Oncogene (1998) 16, 2687 ± 2693  1998 Stockton Press All rights reserved 0950 ± 9232/98 $12.00 http://www.stockton-press.co.uk/onc SHORT REPORT ANA, a novel member of Tob/BTG1 family, is expressed in the ventricular zone of the developing central nervous system

Yutaka Yoshida1, Satoru Matsuda1, Naoko Ikematsu1, Junko Kawamura-Tsuzuku1, Johji Inazawa2, Hisashi Umemori1 and Tadashi Yamamoto1

Department of 1Oncology and 2Human Genome Center, The Institute of Medical Science, The University of Tokyo, 4-6-1 Shirokanedai, Minato-ku, Tokyo 108, Japan

Using a polymerase chain reaction-mediated cloning factors are implicated in the neural commitment and procedure, we have identi®ed a novel member, termed di€erentiation (Lee, 1997). Then a question arises: how ANA (from Abundant in Nueroepithelium Area), of Tob/ are the extracellular signals transmitted to the nucleus BTG1 family of antiproliferative . Molecular in the cells? cloning and analysis of cDNAs revealed that the human The tob , together with the and pc3/tis21/ and mouse ANA encoded a of 252 amino acids. genes, forms a new antiproliferative gene family The amino-terminal half of ANA was homologous to the (Matsuda et al., 1996; Rouault et al., 1992, 1996; previously characterized antiproliferative gene products, Bradbury et al., 1991; Fletcher et al., 1991). Both the BTG1, PC3/TIS21/BTG2, and Tob. The human ANA BTG1 and Tob suppress growth of NIH3T3 gene was localized at 21q11.2-q21.1. ANA cells when overexpressed (Matsuda et al., 1996; was expressed in a variety of tissues and cell lines, its Rouault et al., 1992). Overexpression of pc3 in expression being high in the ovary, testis, prostate, NIH3T3 and PC12 cells also leads to inhibition of thymus, and lung. Further analysis revealed that ANA their proliferation (Mantagnoli et al., 1996). Moreover, expression was high in the ventricular zone of the BTG2/TIS21/PC3 inactivation in ES cells leads to the developing central nervous system. Finally, overexpres- alteration of DNA damage-induced G2/M arrest sion of ANA impaired serum-induced progres- (Rouault et al., 1996). However, the underlying sion from the G0/G1 to . In conclusion, ANA is a mechanism(s) of the inhibition remains to be estab- fourth member of the Tob/BTG1 family that might play lished. roles in neurogenesis in the central nervous system. To further understand the function of the tob family genes, we searched for a new member of this family Keywords: Tob/BTG1 family; antiproliferative activity; using a PCR-mediated cloning procedure. Here we ventricular zone; cartilage show identi®cation and characterization of a new member of the tob family gene, termed ANA. Our data suggest that ANA is involved in the regulation of the development of the central nervous system. The In the developing mammalian central nervous system, molecular mechanism by which the cytoplasmic neural precursor cells dividing in the ventricular zone antiproliferative family protein is involved in neural determine their fate to become or glial cells. commitment is discussed. The di€erentiation of neural precursor is regulated, at least in part, by interactions between neural precursor Molecular cloning and characterization of a novel tob cells and their environments. For example, basic related gene ®bloblast growth factor (bFGF) participates in de®ning rostro-caudal identity of the neural tube We previously showed that the human tob mRNA is (Lumsden and Krumlauf, 1996). 2.3 kb long. The tob cDNA probe also detected other (Shh) and members of the transforming growth mRNA species with weak signals in some cell lines factor-b (TGF-b) family in¯uence ventral and dorsal such as FL18 and Daudi cells (Matsuda et al., 1996), features of the development in the caudal neural tube, suggesting the presence of tob-related genes. To and recombinant Shh induces ¯oor plate cells and identify a novel tob-related gene(s), we employed the motor neurons in neural plate explants (McKay, 1997; reverse transcription and polymerase chain reaction Roelink et al., 1995; Marti et al., 1995). In addition to (RT ± PCR)-mediated cloning procedure. Using RNAs extracellular signals, several region- and cell-type- from FL18 cells and degenerated oligonucleotide speci®c transcription factors are involved in neural primers corresponding to the amino acid sequences development: the products of the Hox, Pax and POU WFPEKP and WVDPYE, which were highly con- genes are involved in the de®nition of speci®c regions served among Tob, BTG1 and PC3 (Matsuda et al., along the rostro-caudal axis of the developing nervous 1996), we obtained nearly 100 clones with insert of system (Kessel and Gruss, 1990; Graham et al., 1991), expected size. Nucleotide sequencing of the cloned and basic helix ± loop ± helix (bHLH) transcription fragments followed by homology search with the Genbank and PIR databases revealed that one of the fragments, a 190-bp fragment, represented a novel Correspondence: T Yamamoto gene. To obtain the entire coding sequence of the gene, Received 23 June 1997; revised 18 December 1997; accepted 18 we screened an oligo-(dT) primed cDNA library December 1997 generated from Daudi cells and the murine testis in Expression of ANA in the neuroepithelium YYoshidaet al 2688 lZAPII. Nucleotide sequencing of the cDNA inserts (AATAAA), a poly (A) tail, and four ATTTA motifs. revealed a 756 bp open reading frame encoding a The ATTTA motif appears in the 3'-untranslated protein of 252 amino acids (Figure 1). The calculated regions of several immediate early genes and proto- molecular weight of the protein product was 28 970. oncogenes, and is thought to be responsible for rapid The gene for this protein was termed ANA from degradation of mRNAs (Shaw and Kamen, 1986). The Abundant in the Neuroepithelial Area. The 3'- homology between the mouse and human ANA genes untranslated region of mouse ANA was rich in A and was high: nucleotide sequence identity was 91%; and T nucleotides and contained a polyadenylation signal the amino acid sequence identity was 93% (Figure 1b).

