Proc. Natl. Acad. Sci. USAVol. 88, pp. 1095-1099, February 1991Cell Biology

Isolation of an additional member of the fibroblast growth factorreceptor family, FGFR-3

(tyrosine kinase growth factor receptor/activation by acidic and basic fibroblast growth factors/K-562 cells/cDNA)

KATHLEEN KEEGAN*, DANIEL E. JOHNSONt, LEWIS T. WILLIAMSt, AND MICHAEL J. HAYMAN*t*Department of Microbiology, State University of New York at Stony Brook, Stony Brook, NY 11794; and tDepartment of Medicine, Howard HughesMedical Institute, Program of Excellence in Molecular Biology, Cardiovascular Research Institute, University of California, San Francisco, CA 95143

Communicated by William J. Lennarz, November 5, 1990

ABSTRACT The fibroblast growth factors are a family ofpolypeptide growth factors involved in a variety of activitiesincluding mitogenesis, angiogenesis, and wound healing.Fibroblast growth factor receptors (FGFRs) have previouslybeen identified in chicken, mouse, and human and have beenshown to contain an extracellular domain with either two orthree immunoglobulin-like domains, a transmembrane do-main, and a cytoplasmic tyrosine kinase domain. We haveisolated a human cDNA for another tyrosine kinase receptorthat is highly homologous to the previously described FGFR.Expression of this receptor cDNA in COS cells directs theexpression ofa 125-kDa glycoprotein. We demonstrate that thiscDNA encodes a biologically active receptor by showing thathuman acidic and basic fibroblast growth factors activate thisreceptor as measured by 45Ca2 efflux assays. These dataestablish the existence of an additional member of the FGFRfamily that we have named FGFR-3.

Acidic fibroblast growth factor (aFGF) and basic fibroblastgrowth factor (bFGF) are members of a family of multifunc-tional polypeptide growth factors that have been shown tostimulate proliferation of cells of mesenchymal, epithelial,and neuroectodermal origin. They also play a role in otherprocesses, including angiogenesis, promotion of differentia-tion of adipocytes and neurons, and would healing (1-3). Inaddition to aFGF and bFGF, five other members of thefibroblast growth factor (FGF) family have been identified.These include the oncogenes int-2 (4) and hst/Kaposi FGF(5), and three others-FGF-5 (6), FGF-6 (7), and keratinocytegrowth factor (8), which may play roles in normal growth anddevelopment.

Recently, the human, chicken, and mouse FGF receptors(FGFRs) have been shown to be members of the tyrosinekinase receptor family (9-15). The FGFRs are predicted toencode proteins with an extracellular domain containingeither two or three immunoglobulin-like domains, a trans-membrane domain, and a cytoplasmic tyrosine kinase do-main. The FGFR was initially purified from chicken embryoas a receptor that bound bFGF (9). The receptor cDNA wascloned by using oligonucleotide probes based on this proteinsequence. The extracellular region of the chicken FGFR waspredicted to encode three immunoglobulin-like domains (9).The gene for human FGFR was shown to encode multipleforms ofthe receptor, including one in which the extracellularregion contains the coding potential for two immunoglobulin-like domains and one with the coding potential for threeimmunoglobulin-like domains (12). These different formsappear to be generated by alternatively spliced mRNAs (12).Two similar forms of the receptor have been isolated frommouse cDNA libraries (13-15). Subsequent characterizationof the protein products of the human and chicken receptor

A

9.5-7.5- A4.4-

.42.4 -.,,1.-

..;s

B 1 G AT CPZ - -

A- '- _---

_ _S

FIG. 1. (A) Northern blot of K-562 RNA hybridized with clone17B. A major mRNA of 4.5 kb and a minor mRNA of 7 kb are seenin these cells. Sizes of molecular size standards, in kb, are indicatedat left. (B) Primer extension of clone 17B. An oligonucleotide thathybridized near the 5' end of clone 17B was hybridized to K-562RNA, extended with reverse transcriptase, and electrophoresed on5% sequencing gel next to a sequencing ladder (lanes G, A, T, andC). The 175-nucleotide extension product is indicated in lane 1 by thearrow.

genes has demonstrated that both aFGF and bFGF bind (12)and stimulate tyrosine kinase activity of these receptors (14,16). The mouse FGFR tyrosine kinase activity is stimulatedby Kaposi FGF in addition to bFGF (15). These results raisethe question of whether all members of the FGF family sharea common cell-surface receptor or interact with distinctreceptors. In this report, we describe the isolation andcharacterization of a gene§ that is very similar to, yet distinctfrom, the previously described FGFR-encoding gene. Fur-thermore, we demonstrate that the protein product of thisgene can be activated by both aFGF and bFGF, thus estab-lishing the existence of an additional member of the FGFRfamily.

