425
Biot;dinica et Biophysica Acta, 1088 (1991) 425-428 © 1991 Elsevier Science Publishers B.V. 0167-4781/91,/$03.50 ADONIS 016747819100111C BBAEXP 90222
BBA Report - Short Sequence-Paper
Primary structure of a human protein kinase regulator protein P e t e r J. N i e l s e n Max Planck Institut ftir lmmunbwlog/e, Fretburg (F R. G.) (Received 14 November 19901
Key words: Protein, 14.3.3; Hydroxylase activator; Protein kinase C inhibitor protein; (Human cDNA)
A human T-cell cDNA is characterized which codes for a protein closely related to a family of proteins alternatively described as co-regulators, of monoamine biosynthesis in neurons and as inhibitors of protein kinase C. The preOi~ed human protein is 28 kDa and has a p! of 4.5. Alignment of putative protein sequences shows strong conservation from Drosophila to man. Several mRNA transcripts hybridizing to the cDNA are seen in human epithelial and T-cell lines and in various mouse tissues.
The 14.3.3 protein was originally identified as an abundant 'brain specific' protein [1]. It has been estimated to be approx. 1% of total cytosolic protein in brain but can be found at lower levels in most other tissues [2]. The 14.3.3 protein has been shown, in conjunction with Ca2+/calmodulin dependent protein kinase type II to activate tyrosine and tryptophan hydroxylases [3]. In neurons, these enzymes are believed to be rate-limiting in the biosynthesis of catecholamine and serotonin, respectively. The same or very similar !4.3.3-1ike proteins have been purified from sheep brain on the basis of their ability to inhibit protein kinase C [4]. Both the purified kinase activator and the kinase-C inhibitor preparations contain heterogeneous polypeptides. Bovine brain 14.3.3 for example, was reported to contain at least seven different proteins [5]. The nature of this subunit heterogeneity is not known but all have been reported to share a common immunological determinant and have very similar but not identical amino acid sequences [5]. The extent to which the different subunits are responsible for different functional or physiological activities of the protein remains to be established. A human cDNA library was constructed by standard techniques using RNA derived from the Jurkat T-cell
The sequence described here has been submitted to the EMBL/ Genbank Data Libraries under the accession number X56468. Correspondence: P.J. Niel~n, Max Planck lnstitut for Immunbiologie. Stiibeweg 51, D-7800 Fre~burg, F.R.G.
line (obtained from Dr. Theophii Staehelin). During the screening of this bank with a cDNA probe coding for a previously uncharacterized nuclear protein (Nielsen, unpublished data), a cDNA clone was fortuitously isolated which proved to be the human equivalent of the so called 'brain specific" regulator or 14.3.3 protein. The nucleotide and deduced protein sequence of this 1.8 kb clone is shown in Fig. 1. One long open reading frame is present extending from the first A U G at position 126 to a translational stop at 861. The encoded protein is predicted to contain 245 amino acids with a molecular mass of 27.8 kDa and a p l of 4.5. Possible polyadenylation signals were seen at 15!4 and 1658 but not at the end of the cDNA, presumably because it is not full length. A computer-aided search revealed a strong similarity with the recently described bovine 14 3.3 protein gene [5]. Within the coding region, the nucleotide sequences are 71% identical. The amino acid sequences (aligned in Fig. 2) show 72% identity and 84% similarity. There are two positions where the human sequence has a two amino acid deletion relative to the bovine sequence. The 5' and 3' noncoding regions of the mRNA's show no relatedness (not shown). The computer search also revealed a match with a portion of a cDNA reported to encode an alternative form of the EGF-receptor from Drosophila [6]. In this alternative form, a 5' 321 bp sequence differing from other EGF-receptor cDNA clones and containing an open reading frame was interpreted to result from alternative splicing. Since the match to the 14.3.3 protein sequence includes only the 321 bp 5' 'alternative exon', it seems likely that this
