3000 Ci/mrM) (UTP) was then added, and the
`nuclear suspension was incubated at 30°C for 30
`minutes, after which time 15 p.1 of DNase 1 (5 pg/
`ml) in 10 mM CaCI2 (5 pLg/ml) was added. After S
`minutes at 30°C, the reaction was made 1 x SET
`(1 percent sodium dodecyl sulfate (SDS), 5 mM
`EDTA, 10 mMtris-HCl, pH 7.4), and proteinase
`K was added to a concentration of 200 ,uglml.
`After incubation at 37C for 45 minutes, the
`solution was extracted with an equal volume of a
`mixture of phenol and chloroform, and the inter-
`hase was again extracted with 100 p.l of lx
`SET. Ammoniun acetate (IOM) was added to
`the combined aqueous phases (original plus
`reextraction) to a final concentration of 2.3M, an
`equal volume of isopropyl alcohol was added,
`and nucleic acid was precipitated (-70°C for 15
`minutes). The precipitate was centrifuged in a
`microcentrifuge for 10 minutes, and the pellet
`was resuspended in 100 IL1 of TE (10 mM tris-
`HCl, I mM EDTA) and centrifuged through a G-
`50 (medium) spin column. The eluate was made
`0.2M in NaOH and after 10 minutes on ice,
`HEPES was added to a concentration of 0.24M.
`Two and one-half volumes of ethanol were then
`added, and the solution containing the precipi-
`
`tate held overnight at -20°C. After centrifuga-
`tion in a microcentrifuge for 5 minutes, the pelet
`was resuspended in hybridization buffer, which
`consisted of [10 mM TES, pH 7.4, 0.2 percent
`SDS, 10 mM EDTA, 0.3M NaCi, lx Den-
`hardt's, and Escherichia coli RNA (250 ,ug/ml)].
`Nitrocellulose filters containing plasmid DNA's
`were prepared with a Schleicher & Schuell Slot
`Blot Apparatus under conditions suggested by S
`and S, except that wells were washed with lOx
`SSC (saline sodium citrate). These filters were
`first hybridized in the hybridization solution
`described above for a minimnum of 2 hours at
`65C. After this preliminary hybridization, the
`ifiters were hybridized to the runoff products in
`hybridization solution for 36 hours. A typical
`reaction contained 2 ml of hybridization solution
`with 1 x l0 cpm/ml. After hybridization, filters
`were washed or 1 hour in 2x SSC at 65'C. The
`filters were then incubated at 370C in 2x SSC
`with RNase A (10 mg/ml) for 30 minutes and
`were subsequently washed in 2x SSC at 37°C
`for 1 hour. Alternatively, after hybridization the
`filters were washed twice for 15 minutes in 0.1
`percent SDS, 2x SSC at room temperature, and
`then washed at 600C (0.1 percent SDS, 0.1 x
`
`SSC) for 30 minutes. Either protocol for proc-
`essing of the filters after hybridization yielded
`the same specificity in signal. Filters were then
`exposed to Kodak XAR film in cassettes con-
`taining Lightening-Plus screens at -70°C for
`various times.
`45. C. Yanisch-Perron, J. Vierra, J. Messing, Gene
`33, 103 (1985).
`46. S. L. McKnight, E. R. Gavis, R. Kingsbury, R.
`Axel, Cell 25, 385 (1981).
`47. M. Groudine and C. Casimir, Nucleic Acids
`Res. 12, 1427 (1984).
`48. We thank many of our colleagues for discussion
`and suggestions during the course of this work;
`Hal Weintraub, Paul Neiman, and Craig Thomp-
`son for comments on the manuscript; Craig
`Thompson for assistance in obtaining lympho-
`cyte preparations; Bill Schubach for plasmid
`pBK25; and Kay Shiozaki for assistance with
`the manuscript. Supported by NIH grants CA
`18282 (M.L.) and CA 28151 (M.L. and M.G.),
`and NSF grant PCM 82-04696 (M.G.), and a
`scholarship from the Leukemia Society of
`America (M.G.)
`30 July 1985; accepted 15 October 1985
`
`RESEARCH ARTICLE
`
`Tyrosine Kinase Receptor with Extensive
`Homology to EGF Receptor Shares
`Chromosomal Location with neu Oncogene
`
`Lisa Coussens, Teresa L. Yang-Feng, Yu-Cheng Liao
`Ellson Chen, Alane Gray, John McGrath, Peter H. Seeburg
`Towia A. Libermann, Joseph Schlessinger, Uta Francke
`thur Levinson, Axel Ullrich
`
`In contrast to v-erbB, which encodes a
`68,000-dalton truncated EGF receptor,
`the neu oncogene product is a 185,000-
`dalton cell surface antigen that can be
`detected by cross-reaction with polyclo-
`nal antibodies against EGF receptor (11);
`neu may itself be a structurally altered
`cell surface receptor with homology to
`the EGF receptor and binding specificity
`for an unidentified ligand.
`Using v-erbB as a screening probe, we
`isolated genomic and cDNA clones cod-
`ing for an EGF receptor-related, but
`distinct, 138,000-dalton polypeptide hav-
`ing all the structural features of a cell
`surface receptor molecule. On the basis
`of its strtictural homology, this putative
`receptor is a new member of the tyro-
`sine-specific protein kinase family. It is
`encoded by a 4.8-kb messenger RNA
`(mRNA) that is widely expressed in nor-
`mal and malignant tissues. We have lo-
`calized the gene for this protein to q21 of
`chromosome 17, which is distinct from
`the EGF receptor locus, but coincident
`with the neu oncogene mapping position
`(12). We therefore consider the possibili-
`ty that we have isolated and character-
`ized the normal human counterpart of
`the rat neu oncogene.
