March 2005
Biol. Pharm. Bull. 28(3) 429—433 (2005)
429
Cloning and Expression of a Brain-Specific Putative UDP-GalNAc: Polypeptide N-Acetylgalactosaminyltransferase Gene Naosuke NAKAMURA,a, 1) Shinya TOBA,a, 1) Mitsuharu HIRAI,a Shinichi MORISHITA,a Tadahisa MIKAMI,C Morichika KONISHI,d Nobuyuki ITOH,d and Akira KUROSAKA*,a,b a
Department of Biotechnology, Faculty of Engineering, Kyoto Sangyo University; b Institute for Comprehensive Research, Kyoto Sangyo University; Kamigamo-motoyama, Kita-ku, Kyoto 603–8555, Japan: c Department of Biochemistry, Kobe Pharmaceutical University; Higashinada-ku, Kobe 658–8558, Japan: and d Department of Genetic Biochemistry, Kyoto University Graduate School of Pharmaceutical Sciences; Yoshida-shimoadachi, Sakyo-ku, Kyoto 606–8501, Japan. Received November 16, 2004; accepted December 14, 2004; published online December 21, 2004 We isolated a rat cDNA clone and its human orthologue, which are most homologous to UDP-GalNAc: polypeptide N-acetylgalactosaminyltransferase 9, by homology-based PCR from brain. Nucleotide sequence analysis of these putative GalNAc-transferases (designated pt-GalNAc-T) showed that they contained structural features characteristic of the GalNAc-transferase family. It was also found that human pt-GalNAc-T was identical to the gene WBSCR17, which is reported to be in the critical region of patients with Williams–Beuren Syndrome, a neurodevelopmental disorder, and to be predominantly expressed in brain and heart. In order to investigate the expression of pt-GalNAc-T in brain in more detail, we first examined that of human pt-GalNAc-T by Northern blot analysis and found the expression of the 5.0-kb mRNA to be most abundant in cerebral cortex with somewhat less abundant in cellebellum. The expression of rat pt-GalNAc-T was investigated more extensively. The brain-specific expression of 2.0-kb and 5.0-kb transcripts was demonstrated by Northern blot analysis. In situ hybridization in the adult brain revealed high levels of expression in cerebellum, hippocampus, thalamus, and cerebral cortex. Moreover, observation at high magnification revealed the expression to be associated with neurons, but not with glial cells. Analysis of the rat embryos also demonstrated that rat pt-GalNAc-T was expressed in the nervous system, including in the diencephalons, cerebellar primordium, and dorsal root ganglion. However, recombinant human pt-GalNAc-T, which was expressed in insect cells, did not glycosylate several peptides derived from mammalian mucins, suggesting that it may have a strict substrate specificity. The brainspecific expression of pt-GalNAc-T suggested its involvement in brain development, through O-glycosylation of proteins in the neurons. Key words drome
N-acetylgalactosaminyltransferase; mucin; O-glycosylation; brain; in situ hybridization; Williams–Beuren syn-
Mucin-type O-glycosylation is one of the most important post-translational modifications of proteins in the cell, and a UDP-GalNAc: polypeptide N-acetylgalactosaminyltransferase (GalNAc-transferase) catalyzes the initial step in the biosynthesis of mucin-type oligosaccharide by transferring GalNAc from UDP-GalNAc to a hydroxyl amino acid on a polypeptide acceptor.2,3) This enzyme is biochemically important because it determines the number and positions of Olinked sugar chains in a protein. Recent studies on the molecular cloning of GalNAc-transferases revealed a large gene family, with 15 isozyme genes cloned to date.4—18) The large number of isozymes in the family suggests that O-glycosylation in the cell is regulated through distinctive sets of isozymes expressed in each tissue. Of the GalNAc-transferase family, GalNAc-T9, previously isolated by us, is particularly interesting in that it is specifically expressed in the brain.12) Although several proteins in the brain are reported to carry mucin-type carbohydrate chains, the involvement of brain-specific isozymes in vivo O-glycosylation has not been reported. Here, we report the cloning of a brain-specific putative GalNAc-transferase (pt-GalNAc-T) gene from human and rat that is most homologous to GalNAc-T9. We also found that human pt-GalNAc-T is identical to the gene, WBSCR17, located in the critical region of patients with WBS,19) which is characterized by a neurodevelopmental disorder caused by a haploinsufficiency of multiple genes in this region.20) Although WBSCR17 is most abundantly expressed in brain, ∗ To whom correspondence should be addressed.
