Proc. Natl. Acad. Sci. USA Vol. 92, pp. 9652-9656, October 1995 Cell Biology

TRPC1, a human homolog of a store-operated channel PAUL D. WEs*, JORGE CHEVESICH*, ANDREAS JEROMIN*, CARLA ROSENBERGt, GAIL STETTENt, AND CRAIG MONTELL* *Departments of Biological Chemistry and Neuroscience, The Johns Hopkins University, School of Medicine, Baltimore, MD 21205; and tDepartment of Gynecology and Obstetrics, The Johns Hopkins University, School of Medicine, Baltimore, MD 21287 Communicated by M. Daniel Lane, The Johns Hopkins University School ofMedicine, Baltimore, MD, July 17, 1995

ABSTRACT In many vertebrate and invertebrate cells, Several signaling mechanisms for store-operated Ca2+ entry inositol 1,4,5-trisphospate production induces a biphasic have been proposed (for review, see refs. 1, 3-5, 11). In one Ca2+ signal. Mobilization of Ca2+ from internal stores drives model, a diffusible messenger is produced or released upon the initial burst. The second phase, referred to as store- depletion of the intracellular Ca2+ pool, and this messenger operated Ca2+ entry (formerly capacitative Ca2+ entry), then activates the plasma membrane Ca2+ channels (12). A occurs when depletion of intracellular Ca2+ pools activates a second model suggests that emptying the intracellular Ca2+ non-voltage-sensitive plasma membrane Ca2+ conductance. stores causes a conformational change in the storage organelle Despite the prevalence of store-operated Ca2+ entry, no and/or its surface , such as the InsP3 receptor, which vertebrate channel responsible for store-operated Ca2+ entry is transmitted to plasma membrane Ca2+ channels either by has been reported. trp (transient receptor potential), a Dro- direct coupling or via the cytoskeleton (13, 14). Despite the sophila required in phototransduction, encodes the only prevalence of store-operated Ca2+ entry, no vertebrate SOC known candidate for such a channel throughout phylogeny. In gene has been identified. Elucidation of the mechanism re- this report, we describe the molecular characterization of a sponsible for store-operated Ca2+ entry would be considerably human homolog of trp, TRPCI. TRPC1 (transient receptor aided by cloning and characterizing the channels. potential channel-related 1) was 40%o identical to A variety of pharmacological and electrophysiological stud- Drosophila TRP over most of the protein and lacked the ies indicate that the Drosophila transient receptor potential charged residues in the S4 transmembrane region proposed to gene (tirp), a required in Drosophila vision, is required for be required for the voltage sensor in many voltage-gated ion function of a non-voltage-gated Ca21 channel with character- channels. TRPCI was expressed at the highest levels in the istics of a SQC (17, 18). Drosophila vision uses the inositol fetal brain and in the adult heart, brain, testis, and ovaries. phospholipid signaling system. In response to light-induced Evidence is also presented that TRPC1 represents the arche- inositol phospholipid hydrolysis, two types of channels, a type of a family of related human proteins. Ca2+-selective and a nonselective cation channel, are activated (for review, see ref. 19). In the tip mutant, an abiformal light Studies in many cell types, including human mast cells, T response is observed using an electroretinogram recording. A lymphocytes, parotid acinar cells, and oocytes, have implicated wild-type electroretinogram is characterized by a sustained corneal negative response during light stimulation. However, the formation of inositol 1,4,5-trisphosphate (InsP3) in a in the tip mutant, the electroretinogram response decays to biphasic rise of cytosolic Ca2+ (for review, see ref. 1). The baseline during prolonged bright illumination (20). Whole-cell initial burst results from the direct action of InsP3 on the InsP3- recordings of Drosophila photoreceptor cells indicate that gated Ca2W channels of intracellular Ca2+ stores. This primary there are light-dependent Na+ and Ca2+ plasma membrane response quickly subsides, followed by a more sustained rise of conductances and that the Ca2+ conductance is greatly re- cytosolic Ca2W concentration due to protracted Ca2+ entry duced in tip photoreceptor cells (17, 18, 21). Although most across