Oct 18, 1984 - ... ACC ATO0Cc AGC CRA CIrrT7IC CATCa" TIM TAGRCA Human ..... translated to a specific activity of I x 108 c.p.m./ug (Rigby et al., 1977).
The EMBO Journal vol.3 no.13 pp.3257-3262, 1984
Isolation and characterization of a human p53 cDNA clone: expression of the human p53 gene
Greg Matlashewski, Peter Lamb, David Pim, Jim Peacock', Lionel Crawford and Sam Benchimoll2 Laboratory of Molecular Virology, Imperial Cancer Research Fund Laboratories, P.O. Box 123, Lincoln's Inn Fields, London WC2A 3PX, UK, 'Division of Biological Research, Ontario Cancer Institute, 500 Sherbourne Street, Toronto, Ontario M4X IK9, and 2Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada Communicated by L. Crawford
A cDNA clone for human p53 cellular tumor antigen has been isolated and characterized. This clone contains the complete 3'-untranslated region and most of the open reading frame for the protein. Nucleotide sequence analysis revealed that p53 mRNA contains an Alu repeat in the 3' -untranslated region. Hybridization selection experiments showed this clone was capable of selectively binding p53 mRNA. In vitro translation of SV80 mRNA resulted in the synthesis of two immunoreactive p53 polypeptide species. Northern blot analysis showed that human p53 mRNA was 2.8 kb in length and was present in cell lines containing high and low levels of p53 protein. There appears to be only a single p53 gene in human cells and Southern blot analysis demonstrated no major genomic rearrangements or amplifilcation of the p53 gene in the transformed cell lines examined. Key words: Alu repeats/cDNA clone/human/mRNA/p53
Introduction A variety of transformed cell lines and tumours of primate and murine origin have elevated levels of the cellular phosphoprotein p53 (for review, see Crawford, 1983). This protein is also found in normal primate and murine fibroblasts (Benchimol et al., 1982; Chen et al., 1983; Thomas et al., 1983). However, the levels of p53 protein in these nontransformed cells are 100-fold less than in their corresponding transformed cells (Benchimol et al., 1982; Thomas et al., 1983). Several human transformed cell lines such as HeLa and EJ contain little or no detectable p53 (Benchimol et al., 1982; Thomas et at., 1983). Thus, although p53 levels are generally elevated in transformed cells, this alone cannot be considered as a marker for transformation. In cells transformed by SV40 or adenovirus, p53 is specifically associated with the viral tumour antigens, SV40 large T and adenovirus Elb 58k protein (Lane and Crawford, 1979; Linzer and Levine, 1979; McCormick and Harlow, 1980; Sarnow et al., 1982). These physical associations increase the half life of p53 from - 15 min to >20 h, thus elevating the cellular concentration of p53 (Oren et al., 1981). The association of p53 with viral proteins necessary for transformation suggests a role for p53 in the transformation process. The precise function of p53 in either normal or transformed cells is not known, but the evidence suggests that p53 plays a role in regulating cell division, probably acting during the entry into the cell cycle (Gl) from the resting state (G) (Milner and Milner, 1981; Mercer et al., 1982, 1984; Milner, 1984; Reich and Levine, 1984). Therefore, altering -
(©D IRL Press Limited, Oxford, England.
the expression of such a protein may disrupt the normal regulation of cell division. Although a considerable amount is known about the expression of p53 mRNA and the genomic organization of p53 genes in mouse cells (Oren and Levine, 1983; Oren et al., 1983; Zakut-Houri et al., 1983; Pennica et al., 1984; Benchimol et al., 1984a), much less is known about human p53. This is primarily because mouse p53 cDNA cross-reacts poorly with human p53 nucleotide sequences. This paper reports the isolation and characterization of a cDNA clone specific for human p53. This cDNA clone contains the entire 3' -untranslated region and most of the open reading frame coding for the p53 protein. Using the cDNA clone as a probe, we have inititated a study of the expression of the human p53 gene. In addition, evidence obtained by in vitro translation and Northern blot analysis is presented, indicating that p53 mRNA in human SV80 cells exists as two distinct species. The human p53 cDNA clone has also been used to compare the genomic organization of the p53 gene in transformed and normal cells.