a b

c d

Figure 1 Cloning and sequence analyses of ANA.(a) The nucleotide sequence and its deduced amino acid sequence of the mouse ANA gene. The nucleotide and amino acid sequences are numbered on the left and right, respectively. The ATTTA motifs and polyA signal in the 3'-untranslated region are underlined by single and double lines, respectively. The mouse and human ANA cDNA sequences have been registered to the DDBJ, EMBL and GenBank databases with the accession number D83745 and D64110, respectively. (b) Sequence alignment of human and mouse ANA. Each vertical line and colon indicate an identical residue and a conservative substitution, respectively. (c) Schematic diagram illustrating the structure of Tob family proteins. The hatched boxes (sequences conserved among Tob family members), black box (nuclear localization signal), PQ (proline and glutamine rich region) and P (proline rich region) are indicated respectively. (d) Alignment of the amino-terminal half of ANA with the corresponding regions of the other members of Tob family. Dashes indicate amino acid residues identical to ANA. Oligonucleotide primers corresponding to the amino acid sequences WFPEKP and WVDPYE were synthesized (the sense primer and antisense primer are CATGAATTCA(T/C)TGGT(A/T)(T/C)CCNGA(A/G)AA(A/G)CC and CAAGAATTCTC(A/G)(A/T)ANGG(A/ G)TCNA(T/C)CCA, respectively). To synthesize cDNAs, total RNAs (10 mg) isolated from FL18 cells were incubated with 100 units of modi®ed reverse transcriptase of Molony murine leukemia virus (Superscript, BRL) following the manufacturer's protocol. The cDNA pool (0.5 ml) was PCR ampl®ed as described previously (Fujimoto and Yamamoto, 1994). PCR products were electrophoresed, checked by the expected size (about 190 bp), and ethanol-precipitated. The products were digested with EcoRI whose recognition sites were tagged to the PCR primers. After digestion, the products were subcloned into the EcoRI site of Bluescript SK-II (Stratagene). Screening of cDNA libraries were performed by using the newly isolated DNA insert with standard protocols (Sambrook et al., 1989) Expression of ANA in the neuroepithelium YYoshidaet al 2689 Inspection of the deduced amino acid sequence of the by staining for the ANA protein, and cells that entered ANA gene product (ANA) revealed its signi®cant the S phase were identi®ed by staining for BrdU homology at the amino-terminal half (residues 1 ± incorporation. While 64 out of 75 (85.3%) cells 115) to the corresponding portions of BTG1 (36% micro-injected with a vector containing the b- identical, Rouault et al., 1992), PC3 (38%, Bradbury et galactosidase cDNA incorporated BrdU, only 61 out al., 1991), Tob (31%, Matsuda et al., 1996), and the of 112 (54.5%) cells micro-injected with the ANA Xenopus B9 protein (57%, Genbank) (Figure 1c and d). cDNA showed BrdU incorporation (P50.05). Typical Therefore, we concluded that ANA is a member of data are shown in Figure 2. Since ANA-dependent Tob/BTG1 family antiproliferative proteins. Finally, inhibition of the cell cycle progression was leaky, we the ANA gene turned out to be the same gene as assumed that the elevated expression of ANA retarded recently reported (Gue henneux et al., 1997). the cell cycle progression through the G1 phase. Consistent with this notion, the degree of the apparent inhibition was a€ected by the length of BrdU ANA impairs cell cycle progression from the G /G to 0 1 incubation time (data not shown). Taken together we S phase conclude that the amino-terminal half of ANA, which To assess the ability of ANA to inhibit cell is conserved among Tob/BTG1 family members, is proliferation, we micro-injected the ANA expression inhibitory for cell cycle progression through the G1 plasmid into NIH3T3 cells that had been brought to phase. quiescence by serum starvation. After cells were stimulated with serum to restart synchronous cell Chromosome assignment of the human ANA gene cycle progression, micro-injected cells were identi®ed The chromosomal localization of the human ANA gene was determined by ¯uorescent in situ hybridization (FISH) using the human ANA cDNA fragment as a