the plasma membrane. Ins.P3 also appears to contribute permeable to Ca2+, the tip-dependent channel is also perme- to the second phase of the response because influx of extra- able to Mg2+, but not to Cs+ or Na+ (17). The tip-dependent cellular Ca2+ is initiated by intracellular application of InsP3 channel is not voltage-dependent and appears to be activated (2). by depletion of the internal Ca2+ stores (17, 22). Although the InsP3 appears to activate the plasma membrane Ca2+ con- cation selectivity suggests that the trp-dependent channel is not ductance indirectly. Many recent experiments indicate that identical to the CRAC channel, the apparent activation upon Ca2+ entry across the plasma membrane is coupled to the depletion of the internal Ca2+ pool indicates that the trp- depletion of intracellular Ca2+ stores by InsP3 (for review, see dependent channel is a SOC (17, 22). refs 1, 3-6) in a process referred to as store-operated Ca2+ The properties of the tip mutant indicate that trp either entry [formerly capacitative Ca2+ entry (7, 8)]t. This plasma encodes a non-voltage-gated plasma or a membrane Ca2+ conductance has been referred to as the Ca2+ protein required for activity ofthe channel. Molecular analyses release-activated Ca2+ current (Iaac), and the channel respon- of the tip gene product indicate that tip encodes a channel. sible for this current is known as the CRAC channel (5, 8). TRP is localized to the plasma membrane of photoreceptor Recent data indicate that there are cells, such as the A431 cells and contains multiple putative transmembrane domains epidermoid cell line, that express channels activated by deple- (15, 16, 23, 24). The overall structure of TRP is reminiscent of tion of Ca2+ stores but with characteristics distinct from those of the CRAC channel These channels have prototypical (10). Abbreviations: InsP3, inositol 1,4,5-trisphosphate; SOC, store- higher single channel conductances than CRAC channels, and operated channel; TRP, transient receptor potential protein; TRPC, the current is preferentially carried by Ba2+ rather than Ca2+. transient receptor potential channel-related protein; cTRP, Caeno- Thus, store-operated channels (SOCs) are not limited to rhabditis elegans TRP; TRPL, TRP-like protein; Icac, Ca2+ release- CRAC channels. activated Ca2+ current. CRAC, Ca2+ release-activated Ca2+ channel. Me generic terms store-operated Ca2+entry and store-operated channel (SOC) were adopted to replace the terms capacitative The publication costs of this article were defrayed in part by page charge entry and capacitative calcium entry channel, respectively, at payment. This article must therefore be hereby marked "advertisement" in the International Conference on Receptor Regulated Calcium Influx accordance with 18 U.S.C. §1734 solely to indicate this fact. at Asilomar, CA, May 14-18, 1995 (for review, see ref. 9). 9652 Downloaded by guest on September 28, 2021 Cell Biology: Wes et al. Proc. Natl. Acad. Sci. USA 92 (1995) 9653 known voltage- and ligand-gated ion channels (for review, see complete sequence of the 2.1-kb HFBEF25 cDNA. In addi- ref. 25). A gene encoding a Drosophila retinal-specific protein tion, we screened a human fetal brain library with the HF- with 39% amino acid identity to TRP, TRP-like (trpl), has been BEF25 clone and isolated several additional cDNAs, the identified in a screen for retinal calmodulin-binding proteins longest of which were 4.4 kb (clone HFBEF25B) and 2.9 kb (26). Recent studies in Sf9 cells suggest that TRPL is a (clone HFBEF25C). The combined sequences of these cDNAs nonselective cation channel stimulated by agonists of G pro- indicated that the mRNA was a minimum of 5.5 kb and tein-coupled receptors (27-29). In addition to TRP and TRPL, encoded an open reading frame of 810 amino acids (Fig. 1). a third TRP family member has been sequenced in another Comparison of the deduced amino acid sequence with the invertebrate, Caenorhabditis elegans, as part of the genome gene and protein data banks demonstrated significant homol- sequencing project (30). Each of the three invertebrate TRPs ogy to all three members of the TRP gene family. We found displays weak homology to known cation-channel proteins and 38-40% identity and 58-62% sequence