Results Approximately 106 colonies of a human cDNA library (Okayama and Berg, 1983) were grown on nitrocellulose and prepared for hybridization to nick-translated insert from plasmids HU7-1 and HU2-6. Plasmids HU7-1 and HU2-6 contain portions of the human p53 genomic coding region free of repetitive sequences (P. Lamb, in preparation). Eight reproducibly hybridizing colonies were picked, colony purified and grown to prepare large amounts of plasmid DNA. The plasmid vector used in the construction of the library is such that the cDNA insert can be cut out with XhoI. In all cases, digestion of the purified plasmid DNAs with XhoI yielded two fragments indicative of an internal XhoI site in the cDNA. The plasmid with the longest insert (designated p102) was chosen for further characterization by nucleotide sequencing. The large XhoI fragment from plasmid 102 is 1.9 kb long and contains 200 bp of vector DNA at the 5' end. The smaller XhoI fragment is 0.6 kb in length and contains - 100 bp of vector DNA at the 3' end. A restriction map and sequencing strategy of the cDNA insert of plasmid 102 is shown in Figure 1 together with the location of the p53-specific genomic DNA fragments used as hybridization probes to screen the library. As expected, complete nucleotide sequence homology was found between the indicated genomic fragments and the corresponding cDNA fragments (data not shown). Most important, extensive sequence homology with a mouse p53 cDNA clone (Benchimol et al., 1984a) was observed, confirming the identity and p53 specificity of plasmid 102 (Figure 2). A single open reading frame was located which extended for 881 nucleotides from the 5' end, sufficient to code for 293 amino acids (Figure 2). In comparison with mouse cDNA, which codes for 390 amino acids, we would calculate that -
3257
G. Matlashewski et al.
HU7-1
HU2-6 421
217 XhoI
Pvul Ncol
t
v v
736
1587
STOPSmal
STOP Sinal
v
--
881
131 177
4
XbaIPvuf Ncol IBam HI,Xhol v44 v
1022
1370
44 v v M 1442 1587' 1790 v
ul PvU v
Xhol Irv
1932
1736
---O-
,-
-
~
~
-44-
-~ ~
~
~
~
.
loobp Fig. 1. Restriction
enzyme map and sequencing strategy of human p53 cDNA 102. The solid thick line represents the cDNA and the solid thin line the pcD vector sequences. The upper open boxes show the position of the corresponding genomic fragments labelled HU2-6 and HU7-l used to screen the library.
300 bp at the 5' end of the open reading frame are not present in plasmid 102. The nucleotide homology between the human and mouse open reading frame is 8307 whereas the inferred amino acid sequence homology in this region is 84%. Many of the amino acid differences between mouse and human are conservative changes. Thus the overall conformation of most of the protein would be expected to be similar. The 3'-untranslated region of human p53 mRNA is different from the mouse with regards to both nucleotide sequence and length. The length of the 3'-untranslated region of human p53 mRNA, as determined from the cDNA sequence, is 1.2 kb. The corresponding region in mouse p53 mRNA is 400 bp in length (Zakut-Houri et al., 1983; Pennica et al., 1984). There are two small conserved regions within the 3' -untranslated region between nucleotides 1448 and 1570 (74%o homology with mouse) and from nucleotide 1937 through to the polyadenylation signal at nucleotide 2056 (67% homology with mouse). An Alu repeat sequence was located in the 3'-untranslated region of human p53 mRNA between the two regions of sequence homology to the mouse noted above. The Alu repeat is in the opposite transcriptional orientation to the p53 mRNA. No Alu sequences are present in mouse p53 mRNA (Zakut-Houri et al., 1983; Pennica et al., 1984; Benchimol et al., 1984a). To determine the size of human p53 mRNA and to compare its size with mouse