| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Oncology, Markers, Clinical Correlates |
Department of Surgical Oncology, Chelsea and Westminster Hospital [Z. A. J. K., S. K. J., N. L-M., H. P., R. Q. W., A. T., A. I., T. G. A-M.], and Advanced Biotechnology Centre, Charing Cross Hospital, Imperial College School of Medicine [A. T., A. I.], London SW10 9NH, England
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Circulating tumor cells could provide a relatively noninvasive and repeatable source of genotypic information that might influence treatment and estimation of prognosis. We selected the p53 gene for analysis, because loss of wild-type p53 protein reduces its efficient involvement in apoptosis (7) and is associated with reduced tumor sensitivity to chemotherapy (8) and radiotherapy (9) as well shorter patient survival (10) . The extent to which these mutations vary within solid tumor deposits and circulating tumor cells from the same patient is unclear. Loss of p53 wild-type protein function results from gene mutations that are clustered within the five most highly conserved p53 gene domains (11) . These mutations can be identified by direct gene sequencing. Multiple mutational analysis permits comparisons between primary colorectal tumors, liver metastases, and circulating tumor cells within the same patient to determine the extent to which similar mutations from each source are identified.
We developed a technique for identifying p53 mutations in tumor cells isolated from the peripheral venous blood of colorectal cancer patients and compared the prevalence and position of these mutations within multiple solid tumor samples from the same patient.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Biopsies of normal colon and liver were obtained from patients with no history of cancer undergoing surgery for nonmalignant conditions such as rectal prolapse, sigmoid volvulus or liver trauma, snap frozen in liquid nitrogen, and stored at -70°C until analysis. Three 7-ml blood samples were collected via a peripheral i.v. cannula into vacutainers containing sodium-EDTA, at the time of anesthesia induction, from patients with no history of cancer undergoing intermediate surgical procedures, such as groin hernia repair. Samples were stored on ice and processed within half an hour of venesection. All patients gave written informed consent to be included in this study, which had been approved by Chelsea and Westminster Hospital Ethics Committee.
p53 Mutational Analysis
Solid Tumor Samples
Samples were taken from diametric edges and the center of primary
tumor and liver metastasis biopsies to produce three discrete tumor
samples per biopsy, which were then separately processed. Histological
confirmation of adenocarcinoma was obtained in all cases.
Circulating Tumor Cell Isolation
RBCs were lysed with red cell lysis buffer (Qiagen, United
Kingdom), retaining all nucleated cells including tumor cells.
Immunomagnetic beads (anti-epithelial Dynabeads, Crawley, Dynal, United
Kingdom) coated with the epithelial specific antibody Ber EP4 were
applied according to the manufacturer’s instructions to select for
epithelial cell surface characteristics found on colorectal cancer but
not on WBCs (14)
. Cell binding to the beads was confirmed
by microscope examination. Several washes ensured that the majority of
nonepithelial cells were removed from the wall of the tubes.
RNA Extraction
RNA was extracted using a modified method of Chomczynski and
Sacchi (15)
either from 10 mg of homogenized tissue or
from epithelial cells isolated from the blood, washed in DEPC-75%
ethanol before being dissolved in DEPC-water, and stored at -70°C
until further analysis. For blood samples, the RNA yield was maximized
by the addition of 10 µl of polyinosinic acid solution (16 g/l)
to each sample. All glassware was rinsed in DEPC-water and autoclaved,
and solutions were made up in DEPC-water. For tissue samples, the
amount of total RNA was determined spectrophotometrically by measuring
absorbance of the sample at 260 nm. The purity of the total RNA was
established by confirming that the 260 nm:280 nm ratio was within the
1.8–2.0 range, indicating that RNA preparations were free of protein
contaminants.
RT-PCR
Solid Tissue Samples.
cDNA was prepared from two µg of total RNA using 100 units of Moloney
murine leukemia virus reverse transcriptase (Life Technologies,
Paisley, United Kingdom), 40 units of RNase inhibitor (Promega,
Southampton, United Kingdom), 8 µl of reaction buffer [250
mM Tris-HCl (pH 8.3), 375 mM KCl, and 15
mM MgCl2], and 10 mM
each deoxynucleotide triphosphate (Promega). Twenty pmol of random
hexamers (Clontech Laboratories, Cambridge, United Kingdom) were added,
and primer annealing was performed by incubation for 2 min at 70°C,
followed by rapid quenching on ice for 5 min. Reverse transcription was
performed by incubating the samples for 1 h at 42°C and then
heating to 94°C for 5 min to inactivate the reverse transcriptase.