similarity between the only in the six putative transmembrane domains. However, the human sequence and TRP, TRPL, and the C. elegans TRP charged residues believed to represent the voltage sensor in the (cTRP) over 624 amino acids (Fig. 1; TRPC1 residues 97-720). S4 membrane-spanning domain of voltage-gated ion channels In addition, there was a second region near the C terminus that see ref. are not conserved. was '30% identical and 60% similar to each of the TRP fatnily (for review, 31) proteins. Overall, the human gene was as similar to the The body of molecular and electrophysiological data argue that invertebrate TRP family as the invertebrate family TRP is a non-voltage-gated plasma Ca2+ channel and suggests members were to each other. The only exceptions were four that TRP might be the Drosophila homolog of one of the SOCs small gaps of 9-34 amino acids in the human protein (begin- characterized electrophysiologically and pharmacologically in ning after residues 158, 425, 499, and 600) relative to Dro- vertebrate cells. Here, we report the molecular characteriza- sophila TRP and TRPL (Figs. 1 and 2D). Three of the four gaps tion of a human homolog of trp, TRPC1, which encodes a occurred in similar positions in cTRP. Due to the high level of product that is '40% identical to the invertebrate TRPs.§ In similarity to the TRP family members, this human gene will be addition, we provide evidence that TRPC1 gene is not the only referred to as TRPCJ (transient receptor potential channel- TRP gene in humans but is the archetypal member of an related 1). additional family of cation channels, some of which may be Domains in TRPC1. The sequence identity between TRPC1 SOCs. and the invertebrate TRP family proteins was significantly higher in the putative cytoplasmic regions (73% residues were MATERIALS AND METHODS identical with at least one other family member) and trans- membrane segments (62%) than in the putative extracellular cDNA and Genomic Library Screening. Genomic and cDNA loops connecting the putative transmembrane domains (31%) clones were isolated by screening filter lifts from AEMBL3 (Fig. 2D). The cytoplasmic region near the N terminus of human genomic and AZAP human fetal brain cDNA libraries TRPC1 (residues 98-368) was most conserved (80%), and the (Stratagene), respectively. domain C-terminal to the membrane-spanning segments was DNA Sequencing. Samples were sequenced using the fluo- 66% identical between TRPC1 (residues 664-720) and at least rescent dideoxynucleotide terminator method of cycle se- one other TRP family member. The lowest conservation in the quencing (32) on a Perkin-Elmer, Applied Biosystems Divi- putative cytoplasmic region was in the two loops connecting sion 373a automated DNA sequencer (33). The composite of the putative transmembrane segments (50%). three cDNAs, HFBEF25, -25B, and -25C, was 5469 bp. The N-terminal region of TRPC1 contained two motifs. The Fluorescent in Situ Hybridization. Slides of banded human first was an ankyrin motif repeated three times in tandem near metaphase spreads were hybridized with biotinylated Ahtrpl-2 the N terminus of all TRP family members (38). A fourth DNA, the hybridization signal was detected, and slides were ankyrin repeat was present in the invertebrate TRPs but not in counterstained all as described (34). TRPC1. The second domain, residues 256-300, was predicted RNA Blot Analysis. RNA blots were purchased from Clon- to form a coiled-coil structure (84% probability) on the basis tech. Fetal RNA was pooled from various developmental of an algorithm designed to identify such motifs (Fig. 2 A, C, stages (19-26 weeks). RNA blots were probed with linearized andD) (36). This coiled-coil region was also predicted to occur HFBEF25 and reprobed with a 2.0-kb B3-actin cDNA. in Drosophila TRP, although with low probability in TRPL and cTRP. In Situ Hybridizations to Tissue Sections. Sections of frozen had lower rat embryos (13-14 weeks of development) were probed with The C-terminal 90 amino acids of TRPC1, which RNA from linearized HFBEF25 identity to the invertebrate TRPs, showed significant homol- 35S-labeled probes, prepared ogy to dystrophin [P(N) = 1.5 x 10-5; residues 718-809; Fig. and hydrolyzed to fragments of "200 bases