with a significant amount also present in heart,19) the expression of its mRNA in the brain still remains to be investigated in detail. Also its gene product has not been biochemically characterized. We, therefore, examined the expression of ptGalNAc-T in the brain by conducting Northern blot and in situ hybridization analyses, and assayed the transferase activity of the recombinant pt-GalNAc-T. MATERIALS AND METHODS Cloning of the Putative GalNAc-Transferase from Human and Rat Homology-based PCR, in combination with 5- and 3-RACE, was carried out using the human GalNAc-T9 sequence, obtaining the full-length rat ptGalNAc-T from brain. Based on the nucleotide sequence found in the NCBI database, a full-length human pt-GalNAcT cDNA was amplified by PCR. Both rat and human cDNAs were cloned into pGEM-T easy vector. Construction of a Recombinant Baculovirus A cDNA fragment for a baculovirus polyhedrin promoter and a signal sequence for the insect secretory protein, glycoprotein67, was isolated by digesting the plasmid vector pAcGP67 (Pharmingen) with EcoRV and BamHI. Synthetic sense and antisense oligonucleotides coding for the FLAG and 6xHis tags were 5-phosphorylated and incubated together at 60 °C for 2 min to obtain a duplex. pFastBacTM1, a transfer vector for the Bac-to-Bac system (Invitrogen), was digested with AccI and blunt-ended with Blunting High (TOYOBO), and
e-mail:
[email protected]
© 2005 Pharmaceutical Society of Japan
430
Vol. 28, No. 3
the product was digested with SacI. All cDNA fragments thus obtained were mixed together and ligated by T4 ligase (Promega), obtaining pFastFHGP67 that codes for the GP67 secretion signal and the FLAG/His tags. Human pt-GalNAcT cDNA deleted of the sequence for the cytosolic domain and the transmembrane region was prepared as outlined previously,21) and inserted into the NotI and XbaI sites of pFastFHGP67. The isolated clone was used for transformation of the host strain E. coli DH10BacTM. The amplification of recombinant baculoviruses and expression of the recombinant pt-GalNAc-T were carried out according to the instructions described in the manual of the Bac-to-Bac® Baculovirus expression system (Invitrogen). Assay for Transferase Activity Three days after the transfection of High Five cells with the recombinant baculoviruses, the conditioned medium was recovered, dialyzed, and mixed with Ni-NTA agarose. After incubation overnight, recombinant pt-GalNAc-T was eluted with 25 mM Tris–HCl buffer (pH 7.2), containing 100 mM NaCl and 500 mM imidazole. The enzyme activity was determined as described previously.21) Northern Blot Analysis For the analysis of the human clone, human brain Multiple Tissue Northern blot II (Clontech) was hybridized with a digoxigenin-labeled probe of human pt-GalNAc-T, and detected as previously described.12) For the analysis of the rat clone, five micrograms of rat total RNA (Origene) was electrophoresed on a 1% agarose gel. RNA was, then, transferred to a positively charged nylon membrane (Roche), and hybridized with a digoxigenin-la-
Fig. 1.
beled rat probe. In Situ Hybridization Saggital sections of Wistar rat embryos (E19.5) were hybridized with 35S-cRNA antisense probe as described previously.22) RESULTS In order to clone genes for the GalNAc-transferase family, homology-based PCR, in combination with 5- and 3RACE, using the nucleotide sequence of human GalNAc-T9, was carried out, and a full-length cDNA clone was obtained from rat brain. Among the cloned GalNAc-transferases, this clone was most homologous to GalNAc-T9 with 77% amino acid similarity. We, then, cloned its human orthologue by PCR using the nucleotide sequence obtained in the database search. The amino acid sequence of the human clone obtained was 98% homologous to that of the rat clone. These clones were referred to as the putative GalNAc-transferases (designated pt-GalNAc-T), since they had structural features conserved in the GalNAc-transferase family, though their transferase activity has not been detected yet as described below. Figure 1 shows the predicted amino acid sequences of human and rat pt-GalNAc-Ts in comparison with human GalNAc-T9. They contained an open reading frame encoding a type II membrane protein consisting of 598 amino acid residues with a 7-amino acid N-terminal cytoplasmic domain, a 20-amino acid transmembrane domain, a 92-amino acid stem region, and a 479-amino acid putative catalytic re-
Amino Acid Alignment of Rat and Human pt-GalNAc-T, and Human GalNAc-T9
The alignment was performed using the pairwise and multiple Clustal W (1.4) method in MacVector. The parameters for the alignment were: slow alignment, open gap penalty10, extend gap penalty0.05, matrixblosum 30, delay divergene10%, and no hydrophile gap penalty. The DXH sequence in the glycosyltransferase 1 motif is outlined. Conserved acidic, histidine, and cysteine residues are indicated by , , and , respectively.
