An Alu-linked repetitive sequence corresponding to 280 amino acids is expressed in a novel bovine protein, but not in its human homologue.

A novel protein harboring a 280-amino acid region from an Alu-linked repetitive sequence (bovine Alu-like dimer-driven family) was isolated from a bovine brain S-100 fraction using monoclonal antibodies against a rat GTPase-activating protein that shares the same epitope. The protein has an apparent molecular mass of 97 kDa (p97). Western blot analysis using extracts prepared from various tissues showed p97 to be predominantly detected in brain and moderately in liver and lung. From sequence analysis of the cDNA encoding p97, it was found that the 840-base pair sequence homologous to a part of the bovine Alu-like dimer-driven family, which has never been shown to be expressed, occurs in the middle of the protein coding region. The protein also contains a pair of intramolecular repeats composed of 40 highly hydrophilic amino acids at the C terminus. Human cDNA homologous to p97 was cloned, and its nucleotide sequence demonstrates that the 840-base pair repetitive sequence and one of the intramolecular repeats are missing. We named p97 bovine BCNT after Bucentaur. These results show that bovine BCNT is a unique molecule and suggest that an analysis of the relationship between bovine bcnt and its human homologue may help further the understanding of gene organization and evolution.

Short and long interspersed repetitive DNA elements, SINE and LINE, are widely distributed in mammals, and a few copies move within the genome as transposable elements. The Alu family is a representative of SINE and exists in nearly 10 6 copies/genome, constituting Ͼ5% of the human genome (1). Recent studies have shown that an insertion of these repetitive sequences into a normal gene sometimes results in hereditary disease (2) or cancer development (3). Furthermore, a repetitive sequence is expressed as a part of proteins (4) and in certain carcinoma cell lines (5). Recently, a complete Alu element was shown to be present in the coding sequence of a human central gene (6). The accumulation of these observations intensifies the question of the biological meaning of repetitive sequences. In bovine, in addition to the Alu-like elements, there is a family of 3.1-kbp 1 repetitive sequences, called the bovine Alu-like dimer-driven family (BDDF), whose 5Ј-and 3Ј-ends are flanked by sequences homologous to the bovine Alu-like sequence (7). During the preparation of monoclonal antibodies (mAbs) against a ras GTPase-activating protein (GAP), GAP1 m (a GAP with two C2 domains and Bruton's tyrosine kinase homology domain) (8), we identified a protein from bovine brain extract whose biochemical characteristics are similar to rasGAP1 m such as molecular mass and affinity for a heparin. Since we observed that ras GTPase activating activity other than that attributable to known rasGAPs might be present in the heparin column chromatography fractions (9), we suspected that this might be a novel protein with a GAPlike structure. To see if this was true, we purified the protein and cloned its cDNA. The cDNA sequence analysis showed that this is a novel protein containing a region homologous to a part of BDDF, but without GAP-like structure. Since a BDDF region has never been shown to exist as a protein, we further characterized p97, including its tissue expression, and compared it with its human homologue. The repetitive sequence was not detected in the human homologue. From the processes involved in the isolation of the new molecule, we named p97 bovine BCNT after Bucentaur.

EXPERIMENTAL PROCEDURES
Preparation of Antigen-Escherichia coli cells containing a plasmid expressing the rat gene, gap1 m , covering 90% of the ORF (from Ile 65 to Ser 847 ) (8) fused to glutathione S-transferase (GST) were grown in LB medium. The fusion protein was induced with 0.1 mM isopropyl-␤-Dthiogalactopyranoside at 25°C for 14 h. The cells were pelleted from 5 liters of culture and lysed with a French press in the presence of protease inhibitors (100 M Pefabloc and 20 g/ml each antipain, leupeptin, pepstatin A, and aprotinin); the lysates were cleared by centrifugation at 15 krpm for 30 min (Beckman JA-20 rotor) at 4°C. The supernatant was applied directly to a column of glutathione-Sepharose (Pharmacia Biotech Inc.), and the adsorbed proteins were eluted with 10 mM glutathione in 50 mM Tris-HCl (pH 9.6) after washing with 50 * A portion of this work was presented in a preliminary report at the 12th Annual Meeting on Oncogenes, Frederick, MD, June 18 -22, 1996. 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.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBank TM /EBI Data Bank with accession number(s) D84513, D84514, and D84515 (bovine brain); D85939 (human placenta); and D86549 (human brain).