by alkaline hydro- 2D]. Included in the dystrophin domain was a region with a lysis as described (35). Selected slides were treated with high probability (62%) of forming a coiled-coil structure photographic emulsion (Kodak NTB2). (residues 760-795; Fig. 2 A, C, and D). The N-terminal 95 amino acids of TRPC1 was not homologous to any protein in RESULTS the data banks. Hydrophobicity analyses indicated the presence of eight Isolation of Human cDNAs from a Fetal Brain Library hydrophobic regions of sufficient length to span the membrane Homologous to trp. To identify a candidate human homolog of (Fig. 2B; ref. 39). However, two of the hydrophobic regions in the Drosophila gene tip we searched the National Center for TRPC1 (the third and seventh), were not predicted to form Biotechnology Information Expressed Sequence Tag data sufficiently long a-helices in all the TRPs to span the mem- base. One sequence (EST 05093; clone HFBEF25) from a brane. None of the remaining six transmembrane segments human fetal brain cDNA library contained an open reading (S1-S6), including the S4 segment, had the positively charged frame with 50% identity to a 48-amino acid portion of the residues that appear to constitute the voltage sensor of many 1275-amino acid TRP sequence [residues 75-122; smallest voltage-gated cation channels. Poisson probability score, P(N), was 0.00047]. To determine Chromosomal Localization of TRPCI. To map TRPCI, whether this human gene was a homolog oftrp, we obtained the fluorescent in situ hybridization on metaphase spreads was performed by using a 17-kb clone, Ahtrpl-2, §The sequences reported in this paper have been deposited in the isolated from a genomic library. Ahtrpl-2 was partially se- GenBank data base (accession nos. X89066, X89067, and X89068). quenced to confirm its identity (data not shown). Hybridiza- Downloaded by guest on September 28, 2021 9654 Cell Biology: Wes et al. Proc. Natl. Acad. Sci. USA 92 (1995)

hTRPC1 MCPGIPGPRAEAAVGTTHPFSSPGAWLGSGSGSGPVGAPPPSPGLPPSWAAMMAALYPSTDL SGASSSSLPSSPSSSSPNEVMALKDVREVKEENT 96 hTRPC1 dTRP DYICQK ., 161 HNIEVG. . DALLHAISEEYVEAVEELLQWEETNHKEG 126 dTRPL VMEV.ETK. DALLHAINAEFVEAVELLLEHEELIYKEG 135 cTRP QQPDIRIGNALLCAIREGVYRLVEVLVNHPNITREML 163 _11

hTRPCI .. .LMM RIQNPYSTI v F L0 CMEE E R 253 dTRP QPYSWUAVDRSKSTF .Y :TI. Ill I 5 T 221 dTRPL EPYSW'QKVOINTAMFA i. 5 .A iC ASi. SREVU T. T F S YSN A CESN T 230 cTRP GDGWSQALDPSEAASAEYSS I I1t5' QL Q N WMs SP 261 11 hTRPC1 M. e= AA .. ErM SU rM,.. K.--. Y =-..N NTS------SD LD K R GL IEMUM,IB VBVU NDI-el tawmjl;fiQBT R. F E-WSN. l F |NT Y 35 1 dTRP W ER Q A S4 A E T RQMVQD TS D. T M 1FNH. .1SHDIWC1G T:T EHTAAIL 317 dTRPL 3W. N An QSEK-E M 1 R1 KEV T!T EAYI .K. Q M SS Y E G Y H L 3 2 4 cTRP iKWl lQR lA FE H K ET1 Lit E=QYSCSCI VIAC KDGNVNDDNIDVWAS. KLS E H L TS' 358 hTRPC1 FFGQMS PCKIITL TL I FFIH G F .. V LEQ LNY VN ...... ElDK.NT 432 dTRP Kos LPE-QASQ.QLEDEtKC..SX...... V tDEGAKF.RPKS MF M GAA RVVQITFELLAFPWMLTMLEIDWRlHE 415 dTRPL YDGLP F SI VDfV.CIAQUAVLCLYCIAQVCE LRTLJLNRK. L AS LFFIF I QRAODDFVRIFGTTRMKKELAEQELRQ 422 CTRP YE GPrP rEIFC I FYYS VFA G QSGIWANFFLYAFLLFL LMERLELvRs. TWATFEDY ...... RYEKGERGGMTRASD 448

hTRPC1 MEALER 1DYE L R. Y ED FiE ES QISo AI F A KV.HN K D ...... FAD DA H'T F 5 2 1 dTRP RESLPGP ELA T EEL A I. D.YlS TWILCAAT WVI'RDLWFRGIOPYF E H 'A 513 dTRPL R TPSKLELIVVMYFY . E EIFAV KSYRNMW F1D L.R HAF IQQD D Q MAY'I E HD A 520 cTRP I C REPATWVESIVFTFTCVVLEEOEEFKRYFRQWWWMRWWe CUISRLSSIITY.Y . REDPYRYTV T Y T:S E E M V 5 4 3 hTRPC1 _sFFMyToSI 3 QB{FLGI3L FL|EDK|TSKEq3YG I FC E4QQNDT ...... HIF IG|iC 608 dTRP UV HI.Fl INPH L I I I IF F.IYTAUGCENEL. . WYAELEMKCYVHLHPDVAOFDDQEKACTIWRRSNLFEgS 6 0 9 dTRPL FflAflKVHLFSINPH _ L VI IV FFIYT AC N. . .WIFAALEISKCYVLPGGEADWGSHGDSCMKWRRDGNLFEESS 616 cTRP FI GEFFAIYLFQGNPYILCVV V .AWYDPNTDViLPGATCKHES:MV ...... S5IADSY 628 I hTRPC1 F YI AHVUIEVTRFEGEREL S AsI V V T K 7 0 6 dTRP QS A S VDLVSf1 DLAGIKS FTR WALLMF rS N L N 1 7 0 5 dTRPL QS AS MVGL DDf. ELfGIKSYTR LLMF S I N LIN I 7 1 2

cTRP LT L SL ITKPEDT. . DVVENHKITQW QGMFIM THCTSI LNN VL V 7 2 4 hTRPC2 Q F TM Y.NEEH.VVEE .DVPQFLVPE RA LlYEIFI IM LNM hIRPC3 TSVV . .. LKEDHKFIENIEYfLYEI TMUYVMLNMIJISINS"Y EIEDDSJV *'6SXlMGKKl 1VL s

hTRPCl TIC5MISSRE SHTMGIVIRQUOSLKEWRNL ...... KI.DE....