426 10 30 50 ~g~gggsctcgc~cgcggccgcggagac~gaaqctetegaggctcctcccgctgeg 90 90 110 g~cggcgctcgccctegctetcctcgecctccgccecggccccggccccggccecgcge 130 150 170 ecgccatggagaagactgagetgltccaglaggceaagctggccglgclggecglgcget M E K T E L I O K A K L & ~ O A E R T 190 210 230 aegacgaeatggccacctgcatgaaggca~gaccgagcagggcgcc~gct~ccaaeg D D M A T C M K J V T E O G A E L S N E 250 290 290 lggagcgcaacctgctctec~ggcctacalgaac~g~cggggg¢cg¢Ig~c¢gcct E R N L L S V A T ~ N V V G G n l S & W 310 330 350 ggagg~catctctagcatcgageaglag~ccgaeacetecgacaaqas~tgcsgetga R V I S S I E Q ~ T D T S D K I L O L I 370 390 410 ttaaggsctategggagaaa~gga~ccgagetgsgatccetctgcaeeaeg~gctgg ~ D y R ~ K V | S E L R S I C T T V L E 430 450 670 latt~tggataaatatttaltageeaatgcaactaatccagaga~lag~¢ttctatc L L D K I L I A N & T N P I S K V F Y L 490 510 530 tgaasatgaagg~gattacttecg~accttgctgaa~tgc~g~tgatcgaa
K H K G D Y F R ~ L A E V A C G D D R K
550 570 590 aaeaaacgatagataaZtcceaaggagettaceaagaggcatttgatataagcaagaaag
~ T I D N S O G A Y Q ~ A F D I S K K E
610 630 650 agatgeaaeccacacacecaatccgcctggggett~ctcttaa~tttct~attttaet M O P T H P I R L G L A L N F S V F Y ~ 670 690 710
atgagattcttaataacccagagcttgcctgcacgctggctaaaacggcttttgatgagg E I L N N P E L A C T L & K T A F D E A 730 950 770 cclttgctgalcttgatacactgaatgalgactcatac~aagaca~accctcatcltgc I A E L D T L N [ D S T ~ D S T L I M O 790 810 830 a~tgcttaglgacaacctaacaetttggacatcagaca~gca~agaaglat~qetg b L R D N L T L W T S D S A G E [ C D & 950 970 890 cggcagaaggggctgaaaactaaatccataeagg~catcett~ttc~teuagaaa
& E G A E N *
910 930 950 ectttttleaeatetccatteettattecacttggatttcctatageaaa~aacceatt 990 990 lOlO cat~atggaatcaa~ttata~cttttcacaetgcagctttgggaaaacttcat 1030 1050 1070 tccttglttt~tt~cttggccttcctg~gca~actgct~igaaaa~atta 1090 1110 1230 atagcttc~tttcatataaacatal~aactcccaaacacttat~lgacjcjactaalaat 1150 1170 1190 ~atetg~atttaa~aatctgaaeea~t~gc~a~gact~ttt~attact~ 1210 1230 1250 gaaaataagaaaat~a~aattacaatttaaag~gtatteeaeataacttcttaattt 1270 1290 1310 ctacattccctcccttactcttcgggg~ttcctttca~aagcaacttttecatgctct 1330 1350 1370 taatgtattccttttta~aggaatccggaa~attagattgaatggaaaagcacttgcc 1390 1410 2430 ltctct~ctaggg~cacaaattgaaatggctcct~atcacataccggag~ctt~g 1450 1470 1490 tatct~ggccaacaggga~ttcctt~ttcactctttatttgctgct~ttaa~tgcc 1510 1530 1550
aacctcccctcccaataaaaattcacttacacctcctgccttt~a~tctg~attcac 15~0
1590
1610
tttactat~gatagaa~agcat~tgctgecagaatacaagcattgctttt~caaat 163C 1650 1690 taaag~gcat~catttcttaatacactagaaaggggaaataaattaaa~acacaa~e 1690 1910 1730 caagtctaaaacttta~acttttcecatgcagattt~gcac~t~gagagg~cca 1750 1770 1790 ~t%~cta~gatt~tatttagaga~tggaccactatt~ttgctaatcattg~ 1810 1830 1850
ct~ag~ cecaaaaaagcctt~gaaaat~tatgccctat~aacagcaqagtaaeata aa
Fig. 1. Human eDNA sequence ~ r 14.3.3 protein. The nucleotide sequence was determined on both strands except ~ r the non-coding re~ons 1230-1440 and 1670-186~3 w ~ c h w e ~ de.trained on one strand o d y . The sequen~ includes the complete open reading ~ a m e be~nning ~ t h the first A U G at nueleotide 126 and ~rminating at position 863. The deduced amino ~ i d sequence is shown below the nucleotide sequence.
