Differential in situ hybridization for determination of mutational specific expression of the p53 gene in human hepatoma cell lines

Pathology

ISSN: 0031-3025 (Print) 1465-3931 (Online) Journal homepage: https://www.tandfonline.com/loi/ipat20

Differential in situ hybridization for determination of mutational specific expression of the p53 gene in human hepatoma cell lines Yue Lin, Lin-Kung Ong & Soh-Ha Chan To cite this article: Yue Lin, Lin-Kung Ong & Soh-Ha Chan (1995) Differential in situ hybridization for determination of mutational specific expression of the p53 gene in human hepatoma cell lines, Pathology, 27:2, 191-196 To link to this article: https://doi.org/10.1080/00313029500169862

Published online: 06 Jul 2009.

Submit your article to this journal

Article views: 5

View related articles

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=ipat20

Pathology (1995), 27, pp. 191-196

DIFFERENTIAL IN SITU HYBRIDIZATION FOR DETERMINATION OF MUTATIONAL SPECIFIC EXPRESSION OF THE p53 GENE IN HUMAN HEPATOMA CELL LINES Y u ~ LIN,* LIN-KUNG ONG* AND SOH-HA CHAN*'~

*Department of Microbiology, Faculty of Medicine and tWHO Immunology Centre, National University of Singapore

Summary The codon 249 mutation specific expression of the p53 ge0e was determined in 7 human hepatocellular carcinoma (HCC) cell lines. Two 20-base oligomers complementary to bases 872-891 of human p53 cDNA with a single nucleotide difference in the third position of codon 249 were end-labelled with biotin-conjugated dATP using terminal deoxynucleotidyttransferase (TdT). The hybridized oligomer was visually detected in situ using streptavidin-alkaline phosphatase (AP) conjugate and AP substrate. Expression of the codon 249 mutant p53 was steady in PLCtPRFt5 and Mahlavu cells (derived from African patients), while Huh4, Huh6, Huh7 and HCC-M cells (derived from Japanese patients) expressed only the codon 249 wild-type p53. The transcripts of the p53 gene were undetectable in Hep3B cells (derived from an American patient). HyN'idizations of the codon 249 specific otigomers were specific to the p53 transcripts, since the cells that expressed p53 gene homogeneously were stained in the cytoplasm only by differential hybridization with a codon 249 specific oligomer; moreover, hybridization with a labelled oligomer non-complementary to the p53 cDNA showed nuclear stainings. Thus, detection of the codon 249 mutant p53 mRNA by differential in situ hybridization is a specific method for studying the mutation-specific expression of the p53 gene in liver cancers at the cellular level, while simultaneously visualizing the cell morphology. The results also support the notion that the p53 gene codon 249 mutation may have etiological implications involving HCC from various geographic areas.

Keywords."In situhybridization,HCC,p53 transcripts,codon249mutation, hepatocarcinogenesis. Accepted 13 December 1994 INTRODUCTION

The human p53 tumor-suppressor gene, located on chromosome 17p13.1,1 encodes a 393-amino acid nuclear phosphoprotein which is involved in the regulation of cell proliferation. Loss of normal p53 function has been associated with" cell transformation in vitro and the development of neoplasms in vivo. 2 This loss may occur either by mutation of one allele of the p53 gene followed by deletion of the remaining wild-type allele or through inactivation of the wild-type protein by oligomerization with the more stable mutant protein 3'4 or by binding to other proteins. 5"6 Mutations of the p53 gene have been observed with a high prevalence in more than one-half of human malignancies derived from epithelial, mesen-