Isolation of mAbs-Monoclonal antibodies against p97 (bovine BCNT) were prepared as described (11) 2 by Western blot screening using bovine brain S-100 separated by 7.5% SDS-polyacrylamide gel electrophoresis (PAGE). Five independent clones were obtained, and all belonged to a subclass of IgG1, , as determined by a monoclonal antibody isotyping kit (Amersham Corp.). Four of the clones were used for further experiments; the fifth was eliminated because it grew slowly.
Preparation of an Anti-p97 (BCNT) IgG Column-Supernatants from serum-free cultures of four hybridoma clones (GIT medium, Nissui, Tokyo) were harvested, and their IgGs were isolated as follows. Precipitates were obtained with 50% saturated AmSO 4 and dialyzed against phosphate-buffered saline. The samples were applied to a column of goat anti-mouse IgG-Sepharose (Zymed Laboratories, Inc., South San Francisco, CA), and IgG fractions were eluted with 0.1 M glycine HCl (pH 2.5) containing 0.5 M NaCl and immediately neutralized with Tris-HCl (pH 8.8). Purified IgGs were concentrated by AmSO 4 precipitation at 70% saturation and dialyzed against 0.2 M NaHCO 3 containing 0.5 M NaCl. The preparation (2.5 mg) was then coupled to HiTrap/ N-hydroxysuccinimide-activated resin (Pharmacia Biotech Inc.) by incubation for 4 h at 4°C on a rotary shaker. The coupling yield was estimated to be 88% by determination of protein concentration (Bio-Rad).
Isolation of p97 (Bovine BCNT)-Detection of p97 (bovine BCNT) was carried out by Western blotting as described previously (12). A bovine brain S-100 fraction (4.4 mg/ml, 300 ml) prepared as described previously (9) was fractionated with AmSO 4 (30 -55%). The precipitate was dialyzed against phosphate-buffered saline and then against buffer A (9) containing 0.2 M NaCl and applied to a heparin column (1 ml of HiTrap, Pharmacia Biotech Inc.). The column was sequentially washed with buffer A containing 0.2 or 0.3 M NaCl, and the antigen was eluted with buffer A containing 0.7 M NaCl but without dithiothreitol. The eluates were applied to an anti-p97 IgG-HiTrap column, washed with buffer A containing 0.2 M NaCl and 10 mM Tris-HCl (pH 8.0) but without dithiothreitol, and then eluted with 0.1 M glycine HCl (pH 2.5). The eluate was immediately neutralized with Tris-HCl (pH 8.8). The samples were concentrated 10-fold with Ultrafree (30,000-Da cutoff; Millipore Corp.). Purity was estimated by subjecting the preparations to two-dimensional gel electrophoresis as described previously (13). The overall yield was ϳ10%, and the degree of purification was ϳ20,000-fold compared with the S-100 fraction, as estimated by densitometric analysis using National Institutes of Health IMAGE software.
Determination of the Partial Amino Acid Sequence of p97 (Bovine BCNT)-Partially purified p97 (bovine BCNT) was separated by 7.5% SDS-PAGE, blotted onto a polyvinylidene difluoride membrane (Immobilon-P, Millipore Corp.), and stained with 0.1% Ponceau S in 1% acetic acid. The band corresponding to p97 (bovine BCNT) was excised and digested with lysyl endopeptidase AP-1 (Wako, Osaka, Japan) for 14 h, and the resulting peptides were separated by reverse-phase HPLC C 8 column chromatography as described previously (14). The major peaks were subjected to a pulse liquid-phase sequencer (Applied Biosystems Model 477A), and the broad bands were analyzed after rechromatography on a C 18 column (Applied Biosystems, Inc.).
