Screen p53 mutations in hepatocellular carcinoma by FASAY: A novel splicing mutation

Screen p53 mutations in hepatocellular carcinoma by FASAY: A novel splicing mutation first-author corresp ‡Corresponding Author ‡Corresponding Author †E-mail:kkhuo@fudan.edu.cn †E-mail:kkhuo@fudan.edu.cn kkhuo@fudan.edu.cn Institute of Genetics, School of Life Sciences, Fudan University, Shanghai 200433, China Institute of Genetics, School of Life Sciences, Fudan University, Shanghai 200433, China Objective: 2 To establish a routine procedure for the detection of p53 mutations in hepatocellular carcinoma (HCC) surgical resections using the FASAY (functional analysis of separated alleles of p53 on yeast) procedure. Methods: 2 p53 status was analyzed by FASAY and cDNA sequencing in 50 cases of HCC. After the extraction of RNA from the frozen tumor and corresponding normal tissues, reverse transcription RT-PCR was carried out using these samples. The assay can detect mutations of p53 mRNA between codons 67 and 347 by the DNA-binding activity of the protein and reveal them as red colonies. Results: 2 Of the 50 specimens, 29 (58%) were positive (mutant) by FASAY. Sequencing analysis confirmed that all 29 FASAY positive tumors harbored mutations, and that no mutations were detectable in any FASAY negative tumors. In 29 p53 mutations, 22 mutations were point missense mutation, 5 were deletions and 2 were splicing mutations. A novel splice mutation on splice donor of intron 6 was reported, which could produce two different mRNAs, respectively using the nearest upstream and downstream recessive splice donor sites. Conclusion: 2 FASAY is a sensitive method for detecting the various types of p53 mutations in HCC, suggesting that the yeast functional assay for the detection of p53 mutations may be essential for elucidating their clinical significance. Hepatocellular carcinoma (HCC) p53 mutation FASAY Splicing mutation INTRODUCTION 1 Hepatocellular carcinoma (HCC) is the sixth most common cancer in new cases per year (626 000) and the third most common cancer resulting in death (598 000) (Parkin et al., 2005 2005 ). China is the country most seriously suffering from HCC, accounting for 55% of total cases and deaths worldwide per year. The clinical behavior of HCC is heterogeneous and difficult to predict, and in patients undergoing resection, recurrence rates can be as high as 50% in two years and life expectancy of as short as twelve months can be possible. Great efforts were put to find an ideal prognostic marker, and the tumor suppressor p53 seems to be the most promising biomark of prognosis in HCC. In an overwhelming majority of clinic studies, the prognostic significance of p53 alterations in HCC relies on the identification of p53 nuclear overexpression by immunohistochemistry (IHC). But IHC would give false-positives from stabilization of wild-type p53 and false-negatives from stop codons, frameshifts and destabilizing mutations (Sjogren et al., 1996 1996 ). Heterogeneity also results from the use of different antibodies, different techniques to prepare sections and different criteria for scoring positives (Chappuis et al., 1999 1999 ). FASAY (functional analysis of separated alleles of p53 on yeast), which was developed by Flaman et al.( 1995 1995 ), can evaluate the functional status of p53 proteins expressed in tumor cells using yeast cells yIG397, leading to detection of significant functional mutation of p53 as a transcriptional factor. The genotype of yIG397 is MATa ade2-1 leu2-3 , 112trp1-1his3-11 , 15can1-100ura3-1 URA3 3×RGC::p CYC :: ADE2 , contains an integrated plasmid with the ADE2 open reading frame under the control of a p53-responsive promoter on URA3 site. In this study, we tested the fidelity of FASAY scoring p53 status of analyzed samples and investigated p53 mutations in HCC. MATERIALS AND METHODS 1 Sample collection 2 Fifty patients who underwent surgical resection for primary HCC at the affiliated Eastern Hepatobiliary Surgery Hospital of the Second Military Medical University between 2002 