INTRODUCTION
Chronic, extensive ulcerative colitis (UC) is well established as a predisposing lesion for the development of colorectal adenocarcinoma1–3). There is compelling evidence that most carcinomas occuring in UC do not arise de novo but are preceded by and evolve from dysplastic mucosa, analogously to the development of sporadic colorectal carcinomas from adenomas4,5) However, the molecular basis of neoplasia in UC is not well understood at the present time.
In sporadic colon cancer, one copy of p53 is often inactivated by point mutation whereas the other copy is frequently deleted6). This inactivation and allelic loss has been known to be a late molecular event in adenoma — cancer sequence.
UC colons with dysplasia, carcinoma or both provide a valuable model for the study of genetic events during tumorigenesis. Histological progression towards cancer occurs in a stepwise fashion in UC (negative — indefinite for dysplasia-dysplasia-cancer)4). The p53 gene mutation has been demonstrated in dysplasia and cancer developed from UC. But there is no report that the mutation of p53 tumor suppressor gene is occured in normal looking mucosa (negative or indefinite for dysplasia).
Recently, there was an interesting report conducted by Robert et al7). They found the chromosomal alterations which precede the histological phenotype of dysplasia with the use of FISH and comparative genomic hybridization. But it is not certain whether the molecular event is preceded in the histological evolution. Therfore, we have conducted this study to determine the mutation of p53 gene which might be occurred in the early stage of tumorigenesis of UC.
MATERIAL AND METHOD
1. Patient and Specimens
Subjects consisted of 16 patients (12 male, 4 female), aged 44.6 on the average, and the period ranged from 1 month to 27 years in this study. Patients showing endoscopic signs of inflammation were classified as normal (N=0), mildly active (N=6), moderlately active (N=2), severe (N=8) according to Sninsky et al8). The 6 pieces of mucosal biopsy specimens in each patient were obtained from the rectal mucosa approximately 15 cm from the anal verge by sigmoidoscopy or colonoscopy. All the patients had no evidence of dysplasia-associated lesions or masses on endoscopic finding. Among the 6 specimens in each patient, 4 specimens were used for DNA extraction, 1 specimen for H&E stain and 1 specimen for immunohistochemical stainning of p53 protein. The normal mucosal specimens (control) for PCR-SSCP were obtained from the colonic mucosa which showed no evidence of inflammation on endoscopic examination.
2. Histological Evaluation
The histopathoiogy of all biopsies taken from UC and normal subjects was blindly evaluated by two gastrointestinal pathologists. Each biopsy specimen was classified as negative for dysplasia, indefinite for dysplasia, low-grade dysplasia, high-grade dysplasia according to the microscopic criteria of the Inflammatory Bowel Disease Dysplasia Morphology Study Group4).
3. DNA extraction
For analsysis of mutational assay of p53 locus, DNA was extracted from 4 specimens of normal and UC mucosa in each patient. Each specimen was put in a bath containing liquid nitrogen and homogenized finely. Powdered tissue product was added with 1 ml of STE (0.1 M Tris-HCL, 40 mM EDTA, 20% sodium dodecyl sulfate) and with proteinase K 10μl (50 mg/ml) in a microcentrifuge tube.
This mixture was reacted in a 55°C tremulous cistern for 10 to 12 hours for digesting the protein. The same volume of phenol/chloroform/isoamyl alcohol (Sigma Chemical Co, Saint Louis Mo. USA) was added to it and centrifuged at 12,000 rpm for 15 minutes. After the centrifuging, the supernatant was taken. This process was repeated 3 times. The supernatant was mixed with 3 M sodium acetate (pH 5.2, 1/10 of the supernatant) and then 100% ethanol were added with the mixture. Finally, genomic DNA was picked up and cleaned with 75% ethanol, dried and dissolved in TE (10 mM Tris pH 8.0, 1 mM EDTA pH 7.2) solution. The DNA content was measured by a spectrophotometer at 260 nm and 280 nm and then used after the confirmation of the absorbancy ratio.
