Downloaded from http://aacrjournals.org/cancerres/article-pdf/55/10/2206/2457268/cr0550102206.pdf by guest on 15 June 2023
`
`[CANCER RESEARCH 55, 2206-2211, May 15, 1995]
`
`Isolation and Sequence Analysis of Human Cadherin-6 Complementary DNA for
`the Full Coding Sequence and Its Expression in Human Carcinoma Cells1
`
`Yutaka Shimoyama, Masahiro Gotoh, Takeo Terasaki, Masaki Kitajima, and Seism»Hirohashi2
`
`5-9-4 Tokodai, Tsukuba 300-26 ¡Y.S.,
`of Japan (JRDC), Tsukuba Research Consortium,
`Hirohashi Cell Configuration Project, ERATO, Research Development Corporation
`5. H.I; Pathofagv Division, National Cancer Center Research Institute, 5-1-1 T-sukiji. Chuo-ku, Tokyo 104 ¡Y.S., M. G., T. T., S. ri.}; and Department
`of Surgery, School of
`Medicine, Keio University, 35 Shinanomachi.
`Shinjuku-ku, Tokyo 160 ¡Y.S., M. K.I, Japan
`
`ABSTRACT
`
`The expression pattern of E- and P-cadherin in human carcinomas has
`been reported by many laboratories. However,
`little is known about
`the
`involvement
`of other cadherin types in human carcinomas.
`cDNA clones
`for a cadherin molecule were isolated from a cDNA library of human
`hepatocellular
`carcinoma cells which lacked E- and P-cadherin expression
`but exhibited cell aggregation activity mediated by an unknown cadherin,
`and they were subjected to sequence
`analysis. The overlapped
`clones
`covered 4315 nucleotides
`and were found to encode a typical cadherin
`molecule
`consisting
`of 790 amino acids. Since the deduced amino acid
`sequence was identical
`to a partially available human cadherin-6 sequence
`except
`for two amino acid residues,
`the clones were considered
`to be
`human cadherin-6
`cDNAs encoding the entire open reading frame. The
`deduced amino acid sequence also showed extremely high homology with
`recently reported rat K-cadherin,
`97% for the putative mature protein,
`suggesting
`that cadherin-6
`is the human counterpart
`of rat K-cadherin.
`Expression of cadherin-6 in various human normal
`tissues and carcinoma
`cells was examined
`by Northern
`blot analysis using a specific probe
`corresponding
`to the signal and precursor
`sequence. Among normal
`tis
`sues examined, brain, cerebellum, and kidney showed strong expression of
`cadherin-6, whereas
`lung, pancreas,
`and gastric mucosa showed weak
`expression. Transcripts
`of cadherin-6 were not detected in normal
`liver,
`whereas
`four of six hepatocellular
`carcinoma
`cell
`lines examined
`ex
`pressed cadherin-6
`abundantly. As reported for rat K-cadherin,
`three
`renal carcinoma
`cell
`lines also expressed cadherin-6
`strongly. The most
`interesting
`finding was obtained for small
`cell
`lung carcinoma
`lines.
`Among 15 of such cell
`lines examined, all of 11 cadherin-6-positive
`lines
`were classified into the classic type, whereas the negative cell lines were all
`of
`the variant
`type. The present
`results
`suggest
`that besides E- and
`P-cadherin, other cadherin molecules are expressed in human cancers and
`are responsible
`for additional biological properties of the carcinoma cells.
`
`INTRODUCTION
`cell-cell adhesion molecule which
`Cadherin is a Ca2+-dependent
`mediates cell-cell binding in a homophilic manner. Cadherin consti
`tutes a gene family and consists of subclasses with individual binding
`specificities.
`Initial studies identified three subclasses, epithelial
`(E-),
`neural
`(N-) and placenta!
`(P-) cadherin, and further characterization
`revealed that they played a key role in many morphogenetic
`events as
`well as in the maintenance of orderly structures
`such as epithelium
`(reviewed in Refs. 1 and 2). We started studying human cadherins,
`focusing on their possible involvement
`in the biological properties of
`cancer, since it was conceivable that metastasis and invasion of cancer
`could not occur without disruption of the mutual adhesion of cancer
`cells (3-6).
`and clinical evi
`To date, a considerable volume of experimental
`dence has been obtained to indicate that dysfunction of E-cadherin, a
`major cadherin molecule expressed in both epithelium and cancer,
`
`Received 12/28/94; accepted 3/15/95.
`The costs of publication of this article were defrayed in part by the payment of page
`charges. This article must therefore be hereby marked advertisement
`in accordance with
`18 U.S.C. Section 1734 solely to indicate this fact.
`1This work was supported in part by a Grant-in-Aid from the Ministry of Health and
`Welfare of Japan for the 2nd Term: Comprehensive
`10-Year Strategy for Cancer Control.
`2 To whom requests
`for reprints should be addressed.
`
`of cancer
`and progression
`to enhanced invasiveness
`least
`leads at
`that
`there are a number of
`(4, 5, 7-12), and it has become apparent
`mechanisms
`responsible
`for
`the E-cadherin dysfunction
`in cancer,
`including loss or reduction of expression, mutations
`(13, 14), and
`aberrations
`of catenins
`(6, 15-22), which connect
`cadherin to the
`cytoskeleton network.