p53 mRNA, poly(A) + RNA was fractionated on a glyoxal agarose gel, transferred to Biodyne A nylon membrane and hybridized with the large XhoI fragment insert of plasmid 102 and a mouse p53 cDNA clone (Benchimol et al., 1984a). The results shown in Figure 3, lane 1 indicate that human p53 mRNA is 2.8 kb in length. The mobility of mouse p53 mRNA, shown on the same gel (Figure 3, lane 2), corresponds to a length of 2.0 kb in agreement with previous reports (Oren et al., 1983; Reich et al., 1983; Benchimol et al., 1984b). Thus, human p53 mRNA is 800 bp longer than mouse p53 mRNA primarily as a result of a much larger 3'-untranslated region. When the washing stringency of a similar Northern blot was reduced, a band of 500 bases became apparent (Figure 3, lane 3). Since the 3' end of the large XhoI fragment contains 100 nucleotides of the Alu sequence (see Figure 1 and 2), this band probably represents Alu RNA polymerase III transcripts which are about the same size in human cells and may contain A-rich regions capable of binding to oligo(dT)cellulose (Jelinek and Schmid, 1982). We also find that p53 mRNA from human SV80 cells consists of two species that can be resolved by agarose gel electrophoresis (Figure 3, lanes 1 and 3). When the p53 protein -
-
-
-
3258
from SV80 cells is fractionated on an SDS-polyacrylamide gel following metabolic labelling and immunoprecipitation, two polypeptides are also observed (Crawford et al., 1981). The nature of the p53 protein doublet observed in many human cells is not understood. It is tempting to speculate that each polypeptide is encoded by one of the mRNA species we observe. To pursue this idea further, plasmid 102 was immobilized on a nitrocellulose filter and used to hybrid select p53 mRNA. Purified p53 plasmid 102 and a control plasmid containing no insert (pAT153) were bound to nitrocellulose filters, and used to select mRNA from total human SV80 poly(A) + mRNA. The filters were washed and the RNA eluted by boiling. Eluted RNA was translated in vitro in a rabbit reticulocyte lysate containing [35S]methionine, followed by immunoprecipitation with anti-p53 serum. The results presented in Figure 4 (lane 3) indicate that plasmid 102 specifically selected mRNA, when translated in vitro followed by immunoprecipitation with immune serum, gave a doublet of -53 000 daltons. We conclude that plasmid 102 is capable of hybrid selecting p53 mRNA that encodes both polypeptide species of the p53 doublet. Because post-translational modifications are not known to occur normally in rabbit reticulocyte lysates, we favour the hypothesis that each polypeptide is encoded separately by one of the two mRNA species which are both capable of specifically hybridizing to plasmid 102. We have re-examined the seven other plasmid clones by restriction enzyme mapping and partial sequence analysis and find no gross differences that might indicate the presence of a second p53 mRNA species in our collection of cDNA clones. Clearly, the nature and significance of the two p53 polypeptides and two p53 RNA species remain to be resolved. We have used insert derived from plasmid 102 as a nucleic acid hybridization probe to detect p53 mRNA in a variety of human transformed cells. Messenger RNA was isolated from SV80, C331 and Raji cells which all contain high levels of p53 protein and from HeLa cells which contain no detectable p53 (Benchimol et al., 1982; Thomas et al., 1983). Northern blot analysis (Figure 5, panel A) shows that p53 mRNA was present in all the cell lines. The SV80 p53 mRNA (lane 1) appears as a single band on this gel which was run for a shorter time than that shown in Figure 3. HeLa cells, which contain no detectable p53 protein, do contain p53 mRNA, (Figure 5, lane 4) although it may be present at a lower level than in the other cell lines. In vitro translation of HeLa and SV80 mRNA followed by immunoprecipitation with anti-p53 serum was performed to determine whether HeLa mRNA could be translated into p53 protein. Anti-p53 serum specifically