Figure 2 E€ect of ANA on cell cycle progression. Serum- deprived NIH3T3 cells micro-injected with the human ANA (a, b and c)orb-galactosidase (d, e and f) expression plasmids were re- stimulated with serum and then stained for ANA (a) and b- galactosidase (d) expression, or BrdU incorporation (b and e). In c and f, cells were stained with DAPI. White arrowheads point to Figure 3 Chromosomal assignment of the human ANA gene by representative cells expressing ANA and b-galactosidase. NIH3T3 FISH. (Left) Metaphase stained with the propidium cells grown on coverslides (1.56104 cells/cm2) were cultured in iodide-labeled ANA probe. Twin-spot signals on the long arm of DMEM containing 0.1% calf serum for 24 h and then micro- chromosome 21 are indicated by arrows. (right) The G-banding injected with the pME-hANA or pME-b-galactosidase (200 mg/ml pattern of the same chromosomes delineated through a UV-2A DNA). After incubating for 12 h in the same conditions, cells ®lter (Nikon). Metaphase chromosomes prepared from the were treated for staining as described previously (Baeg et al., peripheral blood lymphocytes of a healthy donor were hybridized 1995). The ANA protein was detected using the anti-ANA with the T5G1 probe as described (Inazawa et al., 1993). The antibodies raised by immunizing rabbits with a peptide, 15 kbp T5G1 DNA probe used for the in situ hybridization was CDRNHWINPHMLAPH, that corresponds to the amino-acid obtained by screening approximately 56105 recombinant phages residues 238 ± 252 of ANA. BrdU was detected with an anti-BrdU of EMBL3 human genomic DNA library with the human ANA monoclonal antibody BU6-4 (Takara) cDNA Expression of ANA in the neuroepithelium YYoshidaet al 2690 probe. As shown in Figure 3, the signals were located mention that frequent loss of heterozygosity (LOH) on on the long arm of chromosome 21, 21q11.2-21.1. chromosome 21q11 is found in the lung and Although it is not known whether ANA is involved in esophageal cancers (Aoki et al., 1994; Sato et al., the genesis of certain cancer(s), it may be worthy to 1994).

a

Kb — pancreas — kidney — skeletal muscle — liver — lung — placenta — brain — heart — peripheral blood leukocyte — colon — small intestine — ovary — testis — prostate — thymus — spleen 9.5 — 7.5 —