and 2003 entered this study. Cases with preoperative radiation or chemotherapy and liver recurrence were not selected. All patients underwent a complete resection of liver tumor. In each case, fragments of the tumor and corresponding normal tissue were obtained and partly immediately snap frozen in liquid nitrogen and stored at −80 °C until further testing for p53 transcriptional activity to prevent degradation of RNA. The remaining material tumor was formalin fixed and paraffin embedded for pathological evaluation. Anatomopathologic evaluation of the paraffin-embedded tumor specimens was determined stage according to Edmondson classifications. RNA extraction and cDNA synthesis 2 Total RNA was extracted using the total RNA isolation kit (Watson, Shanghai, China) according to the manufacture’s recommendations of 5~10 mg of tissue ground with a microfuge pestle. The quality of the RNA was visually estimated by the presence of the 18S and 28S rRNA bands upon electrophoresis with ethidium bromide stain. The RNA was reverse transcribed, using MMLV reverse transcriptase (Promega, Shanghai, China), into single stranded cDNA using an oligop (dT) primer. The reaction was incubated at 65 °C for 5 min, and iced 5 min to allow the primer to anneal and then at 42 °C for 60 min for reverse transcription. To minimize interference with subsequent applications, the MMLV reverse transcriptase was then denatured by heating at 75 °C for 15 min. The reaction products were stored at −80 °C. Amplification of p53 cDNA by PCR 2 The p53 cDNA was amplified by PCR using the following primers (Invitrogen, Shanghai, China). Human specific primers P3: 5′-ATT-TGA-TGC-TGT-CCC-CGG-ACG-ATA-TTG-AA(S)C-3′, P4: 5′-ACC-CTT-TTT-GGA-CTT-CAG-GTG-GCT-GGA-GT(S)G-3′, where (S) represents a phosphorothioate linkage. PCR was performed for 35 cycles of 94 °C for 30 s, 65 °C for 60 s and 72 °C for 60 s using a thermal cycler that had been preheated to 95 °C. The PCR mix was allowed a final extension at 72 °C. Two microlitres of the cDNA reaction product was amplified in 20 μl of Pfu buffer (Strategene, Shanghai, China) plus 1.25 units of Pfu DNA polymerase, 100 ng of primers, 10% dimethyl sulphoxide, and 50 mol/L dNTPs. PCR controls included (1) negative control incubation without template DNA and (2) positive controls included plasmids pLS76 known to contain the p53 cDNA sequence. PCR reaction product (5 μl) was then run on a 1.5% agarose gel, alongside a molecular weight marker, to check the PCR product. Vector preparation for transformation of yeast 2 The plasmid pRDi22 (Waridel et al., 1997 1997 ) was obtained from Dr. R. Iggo (I.S.R.E.C. Switzerland). One microgram of plasmid was cut by restriction digest using Stu I and Hin dIII. The linearised vector was then dephosphorylated using calf intestinal alkaline phosphatase (Takara, Shanghai, China) at 37 °C for 30 min. The linearised vector was gel purified using the gel DNA isolation kit (Vitagene, Shanghai, China) and redissolved in TE buffer (pH 8.3). Yeast transformation and growth. The yeast strain yIG397 was obtained from Dr. R. Iggo (I.S.R.E.C. Switzerland) and was routinely cultured on complete medium (yeast extract, peptone, dextrose: YEPD), supplemented with adenine at 10×excess concentration (200 g/ml), to avoid selection of spontaneous suppressors of the endogenous ade2 locus. The resulting colonies were all white. Any red colonies would have been discarded. Transformation of competent cells 2 The TE/LiAc/PEG method (Gietz et al., 1992 1992 ) was used to make the yeast competent, and was also used for the transformation. Linearised vector DNA (50 ng), 5 μl of PCR product, 5 μl ssDNA (10 mg/ml) (Sangon, Shanghai, China) were used for the transformation. The components were mixed by vortexing vigorously. The cells were then incubated for 30 min at 30 °C, and again vortexed at 15 min intervals during the incubation, then heat-shocked for 15 min at 40 °C. Cells (100 μl) were plated on synthetic minimal media minus leucine, left to grow for 2~4 d at 30 °C, and the number and