Identification of the cDNA for p97 (Bovine BCNT) by PCR and RACE-To carry out the PCR amplification of a part of the cDNA encoding p97 (bovine BCNT), bovine brain cDNA was obtained from CLONTECH. The primers used for PCR amplification of the DNA corresponding to peptide A (TTRPFRVTNEEDATNEEA) in Fig. 2 were AARACNACNMGNCCNTT and TCYTCRTTNGTNGCRTC. Then, using primers TCCTCTTCATTAGTCACTGAA (corresponding to the sequences determined as described above) and TTYTNACNCARCARGG (corresponding to peptide B (SPLAGEEVRFLTQQGRLSGR)), a 1-kbp PCR product was isolated. For the RACE method, mRNA was isolated from an adult bovine brain using a Fast-Track mRNA purification kit (Invitrogen). cDNA was made with SuperScriptII (Life Technologies, Inc.). A Marathon cDNA amplification kit (CLONTECH) was used to obtain the adaptor-ligated product.
Isolation of the cDNA Encoding p97 (Bovine BCNT) by Library Screening-To isolate the phage clones, we screened 4 ϫ 10 5 independent phage plaques of the cDNA library from bovine brain (15) using 5Ј-32 P-labeled synthetic 53-mer oligonucleotides (both sense and antisense) labeled with [␥-32 P]ATP (6000 Ci/mmol; DuPont NEN). Specific activity was 5.6 ϫ 10 7 cpm/pmol, and hybridization was carried out in the presence of 20% formamide at a concentration of 4 ϫ 10 6 cpm/ml as described previously (16). All nucleotide sequences were determined with an automated DNA sequencer (ALF II or RED, Pharmacia Biotech Inc.) using a Thermo Sequenase fluorescent-labeled primer cycle sequencing kit (Amersham Corp.).
Isolation of the Human Homologue of Bovine BCNT-Three human ESTs (GenBank TM accession numbers H21780, H89000, and N39726) (17) and a mouse EST (GenBank TM accession number W83351) were obtained from Research Genetics, and their full-length sequences were determined. 30-mer oligonucleotide DNAs corresponding to these sequences and human placental mRNA (CLONTECH) were used for 5Јand 3Ј-RACE. To isolate the p97 (BCNT)-related molecule independently, the 5Ј-region of p97 (bovine BCNT, nucleotides 10 -728 of the full-length cDNA) was labeled with [␣-32 P]dCTP (6000 Ci/mmol; Amersham Corp.) by a random priming method and used as a probe for screening a human brain cDNA library (Stratagene).
Epitope Mapping of Anti-bovine BCNT-A Novatope kit (Novagen) was used for epitope mapping. A plasmid containing the full-length cDNA of bovine BCNT was fragmented to an average size of 300 bp and used for screening with anti-p97 (BCNT) mAbs. Seven positive clones were obtained among ϳ9 ϫ 10 3 bacterial colonies. The sequence common to all clones was determined as the possible epitope for anti-p97 antibodies.
Southern Blotting of the Bovine Genome-Bovine genomic DNA was prepared from liver and digested with various restriction enzymes. Ten micrograms of DNA/lane was electrophoresed on a 0.7% Tris borate/ EDTA-agarose gel and blotted onto a nylon membrane. The membrane was prehybridized at 68°C in hybridization buffer (6 ϫ SSC, 5 ϫ Denhardt's solution, 0.1% SDS, 0.1 mg/ml salmon sperm DNA) and probed with a 32 P-labeled 317-bp HinfI-AflII (1480 -1796) fragment encompassing most of the IR sequences. Final washing was performed at 66°C with 0.1 ϫ SSC, 0.1% SDS.