Human p53 cDNA 7
22
C AAA ACC TAC CAG Lys 'Mr Tyr Gin T
37
GOC AUG TAC OCaT I'I Cor Ci0C 1TM TG Ciy 5er Tyr dly Phe Arq Leu Gly Phe Eeu A
T
AC His
C
Asn 112
C
127
RAG TC CARlnCIT AETOBET PhFP50 Css Gin Lao Layss CRA C G G
142
ACC lIt CCT OMG CAG C7C TOO Thr Cys Pro Vai Gin Lecu Trp
Ala
G
52
MTr OCG
CAT His G
5er
RC
Thsr
Gin
L*U
O
67629
C
CCrC
GCC
Oit
IThr
C
So-rT
Pro Ala C Pro
AlTG CC Alt TAG MET Ala rie Tyr
AAG CAG
167
C
337
352
362
367
427
442
CCRA
G
T
457
472
Cit AGC Alt Alt AGA TG CAR aAG ltC ACT Lao Thr tIe tic Ilhr Leu CIL Asp 5cr Ser T
407
307
Glu
Human MDU5
TOT RAGC AGT Cs's RAn nor
Human
GAG
Tyr Pro
502
GAG
T
CHouse
517 GMC COT CT TOT 0CC Human
Glu Val Arg Val Cys Ala
G
G
202
OCR AAT 1TW COT OM Gis' Asn lAu Arg Vol1 TA CCC 412
GOCT RAT CTA CT OGR COC RAGC AOC Ti? Lo-uT Lao GiY Arg Asn nor Phe C
House
OHouse
397
Gly Asn
fm
Lys
T
Lys
TAT 1TI GAT GAC AGA AAC ACT 7ST CGR CAT ACT CIT GIS Cit CCC TAT GAG CCG CT GAG CIT (2CC TE GAG ITT ACC ACt Alt CAC TAG AAG TAG AT Tyr lau Asp Asp Arg RAn TAr PMm Arg Ilis Ser Val Vai Vai Pro Tyr Glu Pro Pro Glu Val Gly 5cr Asp Cys Thr 'Mr tie His Ty'r Ann Tyr MET C A G CG C C C A T A C cc G A G T Gin Glu Ala Glu Tyr LYS tie TCC TMC ATG 2CC (OC ATG RAGC COG AUG CCC ATC nor Cys MET Gly Gly MET Asn Arg Arg Pro tie
LoU RAn
CAC Human TMR Gin CRG HisB Ly5s AGin Sor
His
217 232 247 262 277 292 Alt ACG GRG MT5 GMC AUG CCC TCC CCC CAC CAT GAG CGC TC CRIt GAT AGCC GAT OCT CI GCCC CCT CCT CRG CAT CiT ATC CGA CCC GRA O. "tr Giu Val Vai Arg Arg Cys Pro Hlis His Giu Arg Cys Ser Asp 5cr Asp :;ly Lku Ala Pro Pro Gin Hlia Lou 1ie Arg Val Glu t G C A C C GT T C G Gly Gly 322
9C7C CtRGRA
A
Tyr
Thir Cys'sMTr
157 172 ACA CCC AGC CCC CC CCC 1St CCC Pro Pro Pro Gly Arg Val Arg Ala T A GT G G T Ala Ala 5cr
5cr
CAOCCG Se~ ~ ~ ~ ~5r
T
C A T 7C2C Ala Lys Ser Val T TG
mmT
GAT 7CC
Ci
Val Asp
Gly
T
HoSuse
Asp 532
547
562
ITM OCT MGC ARC GAC CCC CCC ARGA C GAR GAG Cy's Pro Gls' Arg Asp Rirg Rigr Tr Glu Glu Glu C
C
T
A
A
592
577
RAAT Cit CCC RAG RAM Asn Leu Arig Ly's Lys T
OM GAG CCT
Giu Pro GAR lT T
A
POhe
CAG CR0 GAG CIT
Gly
His His CT T Glu Val Leu Cys Pro
Glu Leu
617
CCC CCA Pro Pro
A
(GM RCC Giy Sor
652 667 602 697 712 RC ltCc TOT CCC CR0 CCA RAG RAG RAAA CCA CiTG GAT OGR GRA TAT 'tW RCC CIT CA0 AlTC CCT O00 CCT GAG CCC 1TI GAG nor nor nor Pro Gin Pro Ly's Ly's Ly's Pro Leu Asp 017 Glu Tyr Phc 'Mr Lecu Gin Ile Arig 017 Arig Glu Arig Phe Glu
C Ala
C R Pro Gin
A
T
CA .ys
G
AR
C
ACT
RAG
'Mr Lys GA Ala
622 CGR CCA Ala A G
Arig
CCC CCt RAG RAG Asn Lau Pro Asn TO C7
637
ACC Human Thr R Duse
Thr Cys 727
ATO, T't CGR GAG Cit RAT
NETr Phe Arig Glu Leu
Asn
C
742 GAG GCC Glu Ala
ITOG
Humian
Leu
RA Huse
Lyss
757 772 707 002 017 832 647 GAR Cit RAG GAT 0CC CR0 OCT 0CC RAG GAG CCA OM2 000 RUC AGO GCT CRC TCC RAGC CAC Cit RAG l RAAR RAG OCT CR0 TO AGC ltC CCC CAT ARA AAA Cit AlT Muman Lou 1.73 Asp Ala Gin Ala GilY iys Glu Pro Gily 017 5cr Arig Ala His Ser 5cr His Leu Lys Ser Ly's Lys Gly Gin Ser Thr 5cr Rig His Lye' Ly's Leu NET GTRA T AGA G TT RAGA T C A C T AC Mo,use Thr Glu Ser His Asp TyrTh
Gb
662 677 092 907 St7 AAG AGAGCAR 0G20 CCT CRC 7ItA CRC TOR. CATOT CCA CIT CIT CIT CCC CRC ItGA CR0CG I"' TAr Glu 01y Pro Asp Ser Asp STOP-------Ko-
G
937
922 CCC ACC CCC
PMm
Alt TOCT
--
CCC
952
ItC CCT GCC ATT ITO OCTr 7iT 000IM
Human
6use
TO Lys Val A
Vol
902 967 TCC RAG C, TM TTK; CAATp
997 1012 TOC Cit AGA AUC ACC CR0 GAG I't CAT
1027
1042
7ITO- Ci' TT CCC 000 OCT CCA CitG RAGC RAG
1057 O GCC IXX ACT OCT 01? TG
ITOn TOG Human ---~~~~~~~~~~~~~Hus e
1072
1067
1102
1117