March 2005
Fig. 2.
431
Expression of Human and Rat pt-GalNAc-T
(a) Northern blot analysis of human pt-GalNAc-T. (b) Northern blot analysis of rat pt-GalNAc-T. (c) In situ hybridization analysis of pt-GalNAc-T in the rat embryo of E19.5. d, diencephalon; CP, cerebellar primordium; DRG, dorsal root ganglion.
gion. They had several characteristics commonly found in the GalNAc-transferase family: 1) a glycosyltransferase 1 (GT1) motif, a conserved sequence commonly found in glycosyltransferases,23) 2) a DXH sequence in the GT1 motif, a putative binding site for a sugar donor and/or a metal ion,24) 3) a Gal/GalNAc-T motif consisting of about 40 amino acid residues,23) 4) conserved acidic, histidine, and cysteine residues,23,25,26) and 5) (QXW)3 repeats, a C-terminal lectinlike domain.27) We, then, investigated the expression of pt-GalNAc-T. First, the expression of human pt-GalNAc-T in the brain was examined by Northern blot analysis. Figure 2a shows that it was strongest in cerebral cortex. The level of expression in cerebellum, occipital pole, frontal lobe, temporal lobe, and putamen was moderate. This is in contrast with human GalNAc-T9, which is expressed most abundantly in cerebellum, and to a lesser extent in cerebral cortex.12) The mRNA expression of rat pt-GalNAc-T was also investigated in adult rats. Figure 2b shows that two distinct transcripts of rat pt-GalNAc-T (a major 2.0-kb transcript and minor 5.0-kb transcript) were exclusively expressed in the brain. It should be noted that a significant amount of human orthologue mRNA is found in the heart as well.19) It may be possible that more sensitive methods such as RT-PCR may detect a small amount expression of rat pt-GalNAc-T in the heart, although Northern blot analysis did not reveal any transcripts. To examine the mRNA expression in rat embryos, sagittal sections of E19.5 embryos were analyzed (Fig. 2c). A strong discrete hybridization signal was detected in the diencephalon (thalamus), cerebellar primordium, and dorsal root ganglion. This demonstrates that the expression of rat pt-GalNAc-T mRNA in rat embryos as well as adult rats was confined to the nervous system. Rat pt-GalNAc-T expression was, therefore, initiated at least by the late embryonic stage, and the rat brain may require continuous ptGalNAc-T expression in the embryonic and adult stages. To examine the mRNA expression in the rat brain in more
Fig. 3.
In Situ Hybridization of Rat pt-GalNAc-T
Rat brain coronal sections were hybridized with 35S-cRNA antisense rat pt-GalNAcT probe. (a) Adult cerebral cortex, (b) adult cerebellum. (c) and (d) Hippocampus (25); (e) and (f) cerebellum (25). (c) and (e) Light field images, and (d) and (f) dark field images. Cx, cerebral cortex; CA1, CA1 region of hippocampus; CA2, CA2 region of hippocampus; CA3, CA3 region of hippocampus; DG, dentate gyrus; t, thalamus; GrDG, granular layer of dentate gyrus; MoDG, molecular layer of dentate gyrus; GrCL, granular cell layer; MoL, molecular layer.