`Tyrosine kinase-type receptor gene and
`complementary DNA. As part of our at-
`tempts to isolate and characterize the
`chromosomal gene coding for the human
`cellular homologue of the viral erbB gp68
`polypeptide, AEV-ES4 erbB sequences
`(2.5-kb Pvu II fragment of pAEV) (13)
`were used as a 32P-labeled hybridization
`probe for the screening of a human geno-
`mic DNA library at reduced stringency
`
`lin (6), PDGF (7), and insulin-like growth
`factor 1 (IGF-1) (8); hence more connec-
`tions may be found between tyrosine
`kinase growth factor receptors and tyro-
`sine kinase oncogene products.
`Comparison of the complete primary
`structure of the human EGF receptor (9)
`with the sequence of the avian erythro-
`blastosis virus (AEV) transforming gene,
`v-erbB (10), revealed close sequence
`similarity; in addition, there were amino
`and carboxyl terminal deletions that may
`reflect key structural changes in the gen-
`eration of an oncogene from the gene for
`a normal growth factor receptor (3, 9).
`Another oncogene, termed neu, is also
`related to v-erbB and was originally
`identified by its activation in ethylnitro-
`sourea-induced rat neuroblastomas (11).
`
`Growth factors and their receptors are
`involved in the regulation of cell prolif-
`eration, and several recent findings sug-
`gest that they also play a key role in
`oncogenesis (1-4). Of approximately 20
`identified oncogenes, the three that have
`been correlated with known cellular pro-
`teins are each related to either a growth
`factor or a growth factor receptor. The B
`chain of platelet-derived growth factor
`(PIDGF) is encoded by the proto-onco-
`gene c-sis (2), the erb-B oncogene prod-
`uct gp68 is a truncated form of the epi-
`dermal growth factor (EGF) receptor (3),
`and the proto-oncogene c-fms may be
`related or identical to the receptor for
`macrophage colony-stimulating
`factor
`(CSF-IR) (4).
`The receptor-related oncogenes are
`members of a gene family in that each
`has tyrosine-specific protein kinase ac-
`tivity, and is associated with the plasma
`membrane (5). Such features are also
`shared by several other polypeptide hor-
`mone receptors, including those for insu-
`1132
`
`Lisa Coussens, Yu-Cheng Liao, Ellson Chen, Alane Gray, Peter H. Seeburg, Arthur Levinson, and Axel
`Ullrich are in the Department of Molecular Biology, Genentech, Inc., 460 Point San Bruno Boulevard, South
`San Francisco, CZornia 94080; John McGrath is currently with the Department of Biology, Massachusetts
`Institute of Technology, Cambridge, Massachusetts 02142; Towia Libermann and Joseph Schlessinger are in
`the Department of Chemical Immunology at the Weizmann Institute of Science, Rehovot 76100, Israel; and
`Teresa L. Yang-Feng and Uta Francke are in the Department of Human Genetics at Yale University School
`of Medicine, 333 Cedar Street, New Haven, Connecticut 06510.
`
`SCIENCE, VOL. 230
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1043 Page 1 of 8
`
`

`
`(14). Clone Xc-erbB/l was isolated; it
`contained a hybridizing 1.8-kb Bam HI
`fragment, which was subjected to DNA
`sequence analysis. The 1838-bp
`se-
`quence contains three complete and one
`partial erbB-homologous exons separat-
`ed by short intervening sequences (Fig.
`1). Comparison of this human gene se-
`quence with our complete cDNA-de-
`rived human EGF receptor protein se-
`quence (9) revealed 32 differences (18.7
`percent) within the
`171 amino acid
`stretch of combined exons, suggesting
`that this gene fragment was not derived
`from the human EGF receptor gene.
`Since this gene may code for an un-
`known tyrosine kinase-type receptor
`that is closely related to the human EGF
`receptor, we named it HER2.
`Northern blot analysis (15) with the
`32P-labeled 1.8-kb HER2 fragment as a
`hybridization probe revealed a 4.8-kb
`mRNA in human term placenta poly(A)+
`RNA, distinct from the 5.8- and 10.5-kb
`EGF receptor mRNA's also present at
`high levels in this tissue (Fig. 2a, lane 1).
`Thus, we had isolated a portion of an
`EGF receptor-erbB-related but distinct
`gene. To obtain its complete primary
`structure, two single-stranded synthetic
`oligonucleotide probes (16) were pre-
`pared from HER2 exon sequence regions
`that differed sufficiently (less than 60
`percent nucleotide sequence homology)
`from EGF receptor DNA sequences
`(Fig.
`1 and 2) and used to screen
`1,
`a term placenta complementary DNA
`(cDNA) library of 2 x 106 independent
`recombinant clones in XgtlO (17). Fifty-
`two clones were isolated; they hybrid-
`ized strongly with both synthetic probes
`and weakly with an EGF receptor cDNA
`fragment (HER64-3) (9) containing the
`homologous region within the tyrosine
`kinase domain. One of these, XHER2-
`436, had the longest cDNA insert (4.5
`kb), consisting of three Eco RI fragments
`(1.4, 1.5, and 1.6 kb).
`The complete cDNA sequence of this
`clone is shown in Fig. 3. The longest
`open reading frame starting with a me-
`thionine codon codes for a 1255 amino
`acid polypeptide (137,828 daltons) and
`contains the 171 residues encoded by the
`four exons in the 1.8-kp HER2 gene Bam
`HI fragment (Fig. 1). This 3765-bp cod-
`ing sequence is flanked by 150 bp of 5'
`untranslated sequence and a TGA stop
`codon, followed by a 627-nucleotide 3'
`untranslated sequence. No stop codon is
`found in the 5' untranslated region. In
`support of our assignment, however, the
`initiation codon at position 151 is flanked
`by sequences that follow perfectly Ko-
`zak's rule (18) for translation initiation.
`The 3' untranslated sequence contains a
`6 DECEMBER 1985
`
`potential poly(A) addition
`signal
`se-
`quence (AATATA) 12 nucleotides up-
`stream from a stretch of 15 adenylate
`residues. We are not certain if this (A)15
`stretch is part of a poly(A) tail or repre-
`sents an internal poly(A) stretch of a
`longer 3' untranslated sequence.