MATERIALS AND METHODScDNA Cloning. A human K-562 cDNA library (from Owen

Witte, University of California, Los Angeles; ref. 17) washybridized under low-stringency conditions [50% (vol/vol)formamide/5x SSC (1x SSC is 0.15M sodium chloride/0.015M sodium citrate, pH 7)/25 mM sodium phosphate, pH6.5/1x Denhardt's solution (lx Denhardt's solution is 0.02%polyvinylpyrrolidone/0.02% Ficoll/0.02% bovine serum albu-min) and salmon sperm DNA at 250 ,ug/ml/10% (wt/vol)dextran sulfate for 18 hr at 38°C]. The entire v-sea gene was[a-32P]dCTP-labeled with a random primer kit (BoehringerMannheim) and used as probe. Filters were washed once for15 min in 1x SSC/0.1% SDS at 42°C and twice for 20 min in

Abbreviations: FGF, fibroblast growth factor; aFGF, acidic FGF;bFGF, basic FGF; FGFR, FGF receptor.iTo whom reprint requests should be addressed.§The sequence reported in this paper has been deposited in theGenBank data base (accession no. M58051).

1095

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

Mar

ch 2

4, 2

020

1096 Cell Biology: Keegan et al. Proc. Natl. Acad. Sci. USA 88 (1991)

0.2x SSC/0.1% SDS at 42°C. Full-length clones were then phate, pH 6.5/1x Denhardt's solution/salmon sperm DNAisolated by screening at high stringency (using the hybrid- at 250 AgIml/l0o (wt/vol) dextran sulfate, 42°C] for 18 hr andization buffer described above at 42°C for 18 hr) with a washed at O.1x SSC/0.1% SDS at 65°C for 2 hr. An RNA[a-32P]dCTP-labeled partial cDNA clone of FGFR-3 as ladder (BRL) was used for molecular size standards.probe. The cDNAs were subcloned into pBSKS+ (Strata- Primer Extension. An oligonucleotide, 5'-GACGGGCAG-gene) and sequenced using the dideoxynucleotide chain- CGTGCCTGC3' beginning 80 base pairs (bp) downstreamtermination method (18). The putative hydrophobic signal from the 5' end of clone 17B was labeled with [,y-32P]ATP andsequence and transmembrane sequence were identified by hybridized to 40 ,ug ofRNA at 30°C in 50% formamide/20 mMKyte and Doolittle hydropathy analysis (19). Pipes, pH 6.5/0.9 M NaCI/1 mM EDTA. Primer extensionNorthern (RNA) Blot Analysis. mRNAfrom K-562 cells was was done as described (20), and the products were electro-

prepared by using the guanidinium hydrochloride/CsCl phoresed on a 5% sequencing gel next to a sequencing ladder.method (20). Ten micrograms of total RNA was loaded on a Transient Expression and Immunoprecipitation. Clone 17Bformaldehyde gel, and a Northern blot was done as described was subcloned into the simian virus 40- based expression(21). The blot was hybridized with a [a-32P]dCTP-labeled vector pMT2 (22, 23). Electroporation into COS cells wasprobe derived from the kinase domain of FGFR-3 at high done at 675 V/cm. Sixteen hours later, cells were treated withstringency [50% formamide/5 x SSC/25 mM sodium phos- increased amounts of tunicamycin (0, 2.5, and 5.0 ,g/ml) for

1 CGCGCGCTGCCTGAGGACGCCGCGGCCCGCCCGCCATGGGCGCCCCTGCCTGCGCCTCGCGCTCTGCGTGGccGTGGccArcGTGGCCGGCGCCTCCTCGGAGTCCTTGGGGACG 120M G A P A C A L A L C V A V A I V A G A S S E S L G T 27

121 GAGCAGCGCGTCGTGGGGCGAGCGGCAGAAGTCCCGGGCCCAGAGCCCGGCCAGCAGGAGCAGTTGGTCTTCGGCAGCGGGGATGCTGTGGAGCTGAGCTGTCCCCGCCGGGGGTGGT 24028 E Q R V V G R A A E V P G P E P G 0 Q E 0 L V F G S G D A V E L S C P P P G G G 67

241 CCCATGGGGCCCACTGTCTGGGTC3GGATGGCACAGGGCTGGTGCCCTCGGAGCGTGTCCTGGTGGGGCCCCAGCGGCTGCAGGTGCTGATGCCTCCCACGAGGACTCCGGGGCCTAC36068 P M G P T V W V K D G T G L V P S E R V L V G P Q R L 0 V L N A S H E D S G A Y 107

361 AGCTGCCGGCAGCGGCTCACGCAGCGCGTACTGTGCCACTTCAGTGTGCGGGTGACAGACGCTCCATCCTCGGGAGATGACGAAGACGGGGAGGACGAGGCTGAGGACACAGGTGTGGAC 480108 S C R Q R L T 0 R V L C H F S V R V T D A P S S G D D E D G E D E A E D T G V D 147