hTRPC1 FRN EI RDL L G FRTSKYAMFYPRN 8 1 0

FIG. 1. Deduced amino acid sequence of TRPC1 and alignment to other TRPs. The running tally of amino acids, shown as single-letter codes, is indicated on the right. Ankyrin repeats and putative transmembrane segments (S1-S6) are indicated by double and single underlines, respectively. Amino acids identical between TRPC1 and at least one other member of the TRP family are indicated with white lettering on a black background, whereas amino acids that are similar (defined as nonidentical amino acids with a score >0 on a BLOSUM-62 scoring matrix) between TRPC1 and at least one other member of the gene family are boxed in grey. The asterisk in the TRPC2 sequence (corresponding to residue 690 in TRPC1) indicates a stop codon. dTRP, Drosophila TRP; dTRPL, Drosophila TRPL. tion was detected in 11 metaphases, in the middle of the long base with the BLAST algorithm using different TRP family arm (q) of . An additional four cells showed members and found two sequences (T67673 and R34716) hybridization signals in the metaphases, previously docu- obtained from clones (66773 and 37199) isolated from human mented by G-banding (Fig. 3). No consistent signal was fetal liver/spleen and fetal brain cDNA libraries, respectively, identified on any other chromosome. We assigned the Ahtrpl-2 with homology to the TRP family [smallest P(N), which was to hybridization signal to 3q22-q24. No likely candidate diseases TRPL, was 3.7 x 10-11 and 2.0 x 10-18, respectively]. Clones map to this position. 66773 and 37199 contained 100- and 85-amino acid regions, Tissue Distribution of TRPC1. To determine which human which were 37-47% and 40-49% identical to various members fetal and adult tissues express TRPC1, we probed RNA blots of the TRP family, respectively (Fig. 1). Due to the homology containing poly(A)+ RNA prepared from various tissues. We to the TRP family, these genes are referred to as TRPC2 and found that TRPC1 was expressed as a single mRNA species TRPC3. TRPC2 may be a pseudogene because a stop codon, that migrated at 5.4 kb (Fig. 4). During fetal development, corresponding to residue 690 in TRPC1, was present in two TRPC1 was expressed at the highest level in the brain and at cDNAs isolated from different cDNA libraries. TRPCI and lower levels in the liver and kidneys (Fig. 4A). In situ hybrid- TRPC2 do not form a gene cluster because TRPC2 mapped to izations (on fetal rats) were consistent with these findings (Fig. llpl5.3-15.A (data not shown). In contrast to TRPC2, TRPC3 5). A strong signal was detected in the telencephalic, dience- appears to represent a second bona fide human member of phalic, and hindbrain regions of the embryonic brain, and the TRP family. weaker signals were detected in the heart and the liver/spleen/ gastrointestinal tract anlagen (Fig. 5). No signals were detected DISCUSSION in embryos labeled with sense control probes (data not shown). In the adult, TRPCI was expressed at the highest levels in the The TRP Family of Genes Is Conserved from Worms to heart, testes, ovaries (Fig. 4 B and C), and in many regions of Humans. Ca2+ influx stimulated by the inositol phospholipid the brain (Fig. 4 C and D). signaling system occurs in many cell types in vertebrates and Identification of Additional Putative Human Homologs of invertebrates (for review, see refs. 1, 3-6, 40). Many studies, TRP. We recently queried the Expressed Sequence Tag data primarily in various human cell lines, indicate that the non- Downloaded by guest on September 28, 2021 Cell Biology: Wes et al. Proc. Natl. Acad. Sci. USA 92 (1995) 9655

A = 0.- A B C D .0 0.6+1 .0co B. 03-. FIG 4. R bl analysiso T-R_C1 (A) Hm fea R ot. 200 400 600 8Oa

0. l 0. wags reprobed with;a 13-actin probe, showing that th re eve_nlym :S0.