`Besides
`the three subclasses mentioned above, many molecules
`with characteristics of cadherin have been identified over the past few
`years. These include the typical cadherins, which possess the same
`domain structure (23-26),
`a truncated-type
`cadherin (27), cadherin-
`related proteins with more than five extracellular
`subdomains
`(28),
`and desmosomal
`cadherins
`(29-31),
`indicating the existence of a
`cadherin superfamily.
`In addition, a PCR technique based on the high
`conservation
`among cadherin
`family members
`has facilitated
`the
`molecular
`cloning of various
`cadherins
`(32-34). Because E- and
`P-cadherin are thus far the only two cadherins
`that have been well
`characterized in cancer,
`it would be of interest
`to determine whether
`other cadherins
`are expressed
`in cancer
`and affect
`its biological
`properties. Before undertaking the present study, we had noticed some
`cancer
`cell
`lines exhibiting
`strong cell aggregation
`activity after
`trypsin/Ca2+ treatment
`(35) without expression of E- and P-cadherin
`
`suggesting that other cadherin molecule(s) were
`(data not shown),
`present
`in them. In this report, we describe the molecular cloning and
`sequence analysis of a cadherin molecule from one of those lines, a
`human hepatocellular
`carcinoma
`cell
`line LÌ21(36), and its unique
`expression pattern in cancer cells.
`
`MATERIALS AND METHODS
`
`PCR. To obtain a cDNA fragment of the unknown cadherin expressed in
`Li21 cells, PCR was performed based on a strategy similar to that described by
`Suzuki et al. (32). The primer pair used was the upstream primer 5'-GGA-
`ATTCAC(ACGT)GC(ACGT)CC(ACGT)CC(ACGT)TA(CT)GA-3'
`and the
`downstream primer GGAATTCTC(ACGT)GC(ACGT)A(AG)(CT)TT(CT)-
`TT(AG)AA-3',
`both of which contained an EcoRl
`recognition site at their 5'
`ends. Poly(A)*3 RNA was prepared
`from C-LÌ21cells
`as described
`by
`Okayama et al. (37). Single-strand cDNA was synthesized using Superscript
`II
`(GIBCO-BRL, Gaithersburg, MD) and subjected as a template to 35 cycles of
`PCR (94°C,55°C,and 72°Cfor 1.5, 2, and 3 min,
`respectively). The PCR
`
`product, about 160 bp long, was purified, digested with EcoRI, and subcloned
`into the EcoRI
`site of
`the pBluescript
`II SK(—) phagemid
`vector. This
`fragment was used as a probe for cDNA cloning as described below.
`Construction
`and Screening
`of cDNA Library. An oligo-dT-primed
`cDNA was synthesized
`from poly(A)+ RNA of LÌ21cells using a cDNA
`
`system plus kit (Amersham, Buckinghamshire, United Kingdom),
`synthesis
`ligated to the arms of AgtlO via EcoRI adaptors, and packaged in vitro using
`the cDNA cloning
`system, AgtlO (Amersham).
`The phage plaques were
`transferred
`to nylon filters. The filters were hybridized
`at 42°Cin a buffer
`
`labeled
`containing 50% formamide, 0.65 M NaCl, and the above PCR fragment
`with [a-12P]dCTP, washed twice with 0.1X SSC-0.1% SDS at 65°Cfor 30
`min, and exposed to Kodak XAR-5 film (Rochester, NY) at -70°C with an
`
`intensifying
`conditions
`
`screen. Positive
`as those above.
`
`clones were plaque purified under
`
`the same
`
`3 The abbreviations used are: poly(A) +, polyadenylated; bp, base pair(s); SCLC, small
`cell lung cancer; N-CAM, neural cell adhesion molecule.
`
`2206
`
`1 of 6
`
`OnCusp
`Ex. 1021
`
`

`

`Downloaded from http://aacrjournals.org/cancerres/article-pdf/55/10/2206/2457268/cr0550102206.pdf by guest on 15 June 2023
`
`SEQUENCE AND EXPRESSION OF HUMAN CADHERIN-6
`
`DNA Sequence Analysis. The inserts of the positive clones were sub-
`cloned into pBluescript
`II SK(-)
`or pUCllS
`(Takara Shuzo, Kyoto, Japan).
`The phagemids were denatured by alkali
`treatment and subjected to sequence
`analysis
`using
`a 7-deaza-GTP
`Sequenase
`version
`2 kit
`(United
`States
`Biochemicals, Cleveland, OH). The DNA sequence was determined on both
`strands successively using oligonucleotide
`primers which anneal
`the vector and
`insert DNA sequences. The nucleotide
`and amino acid sequences were ana
`lyzed using the Gene Works software package (IntelliGenetics, CA).