clone represents a hybrid cDNA generated during cloning of the eDNA's. A comparison of the primate, bovine, ovir.e and insect sequences is shown in Fig. 2. The human and Drosophila sequences are .nore similar to each other than either is to the bovine sequence. Both amino acid gaps in the human sequence are also present in the Drosophila sequence. The simplest interpretation of this observation is that the human and Drosophila sequences represent equivalent genes in these two species and the bovine sequence corresponds to a different member of the 14.3.3 family. The expression of 14.3.3 sequences was examined in several cel~ types from human and mouse origin. Expression was seen in both the Jurkat human T-cell lymphoma (Fig. 3, lane T) and the HeLa human carcinoma line (not shown). In both cases, two strong bands were seen of estimated sizes 1.9 kb and 2.1 kb and a weak band was seen at 1.7 kb. When total RNA from several mouse tissues was examined, two bands of 1.7 and 1.9 kb were seen. Based on densitometric scanning of appropriately exposed autoradiograms and correcting for differences in RNA loading using the hybridization signal for ribosomal protein S12 mRNA, 14.3.3 levels in !ung and kidney are estimated to be 5-fold and 2-fold less, respectively, relative to that found in brain. Liver and spleen mRNA levels are estimated to be at least 10-50-fold less than brain. Whether the various mRNA species which hybridize to 14.3.3 e D N A are derived from the same or related genes is not known. Hybridization of human 14.3.3 e D N A to human and mouse genomic D N A digested either with BamHl or EcoRI revealed bands ranging in length from 4.4 to 25 kb. The aggregate length of the EcoRl-derived fragments is roughly 60-70 kb, suggesting either one gene with large introns or a gene family. Protein sequence data [4,5] are compatible either with several genes or with a complicated alternative sphcing of one gene. A more detailed analysis of ~he genomic sequences is necessary to distinguish between these possibilities. The technical assistance of Birgit Groetschel is gratefully acknowledged.
427 1 Hum1433 Bov1433, Droegfrc Ovi1433a Ovi1433b Ovi1433c
Hum1433 Bov1433n Droegfrc
MEKTE) IQK
MGDREQ. L. R
AMLAEQAEPY r
DDMATCMKAV
R . . . . . . . . . . .
SA . . . . . . REA
.D.S..V ............... ,D.N..V . . . . . . . . . . . . . . . MDDRED. VYQ . . . . . . . . . . . 51 NWGGRRSAW
RVISSIEQKT
.S
TEQGAELSN£
LNEP. . . D . . . . . . . . . .T.V . . . . . . . . . . . . . . .
£. --DTSDKKLQ
LIKDYREKVZ
i00 SELRSICTTV
.... A...S ...........
MA.GNE...E KV.A . . . . I .
K..ETV.ND.
....
--EA.AR.Q.
K.
A.
, .S . . . . . . . . . . .
IARK
....
~.,
101 Hum,433 Bov1433, D r o e g f rc Ovi1433d Ovi1433e Ovi1433f
50 ERNLLSVAyZ
LELLDKYLIA NAT--NPESK VFYLKMKGDY .A .... F . . K . C N D F Q Y . . . . . . . . . . . . . .G . . . . . . . P K. S--. . . . . . . . . . . . . . . ....
.... ....
FRYLAEVA:G Y ....... S. Y... Y ..... F.T. F . . . . . AQ Y ....... A,
, .E.
,YE.