4.4 —

2.4 — ANA 1.4 —

β-actin

b c E12 E14 E16 E18

ANA — MDA-MB 231 — CA922 — MT2 — IM9 — NA — HSG — MGH — TE1 — ZR75-1 — MRK

ANA

β-actin

β-actin

Figure 4 Expression of the ANA mRNA. The expression of the ANA mRNA in human tissues (a), various human cell lines (b), and mouse developing brain (c) were investigated by RNA blot hybridization. Poly(A)+ RNA (2 mg) prepared from various cancer cell lines and total RNAs (20 mg) from developing brain at indicated days postcoitum were electrophoresed on 0.8% agarose gels containing 2.2 M formaldehyde. The fractionated RNAs were transferred to Hybond-N membranes (Amersham). The ®lters containing poly(A)+ RNA (2.0 mg for each tissue) of various human and murine tissues were purchased from Clontech. The ®lters were hybridized with the RT ± PCR product (190 bp) fragment of human ANA), the 0.5 kbp StuI±KpnI fragment of mouse ANA or a b-actin probe labeled with a-32P-dCTP by random priming. The RNA sizes in kilobases are shown on the left (a) Expression of ANA in the neuroepithelium YYoshidaet al 2691 Expression of the ANA mRNA in various cell types Northern blot hybridization analysis showed that the 1.5 kb ANA mRNA was expressed in various tissues (Figure 4a), its expression level being relatively high in the testis, lung, and ovary and low in the skeletal muscle and spleen. The expression pattern was virtually identical between human and mouse ANA (data not shown). Among human cancer cell lines, the ANA mRNA was expressed relatively high in IM9 lymphatic cells, CA922 oral tumor cells, and ZR75-1 breast cancer cells, and weak in MRK breast cancer cells, MT2 cancer cells, and MGH bladder cancer cells (Figure 4b). The ANA mRNA expression did not appear to be restricted to speci®c types of cancers. It is not known whether the levels of the ANA mRNA in various cell lines correlate with the expression levels of the protein product, because the endogenous proteins were hardly detected with anti-ANA anti- bodies.

Expression and subcellular localization of the ANA protein Rabbit polyclonal antibodies were raised against synthetic peptides corresponding to amino-acid residues 238 ± 252 of ANA. In an immunoblotting experiment, the puri®ed antibody detected a 33 kDa protein in the lysates of HEK293T cells transfected with the ANA expression plasmid pME-hANA (data not shown). The 33 kDa ANA protein was also identi®ed in anti-ANA immunoprecipitates from lysates of the pME-hANA-transfected HEK293T cells that had been metabolically labeled with 35S- methionine and 35S-cystein (data not shown). The expression level of the endogenous ANA protein was extremely low as compared with the overexpressed one. Immuno¯uorescence microscopy showed that ANA was predominantly localized in the cytoplasm of the pME-hANA-transfected 293T cells (data not shown). Cytoplasmic localization of ANA was also observed in NIH3T3 cells micro-injected with pME-hANA (Figure 2a). The results are consistent with the fact that ANA, unlike Tob, does not carry a putative nuclear localization signal.