colour of the yeast colonies were noted. Control transfections included transfection with linear vector only, pLS76 whole vector only, and PCR product only. Untransfected cells were also grown on leucine plus and leucine minus media. PCR products from plasmids with known p53 status i.e. wild-type and known mutants were also transfected. Sequence analysis 2 Sequencing analysis was carried out using automated ABI 3730 apparatus (Applied Biosystems, Shanghai, China). cDNA, synthesized by the polymerase chain reaction protocol described above, was directly sequenced using the following oligonucleotide primers: 5′-ATT-TGA-TGC-TGT-CCC-CGG-ACG-ATA-TTG-AAC-3′, 5′-TAC-TCC-CCT-GCC-CTC-AAC-AAG-ATG-3′, 5′-TTG-CGT-GTG-GAG-TAT-TTG-ATG-AC-3′, and 5′-ACC-TTT-TTG-ACT-TCA-GGT-GCT-GGA-GTG-3′ in both directions. The cycle sequencing reactions were carried out using Taq Dye Deoxy Terminator Cycle Sequencing kit according to the manufacturer’s protocol. Plasmid containing p53 cDNA, recovered from single red yeast colonies, was similarly sequenced following this protocol. For each patient sample, 5 colonies were sequenced. RESULTS 1 Determination of the production of RT-PCR for p53 gene 2 The production of PCR using P3, P4 as primers and vector pLS76 (Flaman et al., 1995 1995 ) as template, which had the wild-type p53 gene CDS, was about 1 kb fragment. The result of RT-PCR coming from the tumor and corresponding normal tissues is shown in Fig. 1 1 . Mark: DL2000 (TaKaRa); 1~8: Different tissue samples respectively; +: pLS76 as positive control; −: ddH 2 O as negative control Identification of p53 statues in tumors and corresponding normal tissue 2 Ade2 − cells grown on medium contained limiting adenine turned red because of the accumulation of an intermediate in adenine metabolism. The strain used for assessing p53 status of yIG397 carrying a reporter with a p53-binding site ribosomal gene cluster (RGC) upstream of the ADE2 gene. When the yeast was transfected with the cDNA from a wild-type p53 gene, the yeast cells express ADE2 , and grow normally as white colonies on minimal synthetic defined agar. Cells transfected with the cDNA from a mutant tumour, failed to transactivate the ADE2 gene and grow to form red colonies. The p53 status can be easily determined by the color of transfected yeast cells (Fig. 2 2 ). But the percentage of yeasts transfected with wild-type cDNA was not 100% white clones, and there were generally ≤10% red colonies as background due to PCR-induced point mutations, alternative splicing of intron 9 affecting the p53 carboxy-terminus. p53 was considered as wild type when < 10% of red or pink colonies were detected (negative FASAY). Above this cutoff value, the FASAY was considered positive. The activity of the p53 mutant was determined by the color of at least 300 colonies per strain. Using this criterion, 50 of the tumors tested contained 29 mutant p53 (58%). The percent in the brackets is ratio of the red clones to total clones on the plate The FASAY results of the corresponding normal tissue varied from 5% to 20% (the data of each sample were not shown). Considering that the tumors tissue may be mixed into the neighboring normal tissue, another cutoff value (25%) was used in the tumor corresponding normal tissues. But according to this cutoff value, no positive FASAY was screened. Though p53 249 Ser mutation induced by aflatoxin B1 (AFB1) (Bressac et al., 1991 1991 ) could be detected in the normal hepatic tissue surrounding tumor or even the patients without HCC (Kirk et al., 2000 2000 ; Huang et al., 2003 2003 ), in these tissues FASAY positive samples were not examined. Sequencing the mutations in p53 2 Sequencing of plasmids rescued from red yeast colonies showed that 22 mutations were point missense mutation, 2 were frameshift deletion mutations, 3 was in-frame deletion and 2 were splicing mutations (Table 1 1 ). The hotspot mutation 249 Ser that occurred 11 times in this study, accounted for 37.9% in total mutations. Edmonson classification The ratio of red colons in total colons