150 DDRKQTIDNS EKKNSWEA. ..
.. ..
Hum1433 Bov1433~
151 QGAYQEAFDI EA. , K . . . E .
SKKEMQPTHP IRLGLALNFS VFYYEILNNP . .EH . . . . . . . . . . . . . . . . . . . . . . Q.A
200 ELACTLAKTA .Q..L...Q.
Hum1433 Bov1433,
201 FDEAIAELDT LNEDSYKDST LIMQLLRDNL TLWTSDSAGE . .D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . QQD.
250 ECDAAEGAEN .AGEGN
Fig. 2. Amino acid ~quence alignment betv,'een 14.3.3-1ike proteins from various spe-ies. The am,no acid sequence deduce~l from the human 14.3.3 eDNA (Hum1433) is aligned with published 14.3.3-1ike proteins deri',ed eilher from t D N A sequences or directly from the protein. ]"he sequences are bovine (Bov1433 rl, Ref. 5), Drosophila (Droegfrc. Ref. 6). and sheep peptides a - f (Ovil1433a-f, Ref. 4). Identical amino acids are indicated by a "." and gaps b', "-'.
A
B M
Lu
Li
B
K
S
T
M
1
2
3
Ill 9.4
lJ
4,4 3.11
-"
1,7
1.1
1.1
°i s~2
~
O
! III,,, w ",'ll"
Fig. 3. Northern and genomic Southern hybndization analysis of 14.3.3 sequences (A) Approx. l0 ,ug of total cellular RNA ,,,,'as separated on 0.9~ agarose gels containing 6~ formaldehyde, transf~'rred to nylon membranes and h~bndt~ed under ,~r~.,~nt ~u~ndition~ 16 × S . ¢ ~ IOx Denhardt's. 0.1~ SDS, 6 4 ° C . 14 h) with the eDNA spanning the coding region from nucleotlde 1 to 1000 ,,vhich v,'as unir'.~rmb ,,~dio|abcUed. Af!Pr hybridization, the filters were washed (two 30 rain changes of 0.2 × SSC. 0.1 q~ SDS. ,12 o C) and expo.,~xl with Kodak XAR-film for 3 days. The same filter was subsequently hybridized with a radiolabelled e D N A for mouse ribosomal protein St2 to corr~:t for differences bet~,een lanes in the amount of RNA loaded. The samples included size markers (M). RNA from the Jurkat human T-cell line (T). and various mouse tissues: lung (Lu). liver (Li), brain (B), kidney (K) and spleen (S). (B) Approx. 15 txg of human genomic DNA (lanes 1 and 2) derived from peripheral blo~l lymphocytes or mouse liver genomic DNA (lane 3) ,,,,-ere digested with BamHl (lanes 1 and 3) or EcoRI (lane 2). electrophoresed in a 0.7~ agarose gel, transferred to a nylon membrane and hybridized under stringent conditions with the ~ame probe described above. Radiolabelled DNA sb. markers were ccelectrophoresed in the rome gel (lane M).
428
References 1 Moore. B.W. and Perez, VJ. (1967) in Physiological and Biochemical Aspects of Nervous Integration (Carlson. FD.. ed.J pp. 343359. Prentice-Hall, Englewood Cliffs. 2 Boston, P.F.. Jackson, P. and Thompson. R.J. (1982) J. Neurochem. 38, 1475-1482. 2 Ichimura. T.. Isobe. T., Okuyama, T.. Yamauchi. T. and Fujusawa, H. (1987) FEBS Left. 219. 79-82.
4 Toker. A., Ellis, C.A., Seller. L.A. and Aitken, A. (1990) Eur. J. Biochem. 191,421-429. 5 Ichimura. T. lsobe, T., Okuyama, T.. TakahashL N., Araki, K, Kuwano. R.. and Takahashi, Y. (1988) Proc. Natl. Acad. Sol. USA 85, 70.~4-7088. 6 Schejter, E.D.. SegaL D., Glazer, L. and Shilo, B.-Z. (1986) Cell 46, 1091-1101.