OCRA OCRAOCR TOGOR OTR OCR CRT AGC AUC TTA GAT l'iT RAG CiT Ii
1132
1147
1162
1177
ACT a CitcAGCAT CTT ITOG ORG ATO TRA GMA Alt- lt I'Mt CR0 ITA RUG C' ACT ETA CRA HUMan
HoAuse .A;CACA 1tT
1192 1207 1222 TAG OTM OGM (CC CRC 1tW AGC OTA CTA AGC AGO GRA OCT CitC
1297
1TT OCR CCT AGC ,tA
1312 CAG ACT OCRICTOR WA
1327
1237
1252
1267
CAG TOT TOR AlT It TOT RAGC 'tW RAG
1342
C ISAT- ATO RRA TRA TOCT RCA TOT 0CC C
1202
CC CRT Alt TOT GRA ATO CCC OCR HumNan
1357 1372 1387 GRA AGC ACC TTT TALT TAG AT OG0 TOT RCA TORA CCC CCT TORA OCT OCT Humban ----
ma~ ~ ~ ~ ~ ~ Nuse,
1402
1417 1432 TO TMl Cit lt CTO ICC CMt OCT 0CC TCC OTM CiT TOT RCA
1507
1522
1447 1462 1477 1492 rT 0CC CR0 CTG CGT AGO TAG AGG CR0 ICC 7tA ACT Cit lT TOO CCC AOC CMA ACC CC TO- TOR Huiaon - -- ~~ 0--T 0 IT A CA ClHouse ~~
1537
~~T
1552
1567
1502
1597
CAA OCT CiT OCT CAR CCT TAG Alt CTR RRA OCR RAT GMt AGC CCR ItC CRC RCC CCC GAG CAT 1TT Alt TOT TOT ATA CAT CAT CCC CAT CCk CCA AGRL CTT CiT TTA Human - OTG 00C -C T C C T C TA CT use --Mo--1612 OCT CR0GI OMCCA
1627
iWTOT1Tr1tm1TT1T
1642 1ST 1ST1T T
1657 1702 1607 1717 --1672 1St1ST 1St1ST GAG ACT000 TOT CiT TOT TKX CCC AUG CCC GAG TOOG ACT0CCCC AltTOG CiT ACT
1732 1762 1747 cx CCC1TCCClTC CCGOCT CCR GM amt CCC CCT CR0 CCT CCC
1037
TMOGCOT Citc AC CC
1792
1052 1067 1097 1002 CCCAGOCCC O tM t KR AG TCC TOO OCT CR0 GCG AlTC CRC CGTOT CR0 CCT CCC ARC
1942 1957 1972 ACC TOO RAG OCT CRA CAT CIT ITA CAT TOCT OCRA 0- CRC Alt lTO C T TItCC C A C TO; 2047
1777
1007
GAkG TAG CCC OCR CCA CR0 CiT CRT0CC ACC ATO0Cc AGC CRA CIrrT7IC
1907 2002 ATTr I-T C- A CCC CAC CCT- lTCCC CIC A C C A
T
1912
1022
CATCa" TIM TAGRCA Human
.- .l~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Huse
1927
CTO CC OCR TTA CRA t' TOR GCCC ACC
It ltC CT-0
C
Htumain
G
Human
2017 2032 =S TAT ATcC RET IT- TAT Alt Human I T "TTA CT A Housc
2062
CALT CitC ITA TTT TAG RATRAAA ACTr TI
CC CCA
Fig. 2. Nucleotide sequence of human p53 cDNA. The nucleotide sequence of the p53 cDNA is given in parallel with the amino acid sequence predicted from the open reading frame. Nucleotide and amino acid differences between the human p53 cDNA and the comparable mouse p53 cDNA (shown below the human p53 cDNA sequence) are noted. The numbers used indicate the positions in the human p53 cDNA. Hyphens denote the sequences which are not in mouse or human p53 cDNA. The boxed region contains the Alu repeat sequences. The AATAAA polyadenylation signal sequence is underlined. 3259
G. Matlashewski et al.
M Kb
1
2
3
1
2
3
-Niaft,
-4
-4
p
53
0
Fig. 4. Hybridization selection with human p53 cDNA 102. Plasmid DNA was bound to nitrocellulose filters and hybridized to 100 Ag of poly(A)+ mRNA from SV80 cells. Selected RNA was translated in vitro in a rabbit reticulocyte lysate in the presence of [35S]methionine followed by immunoprecipitation and analysis by SDS-polyacrylamide gel electrophoresis. Lane 1, control translation containing no mRNA and immunoprecipitated with anti-p53 serum. Lanes 2 and 3, polypeptides immunoprecipitated with anti-p53 serum from translation of mRNA selected by a control plasmid pAT153 without insert or p53 plasmid 102, respectively. The arrows denote the position of p53 polypeptides.
Fig. 3. Northern blot analysis of human and mouse p53-specific mRNAs. The filter was hybridized simultaneously with radiolabelled large XhoIXhoI fragment (Figure 1) and a plasmid containing mouse-specific p53 cDNA (Benchimol et al., 1984a). Lane 1 and 2, 10 yg of human SV80 and 10 ytg of mouse SVA3IE7 poly(A)+ mRNA, respectively. Lane 3, 10 Ag of SV80 poly(A)+ mRNA hybridized with radiolabelled large XhoI-XhoI fragment and washed using reduced stringency conditions. Lane M, HindlII and EcoRI digested lambda DNA fragments were run on the same gel and stained with ethidium bromide to act as mol. wt. markers.