Northern Blotting of p97 (Bovine BCNT)-Five micrograms of bovine brain poly(A) ϩ mRNA prepared as described above were separated and blotted onto nitrocellulose as described previously (18). The filter was probed with the 5Ј-region of p97 (bovine BCNT) DNA (nucleotides 1-697 of the cDNA) labeled with [␣-32 P]dCTP (3000 Ci/mmol) by a random priming method at 42°C for 14 h and then washed under stringent conditions (0.1 ϫ SSC, 0.1% SDS at 68°C for 80 min).
Expression of p97 (Bovine BCNT)-p97 (bovine bcnt) gene fragments were joined to compose the whole ORF and 5Ј-untranslated region. The full-length cDNAs were ligated into expression vectors pcDNAI (Invitrogen) and pCEV18 (a gift from Dr. K. Maruyama), a modified pCEV4 vector that has a pUC replication origin (19). The constructed plasmids were transiently transfected into COS-7 cells using liposomes (LipofectAmine, Life Technologies, Inc.), and the total cell extracts were prepared 24 h after transfection and analyzed by Western blotting with anti-p97 (bovine BCNT) mAbs.
Examination of p97 (Bovine BCNT) Expression in Various Tissues-Organs were freshly isolated, cut into small pieces, and frozen immediately. The frozen samples were sliced with a razor blade on dry ice, and these pieces were homogenized in extraction buffer (1% SDS, 1 mM EDTA, 10 mM Hepes/NaOH (pH 7.4)) supplemented with protease inhibitors (100 M Pefabloc and 20 g/ml each antipain, leupeptin, pepstatin A, and aprotinin) with a Dounce homogenizer. The extracts were then boiled immediately for 5 min and subjected to sonication and centrifugation at 10,000 ϫ g for 10 min. The protein concentrations in the supernatants were estimated using a BCA kit (Pierce). Constant amounts of protein were separated by 12.5% SDS-PAGE and subjected to Western blotting.
Miscellaneous-Protein was detected by Coomassie Brilliant Blue or silver (Bio-Rad) staining, depending on the amount of protein. Bovine brain was obtained from a local slaughterhouse.

RESULTS
Purification of Bovine BCNT-Although a GST fusion protein of rat GAP1 m was used for immunization, the five isolated mAbs recognized a 97-kDa protein in bovine brain (bovine BCNT) more strongly than purified rat GAP1 m (data not shown). Both proteins share similar molecular characteristics such as molecular mass on SDS-PAGE and heparin affinity. To examine the relationship between BCNT and GAP1 m , we isolated the BCNT protein. As shown in Fig. 1, while the isolated preparations showed heterogeneity with respect to both isoelectric point and molecular mass on two-dimensional gels, the silver staining patterns and Western blots were quite similar to each other, indicating high purity of the protein. Therefore, these bands were blotted onto polyvinylidene difluoride filters and excised for lysyl endopeptidase digestion followed by HPLC. Amino acid sequences of the separated peptides were determined and found to represent a novel protein. Thus, we went on to clone the cDNA of bovine BCNT.
Cloning of the cDNA Encoding Bovine BCNT-We performed PCR on bovine brain cDNA using degenerate primers corresponding to the determined amino acid sequence and obtained a 1-kbp fragment. Then, the rest of the gene was obtained using the 5Ј-and 3Ј-RACE method. The ORF deduced from the DNA sequence consists of 592 amino acids and includes all the determined amino acid sequences of the peptides derived from the digests of purified BCNT (Fig. 2). To eliminate PCR artifacts, we also isolated five cDNA clones from a bovine brain phage library by screening with 53-mer oligonucleotides corresponding to a partial peptide sequence. Nucleotide sequence analysis revealed that the RACE product and all phage clones share the same part of the ORF, but one clone resulted in a shorter ORF (Fig. 3). We used the longest ORF for further analysis.