chymal, hematopoietic, and lymphoid tissues. 7 Most p53 gene mutations are missense and so give rise to an altered protein. These mutations are most frequently dispersed over a conserved 200-codon region of the gene. The most striking p53 mutational spectrum found in human cancers is that of hepatocellular carcinomas (HCC), particularly from southeast Asia (including southern China and Vietnam) and southern Africa (South Africa and Mozambique) where hepatitis B virus (HBV) infection and aflatoxin B1 (AFB1) exposure are major risk factors. 8 Over 90°7o of p53 mutations are G to T transversions in the third nucleotide of codon 249. 9-a3 The presence of this specific G to T transversion is consistent with the occurrence of DNA changes induced by AFB1 as the mutagenic metabotite forms NT-deoxyguanosine adducts which induce primarily G to T transversions and G to A transitions, a4 This mutational specificity has etiological implications in hepatocellular carcinogenesis in certain regions of the world. Many groups have proposed and employed the p53 gene codon 249 mutation as a genetic biomarker which may be informative in the elucidation of the complex etiology of HCC. Current attempts to detect this specific mutation have been made by using polymerase chain reaction (PCR) and restriction enzyme analysis, TM ~5 PCR and DNA sequencing analysis, 9'12 or PCR-single-strand conformational polymorphism (PCR-SSCP) and direct DNA sequencing analysis.a3"16 As these methods are time consuming and instrument dependent, we have sought to develop a modified in situ hybridization technique for rapid detection of the specific mutation at the transcriptional level in HCC. In situ hybridization with oligonucleotide probes has been designed to distinguish between closely related mRNAs.lr" 18 Utilizing differential in situ hybridization with synthetic oligomers complementary to human p53 cDNA, various conditions were maximized with the hepatoma cell lines in which the codon 249 sequences are known. A total of 7 HCC cell lines were studied for the codon 249 mutation specific expression of the p53 gene. Our data indicate that this method is specific and rapid, visualizing cell morphology and signal simultaneously. Expression of the codon 249 mutant p53 is common in the cell lines (Mahlavu and PLC/PRF/5) derived from African patients but not in those (Huh4, Huh6, Huh7 and HCC-M) from Japanese.

192

Pathology (1995), 27, April

LIN et al.

MATERIALS AND METHODS Cell lines. Human HCC cell lines HuM, 19 Huhr,2O Huh7,21 Hep3B,22 HCC-M, 2~ Mahlavu 24 and PLC/PRF/52s were grown in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal calf serum and 0.001% penicillin/streptomycinin the presence of 5% COz at 37°C. Oligo-probes. Three oligodeoxynucleotide probes (HE7ASW, HE7ASM and HE7SWP) were prepared with an Applied Biosystems DNA synthesizer. HE7ASW and HE7ASM are both 20-oligomer probes complementary to positions 872-891 of the nucleotide sequence of the human p53 cDNA 26 with a single nucleotide difference at codon 249 of the gene. The change of the single nucleotide, resulting in the annealing temperature (T~,n= 4 × (G + C) + 2 × (A + T) - 5 °C) difference between the 2 oligomers, was positioned near the centre of the sequence to maximize the thermal instability of the mismatching hybridization. 17 HE7ASW (5 '-GAGGATGGGCCTCCGGTTCA-3 ', Tann=61 °(2) was used to detect the codon 249 wild-type p53 mRNA (249w~), while HE7ASM (5 '-GAGGATGGGACTCCGGTTCA-3 ', T,n. = 59°C) was for the codon 249 mutant p53 mRNA (249m9 detection. HE7SWP (5 ' - C A T C A T C A C A C T G G A A G A C T C C - 3 ', Tann=61 °C), a non-complementary 22-oligomer probe which also spans a short region near codon 249 but does not bind to the target mRNA, was used as a control for non-specific probe binding. Each oligomer (50 pmol) was labelled at the 3' end with terminal deoxynucleotidyltransferase (TdT) and either 100/~Ci of [~r-SzP]dATP (Amersham, sp. act. 3000 Ci/mmot) or 2 nmol biotin-7-dATP (Gibco BRL) using DNA Tailing Kit (Boehringer Mannheim). Northern analysis. RNAs were isolated from Huh7 and Mahlavu cells by the guanidinium thiocyanate extraction procedureY Ten/~g of total RNA from each preparation were electrophoresed on a 1% agarose gel under denaturing conditions, zs After electrophoresis, RNA was transferred onto a Hybond-N filter (Amersham) in 20 × SSPE (3 M NaC1, 200 mM NaH2PO,, 20 mM EDTA). The filter was irradiated under UVillumination for 5 mins and prehybridized at 42 °C for 4 hrs in a solution of 5 × Denhandt's (0.02% ficoll, 0.02% polyvinyl pyrollidone, 0.0207o BSA), 5 ×SSPE, 50% deionized formamide, 0.1% SDS, and 100 gg/mL denatured salmon sperm DNA. Hybridization was carried out with 32P-labelled oligomer probe overnight at 42 °C. The filter was then washed for 15 mins each with 2×SSPE/0.1% SDS and 1 ×SSPE/0A% SDS at 42°(2 and 5 mins with 0.5 ×SSPE/0.1% SDS at T~,, of the oligomer and 15 rains with 0.5×SSPE/0.1% SDS at room temperature (RT). After drying, the filter was exposed to Kodak X-OMT film at - 7 0 ° C . In Situ hybridization. Subconfluent cells grown on coverslips under standard culture condition were washed with phosphate-buffered saline (PBS) for 3 rains and fixed in 2% paraformaldehyde for 10 mins. The cells were then washed twice with PBS and treated with 0.1% TritonX 100 for 10 mins.~Following treatment, the cell samples were washed with PBS and post-fixed in 2% paraformaldehyde for another 10 mins. The coverslips were air dried and stored at - 2 0 °C. All solutions, except Tris buffers, were treated with 0,107o diethy~ pyrocarbonate (DEPC) and autoclaved. Prior to hybridization, the coverslips were warmed to RT, washed with PBS and incubated in 2 x SSC (300 mM NaC1, 30 Mm Na3 citrate) at RT for 10 rains. The cell samples were then directly hybridized facedown on a glass slide overnight at RT with 50 ~L of a mixture containing 5007o deionized formamide, 1007o dextran sulfate, 2 x S S C , 1 x Denhardt's solution, 5 mg/mL denatured salmon sperm DNA, and 20 ng/gL biotin-labelled oligo-probe. After hybridization, the coverslips were washed face-up in a 24-well tissue culture plate for 1 hr each with 2 x S S C and l xSSC at RT and 15 mins each with 0.5 xSSC at T~n~ (of each oligomer) and 0.1 x SSC at RT. To detect biotin-labelled probe, the cell samples were washed with Buffer A (100 mM Tris-HC1, pH 7,5 and 150 mM NaC1 and incubated for 30 min at RT in Buffer A containing 2% normal sheep serum and 0.3% Triton-X 100 and 1 hr in Buffer A containing I% normal sheep serum, 0.3% Triton-X 100 and 1 ;ag//xL streptavidin-alkaline phgsphatase (S-AP). Unbound S-AP was removed by washing the coverslips for 30 mins with Buffer A and