detail, coronal sections of the adult brain were analyzed by in situ hybridization with a 35S-labeled antisense or sense cRNA probe. Rat pt-GalNAc-T was preferentially expressed in CA1, CA2, and CA3 in the hippocampus, intermediate layers of the cerebral cortex, and thalamus (Fig. 3a). It was also strongly expressed in the cerebellum (Fig. 3b). We, then, examined microautoradiographic images of rat brain sections subjected to in situ hybridization. In these sections, cells in the tissues were located by Nissl staining. Strong expression of rat pt-GalNAc-T mRNA was observed in CA3 and the granular layer of the dentate gyrus in the hippocampus (Figs. 3c, d), and the granule cell layer in the cerebellum (Figs. 3e, f). Figure 4 shows high magnification images of the in situ hybridization sections. With Nissl staining of brain sections, glial cells appeared as small intensely stained (dark) cells. In contrast, neurons were generally larger and less intensely stained (lighter) owing to their volume.28) Analysis of the sections clearly revealed that the grains of the hybridized antisense probe were exclusively associated with neurons (Figs. 4a, b). These observations demonstrated that rat ptGalNAc-T was expressed in neurons, but not in glial cells, in certain regions of the brain. Finally we assayed the catalytic activity of human ptGalNAc-T. For this, a truncated pt-GalNAc-T, which lacks the cytoplasmic tail and transmembrane region, but has FLAG and 6xHis tags at the N-terminus, was expressed in insect cells. The recombinant proteins in the culture medium were recovered and purified with Ni-agarose, and then assayed for the activity. Recombinant pt-GalNAc-T, however, did not glycosylate peptides derived from mammalian mucins, such as MUC1a (AHGVTSAPDTR), MUC5AC (GTTPDPVPTTG), and MUC7 (TTAAPPTPSAG), while recombinant rat GalNAc-T1 did (data not shown). Judging
432
Fig. 4. High Magnification (40) of in Situ Hybridization of Rat ptGalNAc-T in Hippocampus Arrowheads and arrows indicate neurons and glial cells, respectively. (a) Sense probe; (b) antisense probe.
from its restricted distribution in neurons, pt-GalNAc-T may not glycosylate typical mucin-type molecules, but catalyze brain-specific O-glycosylation. DISCUSSION We cloned cDNAs from human and rat, which are most homologous to GalNAc-T9. These clones (pt-GalNAc-Ts) contained the structural features characteristic of GalNActransferases. Hence, it is expected that they encode active GalNAc-transferases, but they did not exhibit biochemical transferase activity when assayed using peptides with typical mucin polypeptide sequences. The failure to detect the catalytic activity of pt-GalNAc-T may result from several causes. First of all, pt-GalNAc-T may be a member of socalled follow-up type isozymes, which requires partial glycosylation of the acceptors in order to recognize them as substrates. In fact, prior glycosylation of the acceptor peptides was reported to be a prerequisite for glycosylation by GalNAc-T4, -T7, and -T10.7,10,13) Secondly, the transmembrane region of pt-GalNAc-T may be involved in the activity. Expression of the recombinant soluble molecules, therefore, may have generated inactive enzymes. Thirdly, pt-GalNAc-T may belong to other glycosyltransferase families involved in the transfer of different monosaccharides. It is, however, unlikely that pt-GalNAc-T catalyzes reactions other than GalNAc transfer, since it has motif structures of GalNActransferase families, exhibiting very high homology (77%) to GalNAc-T9. Finally, judging from their restricted expression in the brain, pt-GalNAc-T may have a narrow substrate specificity for catalyzing brain-specific glycosylation. Peptides containing typical mucin-like sequences may not func-
Vol. 28, No. 3
tion as acceptors. In fact, there are several examples of O-glycosylated glycoproteins in the brain. Tenascin-R is a brain glycoprotein with a disialylated core-1 structure, Siaa 2→3Galb 1→ 3(Siaa 2→6)GalNAc, as a major carbohydrate chain, and is expressed in the spinal cord, cerebellum, hippocampus, and olfactory bulb, colocalizing with rat pt-GalNAc-T except in the olfactory bulb.29,30) Chromogranin A (CgA) is also found in the brain and spinal cord, as well as in the endocrine and immune systems.31) In situ hybridization of CgA mRNA in rat brain sections revealed the expression to be strongest in the pyramidal cell layer of the hippocampus and the subiculum,31) where rat pt-GalNAc-T is abundantly expressed as well. b -Amyloid precursor protein (APP) is another example of an O-glycosylated protein in the brain.32) Both CgA and APP are known to accumulate in extracellular b -amyloid plaques in Alzheimer’s disease.31,33) CgA is also reported to accumulate