on SD-L plate The status of p53 is determined by the red clones ratio with 10% cutoff; WT: Wild type; M: Mutant Two different splicing products: 6I+ and Δ7E (6 intron partly retention and 7 exon partly deletion). Δ: Delete; MD: Mutant splice donor; I7: Intron 7; I6: Intron 6; ?: Exact mutant sites need sequencing gDNA No. Age/sex Stage a a HBV FASAY (%) b b p53 statue c c DNA sequence Consequence 8 Wild type T3 49/M III + 5 WT \ \ T6 58/M III + 9 WT \ \ T11 55/M III + 10 WT \ \ T12 56/M III + 10 WT \ \ T14 58/M III + 9 WT \ \ T15 59/M III + 10 WT \ \ T18 62/M III + 7 WT \ \ T19 63/M III + 10 WT \ \ T20 54/M II − 6 WT \ \ T21 57/F II + 8 WT \ \ T24 80/M I − 3 WT \ \ T25 36/M I + 4 WT \ \ T30 47/F II − 4 WT \ \ T33 69/M II − 9 WT \ \ T34 57/M II + 10 WT \ \ T37 84/M II − 3 WT \ \ T38 42/M II + 4 WT \ \ T41 47/M I + 3 WT \ \ T44 65/M II + 7 WT \ \ T48 75/M I + 2 WT \ \ T49 56/M I + 9 WT \ \ T50 40/M I + 7 WT \ \ 8 Missense mutation T1 50/M III~IV − 93 M 747 G→T 249 R→S T7 50/M III + 99 M 747 G→T 249 R→S T10 65/F III + 97 M 736 A→G 246 M→V T13 57/M III + 92 M 157 G→T 157 V→F T16 60/M III + 87 M 747 G→T 249 R→S T17 61/M III + 68 M 747 G→T 249 R→S T22 37/M I + 97 M 747 G→T 249 R→S T23 63/M I − 58 M 584 T→A 195 I→N T27 55/M II + 65 M 747 G→T 249 R→S T28 50/M I + 79 M 747 G→T 249 R→S T29 40/M II + 79 M 747 G→T 249 R→S T31 43/M II + 81 M 329 G→T 110 R→L T32 62/M I + 69 M 747 G→T 249 R→S T35 47/M I + 31 M 844 C→T 282 R→W T36 59/M II − 67 M 817 C→A 273 R→S T39 53/F II + 80 M 747 G→T 249 R→S T40 61/M I + 79 M 535 C→A 179 H→N T42 47/M II + 95 M 747 G→T 249 R→S T43 73/M II − 40 M 438 G→T 146 W→STOP T46 35/M II + 81 M 422 G→A 141 C→Y T47 52/F III + 75 M 484 A→T 162 I→F 8 Frameshift mutation T2 65/M III + 28 M 373 Δ1 bp Truncated 168 T8 46/M III + 80 M 186 Δ1 bp Truncated 121 8 In-frameshift mutation T5 43/M III + 91 M 791~793 Δ3 bp ΔL T26 45/M II + 74 M 313~375 Δ62 bp ΔGS~CT T45 51/M II + 64 M 720~734 Δ15 bp ΔSCMGG 8 Splicing mutation T4 42/M III + 29 M I7 MD gDNA? Δ8 exon T9 42/M III + 60 M I6 MD AG→TG Multiple d d In the 5 deletion mutations, the sample T2 (deletion of T in 124 codon) and T8 (deletion of A in 62 codon) both caused frameshifts and premature truncation of p53 translation at codon 168 and 121 respectively; the mutations of the sample T5, T26, T45 all caused in-frame deletion, missing several amino acids in final proteins. For the two splicing mutations, in the sample T9, as the splice donor site of intron 6 AG mutated into TG (Fig. 3a 3a ), there existed two different splicing products: one that used the nearest upstream AG as splice donor had been inserted 49 bp sequence of 6 intron into the final mRNA (Fig. 3a 3a ); the other that used the nearest downstream AG as splice donor therefore deleted a 47 bp sequence of 7 exon in mRNA (Fig. 3b 3b ); both of them caused premature truncation of proteins, stopping at codon 243 and 246. In 6 clones randomly chosen for sequencing, 5 of them were the first splicing forms and only one was the second splicing form, suggesting that the upstream one may be more preferable for splicing than the downstream one. In another sample T4, the entire exon 8 was skipped, which was reported before (Holmila et al., 2003 2003 ). DISCUSSION 1 The p53 gene mutation is one of the commonest genetic alterations found in human cancers. The majority of these mutations are missense point mutations that are scattered along the entire p53 gene (MIM191170). The high degree of heterogeneity of p53 mutations and their heterogeneous distribution have led to the development of various screening procedures in order to increase the speed, sensitivity and specificity of p53 mutation detection. Compared with IHC, genomic sequencing, which usually detect exon 5~8 and a single from normal tissue can easily mask the mutant peak, and single-strand conformation polymorphism (SSCP), FASAY was proved to reach the highest sensitivity (Meinhold-Heerlein et al., 2001 2001 ; Watanabe et al., 2004 2004 ), due to the fact that a larger region exon 4~10 of p53 gene was tested, mutations can be