The final volume of the cDNA reaction sample was 40 µl, and this was
increased to 80 µl by adding DEPC-H2O. Twenty
µl of this cDNA solution were then used for the first round of nested
PCR on all samples using Advantage-HF 2 (High-Fidelity) PCR kit
(Clontech, Palo Alto, CA), which contains Advan Taq DNA
polymerase and a proof-reading polymerase and TaqStart Antibody.
Advantage-HF 2 is designed to achieve high fidelity during PCR
amplification and was developed for mutational analysis
(16, 17, 18)
. The kit was applied according to the
manufacturer’s recommendations using only the high fidelity buffer,
resulting in a final PCR volume of 50 µl. Two sets of specifically
designed primers (selected using MacVector) spanning the p53
coding region between exons 4 and 9 were used. The first step used
primers outside this region (exon 3 forward and exon 10 reverse). The
primer sequences were: exons 3–6, sense primer
CCCTCTGAGTCAGGAAACATTTTC and antisense primer AGTGGATGGTGGTACAGTCAGAGC
(size, 665 bp); and exons 5/6–10, sense primer GCAGCTGTGGGTTGATTCCACA
and antisense primer GCCTGGGCATCCTTGAGTTC (size, 669 bp). In the second
step, 2 µl of the first PCR product were removed and reamplified
using two sets of overlapping inner primers, which ensured that all of
the conserved region of the p53 gene was amplified. The
primer sequences were: exons 4–6, sense primer TGTCCCCGGACGATATTGAAC
and antisense primer TTCCTTCCACTCGGATAAGATGC (size, 465 bp); and exons
5/6–9, sense primer GCTCAGATAGCGATGGTCTGGC and antisense primer
TCTCGGAACATCTCGAAGCG (size, 484 bp). The PCR conditions for the
first reaction were 94°C for 1 min for one cycle, followed by 20
cycles at 94°C for 30 s and 68°C for 4 min. The final
extension step was 68°C for 3 min. Conditions for the second round of
PCR were similar to the first but with 28 cycles of amplification. All
PCR products were directly sequenced.
Blood Samples.
The RT-PCR procedure followed for blood samples was the same as for
tissues, with the same number of first- and second-round PCR cycles,
except that the entire RNA sample was reverse transcribed and the final
volume of 40 µl of cDNA reaction sample was left undiluted. In the
second round of nested PCR, 6 µl of the first PCR product were used.
This policy was adopted because of a lower RNA content with blood
samples compared with solid tissue.
Each RT-PCR run included a positive control produced from either the HT29 or HT115 colorectal cancer cell line and a negative control where target cDNA was substituted by nuclease free water. The Advantage-HF kit used for the PCR contained a 2-kb DNA template that was included as a further positive control in each reaction. The quality of cDNA was assessed using primers for the ubiquitous housekeeping gene ß-microglobulin. To assess RNA integrity and quality, formaldehyde RNA gels were run at differing intervals on randomly selected samples.
Automated Sequencing
All of the PCR products were analyzed by ABI Prism 373 fluorescent
dye terminator sequencing (PE/Applied Biosystems, Foster City,
CA). To ensure accuracy, PCR products were restricted to <500
bp in size, which is within the sequencing range of the ABI prism 373.
The operator undertaking sequencing was unaware of the source of any
sample.
Internal Assay Controls
Tissue or blood samples from no-cancer controls were included in
each PCR run through to the sequencing stage, alongside samples from
cancer patients. The analysis was repeated from the cDNA stage in 60%
of samples. Analysis of samples showing p53 mutations was
repeated up to three times to ensure consistency, and results were
invariably identical.
