharmacol. 44, 242-246.

13. Willet, F., Curzen, N., Adams, J. & Armitage, M. (1992) Br.Med. J. 304, 1415.

14. Bateman, D. N. (1993) Lancet 341, 221-223.15. Ottervanger, J. P., Paalman, H. J. A., Boxma, G. L. &

Stricker, B. H. Ch. (1993) Lancet 341, 861-862.16. Chomczynski, P. & Sacchi, N. (1987) Anal. Biochem. 162,

156-159.17. Hamblin, M. W. & Metcalf, M. A. (1991) Mol. Pharmacol. 40,

143-148.18. Nguyen, T., Marchese, A., Kennedy, J. L., Petronis, A.,

Peroutka, S. J., Wu, P. H. & O'Dowd, B. F. (1993) Gene 124,295-301.

19. Adham, N., Kao, H.-T., Schechter, L. E., Bard, J., Olsen, M.,Urquhart, D., Durkin, M., Hartig, P. R., Weinshank, R. L. &Branchek, T. (1993) Proc. Natl. Acad. Sci. USA 90, 408-412.

20. Amlaiky, N., Ramboz, S., Boschert, U., Plassat, J.-L. & Hen,R. (1992) J. Biol. Chem. 267, 19761-19764.

21. Lovenberg, T. W., Erlander, M. G., Baron, B. M., Racke, M.,Slone, A. L., Siegel, B. W., Craft, C. M., Burns, J. E., Daniel-son, P. E. & Sutcliffe, J. G. (1993) Proc. Natl. Acad. Sci. USA90, 2184-2188.

22. Peroutka, S. J. & McCarthy, B. G. (1989) Eur. J. Pharmacol.163, 133-136.

23. Schoeffter, P. & Hoyer, D. (1989) Naunyn-SchmiedebergsArch. Pharmacol. 340, 135-138.

24. Buzzi, M. G., Moskowitz, M. A., Peroutka, S. J. & Byun, B.(1991) Br. J. Pharmacol. 103, 1421-1428.

25. Schoeffier, P., Waeber, C., Palacios, J. M. & Hoyer, D. (1988)Naunyn-Schmiedebergs Arch. Pharmacol. 337, 602-608.

26. Murphy, T. J. & Bylund, D. B. (1989) J. Pharmacol. Exp. Ther.249, 535-543.

27. Schlicker, E., Fink, K., Gothert, M., Hoyer, D., Molderings,G., Roschke, I. & Schoeffter, P. (1989) Naunyn-SchmiedebergsArch. Pharmacol. 340, 45-51.

28. Molderings, G. J., Werner, K., Likungu, J. & Gothert, M.(1990) Naunyn-Schmiedebergs Arch. Pharmacol. 342, 371-377.

29. Bruinvels, A. T., Landwehrmeyer, B., Moskowitz, M. A. &Hoyer, D. (1992) Eur. J. Pharmacol. 227, 357-359.

30. Ruffolo, R. R. & Waddell, J. E. (1982) Br. J. Pharmacol. 77,169-176.

31. Thomas, R. F. & Liggett, S. B. (1993) Mol. Pharmacol. 43,343-348.

32. Bach, A. W. J., Unger, L., Sprengel, R., Mengod, G., Pala-cios, J., Seeburg, P. H. & Voigt, M. M. (1993) J. Recept. Res.13, 479-502.

33. Hamblin, M. W., McGuffin, R. W., Metcalf, M. A., Dorsa,D. M. & Merchant, K. M. (1992) Mol. Cell. Neurosci. 3,578-587.

34. Humphrey, P. P. A. & Feniuk, W. (1991) Trends Pharmacol.Sci. 12, 444-446.

35. Ferrari, M. D. & Saxena, P. R. (1993) Trends Pharmacol. Sci.14, 129-133.

Neurobiology: Rebeck et al.

Dow

nloa

ded

by g

uest

on

July

26,

202

1