i¸

?

Fig. 1 Northern blot analysis of specific p53 transcripts in Mahlavn and Huh7 cell lines using radio-labelled oligonucleotide probes. (a) Hybridization with the codon 249 mutation specific probe HE7ASM. (b) Hybridization with the codon 249 wild-type specific oligomer HE7ASW. Lanes A and B refer to Mahlavu and HuhT, respectively.

10 mins with Buffer B (100 mM Tris-HC1, pH 9.5, I00 mM NaCI and 50 mM MgCt~). The eolour was developed using alkaline phosphatase substrate solution containing nitroblue tetrazolium chloride (NBT) and 5-bromo-l-chloro-3-indolylphosphate (BCIP) as described elsewhere. 29 The reaction was stopped in a solution of 10 mM Tris-HC1, pH 7.5 and 1 mM EDTA. The coverslips were then dehydrated in ethanol and mounted face-down on a microscope glass slide with Permount medium. The presence of the target p53 mRNA was visualized as localized purple blue staining in the cytoplasm of the cells. The intensity of the colour was graded and used as an indicator of the expression of p53 mRNA. Each set of in situ hybridization studies included control samples which were hybridized without probe.

RESULTS Expression o f the codon 249 mutant p53 m R N A in Mahlavu cells The steady state level of p53 specific mRNA was detected in cell lines Mahlavu and Huh7. 3°'31 A specific G to T mutation at codon 249 of the p53 gene has been found in Mahlavu 9 but not in Huh7. 3a The 32p-labelled oligometric cDNA probes were used to determine the specificity of differential hybridization of the codon 249 mutant or wild-type p53 mRNA. Figure 1 shows the specific hybridization of each oligomer probe to cellular RNA bound to nylon filters. In hybridization with the oligomer HE7ASM (249mt), the 2.5 kb p53 transcripts were observed only in Mahlavu cells (Fig. la), whereas the probe HE7ASW (249wt) only hybridized to the p53 mRNA in Huh7 cells when the same blot was reprobed (Fig. lb). The non-complementary probe HE7SWP showed no hybridization signal in the blot (data not shown). The data suggest that Mahlavu cells express the codon 249 mutation specific p53 mRNA and that the

