givtnwXsXchPeoPtopobsb

ghhdtccdtdv

Ea

3t

All articles available online at http://www.idealibrary.com on

GA0CA

SHORT COMMUNICATION

Identification of a Novel Protocadherin Gene (PCDH11) on theHuman XY Homology Region in Xq21.3

Kenichi Yoshida and Sumio Sugano1

Department of Virology, Institute of Medical Science, The University of Tokyo, 4-6-1 Shirokanedai, Minato-ku, Tokyo 108-8639, Japan

Received August 10, 1999; accepted October 25, 1999

lian Pcdhs are divided into subgroups by the similarityomc

iweiir1Pog(em

nPflmp

iBcBimqmATGiNA

i

Protocadherins (Pcdhs) are members of the rapidlyrowing cadherin superfamily and are thought to benvolved in cell–cell recognition in the central ner-ous system. Using human BH-Pcdh cDNA, we re-rieved a homologous gene from the database. Theew gene (Pcdh-X, HGMW-approved symbol PCDH11)as present on a genomic clone of human chromosome(clone bWXD306), between two sequence tagged

ites, sWXD1362 and 221. Pcdh-X therefore maps to theY homology region in Xq21.3. The open reading frameonsists of 1021 amino acids (aa) including seven cad-erin repeats (EC1-7) in the extracellular domain. Thecdh-X gene consists of at least three exons; the firstxon encodes the 5*-untranslated region, EC1, and halff EC2, the second exon encodes the remainder of thecdh-Xa, and the third exon encodes the cytoplasmic

ail of Pcdh-Xb and its 3*-untranslated region. The sec-nd exon has an alternative splice site that is used toroduce two isoforms with different cytoplasmic tailsf 10 (Pcdh-Xa) or 14 amino acids (Pcdh-Xb). Northernlot analysis revealed an approximately 6.0-kb tran-cript expressed in human and mouse fetalrain. © 1999 Academic Press

The cadherin superfamily can be divided into tworoups, classic and nonclassic types (10, 11). Protocad-erins (Pcdhs) are the major nonclassic type of cad-erin superfamily and are thought to have potential foretermining the diversity of cell–cell interactions inhe brain (10–12). The extracellular domain of Pcdhsonsists of more than six cadherin repeats, and theytoplasmic domain lacks the b-catenin binding siteescribed for classic cadherins (7). Pcdhs are thoughto be evolutionarily older than classic cadherins. In-eed, Pcdhs are isolated from invertebrates as well asertebrates (10). Recent work revealed that mamma-

Sequence data from this article have been deposited with theMBL/GenBank/DDBJ Libraries under Accession Nos. AB026187nd AB026188.

1 To whom correspondence should be addressed. Telephone: 181--5449-5286. Fax: 181-3-5449-5416. E-mail: ssugano@ims.u-okyo.ac.jp.

enomics 62, 540–543 (1999)rticle ID geno.1999.6042

540

888-7543/99 $30.00opyright © 1999 by Academic Pressll rights of reproduction in any form reserved.

f their cytoplasmic domains, but each individualember has a characteristic expression pattern in the

entral nervous system (5, 6, 8).The genomic organization of the cadherin superfam-

ly, especially for classic and desmosomal cadherins, isell characterized. In general, they consist of 15–17xons and exhibit a remarkable degree of conservationn intron position (2). Classic cadherin genes have twontrons in the region corresponding to each cadherinepeat, so that the extracellular domain is divided into0 exons. In contrast, the extracellular domain of thecdhs, Pcdh1, Pcdh2, and Pcdh8, are encoded by onlyne exon (8, 9). Recently, 52 different human Pcdhenes were identified in clusters on chromosome 5q3112). The extracellular and transmembrane domains ofach Pcdh are encoded by one exon, and the cytoplas-ic domain is encoded by three small exons.Here we report the identification and genomic orga-