`Northern Blot Analysis. Total RNA was prepared from various human
`tissues and cancer cell
`lines by acid guanidinium thiocyanate-phenol-chloro-
`form extraction (38), and poly(A)+ RNA was isolated using oligo-dT latex
`beads
`(01igotex-dT30
`Super; Daiichikagaku,
`Tokyo,
`Japan). Human brain,
`cerebellum,
`and pancreas
`poly(A)+ RNAs were purchased
`from Clontech
`(Palo Alto, CA). Poly(A)+ RNAs (2 ¿ig/lane)were separated in 1% agarose/
`formaldehyde gels,
`transferred to nitrocellulose
`filters, and hybridized at 42°C
`
`and 0.65 M NaCl with a specific PCR
`in a buffer containing 50% formamide
`probe described
`below. To avoid cross-hybridization
`with other
`cadherin
`subclasses,
`a 182-bp nucleotide
`sequence
`(positions
`116-297
`in Fig. 1) en
`coding the signal peptide and precursor
`region, where significant homologies
`have not been reported among different
`subclasses, was amplified by a PCR-
`labeling procedure (13) and used as a probe. We confirmed that this probe did
`not detect either E- or P-cadherin (data not shown). The filters were washed
`twice with 2X SSC, 20 mM sodium phosphate buffer, 0.06% sodium PP¡,and
`0.05% SDS at 65°Cfor 30 min and exposed to XAR-5 film at -70°C with an
`
`intensifying screen.
`
`RESULTS
`
`Bl-1, proved to contain the entire open reading frame. Although Bl-1
`also had a poly(A)
`tail at the 3'-end,
`the addition site was different
`from that of AL1-1. Fig. 1 shows the two clones schematically with
`representative
`restriction endonuclease
`sites.
`is shown
`The overlapping nucleotide sequence, 4315 bp in length,
`in Fig. 2. The open reading frame begins with an ATG codon at
`position 121-123,
`terminates at a TAA codon at position 2491-2493,
`and consists of 2370 nucleotides encoding 790 amino acids. A polya-
`denylation signal, ATTAAA,
`is identified at position 3101-3106
`for
`clone Bl-1, but no obvious signal
`is present for ALI—1.The deduced
`amino acid sequence
`contains
`two hydrophobic
`sequences
`corre
`sponding to the signal peptide and transmembrane
`domain,
`the long
`extracellular domain, and the short cytoplasmic
`domain, which are
`structures common to typical cadherins
`(1). This polypeptide
`is ex
`pected to undergo posttranslational modifications
`including proteo-
`lytic cleavage at amino acid position 53-54 (39) and to be expressed
`at the cell surface as the mature protein, which is 737 amino acids long
`with 5 consensus
`sites for N-linked glycosylation
`(40).
`Comparison with Other Cadherins. A homology search with
`cadherin sequences
`reported previously revealed that
`this cadherin
`shows striking identity with two cadherin sequences. One of these is
`cadherin-6, which has been cloned from a human fetal brain library
`and partially sequenced (32). The amino acid sequence of the present
`cadherin from position 377 to 790 completely agrees with the reported
`424 amino acids of cadherin-6,
`except
`for only two amino acid
`residues (Val at position 421 and Thr at position 425 are both He in
`cadherin-6),
`strongly suggesting that the clones presented here encode
`cadherin-6. To avoid confusion,
`the cadherin cloned in the present
`study will be referred to as cadherin-6 hereafter. The other cadherin
`highly homologous with this cadherin-6 is rat K-cadherin, which was
`recently isolated from a rat renal cell carcinoma library (41). Homol
`ogies of cadherin-6 with rat K-cadherin
`and other known human
`cadherins are summarized in Table 1. Homology between cadherin-6
`and rat K-cadherin is extremely high, being 97% for the putative
`mature protein,
`as compared with those between cadherin-6
`and
`known human cadherins. Even for the signal and precursor sequences,
`where each cadherin subclass shows a characteristic sequence,
`the two
`cadherins are markedly homologous,
`suggesting that cadherin-6 is the
`human counterpart of rat K-cadherin.
`Comparison of cadherin-6 with other human cadherin subclasses
`reveals that
`there are two groups of cadherin on the basis of resem
`blance to cadherin-6. One group includes cadherin-8,
`-11 and -12,
`which exhibit over 60% homology with cadherin-6, and when aligned
`with cadherin-6,
`few gaps are required. The other group includes E-,
`N- and P-cadherin, and cadherin-4 and -5, which show lower homol
`ogy, 30-40%, with cadherin-6. Cadherin-13, which lacks the trans-
`membrane
`and cytoplasmic
`regions,
`also shows
`lower homology,
`30%, for the extracellular part.
`in Human Tissues and Carcinoma
`Expression of Cadherin-6
`Cell Lines. To examine cadherin-6 expression by RNA blotting with
`a
`out cross-hybridization
`with other cadherin
`species, we chose
`182-bp fragment corresponding to the signal and precursor sequences
`as a probe. Fig. 3 shows cadherin-6
`expression
`in LÌ21cells and
`various normal human tissues. Cadherin-6 transcripts were detected as
`four bands intensely in LÌ21cells, brain, cerebellum, and kidney and
`faintly in lung, pancreas, and gastric mucosa, with approximate mo
`lecular sizes of 9, 4.1, 3.4, and 2.7 kb. This result
`is inconsistent with
`the finding by Xiang et ai.