* *

481 ACAGGGGCCCCTTACTGGACACGGCCCGAGCGGATGGACAAGAAGCTGCTGGCCGTGCCGGCCGCCAACACCGTCCGCTTCCGCTGCCCAGCCGCTGGCMCCCCACTCCCTCCATCTCC 600148 T G A P Y W T R P E R M D K K L L A V P A A N T V R F R C P A A G N P T P S I S 187

601 TGGCTGAGACGGCAGGGAGTTCCGCGGCGAGCACCGCATTGGAGGCATCAGCTGCGGCATCAGCAGTGGAGCCTGGTCATGGAAGCGTGGTGCCCTCGGACCGCGGCAACTACACC 720188 W L K N G R E F R G E H R I G G I K L R H Q Q W S L V M E S V V P S D R G N Y T 227

721 TGCGTCGTGGAG4C0GTTTGGCAGCATCCGGCAGACGTACACGCTGGACGTGCTGGAGCGCTCCCCGCACCGGCCCATCCTGCAGGCGGGGCTGCCGGCCACCAGACGGCGGTGCTG840228 C V V E N K F G S I R Q T Y T L D V L E R S P H R P I L Q A G L P A N Q T A V L 267

841 GGCAGCGACGTGGAGTTCCACTGCAGGTGTACAGTGACGCACAGCCCCACATCCAGTGGCTCAAGCACGTGGAGGTGAACGGCAGCAAGGTGGGCCCGGACGGCACACCCTACGTTACC 960268 G S D V E F H C K V Y S D A 0 P H I Q W L K H V E V N G S K V G P D G T P Y V T 307

961 GTGCTCAAGACGGCGGGCGCT1CACCACCGACAAGGAGCTAGAGGTTCTCTCCT0TGCACMCGTCACCTTTGAGGACGCCGGGGAGTACACCTGCCTGGCGGGCAtTCTATTGGGTTT1080308 V L K T A G A N T T D K E L E V L S L H N V T F E D A G E Y T C L A G N S I G F 347

1081 TCTCATCACTCTGCGTGGCTGGTGGTGCTGCCAGCCGAGGAGGAGCTGGTGGAGGCTGACGAGGCGGGCAGTGTGTATGCAGGCATCCTCAGCTACGGGGTGGGCTTCTTCCTGTTCATC 1200348 S H H S A W L V V L P A E E E L V E A D E A G S V Y A G I L S Y G V G F F L F I 387

1201 CTGGTGGTGGCGGCTGTGACGCTCTGCCGCCTGCGCAGCCCCCCCAAGAAGGCCTGGGCTCCCCCACCGTGCACAGATCTCCCGCTTCCCGCTCAGCGACAGGTGTCCCTGGAGTCC 1320388 L V V A A V T L C R L R S P. P K K G L G S P T V H K I S R F P L K R Q V S L E S 427

1321 4CGCGTCCATGAGCTCC4CACACCACTGGTGCGCATCGCAAGGCTGTCCTCAGGGGAGGGCCCCACGCTGGCCATGTCTCCGAGCTCGAGCTGCCTGCCGACCCCAATGGGAGCTG1440428 N A S N S S N T P L V R I A R L S S G E G P T L A N V S E L E L P A D P K W E L 467

1441 TCTCGGGCCCGGCTGACCCTGGGC1GCCCCTTGGGGAGGGCTGCTTCGGCCAGGTGGTCATGGCGGAGGCCATCGGCATTGACAGGACCGGGCCGCCAGCCTGTCACCGTAGCCGTG1560468 S R A R L T L G K P L G E G C F G Q V V M A E A I G I D K D R A A K P V T V A V 507

+ + +

1561 18GATGC0GAGACGATGCCACTGACAAGGACCTGTCGGACCTGGTGTCT0GAGATGGAGATGATGAAGATGATCGGGAACACAAAACATCATCAACCTGCTGGGCGCCTGCACGCAG1680508 K M L K D D A T D K D L S D L V S E M E M M K M I G K H K N I I N L L G A C T Q 547

1681 GGCGGGCCCCTGTACGTGCTGGTGGAGTACGCGGCCAAGGGTAACCTGCGGGAGTTTCTGCGGGCGCGGCGGCCCCCGGGCCTGGACTACTCCTTCGACACCTGCMGCCGCCCGAGGAG 1800548 G G P L Y V L V E Y A A K G N L R E F L R A R R P P G L D Y S F D T C K P P E E 587

1801 CAGCTCACCTTCAAGGACCTGGTGTCCTGTGCCTACCAGGTGGCCCGGGGCATGGAGTACTTGGCCTCCCAGAAGTGCATCCACAGGGACCTGGCTGCCCGCAATGTGCTGGTGACCGAG 1920588 Q L T F K D L V S C A Y Q V A R G M E Y L A S 0 K C I H R D L A A R N V L V T E 627