Cell Spiking Studies
The colonic cancer cell line HT 29 was used to test the in
vitro sensitivity of this method in detection of tumor cell
p53 mutation in blood. Initial analysis confirmed a single
arginine- to-histidine (CGT
CAT) mutation at codon 273 in triplicate
HT29 cell samples. Seven-ml whole blood samples from healthy donors
were "spiked" with between 20 and 1 x
104 HT29 cells. The codon 273 mutation was
consistently detected in triplicate experiments to a dilution of 50
HT29 cells in the 7-ml whole blood sample. Nonspiked control blood
samples, which were included with each experiment, were negative on
sequencing for p53 mutation.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
p53 Mutations
p53 mutation was detected in 19 (46%) patients
(Dukes’ stage A, 3 of 5 patients; Dukes’ stage B, 3 of 9; Dukes’
stage C, 2 of 9; metastases, 11 of 18). The source, site, and pattern
of the p53 mutations detected are shown in Table 1
.
|
), identical p53 mutations
were found in all six samples examined (three from each metastasis
biopsy) from the two discrete liver metastases within the same patient.
p53 mutation was not identified in any of the third
patient’s liver metastasis biopsies.
p53 Mutations in Peripheral Blood from Colorectal Cancer
Patients
Eight p53 mutations were identified in cells isolated
from peripheral blood of eight colorectal cancer patients (Dukes’
stage A, 0 of 5 patients; Dukes’ stage B, 1 of 9; Dukes’ stage C, 2
of 7; metastases, 5 of 20). These peripheral blood mutations were
invariably identical to those found in corresponding solid tumor
samples.
p53 Mutation in Corresponding Primary Tumors, Liver Metastases, and
Peripheral Blood from Colorectal Cancer Patients
Five of the 41 patients had synchronous liver metastases that were
biopsied at the time of primary tumor excision. Four p53
mutations were identified in solid tumor samples from four of these
patients (cases 9, 14, 15, and 19; Table 1
). In all four cases, the
pattern of p53 mutation was identical in all six samples
(three samples per biopsy) from matching primary tumor and liver
metastasis biopsies. No p53 mutation was identified in any
biopsies from the fifth patient. The same mutation as in the matching
primary tumor and liver metastasis was also identified in peripheral
blood from two of these synchronous liver metastasis patients (cases 9
and 19; Table 1
), whereas no peripheral blood p53 mutation
was identified in the remaining two patients with primary tumor and
liver metastasis p53 mutations.
"No-Cancer" Control Tissue and Peripheral Blood
p53 mutations were not identified in any of the three
samples per biopsy examined from 12 normal colon and 3 normal liver
biopsies. Examination of triplicate peripheral blood samples from 10
"no-cancer" control patients also did not reveal p53
mutations. All control blood samples gave p53 amplification,
but sequencing revealed only wild-type p53.
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
5' exonuclease activity that can remove mispaired bases
(proofreading activity; Refs. 21
and 22
). To
reduce the risk of transcription error, we used a high fidelity
polymerase with high fidelity buffer and minimized the number of nested
PCR cycles (13)
. It is not clear whether sampling variation could result in a tumor p53 mutation being overlooked where only a single tumor sample is analyzed (23) . Our results with multiple tumor samples suggested a 46% rate of p53 mutation in colorectal carcinoma that is consistent with previous estimates based on p53 gene sequencing of single colorectal cancer samples (24) . We found that where tumor p53 mutation was identified, this was invariably detected in all samples from the same solid tumor and in other tumor metastases within the same patient. This suggests that a single colorectal cancer biopsy is capable of providing reliable p53 gene mutational information.
Because cancers are thought to develop by clonal evolution (25) , the reason for such an apparently monoclonal population within each of our patients is not clear. One previous study (26) has reported discordant p53 mutations between primary colorectal cancer and lymphatic metastases, but this study did not use high fidelity Taq polymerase for PCR, and there may have been an increased risk of transcription error. Other smaller studies (27 , 28) have reported results that are consistent with those in the present study, whereas an additional study (29) also suggested similar results in colorectal but not breast cancer. The present method could not determine whether p53 wild-type tumor cells were also present in p53 mutated carcinomas, and an advantageous p53 mutation within a nonmutated or previously p53-mutated clone might arise. The resulting tumor cell clone bearing different p53 mutations would not be inconsistent with our findings, if the clone size was below the detection threshold. However, clonal divergence in colorectal cancer is thought to occur earlier in the adenoma-carcinoma sequence (30) than the relatively late metastatic stage involved in the present study, and the p53 mutant clone we detected is likely to have predominated over other p53-mutated clones. Thus one explanation for the monoclonal and single p53 mutational pattern detected is that after one p53 mutation, further p53 mutation did not provide additional tumor advantage because p53 mutational benefits had already been gained.