ANA expression during mouse development Figure 5 Expression of ANA during mouse embryogenesis. In situ hybridization was performed on sections of mouse whole Although the ANA expression was low in the mouse embryos and brains as described below. The antisense ANA RNA was used as a probe for a±e.(a) A saggital section of E12.5 adult brain, we found that it was expressed at a embryo. (b) A saggital section of E14 embryo. The arrow signi®cant level in the developing brain. As shown in indicates Meckel's cartilage (Mc). (c) A transverse section Figure 4c, the ANA mRNA level was high in the through the head of E10.5 embryo. (d) A transverse section brain at 12th day postcoitum (E12), when neurogen- through the head of E12.5 embryo. (e) A transverse section esis was actively occurring. To determine the cell types through the spinal cord of E12.5 embryo. Counterstains were carried out with cresylviolet. Abbreviations are as follows: (tv) that expressed the ANA mRNA during mouse indicates third ventricle, (fv) fourth ventricle, (lv) lateral ventricle, embryogenesis, we carried out in situ hybridization (telv) telencephalic vesicle, (liv) liver, (vz) ventricular zone, and experiments using histological sections of the whole (mz) marginal zone. Bars in a, and b, c, d, and e are 1, 0.2, 0.5 embryo and brain from E10.5 through E14.0. ANA and 0.1 mm, respectively. Embryos were collected at indicated times, ®xed in 4% paraformaldehyde, cryoprotected in 20% expression was detected in several tissues of the E12.5 sucrose, embedded in OCT (Tissue-Tek, Miles Inc.) and stored at embryo, being prominent throughout all the ventri- 7808C until analysis. The sections (14 mm) were cut on a cular zone (Figure 5a). In the E14 brain, however, cryostat. Specimens were hybridized with riboprobes labeled with expression of ANA was much regressed (Figure 5b). a-35S-labeled UTP as described previously (Umemori et al., 1992). The results are consistent with the Northern blot data The pBluescript plasmid containing the mouse ANA cDNA (encompassing nucleotides 1 ± 690, Figure 1a) linearized with that the ANA expression is higher in the E12 brain EcoRI was transcribed by T7 polymerase to produce sense and than in the E14 brain (Figure 4c). Furthermore, antisense products Expression of ANA in the neuroepithelium YYoshidaet al 2692 higher magni®cation analysis of the E12.5 spinal cord Implications demonstrated that the cells expressing high amounts of the ANA mRNA were present in the ventricular The present data suggest that ANA is important in zone, where neuroepithelial cells proliferate and the proliferation of the neuroepithelial cells in the commitment to a speci®c neural phenotype proceeds ventricular zone. Alternatively, ANA may play a role (Figure 5e). In contrast, in the intermediate zone and in the growth arrest of the neural precursor cells by its the marginal zone, ANA expression was very low transient expression in the cells of the ventricular zone (Figure 5e). Some cells that are no longer participating right at the time when the commitment of the in DNA synthesis and mitosis migrate laterally to precursor cells into the neural or glial lineage takes form an intermediate layer where cells di€erentiate place. This agrees with the observation that ANA is into neurons or glia (Gilbert, 1994). The expression of associated with antiproliferative activity. It is also the ANA mRNA in the ventricular zone and its possible that ANA is actively involved in the suppression in the intermediate zone suggest that commitment of neuroepithelial cells to speci®c neural ANA plays a role in the event that occurs in the cell types. In addition, ANA may be actively involved ventricular zone. in the formation of the cartilages. It is interesting to Interestingly, the pattern of the ANA expression is speculate that ANA mediates di€erentiation signals quite similar to that of pc3 (Iacopetti et al., 1994), from cell surface to the nucleus. suggesting that ANA and PC3 may have an over- lapping role during neurogenesis. However, unlike pc3, ANA mRNA expression was not con®ned to the ventricular zone. As shown in Figure 5b, high level of ANA mRNA expression was detected in the Acknowledgements Meckel's cartilage and cartilages of the ribs of the We thank J Fujimoto, Y Shiio, T Kohno, J Yokota and M Nakafuku for technical supports and valuable discussions. E14.0 embryo. Weak expression of the ANA mRNA We also thank O Minowa for teaching us to manipulate was also observed in the cartilages at E12.5 (data not the mouse embryo, M Ebisawa and F Matsushita for shown). In mammalian embryos, cranial neural crest technical advice in microscopic studies, S Sakakibara and cells migrate before the neural tube is closed (Johnston H Okano for providing us with Northern blot ®lter, M et al., 1985) and give rise to the facial mesenchyme Ohsugi for providing us Daudi cells and mouse testis (Tan and Morris, 1985). The mesenchyme further library and Y Takahashi, A Tanaka for critical reading of di€erentiates into facial connective tissues including the manuscript. EMBL3 human genomic library (L1018) the Meckel's cartilage. The rib cartilages are produced were obtained from the Japanese Cancer Research from mesenchyme of mesoderm origin. Though their Resources Bank. This work was supported in part by a origins di€er from each other, cells in the both grant for Advanced Cancer Research from the Ministry of Education, Science, Sports and Culture of Japan, the cartilages may be involved in ossi®cation and bone Program for the Promotion of Fundamental Studies in formation. Expression of ANA in embryonic cartilages Health Sciences of the Organization for Drug ADR Relief, may suggest that ANA also plays a role(s) in the R&D Promotin and Product Review of Japan, and a grant proliferation and/or chondrocytic di€erentiation of the from the Research Fellowships of the Japan Society for the cartilage cells. Promotion of Science for Young Scientists.