200 400 600 tla no_- 80( C THPUl ....- , ..M-1IM mm11 I I111mmm 11 Im 11 L]11 FIG. 4. RNA blot analysisof TRPC1. (A) Human fetal RNA blot. =.!!!!!!!!!!ll Hi,Sfffflli-i I (B) Human adult multiple tissue RNA blot m. (C) Human adult dTRi II ...... n.gi. mil I1NEl"M IM. mmi1I11111mmI I II1m IIi*i;I I multiple tissue RNA blot ne.(D) Human adult brain multiple tissue dTRPL C~m _ I W A RNA blot. The signal migrated at 5.4 kb on each RNA blot. Each filter ffp m was reprobed with a ,B-actin probe, showing that the lanes were evenly cTRP MM- i -MM IIn loaded (data not shown). amn., Amygdala; c.c., corpus callosum; c.n., I 1--II - caudate nucleus; hip., hippocampus; p.b.l.; peripheral blood leuko- 0 200 400 600 800 1200 cytes; s.i., small intestines; s.m., skeletal muscle; s.nig., substantia nigra; s.nuc., subthalamilc nucleus; w.b., whole brain. D 31% tially activated by depletion of Ca2+ from the internal stores 62% [T (15, 18, 22). Because Ca2+ channels with electrophysiological 73% Ac characteristics reminiscent of TRP have been observed in many human cells, it is possible that one or more human N homologs of TRP mediate these conductances. 80% 50% 66% TRPC1 appears to be a homolog ofDrosophila TRP because cyt. N-term. cyt. loops cyt. C-term. it is "40% identical to the invertebrate TRPs over 624 amino FIG. 2. Putative domain structure of the TRP family. (A) Predicted acids. This region contains the ankyrin repeats and the six coiled-coil characteristic of TRPC1. The analysis was done, with a putative transmembrane domains. Furthermore, TRPC1 was window size of 28 amino acids (36). (B) The Kyte-Doolittle hydro- as similar to the Drosophila and C. elegans TRP family mem- philicity profile of TRPC1 was generated with a window size of 12 bers as the invertebrate TRPs were to each other. A critical amino acids. (C) Putative domain structure of the TRP family. feature shared between TRPC1 and the other TRP family Identical shading reflects sequence conservation among all members members is the absence of charged residues at every third or (dTRP, Drosophila TRP; dTRPL, Drosophila TRPL). Divergent do- fourth position in the putative S4 transmembrane segment mains are indicated with dissimilar shading patterns. The amino acid scale refers to TRPC1. ank, Ankyrin repeat; cc, coiled-coil domain; proposed to be necessary for the voltage sensor in many 1-6, putative transmembrane regions. (D) Putative domain structure voltage-gated ion channels. The absence of these charged and topology of TRPC1. Percent identity between TRPC1 and at least residues is consistent with the observations that Drosophila one other member of the TRP family is indicated: 31%, extracellular TRP and TRPL are not voltage gated and suggests that TRPC1 loops; 62%, transmembrane segments (S1-S6); 73%, conserved in- is also not voltage gated. The identification of the TRPCI gene tracellular regions (amino acids 97-738); 80%, conserved cytoplasmic demonstrates that the TRP ion-channel family is conserved (cyt.) N terminus; 50%, cytoplasmic loops; 66%, conserved cytoplas- throughout the animal kingdom, from primitive metazoan mic C terminus. A putative pore loop is indicated between transmem- brane segments S5 and S6 (for review, see ref. 37). This region was organisms, such as nematodes, to humans. treated as an extracellular loop when considering percent identities. Conservation of the Ankyrin Repeats in the TRP Family Arrowheads indicate small gaps present in TRPC1 relative to some of Suggests a Critical Role. Ankyrin is an adaptor molecule the other TRP family members. dys, Dystrophin