precipitated p53 from both SV80 and HeLa in vitro translation products (Figure 5, panel B, lanes 7 and 9) whereas the non-immune serum did not (lanes 6 and 8). As an internal control, monoclonal antibody PAb419 was used to immunoprecipitate SV40 large-T antigen from SV80 translation products. This confirms that HeLa p53 mRNA is translationally functional. Southern blot analysis was performed on genomic DNA isolated from normal foetal liver tissue, SV80, Raji, C331 and HeLa cells to establish if p53 genes are rearranged or amplified in these human transformed cells. The DNA samples were cut with EcoRI or PvuII, the fragments separated by electrophoresis and then blotted onto nitrocellulose filters. Blots were hybridized with the NcoI/XhoI fragment located between nucleotides 177 and 1736 of plasmid 102 (Figure 1). This fragment was chosen because it contained no GC polylinker or SV40 DNA sequences although it does contain some Alu sequence. Under high stringency 3260
hybridization and washing conditions only specific bands are apparent (Figure 6, panel A). With the enzymes used, all the cell lines show similar restriction fragment patterns and band intensities. Strongly hybridizing and reproducible bands were obtained at 18 kb for the EcoRI-digested samples and 4.0 kb and 1.6 kb for the PvuII-digested samples. Faintly hybridizing bands at 2.0 kb are not reproducibly observed and are most likely not p53 specific. Thus, there is no detectable gross alteration in the genomic organization of the p53 gene in these transformed cells. The simplicity of the Southern blot pattern indicates that the human p53 gene does not belong to a multigene family. In addition, the appearance of only one hybridizing 18-kb EcoRI fragment suggests that there is only a single human p53 locus. This is in contrast to the mouse where a processed pseudogene is contained within an additional 3.3-kb EcoRI fragment (Zakut-Houri et al., 1983; Benchimol et al., 1984a). The appearance of two PvuII genomic fragments is consistent with this idea since the probe used in this experiment contains a PvuII site. The same blot was re-hybridized using the small XhoI fragment under the same conditions and this resulted in a smearing pattern with few discernable bands (Figure 6, panel B). Since the small XhoI fragment contains 200 bp of Alu sequence (Figure 2), this smear is presumably due to the Alu repetitive sequences within genomic DNA. Discussion The human p53 cDNA clone we have isolated covers most of the open reading frame for p53 and contains the complete 3 '-untranslated region. The amino acid sequence of human p53 predicted from the cDNA sequence reveals strong homology (84%) with the mouse p53 primary amino acid sequence. The human p53 cDNA contains an Alu repeat sequence which is not present in mouse p53 cDNA (ZakutHouri et al., 1983; Pennica et al., 1984; Benchimol et al., 1984a). Alu sequences have previously been reported in human mRNA but the reason, if any, for their presence is not known (Jelinek and Schmid, 1982). The Alu sequence interrupts a region in human p53 mRNA which shows homology to mouse p53 mRNA and the structure of human p53 mRNA is considerably different from that of a mouse p53 mRNA,
Human p53 cDNA
A. 1
B. 5 6 7 8 9
3 4
2
Large T-
p53 mRNA--b
_
=p53 protein
jw
Fig. 5. Panel A. Northern blot analysis of p53 mRNA from transformed cell lines. The lanes contained 10 Ag of poly(A)+ mRNA from SV80 cells (lane 1), Raji cells (lane 2), C331 cells (lane 3) and HeLa cells (lane 4). Panel B. Detection of p53 protein synthesised in vitro. SV80 mRNA was translated as described for Figure 4 and immunoprecipitated with: anti-SV40 large-T monoclonal antibody PAb4l9 (lane 5), non-immune serum (lane 6), and anti-p53 serum (lane 7). HeLa mRNA was treated identically and immunoprecipitated with: non-immune serum (lane 8) and anti-p53 serum (lane 9).
whereas the actual coding sequence of the p53 protein is very similar. In vivo labelling of p53 protein in SV80 cells with [35S]methionine followed by immunoprecipitation reveals the presence of two species of p53 with different apparent mol. wts. (Crawford et al., 1981; Thomas et al., 1983). This may be due to post-translational modifications on the same primary amino acid sequence or to their having different but similar primary amino acid sequences. In this study, immunoprecipitates of p53 synthesised in vitro from total SV80 mRNA or hybrid-selected mRNA using plasmid 102 also show two mol. wt. species of p53. This, as well as the Northern blot data, presents evidence that p53 mRNA also consists of two species. This provides a possible explanation for the existence of two polypeptide species, each being derived from one of the mRNA species. Northern blot analysis on mRNA isolated from the various transformed cell lines we used shows that they all contain p53 mRNA. The p53 protein levels in these cell lines vary by > 100-fold (Benchimol et al., 1982; Thomas et al., 1983), yet this is not reflected at the mRNA level so the amount of p53 protein in these cells must be regulated post-transcriptionally. Similar observations have been made in murine cell lines (Oren et al., 1981; Benchimol et al., 1984b). It is of particular interest that, although HeLa cells contain no detectable p53, they do contain p53 mRNA which is translationally active. Southern blot analysis of the structural organization of the human p53 gene in normal foetal liver DNA and DNA from various transformed cell lines revealed no rearrangements or amplifications. Previous Southern blot analysis of p53specific sequences in mouse cellular DNA using EcoRI digestions showed two fragments, one of 18 kb and another of -3.3 kb (Oren et al., 1983; Benchimol et al., 1984a). Only a single 18-kb EcoRI fragment is apparent on Southern blots of human genomic DNA, suggesting the presence of a single locus and our data with PvuII is consistent with this idea. In conclusion, the structure of human p53 mRNA is different from the structure of mouse p53 mRNA in the
Xho,I
Ncol
Xhol
Xhol v
I
5
177
A M
1 2 3 4 -:
3
1736
B 6 7 8 9 10
1-
2
3 4 5 6
7
8 9 10
Kb
_,
23 9 4
w
6
4
4 3 Vp
:A i q
2 3 2 V0
0 5
Fig. 6. Panel A. Southern blot analysis of genomic DNA from normal foetal liver and transformed human cell lines. DNA (10 Lg) from normal foetal liver (lanes 1 and 6), SV80 cells (lanes 2 and 7), Raji cells (lanes 3 and 8), C331 cells (lanes 4 and 9) and HeLa cells (lanes 5 and 10) were digested with either EcoRI, lanes 1-5, or PvuIl, lanes 6-10. An NcoIXhoI fragment located between nucleotides 177 and 1736 as shown in Figure I was nick translated (108 c.p.m./4g) and used as the hybridization probe. Lane M, HindII-digested lambda DNA mol. wt. markers. Panel B. Same sample arrangement as in Panel A using as the hybridization probe (108 c.p.m./4g) an XhoI-XhoI fragment located from nucleotide 1736 through to the XhoI site within the vector as shown. 3261