ization of a novel member of the Pcdh family, namedcdh-X.2 In contrast to the other member of the Pcdh

amily, Pcdh-X consists of two exons in the extracellu-ar domain. The Pcdh-X gene was mapped to the hu-

an XY homology region in Xq21.3 where none of therevious identified Pcdhs maps.This novel member of the Pcdh family was identified

n a database search using BH-Pcdh-a cDNA (Gen-ank Accession No. AB006755), with chromosomal lo-alization to 4p15 (13). A mRNA entry (1680 bp, Gen-ank Accession No. U79247), named 23599, from an

nfant brain showed significant homology (approxi-ately 57%) with BH-Pcdh-a cDNA. This same se-

uence was present on a genomic clone of human chro-osome X, bWXD306 (GenBank Accession No.C004388), confirming that this was a new Pcdh gene.he 23599 mRNA sequence was compared with theenBank human expressed sequence tag (dbEST) us-

ng BLAST (1). Only three ESTs (GenBank Accessionos. F13304; F10902, isolated from infant brain; andA224524, isolated from the NT2 neuronal precursor)

2 The HGMW-approved symbol for the gene described in this papers PCDH11.

wm

fPacp(t8Pa2sop1tsPO

PBt

dipmbAmot1XiiXc

cw

ds13GCap

541SHORT COMMUNICATION

ere exactly identical to the 59 or 39 ends of the 23599RNA (Fig. 1).To obtain the 59 and 39 ends of the open reading

rame (ORF) of the 23599 mRNA, we carried out RT-CR using human fetal brain poly(A)1 RNA (Clontech)nd primers designed from the bWXD306 sequence. ADNA sequence of 4603 bp was deduced from overlap-ing PCR products and contained an ORF of 3063 bpGenBank Accession No. AB026187) (Fig. 1). The pu-ative start codon for mRNA translation is at nt 846–48. The stop codon TAA is at nt 3909–3911. Thus,cdh-Xa encodes a putative polypeptide of 1021 aminocids (aa). The predicted aa sequences begin with a5-aa hydrophobic stretch, suggesting that this is theignal peptide of the protein. The extracellular domainn the N-terminal side consists of seven cadherin re-eats (EC1–7) as is the case for BH-Pcdh and Pcdh1 (7,3). A 24-aa hydrophobic stretch corresponds to theransmembrane domain (3). The cytoplasmic domainhows fairly high homology to that of BH-Pcdh andcdh1 but not to other members of the Pcdh family.verall, the identity of the deduced aa sequence of

FIG. 1. Genomic organization and cDNA structure for Pcdh-X.epicted by solid boxes to indicate the Xq21.3 locus. The ORFs for Pcolid lines indicate RT-PCR products. Arrows with dotted lines indicmin at 55°C, and 1 min at 72°C. Primers used were 59-CTGGTGGT

9, 59-GCACCATTGTTCCCAGCAAC-39, 59-CACTGGCATGAATGCACCTCTCGA-39 as sense primer and 59-CTACGTCAGGATCAACAGAGGTACCTCG-39, 59-GTCATACTGCCATCTACAGAC ATGGTCntisense primer for Pcdh-Xa; 59-TACCTGTGTCCGTACACACCA-39rimer for Pcdh-Xb.

cdh-Xa was approximately 42% to that of humanH-Pcdh-a and Pcdh1 (Fig. 2). These three members of

he Pcdh family therefore constitute a novel subfamily.Among the Pcdh family, Pcdh1, BH-Pcdh, and 52

ifferent recently identified Pcdhs are known to havesoforms generated by alternative splicing in the cyto-lasmic domain (7, 12, 13). Using the cytoplasmic do-ain of Pcdh-Xa, we searched the human dbEST li-

rary and identified a variant EST entry (GenBankccession No. AA730407 isolated from Ewing sarco-a). Its 59 end corresponded to the cytoplasmic domain