`(41)
`that K-cadherin was scarcely ex
`pressed in adult rat tissues and also the report by Suzuki et al. (32) that
`brain expressed cadherin-6 very faintly.
`Interestingly,
`expression of
`the 9-kb cadherin-6 transcript was very faint in LÌ21cells and kidney
`but
`intense in brain and cerebellum,
`and expression
`of the 2.7-kb
`transcript was also faint in cerebellum and kidney,
`indicating that each
`2207
`
`Amplification and Isolation of a Cadherin cDNA Fragment. A
`PCR method was used to isolate a cDNA fragment of an unknown
`cadherin expressed in LÌ21cells using a set of two degenerate oligo
`nucleotide primers encoding highly conserved sequences
`in the cyto-
`plasmic domain of known cadherins
`(32), and a fragment about 160
`bp long was obtained (data not shown). This fragment was subcloned
`into pBluescript
`II SK(-)
`and subjected to sequence analysis. The
`DNA sequence consisted of 155 nucleotides
`and showed significant
`homology with reported cDNA sequences of known cadherins
`(data
`not shown).
`cDNA Cloning and Sequencing. The AgtlO cDNA library con
`structed from Li21 poly(A)+ RNA was screened with the 155-bp
`fragment. Approximately
`4 X IO4 recombinants were screened, and
`only one positive clone, designated ALI-I, was isolated. This clone
`contained a cDNA insert of 3787 bp with the poly(A) sequence, but
`comparison with cDNA sequences of known cadherins
`revealed that
`it lacked the translation initiation codon and the following nucleotides
`encoding the signal peptide, precursor
`region, and most of extracel
`lular domain 1. In order
`to identify this sequence,
`the library was
`screened again using a 321-bp nucleotide
`located at
`the 5'-end of
`AL1-1 (positions 650-970
`in Fig. 1) amplified by PCR as a probe.
`Approximately
`2 X IO5 recombinants were rescreened,
`and three
`
`positive clones were isolated. Among these clones, one, designated
`
`B
`
`E
`
`IiXIH
`
`I
`
`B\Y—
`
`P
`
`•-—
`
`j
`
`1 Kb
`
`AL1-1
`B1-1
`
`EC
`
`CP
`
`H
`
`and Bl-1, and their restriction map wilh a schematic
`Fig. 1. Two cDNA clones, ALI-I
`drawing of cadherin structure. The box indicates the coding region. Mulched, solid, and
`clear areas represent
`the signal and precursor sequence,
`the transmembrane domain, and
`the extracellular
`(EC) and cytoplasmic (CP) domains,
`respectively. B, ßamHI;£.EcoRl;
`H. Hindin-, P. Pst\; X, Xbal.
`
`2 of 6
`
`OnCusp
`Ex. 1021
`
`

`

`SEQUENCE AND EXPRESSION OF HUMAN CADHERIN-6
`
`1
`ATGAGAACTTACCGCTACrrCTTGCTGCTCTITTGGGTGGGCCAGCCCTACCCAACTCTC^
`121
`1 MRTYRYFLLLFWVGOPYPT
`
`LSTPLSKRTSGFPAKKRALEL
`F
`N
`TCTGGAAACAGCAAAAATGAGCTGAACCGTTCAAAAAGGAGCTGÃ(cid:141)^TGTGGAATCAGTTCTTTCTC^
`241
`41SGNSKNELNRSKRSWMWNQFFLLEEYTGSDYQYVGKLHSD
`AR
`S
`
`À
`
`R
`
`Downloaded from http://aacrjournals.org/cancerres/article-pdf/55/10/2206/2457268/cr0550102206.pdf by guest on 15 June 2023
`
`81QDRGDGSLKYILSGDGAGDLFIINENTGDIQATKRLDREE
`
`601
`161
`
`AAGGAGGTTTA
`KEVYTATVPEMSDVGTFVVQVTATDADDPTYGNSAKVVYS