IN SITU DETECTION OF MUTANT p53 rnRNA

193

L%

Fig. 2

In situ determination of specific p53 transcripts in Mahlavu and Huh7 cell lines using biotin-labelled oligomer probes. Different staining

results were observedby in situ hybridization:(a) no staining with HE7ASW (249-~);(b) cytoplasmicstaining with HE7ASM (249m');(c) cytoplasmic staining with HE7ASW (249'~); (d) no staining with HE7ASM (249m~);(e) no staining without probe; (f) nuclear staining with HE7SWP (a non-complementaryprobe). (a), (b), (e), and (f) refer to Mahlavu; (c) and (d) refer to Huh7. (In situ hybridization,originalmagnification x 400.)

oligometric probes are specific for determination of closely related p53 mRNA.

In situ hybridization In order to visualize the hybridization of the probes to the specific p53 mRNAs in the 2 hepatoma cell lines in situ, each oligomer probe was labelled with biotin for enzymetric colour detection and conditions of differential hybridization and colour detection were modified to both preserve the morphology o f cells and maximize colour detection o f the labelled probe. Cell fixation and treatment were important for optimal cell morphology and

colorimetric signal detection. The common practice using 4% paraformaldehyde for cell fixation and proteinase K treatment failed to provide optimal morphology of the tumor cells, resulting in poor signals. Therefore, 2% paraformaldehyde fixation and 0.1% Triton-X 100 treatment were used in all subsequent in situ hybridization studies. The biotin-labelled probes HE7ASM (249 mr) and HE7ASW (249w0 were used for enzymetric colour detection with streptavidin-AP conjugate. The appearance of a light to dark cytoplasmic and light nucleolar purple blue colour signals depended upon the extent of hybridization,

194

LIN

Pathology (1995), 27, April

et al.

TABLE 1 Detection of the codon 249 mutant p53 mRNA by in situ hybridization HCC cell lines Mahlavu PLC/PRF/5 HepJB I--Iuh4 Huh6 Huh7 HCC-M

Oligo-probes HE7ASM

HE7ASW m

-t-

+ m

+ + + +

No probe

Patient origin South Africa Mozambique America* Japan Japan Japan Japan

*American black male.

probe level and streptavidin binding. Although direct hybridization and washing followed standard protocols, the washing temperature, times and stringency described in Materials and Methods were essential for reduced background binding. Under these conditions, non-specific hybridization of each probe and binding of streptavidin were minimal (Fig. 2). Maximum signal-to-noise ratio was obtained by stringent washes of the cell preparations at the annealing temperature for 15 mins after hybridization, room temperature incubation with streptavidin-AP conjugate in darkness for 2 hrs and colour development with NBT/BCIP (nitroblue tetrazolium chloride/5-bromo1-chloro-3-indolyl phosphate) for 30 mins. The specificity of in situ hybridization was demonstrated by differential hybridizations in the examined cell lines. Mahlavu cells,

known to carry the p53 codon 249 (G to T) point mutation, exhibited strong cytoplasmic staining after incubation with streptavidin-AP conjugate only in the presence of the 249mt probe HE7ASM (Fig. 2a, b), while the same staining pattern in Huh7 cells with the codon 249 wild-type p53 mRNA was only seen in hybridization with the 249 ~t oligomer HW7ASW (Fig. 2c, d). This is consistent with Northern blot analysis described earlier. Moreover, the lack of colour was observed in the cell preparations incubated with streptavidin-AP conjugate in the absence of labelled probe (Fig. 2e) and nuclear staining in the presence of the non-complementary probe HE7SWP (Fig. 2f) in parallel hybridizations. Thus, the appearance of cellular staining after in situ hybridization is due to specific hybridization of each oligomer probe to the p53 mRNA.