detected in the presence of large amounts of normal tissue because hundreds of clones are examined from each sample and the simple red vs white read-out means that mutations are not easily overlooked (Chappuis et al., 1999 1999 ). The FASAY detected 29 mutations in the 50 tumors investigated (58%), and the sequencing analysis confirmed that all 29 FASAY positive tumors harbored mutations, with no mutations detected in any FASAY negative tumors. In these mutations, the non-missense mutations including frameshift, in-frame deletion and splice mutation took place 7 times, accounting for 24% in total mutations detected by FASAY in HCC. In contrast, less than 20% of p53 mutations in the UMD database ( http://www.umd.be:2072/ http://www.umd.be:2072/ ) were non-missense mutations, with splice mutations in the p53 gene described as rare events that occur at a frequency of less than 1% (Holmila et al., 2003 2003 ). Using RNA as starting material for a test, relatively rare mutant sequences can be detected by FASAY (Chappuis et al., 1999 1999 ), partly due to the reason mentioned above. Recently, several reports about novel splicing mutations of p53 in other cancers were all detected by FASAY assay (Chappuis et al., 1999 1999 ; Verselis et al., 2000 2000 ; Holmila et al., 2003 2003 ). Geographic and ecologic studies of HCC are limited by the fact that regions with a high incidence of HCC frequently have a population that is simultaneously exposed to AFB1 and HBV (Kirk et al., 2000 2000 ). China has a population with a high incidence of HCC, chronic HBV infection, and a high exposure to AFB1 (Hsu et al., 1991 1991 ; Parkin et al., 2005 2005 ). Among patients with HCC, we did not observe any differences in chronic infection with HBV (based on HBsAg status) between those patients with HCC who were positive and negative for p53 mutation. In this study, the patients with HBV positive (based on HBsAg status) were 41, in which 16 were p53 wild type (16/21) and 25 p53 were mutants (25/29), P > 0.05. On the other hand 249 Ser , as a “hotspot” mutation for HCC, in our study the frequency percent was 37.9% (11/29), and no statistically significant correlation was found between this mutation and various clinical pathologic characteristics. We have established a routine procedure for the detection of p53 mutations in HCC samples using the FASAY procedure. The results of the present study demonstrated that FASAY is a sensitive method for detecting various p53 mutations. Additional large-scale studies are needed to determine whether p53 status may be associated with tumor progression and chemosensitivity, and whether p53 mutation have independent prognosis value in China. References B Bressac , M Kew , J Wands , M Ozturk . Selective G to T mutations of p53 gene in hepatocellular carcinoma from southern Africa . Nature , 1991 . 350 ( 6317 ): 429 - 431 . 1672732 DOI: 10.1038/350429a0 http://doi.org/10.1038/350429a0 . Selective G to T mutations of p53 gene in hepatocellular carcinoma from southern Africa PO Chappuis , A Estreicher , B Dieterich , H Bonnefoi , M Otter , AP Sappino , R Iggo . Prognostic significance of p53 mutation in breast cancer: frequent detection of non-missense mutations by yeast functional assay . Int J Cancer , 1999 . 84 ( 6 ): 587 - 593 . 10567903 DOI: 10.1002/(SICI)1097-0215(19991222)84:6<587::AID-IJC8>3.0.CO;2-8 http://doi.org/10.1002/(SICI)1097-0215(19991222)84:6<587::AID-IJC8>3.0.CO;2-8 . Prognostic significance of p53 mutation in breast cancer: frequent detection of non-missense mutations by yeast functional assay JM Flaman , T Frebourg , V Moreau , F Charbonnier , C Martin , P Chappuis , AP Sappino , IM Limacher , L Bron , J Benhattar , 等 . A simple p53 functional assay for screening cell lines, blood, and tumors . Proc Natl Acad Sci USA , 1995 . 92 ( 9 ): 3963 - 3967 . DOI: 10.1073/pnas.92.9.3963 http://doi.org/10.1073/pnas.92.9.3963 . A simple p53 functional assay for screening cell lines, blood, and tumors D Gietz , A St Jean , RA Woods , RH Schiestl . Improved method for high efficiency transformation of intact yeast cells . Nucleic Acids Res , 1992 . 