in patients with Parkinson’s disease and Pick’s disease.34—36) O-Glycosylation of a -synuclein is reported as well.37) a -Synuclein is a major component of intracellular fibrillary aggregates and implicated in the pathogenesis of Parkinson’s disease.38) Aberrant O-glycosylation may cause the conformational change in a -synuclein, resulting in the deposition of abnormal filaments. Although the relationship of pt-GalNAc-T with these pathological conditions is not clear, it is possible that the deposition of these proteins in the brain is related to altered O-glycosylation. Some of these glycoproteins colocalize with pt-GalNAc-T in the brain, thus raising the possibility that they might be endogenous substrates of pt-GalNAc-T. Studies using synthetic peptides derived from the potential O-glycosylation sites of these glycoproteins are in progress. Human pt-GalNAc-T was found to be located in the WBS critical region, and defined as WBSCR17.19) There are several genes in this region, and concomitant deletion of several WBSCR genes, not a single gene, is possibly involved in the pathogenesis of WBS.20,39) It is also reported that in WBS patients, the brain is significantly smaller, and the cerebellum is enlarged relative to the cerebrum.40) The strong expression of both rat and human pt-GalNAc-Ts in these regions, together with the continual expression of the rat clone throughout the embryo and adult brain, suggests essential roles in the normal development of the brain. Therefore, the deletion of ptGalNAc-T together with other WBSCR genes may be related to the developmental disorder found in WBS. Acknowledgments This work was in part supported by Grants-in-Aid for Scientific Research from the Japan Society for the Promotion of Science, and the Foundation for Bioventure Research Center from the Ministry of Education, Culture, Sports, Science, and Technology, Japan. REFERENCES 1)
These authors contributed equally to this manuscript. The nucleotide sequence presented in this paper will appear in the DDBJ, EMBL and GenBank Nucleotide Sequence Databases under accession number AB040671. 2) Elhammer Å. P., Kézdy F. J., Kurosaka, A., Glycoconj. J., 16, 171— 180 (1999). 3) Clausen H., Bennett E. P., Glycobiology, 6, 635—646 (1996). 4) Homa F. L., Hollander T., Lehman D. J., Thomsen D. R., Elhammer A.
March 2005
5) 6) 7) 8) 9)
10)
11) 12) 13)
14)
15) 16)
17)
18)
19) 20)
P., J. Biol. Chem., 268, 12609—12616 (1993). White T., Bennett E. P., Takio K., Sørensen T., Bonding N., Clausen H., J. Biol. Chem., 270, 24156—24165 (1995). Bennett E. P., Hassan H., Clausen H., J. Biol. Chem., 271, 17006— 17012 (1996). Hagen F. K., TenHagen K. G., Beres T. M., Balys M. M., VanWuyckhuyse B. C., Tabak L. A., J. Biol. Chem., 272, 13843—13848 (1997). TenHagen K. G., Hagen F. K., Balys M. M., Beres T. M., VanWuyckhuyse B., Tabak L. A., J. Biol. Chem., 273, 27749—27754 (1998). Bennett E. P., Hassan H., Mandel U., Hollingsworth M. A., Akisawa N., Ikematsu Y., Merkx G., vanKessel A. G., Olofsson S., Clausen H., J. Biol. Chem., 274, 25362—25370 (1999). Ten Hagen K. G., Tetaert D., Hagen F. K., Richet C., Beres T. M., Gagnon J., Balys M. M., VanWuyckhuyse B., Bedi G. S., Degand P., Tabak L. A., J. Biol. Chem., 274, 27867—27874 (1999). White K. E., Lorenz B., Evans W. E., Meitinger T., Strom T. M., Econs M. J., Gene, 246, 347—356 (2000). Toba S., Tenno M., Konishi M., Mikami T., Itoh N., Kurosaka A., Biochim. Biophys. Acta, 1493, 264—268 (2000). Hagen K. G., Bedi G. S., Tetaert D., Kingsley P. D., Hagen F. K., Balys M. M., Beres T. M., Degand P., Tabak L. A., J. Biol. Chem., 276, 17395—17404 (2001). Schwientek T., Bennett E. P., Flores C., Thacker J., Hollmann M., Reis C. A., Behrens J., Mandel U., Keck B., Schafer M. A., Haselmann K., Zubarev R., Roepstorff P., Burchell J. M., Taylor-Papadimitriou J., Hollingsworth M. A., Clausen H., J. Biol. Chem., 277, 22623—22638 (2002). Guo J. M., Zhang Y., Cheng L., Iwasaki H., Wang H., Kubota T., Tachibana K., Narimatsu H., FEBS Lett., 524, 211—218 (2002). Zhang Y., Iwasaki H., Wang H., Kudo T., Kalka T. B., Hennet T., Kubota T., Cheng L., Inaba N., Gotoh M., Togayachi A., Guo J., Hisatomi H., Nakajima K., Nishihara S., Nakamura M., Marth J. D., Narimatsu H., J. Biol. Chem., 278, 573—584 (2003). Wang H., Tachibana K., Zhang Y., Iwasaki H., Kameyama A., Cheng L., Guo J., Hiruma T., Togayachi A., Kudo T., Kikuchi N., Narimatsu H., Biochem. Biophys. Res. Commun., 300, 738—744 (2003). Cheng L., Tachibana K., Iwasaki H., Kameyama A., Zhang Y., Kubota T., Hiruma T., Kudo T., Guo J. M., Narimatsu H., FEBS Lett., 566, 17—24 (2004). Merla G., Ucla C., Guipponi M., Reymond A., Hum. Genet., 110, 429—438 (2002). Tassabehji M., Hum. Mol. Genet., 12, R229—237 (2003).