One important technique-related determinant of sensitivity to circulating tumor cell p53 mutation is removal of hematogenous cells to prevent the tumor cell mutant p53 signal being overwhelmed by wild-type p53 from WBCs. We found that multiple washes, at the stage of tumor cell binding with immunomagnetic beads, were essential to clear the beads of WBCs, the wild-type p53 signal of which could suppress any mutant p53 cell signal. Reconstitution experiments using cloned viral material suggest that the direct sequencing method is capable of detecting minor sequence variants when present in as little as 10% of the total viral population, but detection of sequence variants is unlikely at lower levels (31) . The finding of wild-type p53 amplification in blood samples from noncancer control patients implies that immunomagnetic bead extraction did not completely clear hematogenous cells. Loss of sensitivity for detection of circulating cancer cells with epithelial characteristics has also been reported where RT-PCR is performed in the presence of hematogenous cell contamination (32) . Thus, improved tumor cell enrichment from blood might increase the sensitivity of the current method. In addition, wild-type p53 amplification in blood samples from noncancer control patients could have been produced by venesection needle-cored epithelial cells (13) or perhaps by other circulating cells with epithelial characteristics (33) .
p53 mutation in circulating tumor cells could only be
identified in a minority (42%) of patients with p53-mutated
solid tumors. A second factor influencing the sensitivity of
circulating tumor cell detection is aggregation of circulating tumor
cells into clumps of varying sizes (34)
. This results in
sample-to-sample variation in detection of tumor cells contained within
different blood samples from the same patient (35)
. This
may explain the finding that p53 mutation in patients with
p53-mutated tumors was identified in only one of three
peripheral blood samples in contrast to all solid tissue samples (Table 1)
taken from each patient.
Single sites of oncogene mutation have been identified previously in peripheral blood from colorectal cancer patients (36) . Our results suggest that this can now be extended to analysis of multiple mutational sites in circulating tumor cells and can be applied to the p53 gene. The versatility and convenience of peripheral blood sampling make circulating tumor cells attractive sources of information about tumor genotype, and the development of techniques based on comparative genomic hybridization (37) could increase the information obtained. This approach might ultimately provide information about tumor sensitivity to treatment and prognosis from a blood test.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 R. Q. W. and H. P. were Stefan Galeski
Research Fellows; S. K. J. and N. L-M. were supported by Colon
Cancer Concern, London, United Kingdom; and Z. A. J. K. was
supported by a research grant from the Trustees of Chelsea and
Westminster Hospital, London, England, United Kingdom. 
2 To whom requests for reprints should be
addressed, at Department of Surgery, Chelsea and Westminster Hospital,
369 Fulham Road, London SW10 9NH, United Kingdom. Phone: 020-8746-8468;
Fax: 020-8746-8231; E-mail: t.allenmersh{at}ic.ac.uk
3 The abbreviations used are: RT-PCR, reverse
transcription-PCR; DEPC, diethyl pyrocarbonate.
Received 5/ 2/00; revised 6/16/00; accepted 6/23/00.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
bacteriophage templates. Proc. Natl. Acad. Sci. USA, 91: 2216-2220, 1994.This article has been cited by other articles:
|
|
 |
|
  A. Russo, V. Bazan, B. Iacopetta, D. Kerr, T. Soussi, N. Gebbia, and for the TP53-CRC Collaborative Study Group The TP53 Colorectal Cancer International Collaborative Study on the Prognostic and Predictive Significance of p53 Mutation: Influence of Tumor Site, Type of Mutation, and Adjuvant Treatment J. Clin. Oncol., October 20, 2005; 23(30): 7518 - 7528. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Fehm, A. Sagalowsky, E. Clifford, P. Beitsch, H. Saboorian, D. Euhus, S. Meng, L. Morrison, T. Tucker, N. Lane, et al. Cytogenetic Evidence That Circulating Epithelial Cells in Patients with Carcinoma Are Malignant Clin. Cancer Res., July 1, 2002; 8(7): 2073 - 2084. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