1921 GACAACGTGATGAAGATCGCAGACTTCGGGCTGGCCCGGGACGTGCACAACCTCGACTACTACAAGAAGACAACCAACGGCCGGCTGCCCGTGAAGTGGATGGCGCCTGAGGCCTTGTTT 2040628 D N V M K I A D F G L A R D V H N L D Y Y K K T T N G R L P V K W M A P E A L F 667

2041 GACCGAGTCTACACTCACCAGAGTGACGTCTGGTCCTTTGGGGTCCTGCTCTGGGAGATCTTCACGCTGGGGGGCTCCCCGTACCCCGGCATCCCTGTGGAGGAGCTCTTCMGCTGCTG 2160668 D R V Y T H 0 S D V U S F G V L L W E I F T L G G S P Y P G I P V E E L F K L L 707

2161 8GGAGGGCCACCGCATGGAC0GCCCGCCMCTGCACACACGACCTGTACATGATCATGCGGGAGTGCTGGCATGCCGCGCCCTCCCAGAGGCCCACCTTCMGCAGCTGGTGGAGGAC2280708 K E G H R M D K P A N C T H D L Y M I M R E C W H A A P S 0 R P T F K 0 L V E D 747

2281 CTGGACCGTGTCCTTACCGTGACGTCCACCGACGAGTACCTGGACCTGTCGGCGCCTTTCGAGCAGTACTCCCCGGGTGGCCAGGACACCCCCAGCTCCAGCTCCTCAGGGGACGACTCC 2400748 L D R V L T V T S T D E Y L D L S A P F E 0 Y S P G G Q D T P S S S S S G D D S 787

2401 GTGTTTGCCCACGACCTGCTGCCCCCGGCCCCACCCAGCAGTGGGGGCTCGCGGACGTGAGGGCCACTGGTCCCCACATGTGAGGGGTCCCTAGCAGCCCTCCCTGCTGCTGGTGCA 2520788 V F A H D L L P P A P P S S G G S R T *

FIG. 2. Sequence of FGFR-3: nucleotide sequence and predicted amino acid (in single-letter code) sequence ofthe longest open reading frameof clone 17B. Bold underlining, hydrophobic sequences; *, cysteine residues; underlining, potential N-linked glycosylation sites; and +,conserved ATP-binding site motif Gly-Xaa-Gly-Xaa-Xaa-Gly (GXGXXG).

Dow

nloa

ded

by g

uest

on

Mar

ch 2

4, 2

020

Proc. Natl. Acad. Sci. USA 88 (1991) 1097

1.5 hr before and throughout a 2-hr [35S]methionine-labelingperiod. Cells were labeled with [35S]methionine at a concen-tration of 150 puCi/ml (1 Ci = 37 GBq). The labeled cells werelysed in 1% Triton X-100/150 mM NaCl/10mM Tris HCI, pH7.5, on ice, precleared with Staphylococcus aureus, and spunat 10,000 x g for 15 min. Immunoprecipitation was performedas described (24) by using rabbit antiserum generated againsta trpE fusion protein expressing amino acids 577-806 ofFGFR-3. Immunoprecipitates were electrophoresed on a7.5% SDS/polyacrylamide gel (25).

45Ca2' Efflux Assays. RNAs were transcribed from linear-ized cDNA templates by using either T7 RNA polymerase(antisense) or T3 RNA polymerase (sense) in the presence of500 gM rRNA-encoding NTP (200 ttM rRNA-encoding GTP)and 500 ,uM 5'-GpppG-3', according to published protocols(12). Oocytes were surgically removed and enzymaticallydispersed by incubation with type II collagenase (1 mg/ml)for 3 hr at room temperature. Individual oocytes wereinjected into the vegetal pole with 50 nl of water or RNAsolution (1 ,ug/,l in water). After injection, oocytes wereincubated at 19°C for 48 hr before performing the 45Ca2'efflux assay; 45Ca2' efflux assays were done as described(12). Briefly, oocytes were incubated with 45CaC12 for 3 hr at19°C and then washed extensively. Groups of five oocyteswere placed in individual wells of a 24-well plate. At 10-minintervals, the medium was removed for counting of radioac-tivity and fresh medium was added. After 40 min, recombi-nant human aFGF and bFGF (a gift from Chiron) were addedto a final concentration of 0.5 nM, and medium collectionswere continued at the specific times.