G. Matashewsld et al.
3' -untranslated region, whereas the human and mouse open reading frames code for very similar proteins. Our data suggest the existence of two p53 mRNA species in SV80 cells. Future studies, including the isolation and characterization of more cDNA clones from SV80 cells are necessary to elucidate the nature and significance of these two mRNA species. The expression of p53 mRNA and the genomic organization of the p53 gene do not appear to be altered in cells that contain high and low levels of p53 protein. This indicates that p53 is regulated in these cells at the post-transcriptional level.
Materials and methods Cells and antibodies The cell lines used were as follows: SV80 cells, a line of SV40 transformed human fibroblasts (Todaro et al., 1966). C331 and HeLa, derived from cervical carcinomas (Auersperg, 1964) and Raji cells derived from a Burkitt lymphoma (Epstein et al., 1966). The pool of human anti-p53 serum was a mixture of sera, collected from five breast cancer patients. The detection and assay of anti-p53 antibodies in these sera have already been described (Benchimol et al., 1982; Crawford et al., 1982). PAb419 monoclonal antibody was L19 of Harlow et al. (1981). Isolation of p53 gene fragment Two genomic segments of human DNA containing sequences from the p53 gene were isolated by cross-hybridization to a mouse p53 cDNA (Benchimol et al., 1984a) and cloned as plasmids pHU2-6 and pHU7-1 (P. Lamb, in preparation). Cloned genomic inserts were gel purified and labelled by nick translation (Rigby et al., 1977). Isolation of human p53 cDNA clone 102 The cDNA library was kindly provided by Hiroto Okayama and was constructed with mRNA from SV40 transformed human fibroblast cells (Okayama and Berg, 1983). About 1 x 106 transformed bacteria were plated out onto nitrocellulose filters. The colonies were replicated, lysed and fixed as described by Hanahan and Meselson (1980). Filters were hybridized with human p53-specific genomic DNA fragments (HU2-6 and HU7-1) using standard procedures (Hanahan and Meselson, 1980). Analysis of nucleic acids by Northern and Southern blotting Total cellular RNA was extracted in the presence of guanidinium isothiocyanate (Chirgwin et al., 1979). RNA samples were chromatographed twice over oligo(dT)-cellulose. Genomic DNA was prepared according to standard procedures (Blin and Stafford, 1976). Hybridization probes were nick translated to a specific activity of I x 108 c.p.m./ug (Rigby et al., 1977). For Northern blots, 10 Ag of polyadenylated RNA was denatured in 1.0 M glyoxal and fractionated on a 1% agarose gel in 10 mM sodium phosphate buffer pH 7.0 then blotted on to Biodyne A nylon membranes. Hybridization conditions were as described by Thomas (1980). For Southem blots, genomic DNA was digested to completion with the appropriate restriction enzyme and fractionated on a 0.8% agarose gel in Trisacetate-EDTA buffer. DNA was blotted onto nitrocellulose filters (Southern, 1975) and hybridized using standard conditions (Maniatis et al., 1982). Hybridization selection and in vitro translation 10 yg of plasmid DNA were linearized by digestion with SaIl, heat denatured and baked onto I cm squares of nitrocellulose filter. Filters were hybridized with 100 Ag of total poly(A) + mRNA, washed and eluted as previously described (Ricciardi et al., 1979). Eluted mRNA was translated in vitro in a rabbit reticulocyte lysate in the presence of [35S]methionine and the products immunoprecipitated with polyclonal anti-p53 serum (Crawford et al., 1982), then subjected to electrophoresis. DNA sequence analysis Nucleotide sequence determinations were carried out by the dideoxy chain termination method (Sanger et al., 1977) in the presence of dideoxynucleoside triphosphates and [35S]thio dATP (Biggin et al., 1983). The resulting DNA products were run on buffer gradient gels (Biggin et al., 1983).
Acknowledgements We are most indebted to Hiroto Okayama for the cDNA library from which
p102 was isolated. We would like to thank Kit Osborn for her help with tissue culture, Peter Goldsbrough for expert technical assistance and Georgina Briody for her help in the preparation of the manuscript. G.M. is a NATO
postdoctoral fellow and S.B. a scholar of the Medical Research Council of Canada. Part of this work was supported by grants from the National Cancer Institute of Canada and the Connaugh Fund to S.B.