Biot;dinica et Biophysica Acta, 1088 (1991) 425-428 © 1991 Elsevier Science Publishers B.V. 0167-4781/91,/$03.50 ADONIS 016747819100111C BBAEXP 90222
BBA Report - Short Sequence-Paper
Primary structure of a human protein kinase regulator protein P e t e r J. N i e l s e n Max Planck Institut ftir lmmunbwlog/e, Fretburg (F R. G.) (Received 14 November 19901
Key words: Protein, 14.3.3; Hydroxylase activator; Protein kinase C inhibitor protein; (Human cDNA)
A human T-cell cDNA is characterized which codes for a protein closely related to a family of proteins alternatively described as co-regulators, of monoamine biosynthesis in neurons and as inhibitors of protein kinase C. The preOi~ed human protein is 28 kDa and has a p! of 4.5. Alignment of putative protein sequences shows strong conservation from Drosophila to man. Several mRNA transcripts hybridizing to the cDNA are seen in human epithelial and T-cell lines and in various mouse tissues.
The 14.3.3 protein was originally identified as an abundant 'brain specific' protein [1]. It has been estimated to be approx. 1% of total cytosolic protein in brain but can be found at lower levels in most other tissues [2]. The 14.3.3 protein has been shown, in conjunction with Ca2+/calmodulin dependent protein kinase type II to activate tyrosine and tryptophan hydroxylases [3]. In neurons, these enzymes are believed to be rate-limiting in the biosynthesis of catecholamine and serotonin, respectively. The same or very similar !4.3.3-1ike proteins have been purified from sheep brain on the basis of their ability to inhibit protein kinase C [4]. Both the purified kinase activator and the kinase-C inhibitor preparations contain heterogeneous polypeptides. Bovine brain 14.3.3 for example, was reported to contain at least seven different proteins [5]. The nature of this subunit heterogeneity is not known but all have been reported to share a common immunological determinant and have very similar but not identical amino acid sequences [5]. The extent to which the different subunits are responsible for different functional or physiological activities of the protein remains to be established. A human cDNA library was constructed by standard techniques using RNA derived from the Jurkat T-cell
The sequence described here has been submitted to the EMBL/ Genbank Data Libraries under the accession number X56468. Correspondence: P.J. Niel~n, Max Planck lnstitut for Immunbiologie. Stiibeweg 51, D-7800 Fre~burg, F.R.G.
line (obtained from Dr. Theophii Staehelin). During the screening of this bank with a cDNA probe coding for a previously uncharacterized nuclear protein (Nielsen, unpublished data), a cDNA clone was fortuitously isolated which proved to be the human equivalent of the so called 'brain specific" regulator or 14.3.3 protein. The nucleotide and deduced protein sequence of this 1.8 kb clone is shown in Fig. 1. One long open reading frame is present extending from the first A U G at position 126 to a translational stop at 861. The encoded protein is predicted to contain 245 amino acids with a molecular mass of 27.8 kDa and a p l of 4.5. Possible polyadenylation signals were seen at 15!4 and 1658 but not at the end of the cDNA, presumably because it is not full length. A computer-aided search revealed a strong similarity with the recently described bovine 14 3.3 protein gene [5]. Within the coding region, the nucleotide sequences are 71% identical. The amino acid sequences (aligned in Fig. 2) show 72% identity and 84% similarity. There are two positions where the human sequence has a two amino acid deletion relative to the bovine sequence. The 5' and 3' noncoding regions of the mRNA's show no relatedness (not shown). The computer search also revealed a match with a portion of a cDNA reported to encode an alternative form of the EGF-receptor from Drosophila [6]. In this alternative form, a 5' 321 bp sequence differing from other EGF-receptor cDNA clones and containing an open reading frame was interpreted to result from alternative splicing. Since the match to the 14.3.3 protein sequence includes only the 321 bp 5' 'alternative exon', it seems likely that this