RESULTS AND DISCUSSIONTo isolate tyrosine kinases expressed in hematopoietic cells,we screened a cDNA library prepared from the humanchronic myelogenous leukemia (CML) cell line, K-562 (17)with the chicken v-sea gene, a receptor-like tyrosine kinase(26) under low-stringency conditions. Among several cDNAsisolated, the nucleotide sequence of one partial cDNA cloneindicated that we had isolated a gene that was only 29ohomologous to v-sea but was much more similar to thepreviously described FGFR (11, 12) and the mouse bek gene(27). A 4.5-kilobase (kb) clone, named 17B, was isolated byscreening the library at high stringency. The size ofthis clonecorresponded to the size ofthe majormRNA from K-562 cells

as shown by Northern blot (Fig. 1A). In addition, primerextension using a primer near the 5' end of the cDNAdemonstrated only 95 base pairs (bp) of additional sequencein the mRNA not present in the cDNA, suggesting that theclone was nearly complete (Fig. 1B).An open reading frame of 806 amino acids was predicted

from the nucleotide sequence of this clone (Fig. 2) and wasfound to be most similar to FGFRs in overall structure andamino acid homology (see Fig. 3A). The predicted initiatingmethionine of clone 17B, although not residing within acontext favorable for strong translation initiation (29), isfollowed by a hydrophobic sequence characteristic ofa signalsequence (30). In addition, this methionine residue is in ananalogous position to the initiating methionine of the FGFR.Also following the same pattern as FGFRs, the extracellulardomain of clone 17B contains three immunoglobulin-likedomains. Constant-type immunoglobulin domains are char-acterized by a cysteine residue followed by a tryptophan 11or 12 residues downstream and a Asp-Xaa-Gly-Xaa-Tyr-Xaa-Cys motif 50-70 residues further downstream (31). Consen-sus sequences for variable-type immunoglobulin domainsconsist of a cysteine residue followed by a tryptophan 16amino acids downstream and a Asp-Xaa-Gly-Xaa-Tyr-Xaa-Cys motif 60-75 residues further downstream (31). Immuno-globulin domains I, II, and III of clone 17B fall into theconstant-type immunoglobulin domain category. Betweenthe first and second immunoglobulin domains is an acidic runof amino acids (Asp-Asp-Glu-Asp-Gly-Glu-Asp-Glu-Ala-Glu-Asp; DDEDGEDEAED in single-letter code) similar toan acidic region found in other FGFRs (Fig. 2). Sevenpotential N-linked glycosylation sites reside in the extracel-lular region. A 25-amino acid hydrophobic sequence charac-teristic of a transmembrane domain lies at amino acids372-396 (Fig. 2).The sequence of the cytoplasmic domain of clone 17B is

typical of tyrosine kinase receptors (ref. 32; Fig. 2). There isa Gly-Xaa-Gly-Xaa-Xaa-Gly (GXGXXG) motif and a con-served lysine residue, both of which are normally found inATP-binding domains and an Asp-Phe-Gly-Leu-Ala-Arg(DFGLAR) sequence, which is found in known tyrosinekinase catalytic domains. The tyrosine kinase domain ofclone 17B has features characteristic of other tyrosine kinasereceptors such as FGFR, platelet-derived growth factorreceptor (33, 34), and colony-stimulating factor 1 receptor(35). The predicted protein sequence ofclone 17B contains an

A

(77) TM Kirqs I T I Trro I 1 1 1 1 1 1i

Hurran FGFR 3 S

B

55 II III TM JM TK 1 KI TK 2

hu FGFR

FGFR-2/bek

Cek 2

23% 22% 48% 70% 28% 44% 82% 43% 91% 38%

144% 2 1 51% 71% 32% 44% 88% 50% 93P 47%

1 8% 37% 76% 87% 68% 73% 9 1% 93% 99% 66%

FIG. 3. (A) Schematic of the human FGFR-3. Shaded box, putative signal sequence; S, immunoglobulin domains with cysteine residues;black box, acidic region; hatched box, transmembrane domain (TM); vertically striped box, kinase domain; and KI, kinase insert region. (B)Amino acid homology between human FGFR-3 and FGFR-1, FGFR-2/bek (36), and chicken Cek-2 sequences. The putative amino acidsequences are broken down into specific regions including the signal sequence (SS), immunoglobulin domains (I, II, and III), transmembranedomain (TM), juxtamembrane region (JM), tyrosine kinase domain (TK 1 and TK 2), kinase insert (KI), and C-terminal tail (C).

C

-5 &I

E t illillilillilit

Cell Biology: Keegan et al.