3262
References Auersperg,J. (1964) J. Natl. Cancer Inst., 32, 135-163. Benchimol,S., Pim,D. and Crawford,L.V. (1982) EMBO J., 1, 1055-1062. Benchimol,S., Jenkins,J.R., Crawford,L.V., Leppard,K., Lamb,P., Williamson,N.M., Pim,D. and Harlow,E. (1984a) in Vande Woude,G.F., Levine,A.J., Topp,W.C. and Watson,J.D. (eds.), Cancer Cells, Vol.2, Oncogenes and Viral Genes, Cold Spring Harbor Laboratory Press, NY, pp.383-391. Benchimol,S., Matlashewski,G. and Crawford,L.V. (1984b) Biochem. Soc. Trans., 12, 706-709. Biggin,M.D., Gibson,T.J. and Hong,G.F. (1983) Proc. Natl. Acad. Sci. USA, 80,3963-3965. Blin,N. and Stafford,D.W. (1976) Nucleic Acids Res., 3, 2303-2308. Chen,S., Blanck,G. and Pollack,R.E. (1983) Proc. Natl. Acad. Sci. USA, 80, 5670-5674. Chirgwin,J.M., Przybyla,A.E., MacDonald,R.J. and Rutter,W.J. (1979) Biochemistry (Wash.), 18, 5294-5299. Crawford,L.V. (1983) Int. Rev. Exp. Pathol., 25, 1-50. Crawford,L.V., Pim,D.C., Gurney,E.G., Goodfellow,P. and TaylorPapadimitriou,J. (1981) Proc. NatI. Acad. Sci. USA, 78, 41-45. Crawford,L.V., Pim,D.C. and Bulbrook,R.D. (1982) Int. J. Cancer, 30, 403-408. Epstein,M.A., Achong,Y.M., Barr,Y., Zajac,B., Henle,G. and Henle,W. (1966) J. NatI. Cancer Inst., 37, 547-559. Hanahan,D. and Meselson,M. (1980) Gene, 10, 63-67. Harlow,E., Crawford,L.V., Pim,D.C. and Williamson,N.N. (1981) J. Virol., 39, 861-869. Jelinek,W.R. and Schmid,C.W. (1982) Annu. Rev. Biochem., 51, 813-844. Lane,D.P. and Crawford,L.V. (1979) Nature, 278, 261-263. Linzer,D.I.H. and Levine,A.J. (1979) Cell, 17, 43-52. Maniatis,T., Fritsch,E.F. and Sambrook,J. (1982) Molecular Cloning: A Laboratory Manual, published by Cold Spring Harbor Laboratory Press, NY. McCormick,F. and Harlow,E. (1980) J. Virol., 34, 213-224. Mercer,W.E., Nelson,D., DeLeo,A.B., Old,L.J. and Baserga,R. (1982) Proc. Natl. Acad. Sci. USA, 79, 6309-6312. Mercer,W.E., Avignolo,C. and Baserga,R. (1984) Mol. Cell. Biol., 4, 276-281. Milner,J. (1984) Nature, 310, 143-145. Milner,J. and Milner,S. (1981) Virology, 122, 785-788. Okayama,H. and Berg,P. (1983) Mol. Cell. Biol., 3, 280-289. Oren,M. and Levine,A.J. (1983) Proc. Natl. Acad. Sci. USA, 80, 56-59. Oren,M., Maltzman,W. and Levine,A.J. (1981) Mol. Cell. Biol., 1, 101-110. Oren,M., Bienz,B., Givol,D., Rechavi,G. and Zakut,R. (1983) EMBO J., 2, 1633-1639. Pennica,D., Goeddel,D.V., Hayflick,J.S., Reich,N.C., Anderson,C.W. and Levine,A.J. (1984) Virology, 134, 477-482. Reich,N.C., Oren,M. and Levine,A.J. (1983) Mol. Cell. Biol., 3, 2143-2150. Reich,N.C. and Levine,A.J. (1984) Nature, 308, 199-201. Ricciardi,R.P., Miller,J.S. and Roberts,B.E. (1979) Proc. Natl Acad. Sci. USA, 76, 4927-4931. Rigby,P.W.J., Dieckmann,M., Rhodes,C. and Berg,P. (1977) J. Mol. Biol., 113, 237-251. Sanger,F., Nicklen,S. and Coulson,A.R. (1977) Proc. NatI. Acad. Sci. USA, 74, 5463-5476. Sarnow,P., Ho,Y.S., Williams,J. and Levine,A.J. (1982) Cell, 28, 387-394. Southern,E.M. (1975) J. Mol. Biol., 98, 503-517. Thomas,P.S. (1980) Proc. Natl. Acad. Sci. USA, 77, 5201-5205. Thomas,R., Kaplan,L., Reich,N., Lane,D.P. and Levine,A.J. (1983) Virology, 131, 502-517. Todaro,G., Green,H. and Swift,M.R. (1966) Science (Wash.), 153, 1252-1254. Zakut-Houri,R., Oren,M., Bienz,B., Lavie,V., Hazum,S. and Givol,D. (1983) Nature, 306, 594-597. Received on 6 September 1984; revised on 18 October 1984