Sequence Analysis-Contrary to our expectations, the BCNT cDNA and gap1 m genes from rat (8) and human (20) showed no significant homology except for a small amino acid sequence as shown below. A data base search of the obtained DNA sequence revealed that BCNT is a novel protein with a calculated molecular mass of 66 kDa. Although its sequence shows no strong homology to any known protein, ϳ840 bp in the center of the ORF are 80% identical to BDDF (7), a repetitive element that has never been shown to exist as a protein. Another feature specific to bovine BCNT is the existence of two repeated sequences (IR) composed of 40 hydrophilic amino acids at the C terminus. A data base search revealed three human EST sequences with ORFs homologous to bovine BCNT as shown in Fig. 2b.
We found two TAATACC sequences present at both boundaries of the 840-bp region (nucleotides 625-631 and 1465-1471) (see Fig. 2), although no obvious splicing consensus sequence is seen. Also, there are three GTCAGG hexamer sequences at nucleotides 515-520, 1608 -1613, and 1728 -1733. The first hexamer occurs at the end of the N-terminal region corresponding to human BCNT (see Fig. 4), and the latter two at the ends of IR I and IR II. These sequences may be involved in the integration of the repetitive sequence.
Human cDNA Related to Bovine BCNT-A data base search revealed that three human ESTs (GenBank TM accession numbers H21780, H89000, and N39726) are highly homologous to bcnt. To determine whether they are human counterparts of bovine BCNT, the full-length cDNA was cloned either by the RACE method or by screening a human brain library using the N-terminal sequence of BCNT excluding the BDDF region. ORFs of the determined full-length cDNA showed 80% identity to bovine BCNT, while the BDDF region present in bovine BCNT was missing, and only one C-terminal repeat was present in the human clone (Fig. 4). There is also a bovine-specific region with ϳ35 amino acids proximal to the BDDF region. Northern blotting of human tissues probed with the 5Ј-region without the repetitive sequence revealed a 1.2-kbp band exclusively (data not shown). These results indicate that a human homologue of bovine BCNT completely lacks the BDDF repetitive sequence.
Epitope Mapping of mAbs against Bovine BCNT-Despite immunization with the GAP fusion protein, the isolated bovine BCNT protein showed no significant homology to GAP1 m . Therefore, we identified the epitope(s) of anti-BCNT mAbs by expressing fragments of bovine BCNT randomly, selecting positive clones with anti-BCNT mAbs, and determining their sequences. Abstracting the common regions from seven positive clones revealed a 13-amino acid sequence (RKQGRLSLDQEEE) as a candidate(s). It is noteworthy that the 13-mer is located in the N-terminal region of BCNT rather than in the repetitive sequence region (Fig. 4). All mAbs recognized the same region, and the alignment of the 13 amino acids and the antigen used for immunization (GST-GAP1 m ) is shown below.

Human BCNT RROGGLSLEEEEE
However, anti-bovine BCNT mAbs did not react with the human GST-BCNT fusion protein at all (data not shown). The fact that both bovine BCNT and rat GAP1 m (but not human BCNT) share the same epitope, strongly suggests that the epitope is around the LDQEE sequence. Northern Blotting and cDNA Expression in COS-7 Cells-To test if the cloned bovine BCNT cDNA encodes the full-length ORF, we carried out both Northern blotting and the transient expression of the cDNA in COS-7 cells. As shown in Fig. 5a, bovine bcnt mRNA included a major band at 3.1 kilobases, which is consistent with the size of the 2.8-kbp cDNA obtained by RACE. Western blot analysis of the total proteins from untreated COS-7 cells or their transfectants with vector alone or plasmid expressing the antisense cDNA showed a 47-kDa band, but not the 97-kDa band (Fig. 5b). On the other hand, the extracts from cells with plasmids expressing the sense cDNA showed the presence of a protein with the same molecular mass as BCNT (97 kDa) from bovine brain extracts on SDS-PAGE (Fig. 5b). These data clearly indicate that the clone covers the full-length ORF. Therefore, despite the apparent difference between the expected molecular mass (66 kDa) and the apparent molecular mass on SDS-PAGE (97 kDa), the cloned cDNA encodes the full-length ORF.