426 10 30 50 ~g~gggsctcgc~cgcggccgcggagac~gaaqctetegaggctcctcccgctgeg 90 90 110 g~cggcgctcgccctegctetcctcgecctccgccecggccccggccccggccecgcge 130 150 170 ecgccatggagaagactgagetgltccaglaggceaagctggccglgclggecglgcget M E K T E L I O K A K L & ~ O A E R T 190 210 230 aegacgaeatggccacctgcatgaaggca~gaccgagcagggcgcc~gct~ccaaeg D D M A T C M K J V T E O G A E L S N E 250 290 290 lggagcgcaacctgctctec~ggcctacalgaac~g~cggggg¢cg¢Ig~c¢gcct E R N L L S V A T ~ N V V G G n l S & W 310 330 350 ggagg~catctctagcatcgageaglag~ccgaeacetecgacaaqas~tgcsgetga R V I S S I E Q ~ T D T S D K I L O L I 370 390 410 ttaaggsctategggagaaa~gga~ccgagetgsgatccetctgcaeeaeg~gctgg ~ D y R ~ K V | S E L R S I C T T V L E 430 450 670 latt~tggataaatatttaltageeaatgcaactaatccagaga~lag~¢ttctatc L L D K I L I A N & T N P I S K V F Y L 490 510 530 tgaasatgaagg~gattacttecg~accttgctgaa~tgc~g~tgatcgaa
K H K G D Y F R ~ L A E V A C G D D R K
550 570 590 aaeaaacgatagataaZtcceaaggagettaceaagaggcatttgatataagcaagaaag
~ T I D N S O G A Y Q ~ A F D I S K K E
610 630 650 agatgeaaeccacacacecaatccgcctggggett~ctcttaa~tttct~attttaet M O P T H P I R L G L A L N F S V F Y ~ 670 690 710
atgagattcttaataacccagagcttgcctgcacgctggctaaaacggcttttgatgagg E I L N N P E L A C T L & K T A F D E A 730 950 770 cclttgctgalcttgatacactgaatgalgactcatac~aagaca~accctcatcltgc I A E L D T L N [ D S T ~ D S T L I M O 790 810 830 a~tgcttaglgacaacctaacaetttggacatcagaca~gca~agaaglat~qetg b L R D N L T L W T S D S A G E [ C D & 950 970 890 cggcagaaggggctgaaaactaaatccataeagg~catcett~ttc~teuagaaa
& E G A E N *
910 930 950 ectttttleaeatetccatteettattecacttggatttcctatageaaa~aacceatt 990 990 lOlO cat~atggaatcaa~ttata~cttttcacaetgcagctttgggaaaacttcat 1030 1050 1070 tccttglttt~tt~cttggccttcctg~gca~actgct~igaaaa~atta 1090 1110 1230 atagcttc~tttcatataaacatal~aactcccaaacacttat~lgacjcjactaalaat 1150 1170 1190 ~atetg~atttaa~aatctgaaeea~t~gc~a~gact~ttt~attact~ 1210 1230 1250 gaaaataagaaaat~a~aattacaatttaaag~gtatteeaeataacttcttaattt 1270 1290 1310 ctacattccctcccttactcttcgggg~ttcctttca~aagcaacttttecatgctct 1330 1350 1370 taatgtattccttttta~aggaatccggaa~attagattgaatggaaaagcacttgcc 1390 1410 2430 ltctct~ctaggg~cacaaattgaaatggctcct~atcacataccggag~ctt~g 1450 1470 1490 tatct~ggccaacaggga~ttcctt~ttcactctttatttgctgct~ttaa~tgcc 1510 1530 1550
aacctcccctcccaataaaaattcacttacacctcctgccttt~a~tctg~attcac 15~0
1590
1610
tttactat~gatagaa~agcat~tgctgecagaatacaagcattgctttt~caaat 163C 1650 1690 taaag~gcat~catttcttaatacactagaaaggggaaataaattaaa~acacaa~e 1690 1910 1730 caagtctaaaacttta~acttttcecatgcagattt~gcac~t~gagagg~cca 1750 1770 1790 ~t%~cta~gatt~tatttagaga~tggaccactatt~ttgctaatcattg~ 1810 1830 1850
ct~ag~ cecaaaaaagcctt~gaaaat~tatgccctat~aacagcaqagtaaeata aa
Fig. 1. Human eDNA sequence ~ r 14.3.3 protein. The nucleotide sequence was determined on both strands except ~ r the non-coding re~ons 1230-1440 and 1670-186~3 w ~ c h w e ~ de.trained on one strand o d y . The sequen~ includes the complete open reading ~ a m e be~nning ~ t h the first A U G at nueleotide 126 and ~rminating at position 863. The deduced amino ~ i d sequence is shown below the nucleotide sequence.