`
`those for EGF and insulin (9, 19). Such
`features are apparent in the hydropathy
`profile (20) comparison (Fig. 4a). On the
`basis of this comparison, and on amino
`acid sequence alignment with the EGF
`receptor (Fig. 4b, region 1), we predict a
`21 amino acid signal sequence (Fig. 4b,
`
`Abstract. A novel potential cell surface receptor of the tyrosine kinase gene family
`has been identified and characterized by molecular cloning. Its primary sequence is
`very similar to that of the human epidermal growth factor receptor and the v-erbB
`oncogene product; the chromosomal location of the gene for this protein is
`coincident with the neu oncogene, which suggests that the two genes may be
`identical.
`
`Comparison of EGF receptor and
`HER2 sequence. As already indicated by
`the v-erbB sequence homology used to
`isolate HER2, the putative HER2 pro-
`tein is very similar in its overall domain
`organization and sequence to the EGF
`receptor. Nevertheless, there are differ-
`ences that are likely to define a specific
`biological role for the HER2 polypep-
`tide.
`The predicted HER2 polypeptide con-
`tains each of the domain features found
`in hormone receptor precursors, such as
`
`1), an amino terminal serine residue, and
`a 632 amino acid putative extracellular
`ligand-binding domain; a highly hydro-
`phobic, 22-amino acid transmembrane
`anchor domain separates the extracellu-
`lar domain from a 580-residue-long car-
`boxyl-terminal
`domain,
`cytoplasmic
`which possesses the highest homology to
`v-erbB and other members of the tyro-
`sine kinase family.
`The 632-amino acid, putative HER2
`ligand binding domain is about 40 per-
`cent homologous with the 621-residue
`
`740
`Lys
`Glu
`Glu
`Ala
`769
`IleProAspGiyGluAsnValLyslieProValAlalleLysValLeuArgGluAsnThrSerProLysAlaAsnLysGluIleLeuAsp
`GGATCCCTGATGGGGAGAATGTGAAAATTCCAGTGGCCATCAAAGTGTTGAGGGAAAACACATCCCCCAAAGCCAACAAAGAAMTCTTAGACGTAAGCCCCTCCACCCTCTCCTGCTAGG
`AGGACAGGAAGGACCCCATGGCTGCAGGTCTGGGCTCTGGTCTCTCTTCATTGGGGMGGGGAGATATGACTCCCGCAAACCTAGACTATTTTTTTGGAGACGGAGCTTGCTCTGTCAC
`CCAGGCTGGAGTGCAGTGGCGTTATCTCGGCTCACTGCAACCTCCACCTCCTGGACTCAAGCGATTTTCATGCCTCAGGCTCCTGAGTAGCTGGGATTACAAGCGCCCGCTAATTTTTTT
`TTTTTTTTTGAGACAGAGTCTCGCTCTGTCACCCAGGCTAGAGTGAAATGGTGCGGTCTCAGCTCAGCCTCCCAGGTTMMAGCGATTCTTCTCCCTCAGTCTCCTGAGTAGCTGGGATTA
`CAGGCGCGAGCCACCACGCCCGGCTAATTTTTGTATTTTTAGTAGAGATGGGATTTCACCATGTTGGCCAGGTTGGTGTCAAMCTCCTGACCTCATGATCCGCCCGCCTCGGCCTCCCAA
`AGTGCTGGGATTACAGGTGTGAGCCACGTGCCCGGCCTAATCTTTGTATTTTTAGTAGAGACAGGGTTTCACCATGTTGTCCAGGCTGGTACTTTGAGCCTTCACAGGCTGTGGGCCATG
`
`Hi
`AspAsn
`770
`Ser
`GluAlaTyrValMetAl aGlyValGlySerProTy
`GCTGTGGTTTGTGATGGTTGGGAGGCTGTGTGGTGTTTGGGGGTGTGTGGTCTCCCATACCCTCTCAGCGTACCCTTGTCCCCAGGAAGCATACGTGATGGCTGGTGTGGGCTCCCCATA
`Cys
`Ile
`HisLysAspAsnile
`Tyr
`Phe
`Ty
`s
`rValSerArgLeuLeuGlyl leCysLeuThrSerThrVaiGInLeuValThrGinLeuMetProTyrGlyCysLeuLeuAspHi sValArgGluAsnArgGlyArgLeuGlySerGinAs
`TGTCTCCCGCCTTCTGGGCATCTGCCTGACATCCACGGTGCAGCTGGTGACACAGCTTATGCCCTATGGCTGCCTCTTAGACCATGTCCGGGAAAACCGCGGACGCCTGGGCTCCCAGGA
`
`831
`Val
`r
`pLeuLeuAsnTrpCysMetGnIlneAlaLys
`CCTGCTGAACTGGTGTATGCAGATTGCCAAGGTATGCACCTGGGCTCTTTGCAGGTCTCTCCGGAGCAAACCCCTATGTCCACAAGGGGCTAGGATGGGGACTCTTGCTGGGCATGTGGC
`
`832
`Thr
`Arg
`Asn
`GlyMetSerTyrLeuGluAspValArgLeuValHisArgAspLeuAlaAlaArgAsnValLeuValLysSer
`CAGGCCCAGGCCCTCCCAGAAGGTCTACATGGGTGCTTCCCATTCCAGGGGATGAGCTACCTGGAGGATGTGCGGCTCGTACACAGGGACTTGGCCGCTCGGAACGTGCTGGTCAAGAGT
`I-CCGT--CT-G--G--C-C---C----A-I Q
`GlyAlaGlu
`Gln
`Glu
`Lys
`883
`Lys
`ProAsnHisValLyslieThrAspPheGlyLeuAlaArgLeuLeuAspIleAspGluThrGluTyrHisAlaAspGlyGlyLys
`CCCAACCATGTCAAAATTACAGACTTCGGGCTGGCTCGGCTGCTGGACATTGACGAG CAGAGTACCATGCAGATGGGGGCAAGGTTAGGTGAAGGACCAAGGAGCAGAGGAGGCTGGGT
`-CAAA------GTGCG--A- (2)
`
`121
`
`241
`
`361
`
`481
`
`601
`
`721
`
`841
`
`961
`
`1081
`
`1201
`
`1321
`
`1441
`
`1561
`
`884
`ValProl leLysTrpMetAlaLeuGl uSerl leLeu
`GGAGTGGTGTCTAGCCCATGGGAGAACTCTGAGTGGCCACTCCCACAACACACAGTTGGAGGACTTCCTCTTCTGCCCTCCCCAGTGCCCATCAAGTGGATGGCGCTGGAGTCCATTCTC
`IleTyr
`910
`His
`ArgArgArgPheThrHisGInSerAspValTrpSerTyrGlyVaI
`CGCCGGCGGTTCACCCACCAGAGTGATGTGTGGAGTTATGGTGTGTGATGGGGGGTGTTGGGAGGGGTGGGTGAGGAGCCATGGCTGGAGGGAGGATGAGAGCTGGGATGGGGAGAATTA