`D
`A
`ATTCTACAGOCSACAGCCCTATTTTTCAGTrGAATC AGAAACAGGTATTATCAAGACAGCTTTGCTCAAC ATGGATCGAGAAAACAGGGAGCAGTACCAAGTGGTGATTC AAGCCAAGGAT
`721
`201ILQGQPYFSVESETGIIKTALLNMDRENREQYQVVIQAKD
`
`241MGGQMGGLSGTTTVHITLTDVHDNPPRFPQSTYQFKTPES
`
`281SPPGTPIGRIKASDADVGENAEIEYSITDGEGLDMFDVIT
`
`321DQETQEGIITVKKLLDFEKKKVYTLKVEASNPYVEPRFLY
`
`361
`
`LGPFKDSATVRIVVEDVDEPPVFSKLAYILQIREDAQINT
`
`A
`
`O
`
`401 TIGSVTAQDPDAARHPVKYSVDRHTDMDRIFNIDSGNGSI
`A
`
`441
`
`FTSKLLDRETLLWHNITVIATEINNPKQSSRVPLYIKVLD
`
`R
`
`i
`
`i
`
`H
`
`HI
`
`A
`
`A
`
`A
`GTCAATGACAACGCCCCAGAATTTGCTGAGTTCTATGAAACTTTTGTCTGTGUUlAAAÅ“AAAGGCAGATCAGTTGATTCAGACCCTaCATGCTGTTGACAAGG^
`1561
`481VNDNAPEFAEFYETFVCEKAKADQLIQTLHAVDKDDPYSG
`
`521
`
`HQFSFSLAPEAASGSHFTIQDNKDNTAGILTRKNGYNRHE
`
`A
`
`561
`
`MSTYLLPVVISDNDYPVCSSTGTVTVRVCACDHHGNMQSC
`
`601
`
`H A
`
`E
`
`A_ L
`
`]HPTGLSTGALVAlLLC::V:iLLV'i'VVLFAALR
`
`R Q R
`
`641KKEPLIISKEDIRDNIVSYNDEGGGEEDTQAFDIGTLRNP
`
`OGCTTTTGATATCGGCACCCTGAGGAATCCT
`
`GAAGCCATAGAGGACAACAAATTACGAAGGGACATTGTOCCCGAAGCCCTTTTCCTACCCCGACGGACTCCAACAGCTCGCGACAACACCGATGTCAGAGATTTCATTAACCAAAGGTTA
`2161
`681EAIEDNKLRRDIVPEALFLPRRTPTARDNTDVRDFINQRL
`KPWRQQS-M
`
`S
`
`721
`
`2401
`761
`
`KENDTDPTAPPYDSLATYAYEGTGSVADSLSSLESVTTDA
`R K M N
`GATCAAGACTATGATTACCTTAGTGACTGGGGACCTCCÄTTCAAAAAGCTrGCAaATATGTATGGAaGAaTGGACAGTaACAAAGACTCCTAATCTGTTGCCTTTTTCA
`DQDYDYLSDWGPRFKKLADMYGGVDSDKDS«
`
`G
`
`G
`
`M
`
`2641 CTTTTTTCTAGTACACTTTTATGAGCTTCCAAGGGGCAMTTTTTATTTTTTAGTGCATCCACriTMCCAAGTCACXCCAACAGGCAGGTGCCGGA
`2761
`C ACTTGTTCTCAGGGCAGCGTGCCCaCTTCCGCTGTCCTGGTGTTTTACTACACTCCATGTC
`AGGTC AGCCAACTOCCCTAACTGTAC
`ATTTC ACAGGCTAATGGGATAAAGGACTGTGC
`
`3121 AGCAACCA^AAACCTAGTACGACTTCATTCCTTCCACTAACTCATAGTTnnTATATCCTAGACTAGACATGCGAAAGTTTGCCTrTGTACCATATAAAGGGOGA
`3241 ATGTTAACCAAGGAAATATATTTTACCATACATTTAAAGTTTTGGCCACCACATGTATCACGGGTCACTTGAAATTCTTTCAGCTATCAGTAGGCTAATGTCAAAATTGTTTAAAAATTC
`3361 TTGAAAGAATTTrCCTGAGACAAATTrTAACTTCTTGTCTATAGTTGTCAGTATTATTCTACTATACTGTACATGAAAGTAGCAGTGTGAAGTACAATAATTCATATTC
`
`3721 AAACAAAAGTGTTTAAGATTGTAATTAAAATGATAGTTGATnTC AAAAGCATTAATTTTTTTTCATTÃœTTTTTAACTTTGCTTTCATGACCATCCTGCCATCCTTGACTTTGAACTAAT
`3841
`3961
`
`4201
`
`TCATTTCCATTTGGGGATATTGTCATATCAÅ“ACATATTTTCTGTTTGGAAACACAC
`
`(A)n
`
`2208
`
`3 of 6
`
`OnCusp
`Ex. 1021
`
`

`

`Downloaded from http://aacrjournals.org/cancerres/article-pdf/55/10/2206/2457268/cr0550102206.pdf by guest on 15 June 2023
`
`SEQUENCE AND EXPRESSTON OF HUMAN CADHERIN-6
`
`selection of the polyade-
`type might show preferential
`tissue or cell
`nylation site. Cadherin-6 mRNA was not detected in poly(A)1 RNA
`
`liver, and colonie mucosa.
`isolated from heart,
`lines was
`Cadherin-6 expression in various human carcinoma cell
`then examined using an approach similar
`to that described above.
`RNA blotting of five hepatocellular
`carcinoma cell
`lines, except
`for
`LÌ21,is shown in Fig. 4A. LÌ22,LÌ23,and LÌ24cells (36) strongly
`expressed 4.1-, 3.4-, and 2.7-kb cadherin-6 transcripts.