4. PCR
0.5 μg genomic DNA was used for the PCR amplification. The used primer sequence was displayed in Table 1. 10 pmol of primer with a 50 μl capacity was used. The Taq DNA polymerase, 10× PCR buffer (Mg2+, 100 mM Tris-HCl pH 8.3, 500 mM KCl, 15 mM MgCl2), and dNTP mixture (2.5 mM) was used. The PCR products were pre-denaturated at 94°C for 5 minutes, annealed at 57°C for 1 minute, at 72°C for 1 minute and 30 seconds, and then at 72°C for 7 minutes. This process was repeated for 35 cycles. This PCR process was done by mastercycler 5330 (Eppedorf, Germany), and PCR products were confirmed with 2% agarose gel (Sigma, Chemical Co.). Gels were stained with ethidium bromide, photographed under UV light and we evaluated the differences in band intensities between normal and tumor DNAs.
5. PCR-SSCP
Single-strand conformational polymorphism (SSCP) analysis was used to detect for mutations with exons 4–8 of the p53 gene. We used modified cold SSCP method which did not require radioisotope and the adjustment of temperature.
15 μl of loading mixture was composed of PCR product (50 ng DNA) with 0.2% glycerol, 1 N NaOH, TBE (4°C) buffer.
The mixture was kept at 75°C for 4 minutes and we put it into ice and rapidly performed electrophoresis in the gel. A 4–20% gradient polyacrylamide gel (8.0 × 8.0 × 0.1cm, 39°C)(Novex, San Dieg, CA) was used. Electrophoresis lasted at 280 volts 25 amphere for 1 hour 45 minutes. The staining of gel was carried out by ethidium bromide (0.5 μg/ml) for 10 minutes. Destaining with the third distilled water was done for 20 minutes, and we observed the band in 340 nm UV box.
6. Silver Sequencing
The 16 μl of the mixture composed of 6% polyacrylamide gel (7 M urea), 30 ng template DNA (PCR products), and DNA sequencing buffer (10 mM Tris-HCl, pH 8.0, 1 mM EDT pH 8.0) was put into the tubes containing GATC annealing buffers and then sequencing of the PCR products was done. Initially, the mixture was predenaturated at 95°C for 2 minutes, denaturated at 94°C for 10 seconds, annealed at 42°C for 30 seconds, at 72°C for 55 seconds, and amplified at 72°C for 55 cycles. After this process, addition of 3 μl of DNA sequencing solution (10 mM NaOH, 95% formide, 0.05% bromophenol blue, 0.05% xylene cyanol) was done. This solution was kept at 70°C for 3 minutes, stuck into ice for 3 minutes. It was electrophoresised in 6% urea gel, with 60 W/1500 V for 1½ hour until bromophenol blue perished. Then, the eletrophoresised gel was kept in stop solution (10% glacial acetic acid) for 20 minutes and cleansed with the third distilled water for 4 minutes. This cleansing process was done 3 times. After cleansing, gel was put in a flask shaker with staining soution (1 g/L silver nitrate, 0.056% formaldehyde) and was shaken slowly for 30 minutes. It was kept in distilled water for 5 seconds and then kept in developing solution (30 g/L sodium carbonate, 0.056% formaldehyde, 2 mg/L sodium thiosulfate) until the band was visible. After visuallization of the band, gel was kept in stop soultion for 10 minutes, and dried at 55°C for 2 to 3 hours. We observed the sequence.
Immunohistochemical analysis
Immunostaining of the p53 protein was performed by the avidin-biotin peroxidase complex method. Briefly, 4–5 μl frozen sections were cut, mounted on polylysine-coated slides, air dried and fixed in cold (4°C) acetone for 10 minutes before being subsequently incubated with:(1) monoclonal antibody 1801 (Ab-2; Oncogene Science. Vector, Burlingame, CA.)(diluted 1:100 overnight at 4°C) which recognizes a denaturation-resistant epitope from amino acid 32 to 79 of the p53 protein;(2) a biotinylated horse antimouse IgG antibody (diluted 1:500, 30 minutes at room temperature); (3) avidin-peroxidase complexs (1 hour). Finally, the slides were developed with 0.5% diaminobenxidine in 0.05 M Tris buffer pH 7.4 containing 0.5% hydrogen peroxide, rinsed in tap water, counterstained with 5% haematoxylin, dehydrated, cleared in xylene and mounted in permanent coverslipping medium. Positive controls were used with two known positive cases of human colon cancer. Negative controls were obtained by replacement of primary antiserum with Tris buffer.
