| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1
AstraZeneca Diagnostics, Gadbrook Park, Northwich CW9 7RA, United Kingdom
2
Department of Surgery, Medical School, University of Newcastle-upon-Tyne, Newcastle-upon-Tyne NE24HH, United Kingdom
3
Department of Surgery, Queen Elizabeth Hospital, Gateshead NE965X, United Kingdom
4
AstraZenecaPharma UK, King’s Court, Wilmslow, Cheshire SK104TG, United Kingdom
5
AstraZeneca Research, Safety of Medicines, Alderley Park, Macclesfield, Cheshire SK104TG, United Kingdom
a
address correspondence to this author at: Renovo Ltd, Manchester Incubator Building, 48 Grafton St., Manchester M13 9XX, United Kingdom; fax 44-161-606-7333
b
address correspondence to this author at: AstraZeneca Diagnostics, Gadbrook Park, Northwich CW9 7RA, United Kingdom
Alterations in the p53 gene have been reported in approximately one-half of human tumors (1). Much debate surrounds the clinical significance of p53 mutations as measured directly or inferred by immunohistochemical staining (2). Many studies have addressed p53 mutational status of tumors with regard to prognosis (3)(4)(5)(6) and optimizing treatment (7)(8) in cancer.
Current recommendations categorize primary breast cancer patients on the basis of tumor size, estrogen receptor status, and histological grade (9). Improvement in the predictive power would aid patient management, particularly in cases where chemotherapy could be predicted to have no benefit. Prognosis and treatment outcomes have correlated with specific p53 mutations, particularly those affecting DNA binding (10)(11)(12). Identification of the particular p53 mutations present in a tumor may therefore have more value than monitoring for any p53 mutation or protein overexpression. Furthermore, to be clinically useful, mutations must be detectable in the presence of significant quantities of "wild-type" (wt) sequences from healthy tissues.
We extracted RNA and DNA in parallel from breast tumors and used real-time ARMSTM allele-specific amplification (13)(14)(15)(16) technology to obtain quantitative data on mutant p53 sequences in both nucleic acid pools.
Oligonucleotides were prepared by Oswel (Southampton, United Kingdom) or AstraZeneca Diagnostics (Abingdon, United Kingdom). Molecular biology-grade reagents were obtained from Sigma unless indicated. ARMS buffer consisted of 1.2 mmol/L MgCl2, 10 mmol/L Tris-HCl, 50 mmol/L KCl, pH 8.3. The reaction buffer was identical to ARMS buffer except that it contained 3.5 mmol/L MgCl2. The cassette dilution buffer consisted of 10 mmol/L Tris-HCl, 50 mmol/L KCl, 1 g/L bovine serum albumin, pH 8.3.
Blood from 50 healthy volunteers, who had given informed consent, was collected in Sarstedt Potassium-EDTA Monovettes by the Clinical Pharmacology Unit at AstraZeneca Pharmaceuticals in accordance with International Committee for Harmonization and Good Clinical Practice guidelines. Breast tumor tissue was obtained from patients at surgery who had received no neoadjuvant therapy. Tumors were snap frozen in liquid nitrogen and stored at -70 °C. Tissue sections were used for subsequent nucleic acid-based analyses.
DNA was prepared from blood via the hot alkaline lysis method (17). RNA was prepared from freshly collected blood via the Trizol method (Life Technologies), according to the manufacturer’s instructions. RNA from 6 mL of blood was stored in 100 µL of RNA secure resuspension solution (Ambion). DNA and RNA were extracted simultaneously from frozen tumor sections using the Trizol-based method (Gibco BRL Life Technologies). RNA extracted from 50 mg of frozen tissue was dissolved in 100 µL of sterile distilled water.
Multiplex PCR was performed on DNA samples derived from blood or tumors (patients 3, 6, and 9) to amplify 477-, 236-, and 415-bp fragments incorporating p53 exons (5/6), 7, and (8/9), respectively. Primers were used at the final concentrations shown:
The PCR mixture contained dNTPs (100 µM; Pharmacia Biotechnology), Taq DNA polymerase (1 U; PE Applied Biosystems), and DNA (2.5 µL) in ARMS buffer (total volume, 25 µL). Reactions were as follows: 1 cycle at 94 °C for 20 min; 40 cycles at 94 °C for 1 min and 60 °C for 1 min; 1 cycle at 72 °C for 10 min.
p53 cDNA was prepared from 5 µL of RNA stock using the Superscript One-Step RT PCR system (Life Technologies) in a total volume of 50 µL. A 680-bp fragment was prepared using forward primer 5'-tgcattctgggacagccaagtctgtga-3' (500 nM) and reverse primer 5'-gaacatctcgaagcgctcacgccca-3' (500 nM). Reverse transcription (RT)-PCRs were as follows: 1 cycle at 50 °C for 30 min; 1 cycle at 94 °C for 2 min; 35 cycles at 94 °C for 30 s, 60 °C for 30 s, and 72 °C for 45 s; 1 cycle at 72 °C for 10 min.
DNA cassettes harboring mutations in either cDNA or gDNA sequence contexts were prepared by the method of Higuchi et al. (18). Upstream half reactions used the following primers:
Downstream half reactions used the following primers:
Nested primer pairs for final cassettes were as follows:
Products were checked for homogeneity on 30 g/L agarose gels and diluted in cassette dilution buffer before real-time ARMS analysis.