`CGGGGCCACCTCAGCATGTGAAGGGAGGGAAGGGGCTGCCTGTGCCCCACCTTGCAGGGTCTGTGCACTTCCCAGGATTA&GGGMAGACCGGGTAGGGTCTGTCTCCTGGCATCACATCT
`1681
`CCCCCTGCTACCTGCCATGATGCTAGACTCCTGAGCAGAACCTCTGGCTCAGTACACTAAAGCTCCCTCTGGCCCTCCCACTCCTGACCCTGTCTCTGCCTTAGGTGTGACTGTGTGGGA
`1801
`GCTGATGACTMGGGGCCAAACCTTACGATGGGATCC
`Fig. 1. Partial sequence of the HER2 gene. A partial Hae III-Alu I genomic library (14) of human
`fetal DNA in X Charon 4A was screened using a radiolabeled 2.5-kb Pvu II fragment of pAEV
`(13) containing coding sequences for the tyrosine kinase domain. Hybridization was as
`described elsewhere (31), except that 30 percent formamide was used at 42°C. Three
`independent clones were isolated which shared a 1.8-kb hybridizing Bam HI fragment. This
`fragment and subsets thereof were isolated, subcloned into M13mplO and M13mpll, and
`sequenced (32). The intron-exon organization was determined by comparison with v-erbB
`sequences (10). Amino acid numbering is based on the complete cDNA sequence shown in Fig.
`3. Nucleotide sequence differences with the human EGF receptor sequence are shown in the
`regions that were used for the design of synthetic oligonucleotide probes 1 (30 nucleotides) and
`2 (21 nucleotides). Amino acid sequence differences with the EGF receptor are shown above the
`HER2 sequence.
`
`1133
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1043 Page 2 of 8
`
`

`
`extracellular EGF binding domain of the
`EGF receptor. This homology includes
`two cysteine-rich subdomains of 26 and
`21 regularly organized cysteine residues
`(Figs. 4a and 2c, subdomains 2 and 3), all
`of which are conserved in the EGF re-
`ceptor. The cysteine residue spacing in
`this region is also homologous with the
`single cysteine-rich domain in the insulin
`receptor oa subunit (19). In contrast,
`HER2 contains only eight potential N-
`linked glycosylation target sites (Asn-X-
`Thr or Ser) as compared to 12 in the
`corresponding region of the EGF recep-
`tor. Only five of these are conserved
`with respect to their relative position in
`each polypeptide.
`The hydrophobic, putative membrane
`anchor sequence located between resi-
`dues 653 and 676 (Fig. 4b, region 4) is
`flanked at its carboxyl terminus by a
`stretch of amino acids of predominantly
`basic character (KRRQQKIRKYTMRR)
`(21), as is found in the EGF receptor
`sequence (9) (Fig. 4b, region 5). This
`region of the EGF receptor contains
`Thr654, which plays a key role in protein
`kinase C-mediated receptor modulation
`(22). A homologous threonine residue is
`embedded in a basic environment in the
`HER2 sequence at position 685 (Fig. 4, a
`and b).
`The region of most extensive homolo-
`gy (78.4 percent) between EGF receptor
`and HER2 (beginning at residue 687)
`extends over 343 amino acids and in-
`cludes sequences specifying the adeno-
`sine triphosphate (ATP) binding domain
`(23) and tyrosine kinase activity (Fig. 4b,
`region 6) (5). This region is also the most
`conserved between v-erbB and EGF re-
`ceptor (95 percent) (9). The collinear
`homology between the EGF receptor-
`erbB and HER2 ceases at position 1032,
`but introduction of gaps into the EGF
`receptor or HER2 sequences reveals
`continued, although decreased, related-
`ness (Fig. 4b, region 7). This sequence
`alignment suggests that the two genes
`evolved by duplication of an ancestral
`receptor gene, and that subsequent nu-
`cleotide sequence divergence in this car-
`boxyl terminal domain led to diverged
`biological roles for the encoded polypep-
`tides.
`The carboxyl terminal domain of
`HER2 is characterized by an unusually
`high proline content (18 percent) and
`4a).
`predominant hydrophilicity (Fig.
`These general features are also found in
`the EGF receptor carboxyl terminal do-
`main with a 10 percent proline content.
`The sequences in this region that are
`found to be conserved are almost exclu-
`sively centered around five tyrosine resi-
`dues, which include the major (Tyr"173)
`1134
`
`and two minor (Tyr"48, Tyr'o") in vitro
`autophosphorylation sites in the human
`EGF receptor (24) (Fig. 4, a and b).