Expression of BCNT in Various Tissues-To gain insight into the biological function of BCNT, we examined the expression of BCNT in various bovine tissues by Western blotting. BCNT was found to be expressed predominantly in brain, moderately in liver and lung, and in small amounts in heart (Fig. 6).
Southern Blotting of the Bovine Genome-Since bovine BCNT includes a part of the repetitive region consisting of 280 FIG. 1. Purity of BCNT from a bovine brain extract. The preparation purified mainly by column chromatography on heparin-HiTrap and anti-BCNT IgG-coupled Sepharose was separated by two-dimensional gel electrophoresis. One gel was subjected to silver staining (a), and the other was subjected to Western blotting with anti-BCNT mAbs (b). amino acids while human BCNT does not, to understand its function, it is essential to clarify whether there is another copy of the bcnt gene without the repetitive sequence present in bovine. As described above, anti-bovine BCNT mAbs that recognize the molecule in a region outside the repetitive sequence detect one band in the bovine extract. To obtain data at the FIG. 2. Nucleotide sequence of BCNT and its deduced amino acid sequence. a, the full-length cDNA of BCNT was constructed by 3Ј-and 5Ј-RACE or was isolated from a bovine brain cDNA library. Numbers start with the ATG start codon (GenBank TM accession number D84513). The peptide sequences determined are underlined. The BDDF region is shaded. The postulated poly(A) addition signal sequence is in italics. Double-lined boxes indicate IR sequences. The TAATACC sequences at the BDDF region borders are boxed. b, IR sequences from human (h), bovine, and mouse (m) are aligned. The residues conserved through all four are shown as consensus sites. genomic level, we performed Southern blotting of the bovine genome cleaved with different restriction enzymes using the 3Ј-region of bovine bcnt excluding the 840-bp repetitive sequence as a probe. The result revealed only one band with BamHI, EcoRI, and HindIII and two bands with BglII, strongly suggesting that there is only one gene (Fig. 7), consistent with the result that PCR carried out on bovine brain cDNA using a set of primers flanking the repetitive region gave no shorter fragment (data not shown).

DISCUSSION
In this paper, we describe the structure and characteristics of a novel protein, BCNT. The BCNT cDNA sequence from bovine brain shows the existence of a repetitive sequence of BDDF in its ORF, which has never been revealed as a protein.
However, this cannot be discounted as a simple PCR artifact because 1) a peptide sequence was found to match this region; 2) library screening resulted in no clone without this region; and 3) PCR carried on bovine brain cDNA using a set of primers flanking this region gave no shorter fragment. Although sev-eral proteins are known to possess a part of the repetitive sequence (4), a sequence as long as that in BCNT is very rare. Recently, we found that the 840-bp repetitive sequence also occurs in sheep and giraffe, but not in human and pig. 3 Therefore, the repetitive sequence associated with the bcnt gene was specific to Ruminantia, and the cDNA difference between human and bovine might be caused at the level of genomic organization. Two 7-nucleotide sequences (TAATACC) and three hexamer repeats (GTCAGG) at both boundaries of the 840-bp and IR sequences might be involved in the integration of the repetitive sequence.