Screening hepatoma cell lines for specific mutant p53 mRNA Using the modified in situ hybridization method, other hepatoma cell lines Huh4, Huh6, HepJB, HCC-M, and P L C / P R F / 5 were examined for the codon 249 mutation specific expression of the p53 gene (Table 1). Hybridization with the oligomer HE7ASM (249mt) showed no colour detection in all but P L C / P R F / 5 cells (Fig. 3a, c), which contain the p53 gene mutation at codon 249 as well. 9 Hybridization with the probe HE7ASW (249 wt) revealed most were positive except P L C / P R F / 5 and HepJB cells (Fig. 3b, d).

b

¸nil¸¸

tl Fig. 3 Examples of determination of specific p53 transcripts in HCC cell lines by in situ hybridization. (a) and (c) show detection with HE7ASM (249mt); (b) and (d)show detection with HE7ASW (249wt). P L C / P R F / 5 line is shown as (a) and (b); Hep3B line in (c) and (d). (In situ hybridization, original magnification x400.)

1N SITUDETECTIONOF MUTANTp53 mRNA 195 DISCUSSION In the present study, we used differential in situ hybridization to identify the codon 249 mutant p53 m R N A in 7 h u m a n h e p a t o m a cell lines. The data demonstrate, for the first time, the localization and level of the specific p53 m R N A in neoplastic hepatocytes. Hybridization was specific to the p53 m R N A , since detection with different oligomer probes showed specific signals. The use of biotin-labelled probe allowed simultaneous visualization of the m o r p h o l o g y and signal due to hybridization with labelled oligomer. An advantage of these specific oligoprobes over radiolabelled probes is the simultaneous visualization of the cellular localization of m R N A and cell m o r p h o l o g y in a permanent record. It also has the potential o f allowing rapid detection of other mutationspecific m R N A s (e.g., mutant/3-globin m R N A in sickle cell anemia) and simultaneous investigation o f other factors (e.g., expression o f proto-oncogenes) that m a y be involved in cell transformation by carcinogens. Expression of the codon 249 mutant p53 was detected by Mahlavu and P L C / P R F / 5 cells by differential in situ hybridization. This is consistent with the previous observations that these two cell lines contain G to T transversions in the third nucleotide of codon 249 o f the p53 gene 9"32 and detectable levels of the p53 specific m R N A and protein. 3°'31 In most tumors, one allele of the p53 gene appears to have been lost via a deletion and the other changed by a point mutation. 33 The fact that Mahlavu and P L C / P R F / 5 cells were only positive to differential hybridization with H E 7 A S M (249 mr) probe indicates that these H C C cell lines are functionally homozygous for the specific codon 249 mutation. It implies that the wild-type allele o f the p53 gene is either transcriptionally silent or has been deleted in these 2 cell lines. As PCR-restriction enzyme analysis and PCR-SSCPD N A sequencing have to date revealed only the mutant allele of the gene in these 2 cell lines (Lin Y, unpublished observations), 9"32 the wild-type allele is most likely to have been deleted altogether. In Hep3B cells, a m a j o r portion of the p53 gene has been reported to be deleted, resulting in the absence of p53 transcripts and p53 protein in the cell line. 3°'31 In the present study, Hep3B ceils were indeed negative to hybridizations with both H E 7 A S M (249 mr) and H E 7 A S W (249 wt) probes. Although the exon-7 sequence of the p53 gene in the cell line is still wildtype, 9 a large deletion has been found to occur after this exon (Lin Y, unpublished observations). The remaining p53 gene in this cell line is likely non-transcribed. Absence of the codon 249 mutation in p53 transcripts was found in H C C cell lines Huh4, Huh6, Huh7 and H C C - M . This agrees with P C R - S S C P - D N A sequencing, where the exon-7 of the p53 gene appears to be intact in these cell lines (Lin Y, unpublished observations). The same result for Huh7 is also seen in the observation by H o s o n o et al. 31 The levels of p53 transcripts were also steady in these cell lines, suggesting that the transcripts are stable. This m a y result in overexpression of p53 protein. We have evidence that accumulation of p53 protein is c o m m o n in these cell lines (Lin Y, unpublished observations). It m a y be due to a mutation in another codon of the gene giving rise to an altered protein with an increased half-life, or to mutation in non-coding