`Three of these tyrosine residues of
`HER2 (positions 1139, 11%, 1248) are
`homologous
`flanked
`by
`sequences
`PQPEYV, ENPEYL, and ENPEYL
`(21), respectively (Fig. 4b, region 7).
`HER2 chromosomal location. In situ
`hybridization of two 3H-labeled HER2
`probes (legend, Fig. Sa) to human chro-
`mosomes resulted in specific labeling at
`bands ql2-)q22 of chromosome 17 (Fig.
`Sa). Metaphase cells (100) were analyzed
`for each probe; 40 percent of cells scored
`for HER2 probe 1 (HER2-1) had silver
`grains over 17ql2--q22 (Fig. Sb). Of the
`209 grains observed, 42 (20 percent)
`were found at this specific region, with
`no other site labeled above background.
`For HER2 probe 2, 36 percent of cells
`had silver grains over the ql2-*q22
`bands of chromosome 17. Of all silver
`grains, 17 percent (42/246) were localized
`to this chromosomal region. A second-
`ary site of hybridization with 3.3 percent
`(8/246) of silver grains was detected at
`bands pl3-ql 1.2 of chromosome 7.
`To test whether this secondary site
`represented cross-hybridization with the
`EGF receptor gene, in situ hybridization
`was carried out with 3H-labeled EGF
`
`a
`
`6.4
`5.5-
`4.8-
`
`b
`
`7.4-
`4.8-
`
`Fig. 2. Northern blot hybridization analysis of
`normal and malignant human tissues. (a) Fetal
`tissues; (lane 1) term placenta, (lane 2) 20-
`week placenta, (lane 3) 20-week liver, (lane 4)
`20-week kidney, (lane 5) 20-week lung, (lane
`6) 20-week brain. (b) Embryonic tumors; (lane
`1) hepatoblastoma, (lanes 2 and 3) Ewing
`sarcoma, (lane 4) rhabdomyosarcoma, (lanes
`5 and 6) neuroblastoma, (lane 7) Wilms' tu-
`mor. Total poly(A)+ RNA was isolated as
`described (33); 4 ,ug per lane was analyzed on
`a I percent formaldehyde-agarose gel. 32p_
`Labeled HER2-1 and HER-2 (legend to Fig. 5)
`were used as hybridization probes under high
`stringency conditions [50 percent formamide,
`5x Denhardt's solution, Sx standard saline
`citrate (SSC), sonicated salmon sperm DNA
`(50 ,ug/ml), 50 ,uM sodium phosphate buffer
`(pH 6.8), 1 mM sodium pyrophosphate, and
`10 jM ATP at 42°C for 16 hours; filters were
`washed three times for 15 minutes at 45°C
`with 0.2x SSC]. The filters were exposed at
`-60°C with a Cronex Lightning Plus intensi-
`fying screen (Dupont) for 7 days. Rat ribo-
`somal RNA's were used as size standards
`(28S, 4.8 kb; 18S, 1.8 kb). RNA sizes are
`given in kilobases.
`
`receptor subclone 64-3. Of 100 cells ex-
`amined, 30 had silver grains at bands
`pl3--qll.2 of chromosome 7 and 3 per-
`cent (5/166) of total grains were found
`over ql2-*q22 of chromosome 17. With
`the other variant probe (HER2-1) no
`grain accumulation was observed at the
`EGF receptor site on chromosome 7.
`Southern blot analysis (25) of DNA
`extracted from nine somatic cell hybrids
`from human and rodent cells confirmed
`the localization of HER2 sequences to
`32P-labeled HER2-1
`chromosome 17.
`and HER2-2 probes were hybridized to
`the same set of Eco RI-digested DNA
`samples. With HER2-1, a 13-kb hybrid-
`izing band was detected in human DNA
`(Fig. Sc, lane 1) and in DNA samples
`from hybrids containing human chromo-
`some 17 (Fig. 5c, lanes 6, 8, 10, and 12).
`Likewise, hybridization of HER2-2 to a
`6.6-kb DNA fragment was observed in
`human control DNA (Fig. Sc, lane 1) and
`in hybrids containing human chromo-
`some 17 (Fig. 5c, lanes 6, 8, 10, and 12).
`Chromosome 17 was the only chromo-
`some with perfect concordant segrega-
`tion; all other chromosomes were ex-
`cluded by two or more discordant hy-
`brids.
`Regional localization to chromosome
`17 was also confirmed by Southern blot
`analysis. In a mouse-human hybrid con-
`taining a rearranged human chromo-
`some 17 with region 17q21-)qter, the
`human HER2 restriction fragments were
`detected (Fig. Sc, lane 4). The HER2
`gene was therefore localized to region
`17q21--qter, in agreement with the local-
`ization made by in situ hybridization.
`Even though a low level of hybridiza-
`tion with probe HER2-2 was seen at the
`site of the EGF receptor gene on chro-
`mosome 7, we were able to show that
`this finding represented cross-hybrid-
`In a control experiment an
`ization.
`EGF receptor probe cross-hybridized to
`the same extent with the HER2 site on
`17q.
`Taken together, the results of the in
`situ and Southern blot hybridizations
`permit the site of the HER2 sequences to
`be further narrowed down to bands
`17q21-q22, with the major peak of silver
`grains at band 17q21.