References

Aoki T, Mori T, XiQun D, Nisihira T, Matsubara T and Kessel M and Gruss P. (1990). Science, 249, 374 ± 379. Nakamura Y. (1994). Genes, Chromosomes & Cancer, 10, Lee JE. (1997). Curr. Opin. Neurobiol., 7, 13 ± 20. 177 ± 182. Lumsden A and Krumlauf R. (1996). Science, 274, 1109 ± Baeg GH, Matsushime A, Kuroda T, Bhattacharjee RN, 1115. Miyashiro I, Toyoshima K and Akiyama T. (1995). Mantagnoli A, Guardavaccaro D, Starace G and Tirone F. EMBO J., 14, 5618 ± 5625. (1996). Cell Growth & Di€erentiation, 7, 1327 ± 1336. Bradbury A, Possenti R, Shooter EM and Tirone F. (1991). Marti E, Bumcrot DA, Takada R and McMahon AP. (1995). Proc. Natl. Acad. Sci. USA, 88, 3353 ± 3357. Nature, 375, 322 ± 325. Fletcher BS, Lim RW, Varnum BC, Kujubu DA, Koski RA Matsuda S, Kawamura-Tsuzuku J, Ohsugi M, Yoshida M, and Herschman HR. (1991). J. Biol. Chem., 266, 14511 ± Emi M, Nakamura Y, Onda M, Yoshida Y, Nishiyama A 14518. and Yamamoto T. (1996). Oncogene, 12, 705 ± 713. Fujimoto J and Yamamoto T. (1994). Oncogene, 9, 693 ± 698. McKay R. (1997). Science, 276, 66 ± 71. Gilbert SF. (1994). Developmental Biology, 4th ed., Sinauer Roelink H, Porter JA, Chiang C, Tanabe Y, Chang DT, Associates: Sunderland, MA, pp. 244 ± 289. Beachy PA and Jessell TM. (1995). Cell, 81, 445 ± 455. Graham A, Maden M and Krumlauf R. (1991). Development, Rouault JP, Rimokh R, Tessa C, Paranhos G, Ffrench M, 112, 255 ± 264. Duret L, Garoccio M, Germain D, Samarut J and Magaud Gue henneux F, Duret L, Callanan MB, Bouhas R, Hayette JP. (1992). EMBO J., 11, 1663 ± 1670. S, Berthet C, Samarut C, Rimokh R, Birot AM, Wang Q, Rouault JP, Falette N, Guehenneux F, Guillot C, Rimokh R, Magaud JP and Rouault JP. (1997). Leukemia, 11, 370 ± Wang Q, Berthet C, Moyret-Lalle C, Savatier P, Pain B, 375. Shaw P, Berger R, Samarut J, Magaud JP, Ozturk M, Iacopetti P, Barsacchi G, Tirone F, Ma€ei L and Cremisi F. Samarut C and Puisieux A. (1996). Nature Genet., 14, (1994). Mech. Dev., 47, 127 ± 137. 482 ± 486. Inazawa J, Saito T, Akiyama T, Aba T and Nakamura Y. Sambrook J, Frisch EF and Maniatis T. (1989). Molecular (1993). Genomics, 17, 153 ± 162. Cloning: A Laboratory Manual. Cold Spring Harbor Johnston MC, Sulik KK, Webster WS and Jarvis BL. (1985). Laboratory Press: Cold Spring Harbor, NY. Teratology, 31, 26A. Expression of ANA in the neuroepithelium YYoshidaet al 2693 Sato S, Nakamura Y and Tsuchiya E. (1994). Cancer Res., Tan S-S and Morriss-Kay G. (1985). Cell Tissue Res., 240, 54, 5652 ± 5655. 403 ± 416. Shaw G and Kamen R. (1986). Cell, 46, 659 ± 667. UmemoriH,WanakaA,KatoH,TakeuchiM,TohyamaM Shiio Y, Yamamoto T and Yamaguchi N. (1992). Proc. Natl. and Yamamoto T. (1992). Mol. Brain Res., 16, 303 ± 310. Acad. Sci. USA, 89, 5206 ± 5210.