Dow

nloa

ded

by g

uest

on

Mar

ch 2

4, 2

020

Proc. Natl. Acad. Sci. USA 88 (1991)

Tunicamycin(ug! ml )

0 25 50

P P P

200-

97-

68-

43-

FIG. 4. Transient expression ofFGFR-3 cDNA in COS cells withand without tunicamycin. Cells were labeled with [35S]methionine inthe presence of increased amounts of tunicamycin (0, 2.5, and 5,ug/ml), lysed, and immunoprecipitated with preimmune sera (P) orantisera recognizing FGFR-3 (I). Molecular mass standards aredenoted at left in kDa.

inserted sequence that splits the tyrosine kinase domain intotwo parts. This kinase insert sequence is 14 amino acids long(residues 576-589), which is the same size as the kinase insertregion of the FGFR (residues 580-593) but is much smallerthan the kinase insert regions ofthe colony-stimulating factor1 and platelet-derived growth factor receptors. The juxta-membrane regions of clone 17B (residues 397-478) andFGFR (residues 396-482) are longer than the correspondingregions of other tyrosine kinases and provide a feature thatdistinguishes this class of growth factor receptors.FGFR and clone 17B appear to encode receptors with

similar structures: they possess three immunoglobulin-likedomains, a transmembrane domain, and a tyrosine kinasedomain. The amino acid homology between FGFR and clone17B in the kinase domain is 87% (Fig. 3). However, the kinaseinsert region, thejuxtamembrane domain, and the C terminusare less conserved, showing 43%, 44%, and 38% homology,respectively (Fig. 3). These less conserved sequences aremost likely to be involved in the specificity of this receptor.

3000 1

2000 -

E0

1000

0

The extracellular region exhibits an overall amino acid iden-tity of 47% when compared with the FGFR, but when it isbroken down into immunoglobulin domains, extracellulardomain I is less conserved, whereas domains II and III arehighly conserved. Extracellular domain I shows the greatestdivergence between the two genes, showing only 22% aminoacid identity (Fig. 3). Recently, two other genes, bek (36, 38)and Cek2 (37), have been shown to exhibit a high degree ofhomology to the FGFR gene and clone 17B (Fig. 3) and arepredicted to encode proteins with similar structural charac-teristics. In addition, bek binds aFGF and bFGF (36), thusplacing it in the FGFR family. In summary, the degree ofamino acid homology between other members of the FGFRfamily and protein 17B demonstrates that we have isolated agene with a predicted structure that has significant similarityto FGFRs.To characterize the protein product of clone 17B, we

assayed transient expression in COS cells. Cells were trans-fected with plasmid containing the 17B cDNA in either thesense or the antisense orientation. Immunoprecipitation wasthen performed with antiserum that recognizes the kinasedomain of the cDNA. A protein of 125 kDa is specificallyimmunoprecipitated in cells expressing the receptorcDNA inthe sense orientation (Fig. 4) but not in the antisense orien-tation (data not shown). A 97-kDa protein is immunoprecip-itated from COS cells transfected with clone 17B whentreated with the N-linked glycosylation inhibitor tunicamy-cin. These data indicate that clone 17B encodes a 125-kDatransmembrane glycoprotein.As stated previously, the degree of amino acid identity

between the FGFR and protein 17B is 47% in the extracellulardomain. The epidermal growth factor receptor and thec-erbB-2 genes show 43% identity in the extracellular domainbut do not share the ability to bind epidermal growth factor(28, 39). Because this would suggest that a high degree ofamino acid identity does not necessarily predict an overlap infactor binding, we asked whether aFGF and bFGF wouldactivate our FGFR-related receptor. To address this ques-tion, we expressed the 17B protein in Xenopus oocytes andmeasured receptor activation by using a sensitive Ca2+ effluxassay (40). This is a rapid functional assay that measures theability ofa receptor to mobilize intracellular Ca2+ stores uponreceptor stimulation. Previous studies have shown that thechicken and human FGFRs bind aFGF and bFGF and areactivated by them in this assay system (12). For our exper-

---a wate r- antisense* FGFR 3-- FGFR 1

Minutes

FIG. 5. bFGF induces 45Ca2+ efflux from Xenopus oocytes injected withRNA encoding FGFR-3. The graph shows 45Ca2+ efflux from oocytesinjected with sense-strand FGFR-3 RNA (V), antisense strand FGFR-3 RNA (*), FGFR-1 (O), or water (o). Injected oocytes were incubatedwith 45CaC12 for 3 hr at 19°C and then washed extensively. Groups of five oocytes were placed in individual wells of a 24-well plate. At 10-minintervals, the medium was removed to count radioactivity, and fresh medium was added. After 40 min, recombinant bFGF was added to a finalconcentration of 0.5 nM, and medium removal was continued at specified times. Each data point represents the average of triplicate wells. Ingeneral, triplicate points did not vary >15%.