f Pcdh-Xa (nt 3853–3878) but the 39 end, containing ahird exon, corresponded to bWXD306 (nt 164,920–65,272), about 3 kb downstream from the end of Pcdh-a, suggesting that it was derived by alternative splic-

ng. RT-PCR analysis determined that this EST is ansoform of Pcdh-X, which was therefore named Pcdh-b. Thus, the cytoplasmic tail of Pcdh-Xa and Pcdh-Xbonsists of 10 and 14 aa, respectively (Fig. 2).The human Pcdh-X exon–intron boundaries were

onfirmed by comparing bWXD306 genomic sequenceith the Pcdh-X cDNA sequence. Three exons and two

en boxes numbered from 1 to 3 represent exons. Three STSs are-Xa and Pcdh-Xb cDNA are represented by open boxes. Arrows with

ESTs. PCR was performed for 30 cycles as follows: 1 min at 94°C,AGTACCTCCAAAGA-39, 59-GATGCCCAGAGCTGAAGAAATGGA-GG-39, 59-GATCCTGGTTGCAGCTGTTG-39, 59-CATCCAAGAACT-GCTG-39, 59-ATGGTCCGCATCCTTATCCGT-39, 59-GTACGGACA-CATCCTTATCCGT-39, 59-AGAATGTCTTCTGAAGGGCTC-39 asense primer and 59-AGCCAGGATGGTCTCAATCTC-39 as antisense

OpdhateCCGACCCGas s

icEtaatseg

wBtiei

o(t

a

542 SHORT COMMUNICATION

ntrons were identified for Pcdh-X; the first exon en-odes the 59-untranslated region, EC1, and the half ofC2, the second exon encodes the rest of Pcdh-Xa, and

he third exon encodes the cytoplasmic tail of Pcdh-Xbnd its 39-untranslated region (Fig. 1). These splicecceptor and donor sequences were found to conform tohe GT-AG rule. The second exon has an alternativeplice site. The intron between exon 1 and exon 2 isspecially long (40,736 bp). Pcdh-X has a distinctenomic organization in the extracellular domain

FIG. 2. Comparison of deduced amino acid sequences of Pcdh-X,nd Pcdh1 (7). The asterisks under the alignment show amino acids

hen compared to other members of the Pcdh family.H-Pcdh is known to have an additional 50-aa inser-

ion in the middle of EC2 (13). The position of the 50-aansertion corresponds to the boundary of exon 1 andxon 2 of Pcdh-X (Fig. 2). This suggests that the 50-aansertion may be another exon of BH-Pcdh.

Pcdh genes are known to constitute a linkage groupn human chromosome 5q31–q33 and 13q14.3–q21.16, 8, 12). Recently, 52 different Pcdh genes were iden-ified in the 5q31 region, and they are organized into

-Pcdh, and Pcdh1. The sequences are from human BH-Pcdh-a (13)nserved among the three proteins.

BHco

tgPsNoTXTf

vdbb(biop

rtS

REFERENCES

1

1

1

1

OaTusw

543SHORT COMMUNICATION

hree closely linked clusters similar to the immuno-lobulin and T cell receptor gene clusters (12). Thecdh-X gene was present between two sequence-taggedites (STSs), sWXD1362 and 221 (GenBank Accessionos. L41965 and L24726, respectively), and included

ne STS, sWXD356 (GenBank Accession No. G11974).hese STSs were mapped to the XY homology region inq21.3 that is homologous to a segment in Yp11.1 (4).o our knowledge, Pcdh-X is the first transcribed gene

rom the XY homology region in Xq21.3.Pcdhs were originally isolated from the central ner-

ous system, and each individual member has a pre-ominant and characteristic expression pattern in therain (7, 11). Northern blot analysis using human fetalrain revealed an approximately 6.0-kb transcriptFig. 3). A faint signal smaller than the major 6.0-kband was also observed. This may correspond to ansoform of Pcdh-X. The same 6.0-kb transcript wasbserved in fetal mouse brain with the human cDNArobe, indicating a high degree of homology (Fig. 3).