20 ( 6 ): 1425 1561104 DOI: 10.1093/nar/20.6.1425 http://doi.org/10.1093/nar/20.6.1425 . Improved method for high efficiency transformation of intact yeast cells R Holmila , C Fouquet , J Cadrane , G Zalcman , T Soussi . Splice mutations in the p53 gene: case report and review of the literature . Hum Mutat , 2003 . 21 ( 1 ): 101 - 102 . 12497643 DOI: 10.1002/humu.9104 http://doi.org/10.1002/humu.9104 . Splice mutations in the p53 gene: case report and review of the literature IC Hsu , RA Metcalf , T Sun , JA Welsh , NJ Wang , CC Harris . Mutational hotspot in the p53 gene in human hepatocellular carcinomas . Nature , 1991 . 350 ( 6317 ): 427 - 428 . 1849234 DOI: 10.1038/350427a0 http://doi.org/10.1038/350427a0 . Mutational hotspot in the p53 gene in human hepatocellular carcinomas XH Huang , LH Sun , DD Lu , Y Sun , LJ Ma , XR Zhang , J Huang , L Yu . Codon 249 mutation in exon 7 of p53 gene in plasma DNA: maybe a new early diagnostic marker of hepatocellular carcinoma in Qidong risk area, China . World J Gastroenterol , 2003 . 9 ( 4 ): 692 - 695 . 12679912 . Codon 249 mutation in exon 7 of p53 gene in plasma DNA: maybe a new early diagnostic marker of hepatocellular carcinoma in Qidong risk area, China GD Kirk , AM Camus-Randon , M Mendy , JJ Goedert , P Merle , C Trepo , C Brechot , P Hainaut , R Montesano . Ser-249 p53 mutations in plasma DNA of patients with hepatocellular carcinoma from The Gambia . J Natl Cancer Inst , 2000 . 92 ( 2 ): 148 - 153 . 10639517 DOI: 10.1093/jnci/92.2.148 http://doi.org/10.1093/jnci/92.2.148 . Ser-249 p53 mutations in plasma DNA of patients with hepatocellular carcinoma from The Gambia I Meinhold-Heerlein , E Ninci , H Ikenberg , T Brandstetter , C Ihling , I Schwenk , A Straub , A Schmitt , H Bettendorf , R Iggo , 等 . Evaluation of methods to detect p53 mutations in ovarian cancer . Oncology , 2001 . 60 ( 2 ): 176 - 188 . 11244334 DOI: 10.1159/000055316 http://doi.org/10.1159/000055316 . Evaluation of methods to detect p53 mutations in ovarian cancer M Parkin , F Bray , J Ferlay , P Pisani . Global cancer statistics, 2002 . CA Cancer J Clin , 2005 . 55 ( 2 ): 74 - 108 . 15761078 . Global cancer statistics, 2002 S Sjogren , M Inganas , GT Norberg , A Lindgren , H Nordgren , L Holmberg , J Bergh . The p53 gene in breast cancer: prognostic value of complementary DNA sequencing versus immunohistochemistry . J Natl Cancer Inst , 1996 . 88 ( 3-4 ): 173 - 182 . 8632491 DOI: 10.1093/jnci/88.3-4.173 http://doi.org/10.1093/jnci/88.3-4.173 . The p53 gene in breast cancer: prognostic value of complementary DNA sequencing versus immunohistochemistry SJ Verselis , JG Rheinwald , JFJr Fraumeni , FP Li . Novel p53 splice site mutations in three families with Li-Fraumeni syndrome . Oncogene , 2000 . 19 ( 37 ): 4230 - 4235 . 10980596 DOI: 10.1038/sj.onc.1203758 http://doi.org/10.1038/sj.onc.1203758 . Novel p53 splice site mutations in three families with Li-Fraumeni syndrome F Waridel , A Estreicher , L Bron , JM Flaman , C Fontolliet , P Monnier , T Frebourg , R Iggo . Field cancerisation and polyclonal p53 mutation in the upper aero-digestive tract . Oncogene , 1997 . 14 ( 2 ): 163 - 169 . 9010218 DOI: 10.1038/sj.onc.1200812 http://doi.org/10.1038/sj.onc.1200812 . Field cancerisation and polyclonal p53 mutation in the upper aero-digestive tract J Watanabe , H Nishiyama , K Okubo , T Takahashi , Y Toda , T Habuchi , Y Kakehi , M Tada , O Ogawa . Clinical evaluation of p53 mutations in urothelial carcinoma by IHC and FASAY . Urology , 2004 . 63 ( 5 ): 989 - 993 . 15135005 DOI: 10.1016/j.urology.2003.11.031 http://doi.org/10.1016/j.urology.2003.11.031 . Clinical evaluation of p53 mutations in urothelial carcinoma by IHC and FASAY 10.1631/jzus.2007.B0081 Project (No. 39980020) supported by the National Natural Science Foundation of China 2006-09-20 2006-12-19 2022-01-05 Copyright © 2007, Journal of Zhejiang University Science 2007 Journal of Zhejiang University-SCIENCE B (Biomedicine and Biotechnology) 02 08