`HER2 expression in normal and malig-
`nant tissues. To obtain further clues re-
`garding the function of this receptor both
`in normal cells and in neoplasms, North-
`ern hybridization analyses (15) were car-
`ried out with several normal human tis-
`sues and randomly collected tumors. A
`hybridizing 4.8-kb mRNA was detected
`in all human fetal tissues analyzed, in-
`cluding term placenta, 20-week placenta,
`liver, kidney, lung, and brain obtained
`SCIENCE, VOL. 230
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1043 Page 3 of 8
`
`

`
`1 ATTGGTGCACGCAGGTGGGCAGGTGGGGGGCGCCCCCCCCGACCCCCGGCTCACGGCACGACAGGCGACGATACC
`1
`10
`20
`30
`40
`50
`MetGluLeuAlaAlaLeu UArgTrpGlyLeuLeuLeuAlaLeuLeuProProGlyAlaAlaSerT'hrGInVal UThrGlyThrAspMetLysLeuArgLeuProAlaSerProGluThrHisLeuAspt4etLeuArgHisLeuTyr
`60
`80
`90
`
`100
`
`110
`
`120
`
`-
`
`130
`
`140
`
`150
`
`1351
`
`160
`170
`180
`........190
`200
`GlyGlyValLeul leGlnArgAsnProGlnLeuUTyrGlnAspThrlleLeuTrpLysAspllePheHisLysAsnAsnGlnLeuAl aLeuThrLeul leAspThrAsnArgSerArgAlaUlsPomePo4tmy
`210
`220
`230
`240
`250
`Gl)ySerArgTWrpGlyGluSerSerGluAspEGlnSerLeuThrArgThrVal
`LysGlyProLeuProThrAspUEHisGluGlnEAlaAlaGlyhrlyProLyIsHisSer
`AlaGlyGiy
`laArg
`751GGTCG
`GGGGGGTTAGT
`AGGCGCCCCGC
`CGGGC
`ATACG CTCGCCACCT
`CCGCATA ~
`CCC
`~~~~~270.
`260
`280
`290
`300
`G1uLeuHisfWroA1aLeuVa1ThrTyrAsnThrAspThrPheG1uSerM4etProAsnProG1uG1yArgTyrThrPheG1yA1aSer
`,AspULeuA1aELeuHisPheAin-iHTsS7eG1yI1 e
`ValThrAl-~Prou
`901 GCFGCMTCCTACAATGACGGTCC CGCTGCCTCAAAAAGGGCAGCATCGGGCGAAATGCCGM
`310
`320
`330
`340
`350
`TyrAsnTyrLeuSerThrAspValGlySer*ThrLeujVal UProLeuHisAsnGlnGluValThrAlaGluAspGlyThrGlnArg*GlIuLys*SerLysProNAlaArgVal*TyrGlyLeuGlyqetGluHlsLeu
`1051TAACACCAGAGGGTCACTGCCCTCCACGGTAACGGAGACAGGI AA
`AGAGCU CGGGTTGCGGAGACCS
`360
`370
`380
`390
`400
`ArgGlValArAlaVlThrSrAlasnIleInGlPheAll-.IMLysLslIePeGlyerLeulaPhLeuPrGSuSrPheAphlyspProlaSersnThAPaPrLeuGnProGuGlneuGanaPPh
`410
`420
`430
`440
`450
`GluThrLeuGluGluIleThrGlyTyrLeuTyrJ leSerAlaTrpProAspSerLeuProAspLeuSerValPheGlnAsnLeuGlnValIlleArgGlyArgIleLeuHisAsnGlyAlaTyrSerLeuThrLeuGlnGlyLeuGlyJle
`GCCGAGGTAAGTCTTCACCGAGCGAACTCTACCGGCTCAACTCGATCGGCATCGAAAGCCTCCCGCCGAGGTGCT
`460
`470
`480
`490
`SW0
`SerTrpLeuGlyLeuArgSerLeuArgGluLeuGlySerGIyLeuAlaLeuIleHisHisAsnThrHisLeuMPheValHisThrVal ProTrpAspGlnLeuPheArgAsnProHisGlnAlaLeuLeuHisThrAlaAsnArgPro
`~~~~~~~520
`510
`550
`GluAspGluUValGlyGluGlyLeuA1loaEHisGlnLeuEA laArgArgAlaLeuLeuGlySerGlyProThrGl nVaI lnerinPheLeuArgGlyGInGluVa1G'uG1luUrgValLeuG lnGlyLeu
`570,............,580
`56
`590
`HlsProGluEGlnProGlnAsnGlySerValThrUPheGlyProGluAlaAspGlnflValAlaEA 1aHisTyrLysAspProProP
`LeuPro
`ProArgGluTyrValAsnAlaArgHis
`VaAAri
`1801 CCCAGGGAGTATGTGAATGCCAGGCACTTGCCACCCTGAGCWAGCCCCAGAATGGCTCAGTGACCTTGACGGAGGCTGACCAG GGCCGCCCACTATAAGGACCCTCCCT~I~C
`610
`620
`640
`650
`ProSerGlyValLysProAspLeuSerTyrlletProlleTrpLysPheProAspGlGluGluGyAla*GlnPro*Prolle S
`ValAspLeuAspAspLysGl ~ProA1aG1uG1nArgA laSerPro
`rHsSer
`1951 -CCCAGCGGTGTGAAACCTGACCTCTCCTACATGCCCATCTGGAAGMTCCAGATGAGGAGGGCGCA*CAGCCTCCCATCAAC
`CCATCCUGTGGACCTGGATGACAGG.CCGCCGAGCAGAGAGCCAGCCCT
`670
`660
`680
`690
`700
`LeuThrSerI leValSerAlaVal ValGlyl leLeuLeuVal Val ValLeuGlyVal ValPheGlylleLeuJlu yArgArgGlnGlnLyslleArgLysTyrmrM4etArgArgLeuLe'uGlnGluThrGluLeuValGluProLeu
`TCTCGGTGTTGCATCTGTGG_GTGTCTGGGGTGGCMGGATCT_CGAATAC_CGGGACTCTGAGGAACGAGcGGTGAGeG_T
`2101CTGCGTC
`710
`720
`730
`740
`750
`ThPrSeGyAarePrAsGeAaGnGeArlyLeAsGuarNueurtyVaPyVoLuGySrsynahGlThVaTALsGyGerplerAsGeGuAnVArsgeroa
`gfic
`~~~~770
`780
`790
`8SW
`
`810
`
`A
`
`830
`
`840
`
`850
`
`880 A
`860
`870
`890
`900