domain. linking a diversity ofplasma membrane proteins to cytoplasmic proteins through a repeated 33-amino acid motif (for review, voltage-gated plasma membrane Ca21 channels responsible for store-operated Ca2+ entry are activated after depletion of di Ca21 from the cytoplasmic storage compartments. However, te: hi the molecular mechanisms underlying this process remain elusive. Molecular and electrophysiological studies have previously defined Drosophila TRP as the first member of a new class of non-voltage-gated Ca2+ channels some of which are poten-

( '410

FIG. 3. Fluorescent in situ hybridization of TRPC1. (A) Fluores- FIG. 5. TRPCJ expression in rat embryos. A 20-lim section of a rat cent in situ hybridization of a partial human metaphase chromosome embryo (embryonic day 13-14) probed with HFBEF25 antisense spread. The TRPC1 gene signals from sister chromatids on both cRNA and developed in emulsion after 12 days. di, Diencephalic 3 (3q22-q24) are indicated with arrows. (B) G-banded region; g, gastrointestinal system; h, heart sac; hi, hindbrain; tel, preparation of the same chromosome spread as in A. telencephalic region. Downloaded by guest on September 28, 2021 9656 Cell Biology: Wes et al. Proc. Natl. Acad. Sci. USA 92 (1995) see ref. 41). In addition, ankyrin repeats occur in a number of 9. Clapham, D. E. (1995) Nature (London) 375, 634-635. plasma membrane proteins and may function in protein- 10. Luckhoff, A. & Clapham, D. E. (1994) Biophys. J. 67, 177-182. protein interactions (41). The domain that is most highly 11. Putney, J. W., Jr. (1993) Science 262, 676-678. conserved among all TRP family members (amino acids 12. Putney, J. W., Jr. (1990) Cell Calcium 11, 611-624. 97-367) contains three ankyrin repeats and a coiled-coil 13. Irvine, R. F. (1990) FEBS Lett. 263, 5-9. 14. Berridge, M. J. (1990) J. Biol. Chem. 265, 9583-9586. region, although an additional ankyrin repeat is present in each 15. Montell, C. & Rubin, G. M. (1989) Neuron 2, 1313-1323. of the invertebrate TRPs (Fig. 3 C and D). Because InsP3 16. Wong, F., Schaefer, E. L., Roop, B. C., LaMendola, J. N., John- receptors expressed in a number of mammalian tissues bind son-Seaton, D. & Shao, D. (1989) Neuron 3, 81-94. directly to ankyrin (42, 43), it is possible that the ankyrin 17. Hardie, R. C. & Minke, B. (1992) Neuron 8, 643-651. repeats of the TRPs also bind to InsP3 receptors. Thus, the 18. Peretz, A., Sandler, C., Kirschfeld, K., Hardie, R. C. & Minke, B. ankyrin repeats in the TRPs may mediate activation of the (1994) J. Gen. Physiol. 104, 1057-1077. channels by enabling the TRPs to respond to a proposed 19. Hardie, R. C. & Minke, B. (1993) Trends Neurosci. 9, 371-376. conformational change in the InsP3 receptor upon depletion of 20. Cosens, D. J. & Manning, A. (1969) Nature (London) 224, Ca2+ from internal stores (for review, see ref. 19). Alterna- 285-287. tively, the ankyrin repeats may mediate a signal transduced 21. Peretz, A., Suss-Toby, E., Rom-Glas, A., Arnon, A., Payne, R. & through the cytoskeleton upon depletion of the Ca2+ pool. Minke, B. (1994) Nature (London) 12, 1257-1267. 22. Vaca, L., Sinkins, W. G., Hu, Y., Kunze, D. L. & Schilling, W. P. Spectrin, which binds both ankyrin and calmodulin, may (1994) Am. J. Physiol. 266, C1501-C1505. undergo a conformational change in response to a rise in 23. Montell, C., Jones, K, Hafen, E. & Rubin, G. (1985) Science 230, cytosolic Ca2+ concentration (44). Ca2+-dependent changes in 1040-1043. spectrin could then lead to TRP activation via binding of 24. Pollock, J. A., Assaf, A., Peretz A., Nichols, C. D., Mojet, M. H., spectrin to the TRP ankyrin repeats. Hardie, R. C. & Minke, B. (1995) J. Neurosci. 