ScorpionsTM primers were produced using either 6-carboxyfluorescein (FAM) or tetrachloro-6-carboxyfluorescein (TET) as the fluorophore (15). Probes were modeled using mfold (19).
Mutation detection used FAM-labeled Scorpions primers and corresponding ARMS primers (175 agc, 5'-cacagcacatgacggaggttgtgacga-3'; 175 cac, 5'-cacagcacatgacggaggttgtgaggga-3'; 245 cgc, 5'-acatgtgtaacagttcctgcatggccc-3') at 400 nmol/L and the TET-labeled quantification Scorpions primer and the corresponding reverse primer at 25 nmol/L. Non-allele-specific primers used to detect mutations in PCR products from genomic DNA were as follows (W = FAM; X = TET; Y = methyl red; Z = hexaethylene glycol):
Non-allele-specific primers used to detect mutations in p53-specific RT-PCR products from RNA were as follows:
The PCR mixture contained dNTPs (0.1 mM), Taq DNA polymerase (1 U), and template in reaction buffer (total volume, 25 µL). Samples were cycled as follows in an ABI Prism 7700 (Applied Biosystems): 1 cycle at 94 °C for 20 min; followed by 50 cycles at 94 °C for 45 s and 60 °C for 45 s. FAM and TET fluorescence was monitored without use of a quenching dye or internal control.
Data were analyzed on the 7700 using the ROX-off setting and threshold cycle (Ct) data processed in MS Excel. The TET Ct was correlated to the corresponding FAM Ct to give a single quantified well-specific data point. For each ARMS test, the three relevant data sets (wt template, 100% mutant template, and sample-derived template) were examined pairwise using a covariance methodology (20).
DNA samples from peripheral blood and tumors gave three clear bands of the expected sizes on PCR amplification (not shown). Samples from blood and tumors gave a single RT-PCR product of the expected size on amplification (not shown).
The allele-specific FAM Ct represents the proportion of mutant template
for the total amount of template indicated by the TET Ct. Thus, plots
of FAM vs TET Cts for sample, wt, and mutant templates could be used to
indicate the proportion of mutant sequences within a sample. Data from
wells containing dilutions of all templates (Fig. 1
) indicated a greater proportion of mutant sequences within the
mRNA as opposed to the genomic DNA pool. The difference in the
detectability of the individual mutations present was statistically
significant in all samples (Table 1
). The estimated percentages of mutant sequences suggested that
the mRNA pool is enriched for mutant p53 sequences by
factors of 
10 000-, 3-, and 10-fold in samples 3, 6, and 9,
respectively.
|
|
Sequence variants need to be present at concentrations of 25% or more to be detected by direct sequencing (21). Detection of somatic mutations in the DNA of neoplastic cells, which may constitute a minority of cells within a sample, requires highly sensitive mutation detection and/or neoplastic cell enrichment (13)(14)(22)(23)(24)(25). The amount of mRNA is not irrevocably tied to cell numbers and thus has the potential to amplify sequences that are relatively overexpressed in a subpopulation of cells.
We simultaneously extracted RNA and DNA from tumor samples and
used ARMS assays to detect mutant p53 sequences at the
genomic and mRNA level. The p53 mutations present within
tumor samples were quantitatively easier to detect in cDNA as opposed
to gDNA pools (Table 1
). An increase in the proportion of mutant
sequences present is the likely reason for the increase in the
statistical difference between sample and wt data sets: comparison of
the data sets from samples with those from wholly mutant templates
demonstrated that the concentrations of mutant sequence were
significantly higher within mRNA pools. The data therefore support the
suggestion that breast cells harboring the mutant p53 gene
have greater steady-state p53 mRNA concentrations than
healthy cells or tumor cells without the p53 mutation
(26). The increased concentration may make direct sequencing
of p53 cDNA a viable method for detecting a significant
fraction of p53 mutations. We estimate that mutant
p53 sequences are enriched by >10 000-fold in the cDNA
compared with the gDNA pools of sample 3 (Fig. 1
and Table 1
). The
resulting 45% concentration of mutant sequence would enable
detection by direct sequencing. Similarly, the increased proportion of
the 175 cac mutation in the p53 mRNA over the
gDNA pool of sample 6 (66% as opposed to 19%) is sufficient to enable
detection by direct sequencing. In many cases, however, the
concentrations of mutations will still be insufficient for unequivocal
assignment of mutational status by direct sequencing of cDNA; the
concentration of the 245 cgc mutation in the cDNA from tumor
9 is still well below that which could be detected by direct
sequencing.
The present study indicates that, by combining sample enrichment and sensitive detection, ARMS-based assays on RNA-derived substrates constitute a powerful method for detecting underrepresented mutant p53 sequences in breast tumor samples and enables detection of low concentrations of p53 missense mutations not otherwise detectable. Additionally, depending on the endogenous p53 mRNA concentrations in relevant samples, mutation detection could serve as a marker for detecting metastases and minimal residual disease.
Acknowledgments
This study was funded by AstraZeneca. We are grateful to Dr. Helen Ambrose and the AstraZeneca Clinical Pharmacology Unit for organizing and collecting blood samples for the wt panels. We thank Dr Simon P. Guy for helping extract RNA and Nikki Childs for oligonucleotide synthesis.
Footnotes
fax 44-1606-493-66, e-mail neil.gibson{at}astrazeneca.com
References
zuker
(accessed throughout
1999)..
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