`ValLeValLySerPrAsnHiValLylIeThAspPeGlyLuAlaAgLeuLuAspIeAspGuThrluTyrisAlaspGTylyMysaAProLeuysrpSetAlaLuGeuSrrgeLuArgAgArgPTTlr
`A
`910
`920
`930
`940
`9s0
`HisGlnSerAspValTrpSerTyrGlyValThrValTrpGluLeuMetThrPheGlyAlaLysProTyrAspGlyl leProAlaArgGlulleProAspLeuLeuGluLysGlyGluArgLeuProGlnProProll eThrll.eAsp
`CAATAGGGATAGTTATGGGGGTAGCMGGCACTAGTGGTCACCGAACCGCTCGAAAGGACGTCCACCCACCATA
`A
`970
`980
`990
`1000
`ValTyrMetlleMetValLysoTrpNetlleAspSerGluAOrgProArgPheArgGluLeuValSerGluPheSerArgMetAlaArgAspProGlnArgPheValVal IleGlnAsnGluAspLeuGlyProAlaSerProLeu
`CAAGTAGTAAWGTATACCGA
`GCAGTCGGGTGGCGATCCCCTGCGGCCCGGMTGCACAATAGCTGCCGCGCCT
`1010
`1020
`1030
`1040
`1050
`
`2851
`
`3001
`
`1060
`
`1070
`
`1080
`
`109
`
`1100
`
`3451
`
`3601
`
`1110
`1120
`1130
`1140
`1150
`LeuProThrHisAspProSerProLeuGlnArgTyrSerGluAspProThrVal ProLeuProSerGluThrAspGlyTyrValAlaProLeuThirlerProGlnProGluTyrValAsnGlnProAspValArgProGlnPropro
`CCAAAGCCACCCAACGTCGGGACCCGACCGCTTAATGTGTCTGCCCGCW CCCGCGATTTACACAAGTGCCACC'C
`1160
`1170
`1180
`1190
`1200
`SerProArgGluGlyProLeuProAlaAlaArgProAlaGlyAl aThrLeuGluArgAlaLysThrLeuSerProGlyLysAsnGlyVal ValLysAspVal PheAlaPheGlyGlyAlaValGluAsnProGluTyrLeuThrProBln
`GCCAAGCCCGCGTGCGCTCGTCATTGAGGCAGCCCCCAGAGAGGTCTAAAGTMGCMGGGGCGGAACCGGATGCCCA
`~~~~~1210
`1220
`1230
`1240
`1250
`17
`
`1
`
`1255
`LeuAspVal ProVaJEND
`GAGGCGGGACGAGCAGCGAAGCTAGGCTAGGGAGAGCTATCGTGATAGGTGAGCCCGCATCCGGACTCAGCGACTT
`3901
`4051
`AGACTCTCTCTATTCAAGCGAGGTCGCCTTGAAGAACCGGATTTTGATTAGCTCCAGGCCAGGCATGTCAACCGTO
`CTCTCAACTGTCGAGCTAGAGTGCGGGGAGGCCMGAGGCAGAAAGGCCTCGGCGCCTAACATCGCCCTAGAGAAC
`4201
`4351 GTTATTCGCGAAATCTCGATMC MGM
`TAAACAAAAACAGGGAGGGTTTGGAGAGGGGGTCTTCCCCC
`4501 TGCATTGCCAAATATATTTTGAAAAAAAAAAAAAA
`Fig. 3. Complete nucleotide and amino acid sequences of HER2 (clone XHER2-436). Synthetic probes 1 and 2 (Fig. 1) were used to screen 2 X 106
`clones of a human placental cDNA library in XgtlO0 (17') as described (9). Fifty-two clones were isolated and characterized by Eco RI restriction di-
`gestion and Southern blot hybridization (25) with the radiolabeled synthetic oligonucleotides (34) and, on a duplicate blot, an EGF receptor Eco
`RI fgragent (HER64-3) as probes. Labeling of oligonucleotides was achieved with 1 unit of T4 polynucleotide kinase (New England Biolabs), 10
`pmol of oligonucleotide and 30 pMol Of ('-_32p]ATP (Amersham) as specified by the supplier. HER64-3 was labeled to high sp~ecific activity (>106
`cpm/,ug) as described (34). Clone XHER2-436 was used for nucleotide sequence analysis (32) of overlapping subclones. Amino acids are
`numbered starting with the initiation methionine (1). Underlining indicates the putative signal sequence (heavy) and the potential poly(A) addition
`signal (fine). Lines on top of the sequence indicate potential glycosylation sites, the black arrow demarcates the EGF receptor threonine 654
`analogue (Thr 686), shading indicates cysteine residues, boxing shows the putative transmembrane region, large open triangles indicate some
`locations ofintrons in the HER2 gene, and the small triangle emphasizes a possible tyrosine autophosphorylation site by homology to Tyr 1173 in
`the EGF receptor sequence.
`6 DECEMBER 1985
`
`1135
`
`PETITIONER'S EXHIBITS
`
`Exhibit 1043 Page 4 of 8
`
`

`
`from a single
`fetus
`2a). Two
`(Fig.
`mRNA's, of 5.4 and 6.4 kb, were also
`detected in term placenta. No cross-
`hybridization with the 5.8-kb and 10.5-kb
`
`EGF receptor mRNA's in
`term placenta
`ler these strin-
`mRNA was observed und
`tions
`gent hyridization
`condil
`(legend,
`Fig. 2). Normal adult h
`uman tissues,
`
`I.,
`
`A M
`
`IA
`
`Ah MibAlA&
`
`I A
`
`A A A
`
`40.6%
`
`78.4%
`
`18.7%
`
`Y (?)
`
`4-Thr654'
`
`III %ill-T
`
`Popyly
`
`I
`
`.