`In contrast, no
`bands were detected in Li7 (42) and HepG2 (43). Consequently,
`four
`of six hepatoma
`lines strongly expressed cadherin-6. Among four
`renal cell carcinoma cell lines examined,
`three expressed cadherin-6
`(Fig. 4ß),which is consistent with the previous
`finding that K-
`cadherin was expressed in one of three human renal cell carcinomas
`(41). Most of the SCLC cell lines examined (11 of 15) also expressed
`cadherin-6 with various degrees of
`intensity,
`although cadherin-6
`mRNA was not detected in four lines, Lu24v, Lul35v, H82, and N417
`(Refs. 44 and 45; Fig. 4C). Besides
`the human carcinoma
`cells
`described here, we examined cadherin-6 expression in various human
`cell
`lines including five colon adenocarcinomas,
`four gastric adeno-
`carcinomas,
`two pancreatic adenocarcinomas,
`and one cell
`line each
`of
`lung adenocarcinoma,
`esophageal
`squamous
`carcinoma,
`vulvar
`squamous carcinoma, cervical carcinoma, bladder carcinoma,
`retino-
`blastoma, and transformed endothelium. Cadherin-6 transcripts were
`not, or only very weakly detected,
`in poly(A) + RNAs from these lines
`
`(data not shown).
`
`DISCUSSION
`
`We have reported here cloning of a human cadherin molecule from
`a cDNA library of a hepatocellular
`carcinoma cell line, LÌ21,which
`does not express either E- or P-cadherin but exhibits cell aggregation
`mediated by an unknown cadherin. The deduced amino acid sequence
`is essentially identical
`to that of partially identified human cadherin-6
`(32), except
`that Val at amino acid position 421 and Thr at 425 are
`both replaced by He in the latter. Since it is unlikely that two different
`genes
`separately encode the two molecules, we conclude
`that
`the
`cDNA clones presented here encode human cadherin-6 and that these
`are the first
`reported clones which cover
`the entire open reading
`frame.
`It has been reported that human N-cadherin cDNAs
`isolated
`from different
`libraries
`show minor differences
`from one another
`(46-48).
`The two amino acid substitutions
`found here may be ex
`plained by polymorphism, mutation, or cloning artifacts.
`It is note
`worthy that, at these two positions, a cDNA clone of rat K-cadherin,
`isolated from a renal cell carcinoma
`library (41), encodes
`the same
`amino acids as ALI-I
`and Bl-1,
`suggesting that the two amino acid
`substitutions may be a characteristic of cancer-specific
`cadherin-6.
`Suzuki et al.
`(32) and Tanihara
`et al.
`(34) have proposed that
`typical
`cadherins
`can be divided into two types on the basis of
`structural similarities:
`type I including E-, N-, P-, B- (23), R- (24), EP-
`(25), and M-cadherin
`(26), and cadherin-4;
`and type II cadherins
`including
`cadherin-5,
`-8,
`-11, and -12. They also suggested
`that
`cadherin-6,
`-7, -9, and -10, for which only restricted sequences
`are
`available, might be classified as type II cadherins (32, 34). The entire
`cadherin-6
`sequence presented
`here shows higher homology with
`cadherin-8,
`-11, and -12 than with other human cadherins
`(Table 1),
`confirming that cadherin-6 should be a type II cadherin. However,
`only cadherin-5 among type II cadherins
`shows a low similarity to
`
`cadherin-6, comparable to the similarity between cadherin-6 and type
`I cadherins (Table 1). This is consistent with the previous finding that
`cadherin-5 shows rather low similarity to other type II cadherins (32),
`suggesting that cadherin-5 is a unique molecule among the type II
`cadherins.
`high similarity to rat
`shows extremely
`Since human cadherin-6
`K-cadherin (41), even in the signal peptide and precursor
`regions,
`it is
`considered to be the human counterpart of rat K-cadherin.
`Interest
`ingly,
`the most diversified region between the two cadherins
`is the
`cytoplasmic domain, where two clusters of amino acids different
`from
`each other are present, whereas the two cadherins are almost
`identical
`in the extracellular
`and transmembrane
`domains with only a few
`conservative
`amino acid changes
`(Fig. 2). In contrast,
`the classical-
`type cadherins, E-, N- and P-cadherin,
`show the highest degree of
`interspecific conservation in the cytoplasmic domain (48-51), which
`is linked to the cytoskeleton via catenins. When each subdomain of
`cadherin-6 is compared with that of other type II molecules,
`the most
`conserved subdomain is not the cytoplasmic domain but the extracel
`lular domain 2 (Table 1). As for the extracellular
`domain of cad
`herin-6,
`three potential sequences
`involved in cell adhesion recogni
`tion, PPI, GAD and HAV, proposed by Blaschuk et al. (52), are absent
`or partially changed in cadherin-6.
`In contrast, possible Ca2+-binding
`sites (53, 54) are well conserved in cadherin-6,
`and four cysteine
`residues involved in the mechanism of cadherin adhesion (55) are also
`conserved in cadherin-6 at amino acid positions 497, 589, 591, and
`600 (Fig. 1). These features are also the case for the rat homologue,
`K-cadherin (41). As mentioned above, cadherin-6 and other
`type II
`cadherins have structural characteristics
`not only similar
`to but also
`different
`from classic cadherins.
`It is thus of considerable
`interest
`to
`investigate whether cadherin-6 and other type II cadherins function in
`the same manner as classic cadherins.
`Analysis of the expression pattern of cadherin-6 revealed in this
`study provided some interesting findings. Both normal kidney and
`renal cell carcinomas expressed cadherin-6.