clone represents a hybrid cDNA generated during cloning of the eDNA's. A comparison of the primate, bovine, ovir.e and insect sequences is shown in Fig. 2. The human and Drosophila sequences are .nore similar to each other than either is to the bovine sequence. Both amino acid gaps in the human sequence are also present in the Drosophila sequence. The simplest interpretation of this observation is that the human and Drosophila sequences represent equivalent genes in these two species and the bovine sequence corresponds to a different member of the 14.3.3 family. The expression of 14.3.3 sequences was examined in several cel~ types from human and mouse origin. Expression was seen in both the Jurkat human T-cell lymphoma (Fig. 3, lane T) and the HeLa human carcinoma line (not shown). In both cases, two strong bands were seen of estimated sizes 1.9 kb and 2.1 kb and a weak band was seen at 1.7 kb. When total RNA from several mouse tissues was examined, two bands of 1.7 and 1.9 kb were seen. Based on densitometric scanning of appropriately exposed autoradiograms and correcting for differences in RNA loading using the hybridization signal for ribosomal protein S12 mRNA, 14.3.3 levels in !ung and kidney are estimated to be 5-fold and 2-fold less, respectively, relative to that found in brain. Liver and spleen mRNA levels are estimated to be at least 10-50-fold less than brain. Whether the various mRNA species which hybridize to 14.3.3 e D N A are derived from the same or related genes is not known. Hybridization of human 14.3.3 e D N A to human and mouse genomic D N A digested either with BamHl or EcoRI revealed bands ranging in length from 4.4 to 25 kb. The aggregate length of the EcoRl-derived fragments is roughly 60-70 kb, suggesting either one gene with large introns or a gene family. Protein sequence data [4,5] are compatible either with several genes or with a complicated alternative sphcing of one gene. A more detailed analysis of ~he genomic sequences is necessary to distinguish between these possibilities. The technical assistance of Birgit Groetschel is gratefully acknowledged.
427 1 Hum1433 Bov1433, Droegfrc Ovi1433a Ovi1433b Ovi1433c
Hum1433 Bov1433n Droegfrc
MEKTE) IQK
MGDREQ. L. R
AMLAEQAEPY r
DDMATCMKAV
R . . . . . . . . . . .
SA . . . . . . REA
.D.S..V ............... ,D.N..V . . . . . . . . . . . . . . . MDDRED. VYQ . . . . . . . . . . . 51 NWGGRRSAW
RVISSIEQKT
.S
TEQGAELSN£
LNEP. . . D . . . . . . . . . .T.V . . . . . . . . . . . . . . .
£. --DTSDKKLQ
LIKDYREKVZ
i00 SELRSICTTV
.... A...S ...........
MA.GNE...E KV.A . . . . I .
K..ETV.ND.
....
--EA.AR.Q.
K.
A.
, .S . . . . . . . . . . .
IARK
....
~.,
101 Hum,433 Bov1433, D r o e g f rc Ovi1433d Ovi1433e Ovi1433f
50 ERNLLSVAyZ
LELLDKYLIA NAT--NPESK VFYLKMKGDY .A .... F . . K . C N D F Q Y . . . . . . . . . . . . . .G . . . . . . . P K. S--. . . . . . . . . . . . . . . ....
.... ....
FRYLAEVA:G Y ....... S. Y... Y ..... F.T. F . . . . . AQ Y ....... A,
, .E.
,YE.