1098 Cell Biology: Keegan et al.

Dow

nloa

ded

by g

uest

on

Mar

ch 2

4, 2

020

Proc. Natl. Acad. Sci. USA 88 (1991) 1099

iments, oocytes were injected with in vitro-transcribed RNAscorresponding to either the sense or antisense of clone 17B.After 2-day incubation, oocytes were preloaded with 45CaC12,stimulated with bFGF, and assayed for levels of 45Ca2+released into the medium (Fig. 5). Addition of 0.5 nM bFGFto oocytes injected with RNA from clone 17B resulted in arapid and large efflux of45Ca2+. Similar results were obtainedupon adding 0.5 nM aFGF (data not shown). Stimulation ofoocytes injected with human FGFR-1 caused a slightly higherpeak of 45Ca2' release that follows a profile similar to thatfrom protein 17B. In contrast, oocytes injected with eitherwater or antisense RNA from clone 17B did not give strongprofiles of efflux. These data indicate that clone 17B encodesa receptor capable of being activated by aFGF and bFGF,thus establishing the receptor as an additional member of theFGFR family. Because of the high degree of homologybetween protein 17B, FGFR, and bek-encoded protein andbecause of its ability to be activated by aFGF and bFGF, wedesignate our protein FGFR-3 and propose that the previ-ously described FGFR/flg protein be named FGFR-1 andbek-encoded protein be named FGFR-2.Both aFGF and bFGF bind and activate the chicken and

human FGFRs (12). One form of the human FGFR that bindsboth factors possesses an extracellular domain consisting ofonly the acidic region at the N terminus followed by immu-noglobulin domains II and III (12). Although the extracellularregion of FGFR-3 is only 47% identical to FGFR-1, there isa high degree of conservation within the acidic domain anddomains II and III; this would suggest that these sequencesare sufficient for binding aFGF and bFGF.We have identified another tyrosine kinase receptor gene

and have shown that the protein product of this gene can bestimulated by aFGF and bFGF. Identification of FGFR-3-,FGFR-1-, and the FGFR-2/bek-encoding genes offers addi-tional evidence for a family of FGFRs. Recently, the se-quence of the chicken gene Cek2 was published (37), and theprotein it encodes has greater amino acid identity to FGFR-3than to either FGFR-1 or FGFR-2/bek and, thus, mayrepresent the chicken homologue of FGFR-3. Although wehave shown that FGFR-3 responds to aFGF and bFGF, wedo not know the identity of the in vivo ligand that interactswith this receptor. Further studies will be needed to deter-mine whether different members of the FGFR family displaydifferent specificities for the known members of the FGFfamily.

We thank L. Morrison, A. Crowe, and W. Schubach for criticalreading of the manuscript, John Lu for technical expertise, J. Lipsickfor assistance with sequence analysis, and K. Donnelly for manu-script preparation. This work was supported by National Institutesof Health Grants R01 HL-32989 and P01 HL-43821 (to L.T.W.) andR01 CA42573 and P01 CA-28146 (to M.J.H.). D.J. was supported byAmerican Heart Association Fellowship 94-1219116.

1. Baird, A., Esch, F., Mormede, P., Ueno, N., Ling, N., Bohlen,P., Ying, S.-Y., Wehrenberg, W. B. & Guillemin, R. (1986)Rec. Prog. Horm. Res. 42, 143-205.

2. Burgess, W. H. & Maciag, T. (1989) Annu. Rev. Biochem. 58,575-606.

3. Rifkin, D. B. & Moscatelli, D. (1989) J. Cell Biol. 109, 1-6.4. Moore, R., Casey, G., Brookes, S., Dixon, M., Peters, G. &

Dickson,C. (1986) EMBO J. 5, 919-924.5. Delli Bovi, P., Curatola, A. M., Kern, F. G., Greco, A.,

Ittmann, M. & Basilico, C. (1987) Cell 50, 729-737.6. Zhan, X., Bates, B., Hu, X. & Goldfarb, M. (1988) Mol. Cell.

Biol. 8, 3487-3495.7. Marics, I., Adelaide, J., Raybaud, F., Mattei, M., Coulier, F.,

Planche, J., Lapeyriere, 0. & Birnbaum, D. (1989) Oncogene4, 335-340.

8. Finch, P. W., Rubin, J. S., Miki, T., Ron, D. & Aaronson,S. A. (1989) Science 245, 752-755.

9. Lee, P. L., Johnson, D. E., Coussens, L. S., Fried, V. A. &Williams, L. T. (1989) Science 245, 57-60.

10. Pasquale, E. B. & Singer, S. J. (1989) Proc. Nail. Acad. Sci.USA 86, 5449-5453.

11. Ruta, M., Howk, R., Ricca, G., Drohan, W., Zabelshansky,M., Laureys, G., Barton, D. E., Francke, U., Schlessinger, J.& Givol, D. (1988) Oncogene 3, 9-15.