`
`"
`
`A i
`
`il I A
`
`A
`
`.
`
`. h
`
`. A.
`
`A I . A.
`
`11A
`
`a
`
`l-
`
`HER 2
`HERI
`3.0r
`
`-24
`
`1
`
`77
`96
`167
`196
`
`267
`295
`
`366
`395
`
`466
`495
`1
`
`566
`595
`11
`
`b H
`
`ER1
`HER2
`
`HER1
`HER2
`
`HERI
`HER2
`HER1
`HER2
`HER1
`HER2
`
`HERI
`HER2
`erbB
`HER1
`HER2
`erbB
`
`including kidney, liver, skin, lung, jeju-
`num, uterus, stomach, and colon, con-
`tained lower but significant amounts of
`the same 4.8-kb mRNA. Because of the
`magnitude of fetal expression, we also
`examined several embryonic
`tumors
`(Fig. 2b); each expressed large amounts
`of the 4.8-kb transcript, although not
`more than that detected in normal fetal
`tissue.
`Thus, it appears that the HER2 gene is
`widely expressed, in both normal adult
`tissues and in several normal fetal tis-
`sues. While detected in most embryonic
`tumors, the HER2 gene was not present
`at higher levels than in fetal tissues; thus,
`the particular level may reflect the state
`of differentiation of a given tumor.
`HER2 structurally characterized as cell
`surface receptor. Using the transforming
`gene of the avian erythroblastosis virus,
`v-erbB, as a hybridization probe, we
`isolated genomic and cDNA sequences
`of an uncharacterized human gene. The
`1255 amino acid polypeptide sequence
`
`F
`
`L Q
`
`ENRT
`
`HA E
`
`K
`
`R
`
`E
`
`F
`
`L
`
`H~ K T
`
`A
`
`MGENNTLV-
`
`A A GHVi;-'HHL
`
`".YG..
`
`TGK
`
`~~~~~~~~e
`
`D
`
`H~
`
`KA~
`
`A
`
`L
`
`I
`G
`AHYKDPP
`MKI
`R
`
`Fl
`
`GRRLLQET
`
`I
`
`2
`
`3
`
`4/5
`
`TQLGTF D FLS QRMFNN'...E
`g
`L
`ASR LEEKK IQ
`MRPSGT GAALLA
`A
`SN
`I
`VQR YD
`KT
`T ERI
`A
`MELAALIRWGLLLALLPPGA STQ--V.iTGTDMKLRLPASPETHLDMLRHLYQGXQVVQGNLELTYLPTNASLSFLQDIQEVQGYVLIAHNQVRQVP
`A ...NVES Q R
`EN Q I
`NMYY NS
`S Y A- K ------K PM N Q
`H A RFSN
`VSSDFLSNMSMDFQ HLGS
`K.-I
`LQRLRIVRGTQLFEDNYALAVLDNGDPLNNTTPVTGASPGGLRELQLRSLTEILKGGVLIQRNPQLVYQDTILWKDIFHKNNQLALTLIDTNRSRA HPg
`D SIPNGS.:.
`SPM KGSR".WGESSED..QSLTRTV'CAGGr-A-RO-KGPLPTD--HEQ-AAGt¢TGPKHSD§LA-LHFNHSGItELHtPALVTYNTDTFESMPNPEGRYTFGAS
`QQ..'.SG..R KS S
`AGE Ni K
`KI I
`K DTKDT P ML
`N
`RE
`I% V R
`PT YQMKDV
`H LVRA.GADSY ME-
`VRK.:'K
`:::.N
`KH KND:S S D
`KK.:
`D
`R
`VV
`I
`:EG
`IGEFKDSLSINAT
`SFTH P
`HI
`VA R
`VTAK.PYNYLSTDVGS .TLV-PLHNQEVTAEDGTQR EK
`KP ARVgYGLGMEHLREVRAVTSANIQEFAG KIFGSLAFLPESFDGDPASNTAPLQP
`DVI SG KNE-YAN IN KK
`ISD
`QE DILK VK
`El
`TKQH QF
`AVVS N TS
`EQLQVFETLEEITGYLYISAWPDSLPDLSVFQNLQVIRGRILHNGAYSLTLQGLGISWLGLRSLRELGSGLALIHHNTHL FVHTVPWDQLFRNPHQALL
`GTSG KTK
`A.2..SPEG#W PE
`IIS
`G NS.>KAT QV
`S..RNVS
`ENSE..IQ..
`AMNI ...T RGP N.IQ..
`RD.
`NDKNL E
`HTANRPEDEVVGEGLAH L
`PEADSGPTQAVNqKPVEERVLQGLPREYVNARHPLPtHPEXQPQNGSVT#FGPEADQEVA
`H. KT.A MGENTLV-
`YA AGH.>.HLHP .. YG>.GPGLE .>TGK
`RSS/
`LLLVVALGIG FS
`F¢VARJPSGVKPDLSYMPIWKFPDEEGA QPIPIN'r-THS'%' LDDKGtPAEORASPLTSIVSAVVGILLVVVLGVVFGILIRRQQKI-RK!RLQE
`ATGM
`A
`RHIV
`RAG11
`
`VS
`
`S
`
`.NKGPGLE
`L:....NHP
`L
`FK I ***
`EA
`LL
`*E
`E
`S DN H C
`A
`K
`ELVEPLTPSGAMPNQAQMRILKETELRKVKVLGSGAFGTVYKGIWIPDGENVKIPVAIKVLRENTSPKANKEILDEAYVMAGVGSPYVSRLLGICLTSTV
`S DN H C
`EA
`FK
`HL
`I
`E
`L
`E
`A
`K
`QLVTQLMPYGCLLDHVRENRGRLGSQDLLNWCMQIAKGMSYLEDVRLVHRDLAARNVLVKSPNHVKITDFGLARLLDIDETEYHADGGKVPIKWMALESI
`LRRRFTHQ