DISCUSSION
Dysplastic lesion is a marker of colon cancer risk in longstanding ulcerative colitis 9–12). So, many investigators have tried to demonstrate the molecular events in dysplastic mucosa. Among these events, the studies for p53 gene have been undertaken actively. As has been previously reported by a number of investigators, p53 alteration is evidently related to certain steps of the multi-step carcinogenesis, because mutations of p53 gene are proven to be present in a variety of types of human cancer, neurofibrosarcoma, brain tumor, osteosarcoma and colon cancer13).
Most tumors undergo inactivation of both copies of p53, one by allelic deletion and the other by point mutation13). This mechanism appears to have been operative in most of the UC associated neoplastic lesions, suggesting that in UC associated neoplasia, as in sporadic colorectal carcinoma, mutation and allelic deletion may occur closely spaced in time5). p53 LOH appears to be a relatively late genetic event in tumorigenesis in sporadic colon cancers14). On the other hand, the timing of p53 mutation appears to be different between the two diseases. Previous work by Baker et al6). indicates that although p53 mutation does precede LOH in sporadic dolon cancer, it is likely to be a relatively late event in the process of tumorigenesis, occurring during the transition from adenoma to carcinoma5,14–16). Allelic deletion of the p53 gene precedes cancer in UC and can be present in its earliest histologically identifiable precursor such as dysplastic mucosa17). And also, p53 gene mutation was demonstrated in non-invasive dysplasia and dysplasia associated lesion or mass in UC18). The recent report7) that chromosomal abnormality is proceded to the histological phenotype in UC dysplasia, this study suggested that molecular event happens earlier than dysplasia in UC dysplasia-cancer sequence.
In this study, 2 patients whose tissue showed low-grade dysplastic mucosa had p53 genetic alteration. Although the number of cases are limited, this finding corresponds with other recent reports that p53 mutation may occur earlier in UC associated colon cancer than in sporadic colon cancer. The different timing of p53 gene alteration in both diseases may potentially illustrate different roles for p53 mutation in various types of cancer. When p53 mutation occurs as a relatively early event, as appears to be the the case in UC, it may an initial growth advantage.
However, when p53 mutation occurs as a be later event, as appears to be the case in sporadic colon cancer, it may contribute to conversion to malignancy19).
Until now, there have been few studies for detecting the p53 genetic alteration in histologically indefinite or negative for dysplasia tissue. Teresa et al19). reported p53 mutation in histologically indefinite or negative tissue in colectomy specimens. However, these specimens were obtained from regions immediately adjacent to histologically abnormal mucosa. Our study showed that 2 patients, whose tissue has no evidence for dysplasia, have p53 genetic alteration. Overall, 4 cases with p53 genetic alterations were demonstrated in our study, two were negative and two were as two low-grade dysplasia. So, p53 gene mutation may appear to be a relatively early event in carcinogenesis in UC.
Interestingly, all p53 gene mutated cases in this study have shown to be a poor responder group with conventional medical treatment. But, non-mutated cases were relatively good responders to medical treatment, except one patient. However, the duration of this study was too short to evautate the possibility that p53 mutated mucosa would develop to the cancerous change. At any rate, we suggested judiciously that poor response to medical treatment may be an additional clinical risk factor of carcinogenesis in UC. Our results showed that the duration of colitis and the extent did not appear to be related with p53 mutation. But the study population was too small to postulate this assumption.
Wang et al. have reported that p53 protein had been overexpressed in 2 non-dysplastic mucosal specimens20). These 2 specimens were obtained from patients with invasive carcinoma. Also, we observed the p53 protein overexpression by immunohistochemical staining in all p53 gene mutated patients. So, our results support this data and may also support the demonstration of p53 mutations which are proven by PCR-SSCP method.
There is a limitation to our study in the collection of specimens. Many other studies have used microdissection to increase the percentage of epithelial cells to be analyzed. We think that if we had used this method, we may have obtained a higher frequency rate of mutation. In conclusion, we suggest judiciously that p53 gene mutation may play a role for the relatively early stage of carcinogenesis, and the factor of clinically poor response to medical treatment may be the predisposing factor of colon cancer.