150 DDRKQTIDNS EKKNSWEA. ..
.. ..
Hum1433 Bov1433~
151 QGAYQEAFDI EA. , K . . . E .
SKKEMQPTHP IRLGLALNFS VFYYEILNNP . .EH . . . . . . . . . . . . . . . . . . . . . . Q.A
200 ELACTLAKTA .Q..L...Q.
Hum1433 Bov1433,
201 FDEAIAELDT LNEDSYKDST LIMQLLRDNL TLWTSDSAGE . .D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . QQD.
250 ECDAAEGAEN .AGEGN
Fig. 2. Amino acid ~quence alignment betv,'een 14.3.3-1ike proteins from various spe-ies. The am,no acid sequence deduce~l from the human 14.3.3 eDNA (Hum1433) is aligned with published 14.3.3-1ike proteins deri',ed eilher from t D N A sequences or directly from the protein. ]"he sequences are bovine (Bov1433 rl, Ref. 5), Drosophila (Droegfrc. Ref. 6). and sheep peptides a - f (Ovil1433a-f, Ref. 4). Identical amino acids are indicated by a "." and gaps b', "-'.
A
B M
Lu
Li
B
K
S
T
M
1
2
3
Ill 9.4
lJ
4,4 3.11
-"
1,7
1.1
1.1
°i s~2
~
O
! III,,, w ",'ll"
Fig. 3. Northern and genomic Southern hybndization analysis of 14.3.3 sequences (A) Approx. l0 ,ug of total cellular RNA ,,,,'as separated on 0.9~ agarose gels containing 6~ formaldehyde, transf~'rred to nylon membranes and h~bndt~ed under ,~r~.,~nt ~u~ndition~ 16 × S . ¢ ~ IOx Denhardt's. 0.1~ SDS, 6 4 ° C . 14 h) with the eDNA spanning the coding region from nucleotlde 1 to 1000 ,,vhich v,'as unir'.~rmb ,,~dio|abcUed. Af!Pr hybridization, the filters were washed (two 30 rain changes of 0.2 × SSC. 0.1 q~ SDS. ,12 o C) and expo.,~xl with Kodak XAR-film for 3 days. The same filter was subsequently hybridized with a radiolabelled e D N A for mouse ribosomal protein St2 to corr~:t for differences bet~,een lanes in the amount of RNA loaded. The samples included size markers (M). RNA from the Jurkat human T-cell line (T). and various mouse tissues: lung (Lu). liver (Li), brain (B), kidney (K) and spleen (S). (B) Approx. 15 txg of human genomic DNA (lanes 1 and 2) derived from peripheral blo~l lymphocytes or mouse liver genomic DNA (lane 3) ,,,,-ere digested with BamHl (lanes 1 and 3) or EcoRI (lane 2). electrophoresed in a 0.7~ agarose gel, transferred to a nylon membrane and hybridized under stringent conditions with the ~ame probe described above. Radiolabelled DNA sb. markers were ccelectrophoresed in the rome gel (lane M).
428
References 1 Moore. B.W. and Perez, VJ. (1967) in Physiological and Biochemical Aspects of Nervous Integration (Carlson. FD.. ed.J pp. 343359. Prentice-Hall, Englewood Cliffs. 2 Boston, P.F.. Jackson, P. and Thompson. R.J. (1982) J. Neurochem. 38, 1475-1482. 2 Ichimura. T.. Isobe. T., Okuyama, T.. Yamauchi. T. and Fujusawa, H. (1987) FEBS Left. 219. 79-82.
4 Toker. A., Ellis, C.A., Seller. L.A. and Aitken, A. (1990) Eur. J. Biochem. 191,421-429. 5 Ichimura. T. lsobe, T., Okuyama, T.. TakahashL N., Araki, K, Kuwano. R.. and Takahashi, Y. (1988) Proc. Natl. Acad. Sol. USA 85, 70.~4-7088. 6 Schejter, E.D.. SegaL D., Glazer, L. and Shilo, B.-Z. (1986) Cell 46, 1091-1101.