`In contrast, although four
`of six hepatocellular
`carcinoma
`cell
`lines examined
`strongly ex
`pressed cadherin-6, and indeed cadherin-6 cDNAs were isolated from
`a hepatocellular
`carcinoma
`library, cadherin-6
`expression was not
`detected in normal
`liver
`tissue. The significance
`of the cadherin-6
`expression in hepatocellular
`carcinoma cells is still unclear;
`therefore,
`it will be important
`in the future to examine cadherin-6 expression in
`surgically resected hepatomas
`together with the morphological
`and
`clinical
`features of the cancers.
`cadherin-6, we
`Since brain and cerebellum strongly expressed
`focused on cadherin-6 expression in SCLC cells, which are known to
`show features of differentiation similar to neuroendocrine
`cells. SCLC
`cells can be subdivided into two types, the classic type and the variant
`type, according to the expression pattern of neuroendocrine
`biomar-
`kers, and separately subgrouped into four types based on their appear
`ance:
`type 1, tightly packed spherical aggregates;
`type 2, relatively
`densely packed aggregates;
`type 3, very loosely adherent aggregates;
`and type 4, cells attached to the substrate (45). As shown in Fig. 4C,
`cadherin-6 expression was detected in 11 of 15 SCLC lines examined.
`Surprisingly, all of these 11 cadherin-6-positive
`cell lines were of the
`classic type and morphologically
`type 1 or 2. In contrast, all four of
`the cadherin-6-negative
`cell
`lines were classified as the variant
`type
`and were morphologically
`type 3 (only Lu24v cells grow as type 2-3;
`
`Fig. 2. Nucleotide sequence (upper} and deduced amino acid sequence (lower) of a cadherin molecule cloned from a Li21 cDNA library. Amino acid sequence is shown in one-letter
`code. The stop codon is marked by an asterisk. The addition site of
`the poly(A)
`tail and the polyadenylation
`signal of clone Bl-1
`are shown by single and double underlining,
`respectively. The putative signal peptide and transmembrane
`region are also indicated by underlining. The solid triangle and clear
`triangles
`show the possible proteolytic cleavage site
`(39) and /V-linked glycosylation
`sites, respectively. Capital and small
`letters below the deduced amino acid sequence indicate amino acid residues of rat K-cadherin (41) and human
`cadherin-6 partially sequenced by Suzuki et al. (32), respectively, which differ
`from the sequence presented here. The nucleotide sequence data will appear
`in the GSDB/DDBJ/
`EMBL/NCBI databases under the accession number D31784.
`
`2209
`
`4 of 6
`
`OnCusp
`Ex. 1021
`
`

`

`SEQUENCE AND EXPRESSION OF HUMAN CADHER1N-6
`
`of
`the expression
`that
`suggest
`Refs. 44 and 45). These findings
`differentiation
`and that
`cadherin-6 is associated with neuroendocrine
`cadherin-6 may be a useful marker
`for distinguishing
`between the
`classic and variant
`types; furthermore,
`cadherin-6 expression may be
`responsible
`for the tighter cell-cell adhesion evident
`in SCLC cell
`lines positive
`for the molecule
`than in those lacking it. A similar
`finding has been reported for the Ca2+-independent
`cell-cell adhesion
`molecule N-CAM, also called cluster 1 SCLC antigen; N-CAM was
`abundantly expressed in classic-type SCLCs, whereas the expression
`
`Table i Homologies between cadherin-6 and other cadherins for separate
`the putative mature protein
`The extracellular domain is divided into five subregions according to Tanihara et al.
`(34). rK, E, N, P, 4, 5, 8, 11, 12, and 13 represent percentage homology between human
`cadherin-6 and rat K-cadherin (41), human E-cadherin (49), N-cadherin (46), P-cadherin
`(50), and cadherin-4,
`-5, -8, -11, -12, and -13 (32, 34),
`respectively. EC, extracellular
`domain; TM,
`transmembrane
`domain; CP, cytoplasmic domain.
`
`regions and
`
`A 1 2345
`
`B
`
`1
`
`2
`
`3
`
`4
`
`C 1 2
`
`34
`
`5
`
`67
`
`8
`
`910
`
`11 12 131415
`
`-
`
`I
`
`i
`
`Downloaded from http://aacrjournals.org/cancerres/article-pdf/55/10/2206/2457268/cr0550102206.pdf by guest on 15 June 2023
`
`0 Because cadherin-13 lacks the transmembrane
`homology in the extracellular domain is shown.
`
`and cytoplasmic
`
`domains, only the
`
`1 234
`
`56
`
`789
`
`10 11
`
`lines (A), renal cell
`carcinoma cell
`Fig. 4. Expression of cadherin-6 in hepatocellular
`carcinoma cell lines (B) and SCLC cell lines (C) examined by Northern blot analysis. A:
`Lane 1. Li7; Lane 2, LÌ22;Lane 3, LÌ23;Lane 4, LÌ24(36, 42); Lane 5, Hep G2 (43). B:
`Lane I, KT1A; Lane 2, KT4; Lane 3. KT12 (57); Lane 4, KT35. C: Lane 1, Lu24; Lane
`2. Lu24v; Lane 3, Lul30; Lane 4, Lul34A; Lane 5. Lul34B; Lane 6, Lul35v; Lane 7,
`Lul39; Lane 8, Lul40; Lane 9. Lul41; Lane JO, Lul43 (44); Lane 11, H69; Lane 12, H82;
`Lane 13, N230; Lane 14, N231; Lane 15, N417 (45). Arrowheads, positions of cadherin-6
`transcripts.