regions of the gene leading to overexpression of the protein, or to complex formation between p53 and other cellular proteins resulting in p53 accumulation. Furthermore, amongst the cell lines studied, Huh4, H u h 6 , Huh7 and H C C - M are derived from Japanese patients, while Mahlavu and P L C / P R F / 5 are from Africa. Our data also support previous studies that the p53 gene codon 249 mutations are c o m m o n in H C C s f r o m South Africa T M and specific regions o f China, 1°'12'~3 but not f r o m J a p a n 16"34 and other countries, ls'3s-37 ACKNOWLEDGEMENTS We are grateful to Drs H u h (Tokyo University, Japan) for Huh4 cells, Doi (Okayama University, J a p a n ) for H u h 6 ceils, N a k a b a y a s h i ( O k a y a m a University, Japan) for Huh7 cells, Morizane (Keio University, Japan) for H C C - M cells, Knowles (Wistar Institute, Pennsylvania, USA) for Hep3B cells, Prozeskey (Medical Research Council, South Africa) for Mahlavu cells, and Alexander (Virus Cancer Research Unit, Johannesburg, South Africa) for P L C / P R F / 5 cells. We thank Dr Benjamin Li (IMCB, NUS) for synthesizing the oligomer probes and Mrs Josephine H o w e for photographic assistance. This work was supported by grant RP920323 f r o m the National University of Singapore. Address Jbr correspondence." Y.L., Department of Microbiology,Faculty

of Medicine, National Universityof Singapore, LowerKent Ridge Road, Singapore 0511.

References 1. Miller C, Mohandas T, Wolf D et al. Human p53 gene localized to short arm of chromosome 17. Nature 1986; 319: 783-4. 2. LevineAJ, Momand J, Finlay CA. The p53 tumour suppressor gene. Nature t991; 351: 453-6. 3. Kraiss S, Quaiser A, Oren M, Montenarh M. Oligomerization of oncoprotein p53. J Virol 1988; 62: 4737-44. 4. Green MR. When the products of oncogenes and antioncogenes meet. Cell 1989; 56: 1-3. 5. Lane DP, Crawford LV. T antigen is bound to a host protein in SV40-transformed cells. Nature 1979; 278: 261-3. 6. ScheffnerM, WernessBA, HuibregtseJM et al. The E6 oncoprotein encoded by human papillomavirus type 16 and 18 promotes the degradation of p53. Cell 1990; 63: 1129-36. 7. HollsteinMC, SidranskyD, VogelsteinB, Harris CC. p53 mutations in human cancers. Science 1991; 253: 49-53. 8. Bosch FX, Mufioz N. Prospects for epidemiological studies on hepatocellular cancer as a model for assessing viral and chemical interactions. In: Bartsch H, HemminkiK, O'Neill IK, eds. Methods for detecting DNA damaging agents in humans: Applications in cancer epidemiologyand prevention. Lyon: International Agency for Research on Cancer, 1988; 427-38. 9. Bressac B, Kew M, Wands J, Ozturk M. SelectiveG-mutation to T-mutation of p53 gene in hepatocellular carcinoma from southern Africa. Nature 1991; 350: 429-31. 10. Hsu IC, Metcalf RA, Sun T et al. Mutational hotspot in the p53 gene in human hepatocellularcarcinomas. Nature 1992; 350: 427-8. 11. Ozturk M, collaborators, p53 mutation in hepatocellular carcinoma after aflatoxin exposure. Lancet 1991; 338: 1356-9. 12. ScorsoneKA, Zhou YZ, Butel JS, Slagle BI.. p53 mutations cluster at codon 249 in hepatitis B virus-positivehepatocellular carcinomas from China. Cancer Res 1992; 52: 1635-8. 13. Li DZ, Cao YQ, He LP et at. Aberrations of p53 gene in human hepatocellular carcinoma from China. Carcinogenesis 1993; 14: 169-73.

196

L I N et al.

Pathology (1995), 27, April

14. Foster PL, Eisenstadt E, Miller JH. Base substitution mutations induced by metabolically activated aflatoxin B1. Proc Natl Acad Sci USA 1983; 80: 2695-8.

26. Zakut-Houri R, Bienz-Tadmor B, Givol D, Oren M. Human p53 cellular tumour antigen: cDNA sequence and expression in COS cells. EMBO J 1985; 4: 1251-5.