ACKNOWLEDGMENT

This work was partly supported by a grant-in-aid for scientificesearch from the Ministry of Education, Science, Sports, and Cul-ure of Japan. K.Y. is a recipient of a fellowship from the Japanociety for the Promotion of Science (PD).

FIG. 3. Expression of Pcdh-X mRNA in human and mouse brain.ne microgram of poly(A)1 RNA was loaded on each lane (lanes 1nd 2 are human fetal and mouse embryo at 20 days, respectively).he cytoplasmic region of the Pcdh-Xa cDNA (nt 2957–3867) wassed as probe. The positions of ribosomal RNAs (28S and 18S) arehown on the right. Loading control was performed by hybridizationith mouse b-actin (bottom).

1. Altschul, S. F., Gish, W., Miller, W., Myers, E. W., and Lipman,D. J. (1990). Basic local alignment search tool. J. Mol. Biol. 215:403–410.

2. Greenwood, M. D., Marsden, M. D., Cowley, C. M. E., Sahota,V. K., and Buxton, R. S. (1997). Exon–intron organization of thehuman type2 desmocollin gene (DSC2): Desmocollin gene struc-ture is closer to “classical” cadherins than to desmogleins.Genomics 44: 330–335.

3. Kyte, J., and Doolittle, R. F. (1982). A simple method for dis-playing the hydropathic character of a protein. J. Mol. Biol.157: 105–132.

4. Mumm, S., Molini, B., Terrell, J., Srivastava, A., and Schles-singer, D. (1997). Evolutionary features of the 4-Mb Xq21.3 XYhomology region revealed by a map at 60-kb resolution. GenomeRes. 7: 307–314.

5. Obata, S., Sago, H., Mori, N., Davidson, M., St. John, T., andSuzuki, S. T. (1998). A common protocadherin tail: Multipleprotocadherins share the same sequence in their cytoplasmicdomains and are expressed in different regions of brain. CellAdhes. Commun. 6: 323–333.

6. Sago, H., Kitagawa, M., Obata, S., Mori, N., Taketani, S., Roch-elle, J. M., Seedling, M. F., Davidson, M., St. John, T., andSuzuki, S. T. (1995). Cloning, expression, and chromosomallocalization of a novel cadherin-related protein, protocad-herin-3. Genomics 29: 631–640.

7. Sano, K., Tanihara, H., Heimark, R. L., Obata, S., Davidson, M.,St. John, T., Taketani, S., and Suzuki, S. (1993). Protocad-herins: A large family of cadherin-related molecules in centralnervous system. EMBO J. 12: 2249–2256.

8. Strehl, S., Glatt, K., Liu, Q.-M., Glatt, H., and Lalande, M.(1998). Characterization of two novel protocadherins (PCDH8and PCDH9) localized on human chromosome 13 and mousechromosome 14. Genomics 53: 81–89.

9. Suzuki, S. T. (1996). Protocadherins and diversity of the cad-herin superfamily. J. Cell. Sci. 109: 2609–2611.

0. Suzuki, S. T. (1996). Structural and functional diversity ofcadherin superfamily: Are new members of cadherin superfam-ily involved in signal transduction pathway? J. Cell. Biochem.61: 531–542.

1. Uemura, T. (1998). The cadherin superfamily at the synapse:More members, more missions. Cell 93: 1095–1098.

2. Wu, Q., and Maniatis, T. (1999). A striking organization of alarge family of human neural cadherin-like cell adhesion genes.Cell 97: 779–790.

3. Yoshida, K., Yoshitomo-Nakagawa, K., Seki, N., Sasaki, M.,and Sugano, S. (1998). Cloning, expression analysis, andchromosomal localization of BH-protocadherin (PCDH7), anovel member of the cadherin superfamily. Genomics 49:458 – 461.