12. Johnson, D. E., Lee, P. L., Lu, J. & Williams, L. T. (1990)Mol. Cell. Biol. 10, 4728-4736.

13. Reid, H. H., Wilks, A. F. & Bernard, 0. (1989) Proc. Natl.Acad. Sci. USA 87, 1596-1600.

14. Safran, A., Aaron, A., Orr-Urtereger, A., Neufeld, G., Lonai,P., Givol, D. & Yarden, Y. (1990) Oncogene 5, 635-643.

15. Mansukhani, A., Moscatelli, D., Talarico, D., Levytska, V. &Basilico, C. (1990) Proc. Natl. Acad. Sci. USA 87, 4378-4382.

16. Ruta, M., Burgess, W., Givol, D., Epstein, J., Neiger, N.,Kaplow, J., Crumley, G., Dionne, C., Jaye, M. & Schlessinger,J. (1989) Proc. Natl. Acad. Sci. USA 86, 8722-8726.

17. Mes-Masson, A.-M., McLaughlin, J., Daley, G. Q., Paskind,M. & Witte, 0. (1986) Proc. Natl. Acad. Sci. USA 83, 9768-9772.

18. Sanger, F., Nicklen, S. & Coulson, A. R. (1977) Proc. Natl.Acad. Sci. USA 74, 5463-5467.

19. Kyte, J. & Doolittle, R. F. (1982) J. Mol. Biol. 157, 105-132.20. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D.,

Seidman, J. G., Smith, T. A. & Struhl, K., eds. (1987) CurrentProtocols in Molecular Biology (Wiley, New York).

21. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989) MolecularCloning:A Laboratory Manual (Cold Spring Harbor Lab., ColdSpring Harbor, NY), 2nd Ed.

22. Kaufman, R. J., Murtha, P. & Davies, M. V. (1987) EMBO J.6, 187-193.

23. Kaufman, R. J. & Murtha, P. (1987) Mol. Cell. Biol. 7, 1568-1571.

24. Knight, J., Beug, H., Marshall, J. & Hayman, M. (1988)Oncogene 2, 317-326.

25. Laemmli, U. K. (1970) Nature (London) 227, 680-685.26. Smith, D. R., Vogt, P. K. & Hayman, M. J. (1989) Proc. Nail.

Acad. Sci. USA 86, 5291-5295.27. Kornbluth, S., Paulson, K. E. & Hanafusa, H. (1988) Mol. Cell.

Biol. 8, 5541-5544.28. Coussens, L., Yang-Feng, T., Yu-Cheng, L., Chen, E., Gray,

A., McGrath, J., Seeburg, P., Libermann, T. A., Schlessinger,J., Franke, U., Levinson, A. & Ullrich, A. (1985) Science 230,1132-1139.

29. Kozak, M. (1986) Cell 44, 283-292.30. Mount, S. M. (1982) Nucleic Acids Res. 10, 459-472.31. Williams, A. F. & Barclay, N. A. (1988) Annu. Rev. Immunol.

6, 381-408.32. Hanks, S., Quinn, A. M. & Hunter, T. (1988) Science 241,

42-52.33. Matsui, T., Heidaran, M., Miki, T., Popescu, N., LaRochelle,

W., Kraus, M., Pierce, J. & Aaronson, S. (1989) Science 243,800-803.

34. Yarden, Y., Escobedo, J. A., Kuang, W.-J., Yang-Feng, T. L.,Daniel, T. O., Tremble, P. M., Chen, E. Y., Ando, M. E.,Harkins, R. N., Franke, U., Fried, V. A., Ullrich, A. &Williams, L. T. (1986) Nature (London) 323, 226-232.

35. Coussens, L., Van Beveren, C., Smith, D., Chen, E., Mitchell,R. L., Isacke, C. M., Verma, I. M. & Ullrich, A. (1986) Nature(London) 320, 277-280.

36. Dionne, C. A., Crumley, G., Bellot, F., Kaplow, J. M., Sear-foss, G., Reita, M., Burgess, W. H., Jaye, M. & Schlessinger,J. (1990) EMBO J. 9, 2685-2692.

37. Pasquale, E. B. (1990) Proc. Natl. Acad. Sci. USA 87, 5812-5816.

38. Hattori, Y., Odagiri, H., Nakatani, H., Miyagawa, K., Naito,K., Sakamoto, H., Katoh, O., Yoshida, T., Sugimura, T. &Terada, M. (1990) Proc. Natl. Acad. Sci. USA 87, 5983-5987.

39. Yamamoto, T., Ikawa, S., Akiyama, T., Semba, K., Nomura,N., Miyajima, N., Saito, T. & Toyoshima, K. (1986) Nature(London) 319, 230-234.

40. Williams, J. A., McChesney, D. J., Calayag, M. C., Lingappa,V. R. & Logsdon, C. D. (1988) Proc. Natl. Acad. Sci. USA 85,4939-4943.

Cell Biology: Keegan et al.

Dow

nloa

ded

by g

uest

on

Mar

ch 2

4, 2

020