The molecular mass deduced from the ORF of the bovine BCNT cDNA is only 66,000 Da, which is less than the apparent molecular mass on SDS-PAGE. However, the apparent size is consistent with that of the protein expressed in COS-7 cells FIG. 4. Schematic presentation of the structural relationship between bovine BCNT and its human homologue and the epitope region. The human cDNA related to bovine BCNT was constructed by 5Ј-and 3Ј-RACE from human placental cDNA (GenBank TM accession number D85939) or was isolated from a human brain cDNA library (GenBank TM accession number D86549). The nucleotide sequences were determined and compared with bovine BCNT cDNA. Percentages indicate the amino acid identity between BCNT and human BCNT (hBCNT). The repeats of two TAATACC sequences and three GTCAGG sequences are shown by black bars. The epitope-related 13-amino acid sequence of bovine BCNT and the corresponding sequence in its human homologue are shown, and its location in bovine BCNT is illustrated. transfected with plasmids expressing the cDNA. The results of Northern blotting show a major band at 3.1 kilobases. The cDNA sequence around the potential initiation Met codon matches Kozak's rule well (21). Sequence comparison with human BCNT cDNA shows that the 5Ј-end of this region is not conserved between the two despite high homology in the postulated ORF regions. Therefore, although post-modification such as phosphorylation may also contribute to the apparent molecular size difference, we conclude that the cDNA codes the full-length ORF. The GST fusion protein of human EST expressed in E. coli migrates around 68 kDa on SDS-PAGE, although it was calculated to be 48 kDa. 3 This result implies that the BDDF homologous region is not the sole reason for the anomalous behavior of bovine BCNT on SDS-PAGE. This tendency has been reported in some hydrophilic proteins (22,23) such as neuromodulin, tau, and calpastatin. Since bovine BCNT is rich in charged amino acids (Glu, 11.2 mol %; and Lys, 8.8 mol %) and is especially rich in Glu in its N-terminal region, it provides another example of abnormal electrophoretic mobility on SDS-PAGE.
It is essential to clarify the function of BCNT and how the repetitive region is related to this function. Therefore, it is critical to evaluate whether there is only one bcnt gene in bovine and whether there is an isomer(s) of BCNT without the repetitive sequence region. Southern blotting of the bovine genome probed with the 5Ј-region of bcnt excluding the repetitive sequence showed that there is one gene. Consistent with this result, all phage clones obtained with the 53-mer nucleotide corresponding to the region other than the repetitive portion included the same 280-amino acid repetitive region, although one clone had a shorter C terminus (Fig. 3). Furthermore, we immunoblotted bovine tissue extracts with anti-bovine BCNT mAbs whose epitopes are located in regions other than the repetitive region. These mAbs detected only one major band, although a more detailed study is required. The expected molecular mass of human BCNT was calculated to be 24 kDa. It is intriguing that anti-BCNT mAbs recognized a 47-kDa protein, but not a 97-kDa protein, in extracts from COS-7 cells (see Fig. 5b) or human neuroblastoma SH-SY cells (24). 3 A 47-kDa protein was also found in rat brain and rat fibroblast 3Y1 cells. 3 Its identification is now under investigation.
To gain insight into the function of BCNT, we examined BCNT expression levels in various tissues and found that bovine BCNT is expressed predominantly in brain. Furthermore, we observed that BCNT is localized in Purkinje cells in bovine cerebellum. 3 Both bovine and human BCNT proteins have many consensus phosphorylation sites for casein kinase II (25), which is localized in nuclei. In fact, both human and bovine BCNT proteins are phoshorylated by casein kinase II. 3 A mouse EST (GenBank TM accession number W83351) shows high homology to human bcnt. This may indicate that BCNT is highly conserved in mammals, but that a repetitive sequence is inserted in the bovine protein. No sequence homologous to bcnt is found in the genomic sequence of Saccharomyces cerevisiae. Mammalian conservation suggests a biologically important role for this family. Although the function of BCNT is not yet clear, an analysis of the relationship between bcnt and its human homologue may shed light on gene development and organization. Bovine DNA digested with the indicated enzymes was separated on a 0.7% agarose gel and transferred to a nylon membrane. The membrane was probed with a 32 P-labeled 3Ј-fragment located downstream of the 840-bp repetitive sequence.
Kawabata and T. Yamamoto for comments on the expression library, T. Honda and M. Mercken for providing SH-SY cell extracts, N. Nomura for providing the data for the DNA sequence homology search, J. Takeuchi for help with naming, and M. Dooley-Ohto for editing the manuscript.