15. Hayward NK, Walker GJ, Graham W, Cooksley E. Hepatocellular carcinoma mutation. Nature 1991; 352: 764.

27. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987; 162: 156-9.

t6. Murakami Y, Hayashi J, Hirohashi S, Sekiya T. Aberrations of the tumour suppressor p53 and retinoblastoma genes in human hepatocellular carcinomas. Cancer Res 1991; 51: 5520-5. 17. Conner BJ, Reyes AA, Morin C et al. Detection of sickle cell/3~globin allele by hybridization with synthetic oligonucleotides. Proc Natl Acad Sci USA 1983; 80: 278-82. 18. Lewis ME, Sherman TG, Watson SJ. In situ hybridization histochemistry with synthetic oligonucleotides: strategies and methods. Peptides 1985; 6:75-87 (suppl. 2). 19. Huh N, Utakoji T. Production of HBs-antigen by two new human hepatoma cell lines and its enhancement by dexamethasone. Japn J Cancer Res 1981; 72: 178-9. 20. Doi I. Establishment of a cell line and its clonal sublines from a patient with hepatoblastoma. Japn J Cancer Res 1976; 67: 1-10. 21. Nakabayashi H, Taketa K, Miyano K et al. Growth of human hepatoma cell lines with differentiated functions in. chemically defined medium. Cancer Res 1982; 42: 3858-63. 22. Aden DP, Fogel A, Plotkin Set al. Controlled synthesis of HBsAg in a differentiated human liver carcinoma-derived cell line. Nature 1979; 282: 615-6. 23. Watanabe T, Morizane T, Tsuchimoto K et al. Establishment of a cell line (HCCM) from a human hepatocellular carcinoma. Int J Cancer 1983; 32: 141-6. 24. Prozeskey O, Brits C, Grabow W. In vitro culture of cell lines from Australia antigen positive and negative hepatoma patients. In Saunders S, Terblanche J, eds. Liver. London: Pitman Medical, 1973; 358-60. 25. Alexander J J, Bey EM, Geddes EW, Lecatsas G. Establishment of a continuously growing cell line from primary carcinoma of the liver. S Afr Med J 1976; 50: 2124-8.

28. Maniatis T, Fritsch EF, Sambrook J. Molecular Cloning: A Laboratory Manual. New York: Cold Spring Harbor Laboratory Publications, 1982. 29. Leafy J J, Brigate D J, Ward DC. Rapid and sensitive colourimetric method for visualizing biotin-labelled DNA probes hybridization to DNA of RNA immobilized in nitrocellulose: bio-blots. Proc Natl Acad Sci USA 1983; 80: 4045-9. 30. Bressac B, Galvin KM, Liang TJ et al. Abnormal structure and expression of p53 gene in human hepatocellular carcinoma. Proc Natl Acad Sci USA 1990; 87: 1973-7. 31. Hosono S, Lee CS, Chou MJ et al. Molecular analysis of the p53 alleles in primary hepatocetlular carcinomas and cell lines. Oncogene 1991; 6: 237-43. 32. Murakami Y, Hayashi K, Hirohashi S, Sekiya T. Detection of aberrations of the p53 alleles and the gene transcript in human tumour cell lines by single-strand conformational polymorphism analysis. Cancer Res 1991; 50: 3356-61. 33. Nigro JM, Baker SJ, Preisinger AC et al. Mutations in the p53 gene occur in diverse human tumour types. Nature 1989; 342: 705-8. 34. Nishida N, Fukuda Y, Kokuryu H et al. Role and mutational heterogeneity of the p53 gene in hepatoeellular carcinoma. Cancer Res 1993; 53: 368-72. 35. Challen C, Lunec J, Warren W e t al. Analysis of the p53 tumoursuppressor gene in hepatocellular carcinomas from Britain. Hepatology 1992; 16: 1362-6. 36. Hollstein MC, Wild CP, Bleicher F et al. p53 mutations and aflatoxin BI exposure in hepatocellular carcinoma patients from Thailand. Int J Cancer 1993; 53: 51-5. 37. Hosono S, Chou M J, Lee CS, Shih C. Infrequent mutation of p53 gene in hepatitis B virus positive primary hepatocellular carcinomas. Oncogene 1993; 8: 491-6.