15, 3747-3760. TRP May Constitute a Multigene Family in Humans. If 25. Jan, L. Y. & Jan, Y. N. (1992) Cell 69, 715-718. human homologs of TRP are SOCs, then TRPC1 would not be 26. Phillips, A. M., Bull, A. & Kelly, L. E. (1992) Neuron 8,631-642. expected to be the only such channel in humans because it was 27. Hu, Y., Vaca, L., Zhu, X., Birnbaumer, L., Kunze, D. L. & not detected in several adult tissues, such as the liver and Schilling, W. P. (1994) Biochem. Biophys. Res. Commun. 201, peripheral blood leukocytes, in which store-operated Ca2+ 1050-1056. entry has been demonstrated. Furthermore, more than one 28. Hu, Y. & Schilling, W. P. (1995) Biochem. J. 305, 605-611. has in humans see 29. Harteneck, C., Obukhov, A. G., Zobel, A., Kalkbrenner, F. & type of SOC been characterized (for review, Schultz, G. (1995) FEBS Lett. 358, 297-300. ref. 1). Consistent with the proposal that humans express 30. Wilson, R., Ainscough, R., Anderson, K., Baynes, C., Berks, M., multiple TRPC genes, we have recently identified a second et al. (1994) Nature (London) 368, 32-38. human gene, TRPC3, which appears to encode a protein with 31. Hille, B. (1991) Ionic Channels ofExcitable Membranes (Sinauer, homology to the other TRPs. It will be interesting to determine Sunderland, MA). the full size of the human TRP family and whether the human 32. McCombie, W. R., Heiner, C., Kelly, J. M., Fitzgerald, M. G. & TRPs have different functional characteristics, as do Drosoph- Gocayne, J. D. (1992) DNA Sequence 2, 289-296. ila TRP and TRPL. 33. Smith, L. M., Sander, J. Z., Kaiser, R. J., Hughes, P., Dodd, C., Connel, C. R., Heiner, C., Kent, S. B. & Hood, L. E. (1986) This work was supported by a grant from the National Eye Institute Nature (London) 321, 674-679. (EY10852) to C.M. P.D.W. is a Howard Hughes Medical Institute 34. Rosenberg, C., Blakemore, K. J., Kearns, W. G., Giraldez, R. A., Predoctoral Fellow. C.R. acknowledges support from the National Escallon, C. S., Pearson, P. L. & Stetten, G. (1992) Am. J. Hum. Institutes of Health (HD24605). Genet. 50, 700-705. 35. Lyford, G. L., Yamagate, K, Kaufmann, W. E., Barnes, C. A., 1. Fasolato, C., Innocenti, B. & Pozzan, T. (1994) Trends Pharma- Sanders, L. K., Copeland, N. G., Gilbert, D. J., Jenkins, N. A., col. Sci. 15, 77-83. Lanahan, A. A. & Worley, P. F. (1995) Neuron 14, 433-445. 2. Bird, G. St. J., Rossier, M. F., Hughes, A. R., Shears, S. B., 36. Lupas, A., Van Dyke, M. & Stock, J. (1991) Science 252, Armstrong, D. L. & Putney, J. W., Jr. (1991) Nature (London) 1162-1164. 352, 162-165. 37. MacKinnon, R. (1995) Neuron 14, 889-892. 3. Berridge, M. J. (1993) Nature (London) 361, 315-325. 38. Bork, P. (1993) Proteins 17, 363-374. 4. Putney, J. W., Jr., & Bird, G. St. J. (1993) Cell 75, 199-201. 39. Kyte, J. & Doolittle, R. F. (1982) J. Mol. Biol. 157, 105-132. 5. Penner, R., Fasolato, C. & Hoth, M. (1993) Curr. Opin. Neuro- 40. Clapham, D. E. (1995) Cell 80, 259-268. biol. 3, 368-374. 41. Bennett, V. (1992) J. Biol. Chem. 267, 8703-8706. 6. Hoth, M., Fasolato, C. & Penner, R. (1993)Ann. N.Y Acad. Sci. 42. Joseph, S. K. & Samanta, S. (1993)J. Biol. Chem. 268,6477-6486. 707, 198-209. 43. Bourguignon, L. Y. W., Jin, H., Iida, N., Brandt, N. R. & Zhang, 7. Putney, J. W., Jr. (1986) Cell Calcium 7, 1-12. S. H. (1993) J. Biol. Chem. 268, 7290-7297. 8. Hoth, M. & Penner, R. (1992) Nature (London) 355, 353-356. 44. Matsudaira, P. (1991) Trends Biochem. Sci. 16, 87-92. Downloaded by guest on September 28, 2021