433 21) 22) 23) 24) 25) 26) 27) 28) 29) 30) 31) 32)
33) 34) 35) 36) 37)
38)
39)
40)
Tenno M., Saeki A., Kézdy F. J., Elhammer Å. P., Kurosaka A., J. Biol. Chem., 277, 47088—47096 (2002). Yamasaki M., Miyake A., Tagashira S., Itoh N., J. Biol. Chem., 271, 15918—15921 (1996). Hagen F. K., Hazes B., Raffo R., deSa D., Tabak L. A., J. Biol. Chem., 274, 6797—6803 (1999). Wiggins C. A., Munro S., Proc. Natl. Acad. Sci. U.S.A., 95, 7945— 7950 (1998). Wragg S., Hagen F. K., Tabak L. A., J. Biol. Chem., 270, 16947— 16954 (1995). Tenno M., Toba S., Kézdy F. J., Elhammer Å. P., Kurosaka A., Eur. J. Biochem., 269, 4308—4316 (2002). Hazes B., Protein Sci., 5, 1490—1501 (1996). Gerfen C. R., Methods Neurosci., 1, 79—97 (1989). Zamze S., Harvey D. J., Pesheva P., Mattu T. S., Schachner M., Dwek R. A., Wing D. R., Glycobiology, 9, 823—831 (1999). Pesheva P., Spiess E., Schachner M., J. Cell. Biol., 109, 1765—1778 (1989). Winkler H., Fischer-Colbrie R., Neuroscience, 49, 497—528 (1992). Palmert M. R., Podlisny M. B., Witker D. S., Oltersdorf T., Younkin L. H., Selkoe D. J., Younkin S. G., Proc. Natl. Acad. Sci. U.S.A., 86, 6338—6342 (1989). Haass C., Selkoe D. J., Cell, 75, 1039—1042 (1993). Weiler R., Lassmann H., Fischer P., Jellinger K., Winkler H., FEBS Lett., 263, 337—339 (1990). Taupenot L., Ciesielski-Treska J., Ulrich G., Chasserot-Golaz S., Aunis D., Bader M. F., Neuroscience, 72, 377—389 (1996). Yasuhara O., Kawamata T., Aimi Y., McGeer E. G., McGeer P. L., Neurosci. Lett., 170, 13—16 (1994). Shimura H., Schlossmacher M. G., Hattori N., Frosch M. P., Trockenbacher A., Schneider R., Mizuno Y., Kosik K. S., Selkoe D. J., Science, 293, 263—269 (2001). Polymeropoulos M. H., Lavedan C., Leroy E., Ide S. E., Dehejia A., Dutra A., Pike B., Root H., Rubenstein J., Boyer R., Stenroos E. S., Chandrasekharappa S., Athanassiadou A., Papapetropoulos T., Johnson W. G., Lazzarini A. M., Duvoisin R. C., Di Iorio G., Golbe L. I., Nussbaum R. L., Science, 276, 2045—2047 (1997). Ewart A. K., Morris C. A., Atkinson D., Jin W., Sternes K., Spallone P., Stock A. D., Leppert M., Keating M. T., Nat. Genet., 5, 11—16 (1993). Jones W., Hesselink J., Courchesne E., Duncan T., Matsuda K., Bellugi U., Dev. Med. Child. Neurol., 44, 688—694 (2002).