`
`It has also been
`type (56).
`in the variant
`tended to be lower
`level
`reported that
`in kidney development,
`the expression
`pattern of K-
`cadherin,
`a homologue of cadherin-6,
`is similar
`to that of N-CAM
`(41). However,
`it remains to be elucidated whether
`the expressions of
`both cadherin-6 and N-CAM are regulated by the same mechanisms
`and whether
`their
`expression
`actually
`affects
`the morphological
`appearance and biological properties of SCLC cells.
`In conclusion, we have isolated a full-length
`human cadherin-6
`cDNA from a hepatocellular
`carcinoma library and demonstrated
`its
`unique expression in cancer cells. However,
`the exact functions and
`roles of cadherin-6 during normal morphogenesis
`and cancer devel
`opment
`remains
`to be elucidated. We hope that
`this
`report will
`promote further studies on the roles of various cadherin molecules
`in
`cancer development
`and spread.
`
`ACKNOWLEDGMENTS
`
`We thank Drs. A. F. Gazdar and T. Sekine for providing the SCLC lines
`H69, H82, N230, N231, and N417 and the renal cell carcinoma
`cell
`lines,
`respectively. We also thank Dr. T. Yoshida for valuable advice.
`
`REFERENCES
`
`Fig. 3. Expression of cadherin-6 in LÌ21cells and normal human tissues examined by
`Northern blot analysis. Poly(A)+ RNA from LÌ21cells (Lane /), human brain (Lane 2),
`cerebellum (Lane 3), heart (Lane 4),
`lung (Lane 5), kidney (Lane 6),
`liver (Lane 7),
`pancreas
`(Lane 8), gastric mucosa (Lane 9), colonie mucosa (Lane 10), and placenta
`(Lane 11) were hybridized with the 182-bp PCR probe. The RNA sample in Lane 6
`appears
`to be degraded
`in comparison with other
`samples, probably because of the
`prolonged
`ischemie
`stale of
`the kidney tissue during surgical
`resection. Bars
`show
`positions of molecular size marker mRNAs of 9.5, 7.5, 4.4, 2.4, 1.4, and 0.24 kb from the
`top.
`
`2210
`
`regulator. Science
`
`1. Takeichi, M. The cadherins: cell-cell adhesion molecules controlling animal morpho
`genesis. Development
`(Camb.), 102: 639-655,
`1988.
`2. Takeichi, M. Cadherin cell adhesion receptors as a morphogenetic
`(Washington DC), 251: 1451-1455,
`1991.
`3. Shimoyama, Y., Hirohashi, S., Hirano, S., Noguchi, M., Shimosato, Y., Takeichi, M.,
`and Abe, O. Cadherin
`cell adhesion molecules
`in human epithelial
`tissues
`and
`carcinomas. Cancer Res., 49: 2128-2133,
`1989.
`4. Shimoyama, Y., and Hirohashi, S. Expression
`carcinomas. Cancer Res., 51: 2185-2192,
`1991.
`adhesion molecule
`intercellular
`5. Shimoyama, Y., and Hirohashi, S. Cadherin
`hepatocellular
`carcinomas:
`loss of E-cadherin expression in an undifferentiated
`cinoma. Cancer Lett., 57: 131-135, 1991.
`6. Shimoyama, Y., Nagafuchi, A., Fujita, S., Gotoh, M., Takeichi, M., Tsukita, S., and
`Hirohashi, S. Cadherin dysfunction in a human cancer cell line: possible involvement
`of loss of a-catenin
`expression in reduced cell-cell adhesiveness. Cancer Res., 52:
`5770-5774,
`1992.
`
`of E- and P-cadherin
`
`in gastric
`
`in
`car
`
`rKENP458111213ECl1002934293138606270322984347374649757878383973127333037575262304982931212941575343285100232417243751485621TM1002950264941647373CP895048494635606271Total97353732363960606430°
`
`5 of 6
`
`OnCusp
`Ex. 1021
`
`

`

`SEQUENCE AND EXPRESSION OF HUMAN CADHERIN-6
`
`Downloaded from http://aacrjournals.org/cancerres/article-pdf/55/10/2206/2457268/cr0550102206.pdf by guest on 15 June 2023
`
`and some cell
`
`against a novel
`J. R. Autoantibodies
`a disease of cell adhesion. Cell, 67:
`
`Amagi. M.. Klaus-Kovton, V., and Stanley.
`epithelial
`cadherin in pemphigus
`vulgaris,
`869-877,
`1991.
`Suzuki, S., Sano. K., and Tanihara. H. Diversity of the cadherin family: evidence for
`eight new cadherins
`in nervous tiss