Clinical Chemistry Applied Biosystems
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Clinical Chemistry 46: 1562-1573, 2000;
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Inganäs, M.
Right arrow Articles by Tooke, N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Inganäs, M.
Right arrow Articles by Tooke, N.
Related Collections
Right arrow Molecular Diagnostics and Genetics
(Clinical Chemistry. 2000;46:1562-1573.)
© 2000 American Association for Clinical Chemistry, Inc.

Articles

Enzymatic Mutation Detection in the P53 Gene

Mats Inganäs1, Sara Byding1,a, Ann Eckersten1, Solveig Eriksson1, Tomas Hultman1, Anneli Jorsback1, Esfir Löfman1, Fariba Sabounchi1, Ulf Kressner2, Gudrun Lindmark3 and Nigel Tooke1

1 Amersham Pharmacia Biotech, SE-751 84 Uppsala, Sweden.

2 Department of Surgery, Uddevalla Hospital, SE-541 80 Uddevalla, Sweden.

3 Department of Surgery, University Hospital of Northern Sweden, SE-901 85 Umeå, Sweden.
a Author for correspondence. Fax 46-18-6121826; e-mail sara.byding{at}eu.apbiotech.com


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Background: The enzymatic mutation detection (EMD) assay uses the bacteriophage resolvase T4 endonuclease VII, which cleaves preformed heteroduplex molecules at mismatch sites, forming two shorter fragments that can be resolved by gel electrophoresis. The method can be used to detect single and multiple base changes, as well as insertions and deletions.

Methods: The sensitivity, specificity, and positional accuracy of mutation detection by EMD with the PASSPORTTM Mutation Scanning Kit were assessed in a blind fashion for three analytical platforms (radioactive detection and automated laser sequencers ALFexpress and ABI PRISM 377). PCR products of 703 bp covering codons 188–393 of the P53 gene were prepared from colorectal tumor samples and analyzed by EMD; the results were compared to data from cDNA sequencing. A 1362-bp PCR product prepared from IL4r gene was used to test detection of multiple base changes in long PCR products.

Results: The sensitivity for detection of mutations using EMD exceeded 90%, and the specificity exceeded 80% on all analysis platforms. The method localized 90% of mutations to within two codons and four codons for automated laser sequencers and detection by radioactivity, respectively. The method detected at least five mismatches in heteroduplexes >1 kb.

Conclusions: The EMD system facilitates efficient detection of genetic variation in fragments exceeding 1 kb irrespective of location and type. The technology is particularly well suited to the detection of mutations in genes frequently mutated at unpredictable locations.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Interest in the detection of genetic variation caused by mutations and single nucleotide polymorphisms (SNPs)1 has increased dramatically. Technically, the two types of genetic variability often are very similar, if not identical, as in the case of single nucleotide variability. The biological consequences of the two types of variability, however, may differ considerably.

Sequencing has been the gold standard for detection of genetic variability. The method has been used, for example, to detect mutations, insertions, and deletions in the P53 gene (the model gene used in this study) in the form of solid-phase sequencing of DNA and cDNA (1)(2)(3) and cycle sequencing (4). Automated sequencing equipment can also be used to specifically detect insertions and deletions (5). Although very accurate in most situations, sequencing generates large amounts of redundant information at considerable cost when used to detect genetic variability at nucleotide frequencies of 1 in every 100–1000 bp. Efficient methods for rapid scanning of genes or exons of genes for the presence of genetic variation, combined with sequencing to generate exact information of the genetic variation, could drastically reduce the cost of analysis.

The number of methods that may be used for detection of genetic variability is growing steadily. Most methods are applicable only in situations where the location and type of variability have been defined. These include the allele-specific oligonucleotide assay (6), the oligo-ligation assay (7), minisequencing (8)(9), TaqMan® (10), InvaderTM (11), pyrosequencing (12), and PCR-restriction fragment length polymorphism analysis (13).

Other methods that can detect genetic variability irrespective of the location and type of aberration are less common and, in general, less reliable. These methods include heteroduplex analysis (14), single strand conformation polymorphism (SSCP) (15), denaturing gradient gel electrophoresis (16), chemical cleavage (17), dideoxy fingerprinting (18), and denaturing HPLC (dHPLC) (19). Some of these methods have been reviewed recently (20).

A new method for detection of genetic variability, enzyme mutation detection (EMD), has recently become available under the trade name PASSPORTTM Mutation Scanning (Amersham Pharmacia Biotech, Uppsala, Sweden). EMD is based on enzymatic cleavage of mismatch-containing heteroduplexes formed by hybridizing PCR products generated from an unknown sample and a wild-type reference. Heteroduplexes are cleaved by the resolvase T4 endonuclease VII, usually in both strands within 3–6 bp 3' from the mismatch (21). Cleavage products are then detected by electrophoretic separation. The lengths of the fragment(s) generated indicate the approximate location of the mismatch, which can be confirmed by sequencing.

EMD can be used for analyzing fragments up to several kilobases in size, depending on the resolution of gel electrophoresis. At least five mismatches can be detected in a single heteroduplex. The method can also detect insertions and deletions involving one or more bases. Furthermore, the high sensitivity of EMD vs many other techniques can be used for detection of low abundance mutations in clinical specimens or to reduce the cost for analysis of SNPs by mixing samples while maintaining the probability of detecting genetic variation (22).

The functional and biochemical properties of T4 endonuclease VII have been studied in great detail (23)(24). Recently, the x-ray structure was determined (25), contributing to the understanding of the structure-function relationship of the enzyme. T4 endonuclease VII has been used for mutation detection (26)(27)(28), but its limited availability has hampered its utilization for routine application.

EMD may serve as a scanning procedure to identify aberrant samples and to get an indication of the approximate location of the aberration. This information can be used to simplify confirmatory sequencing in a limited region of the sample sequence. The use of EMD as a scanning method can therefore substantially reduce the amount of sequencing work, depending on mutation prevalence and the length of the fragment that is analyzed. The success of this strategy, however, depends on the ability of EMD to detect aberrant samples with high sensitivity and specificity.

This study was undertaken to determine whether PASSPORT has the sensitivity, specificity, and positional accuracy required of a scanning procedure to be used before confirmatory sequencing of samples containing mismatches caused by mutations. As a model system, tumor samples from patients with colorectal cancer were analyzed using PASSPORT on three analytical platforms (ABI PRISMTM 377, ALFexpressTM, and radioactive detection) for the presence of aberrations in a fragment of the P53 gene comprising codons 188–393. These data were compared with data generated previously by cDNA sequencing (2), and the accuracy in terms of sensitivity, specificity, and positional accuracy was determined. All comparisons were performed in a blind fashion. This approach also allowed an assessment of the accuracy of cDNA sequencing data generated by independent PASSPORT analysis on three analytical platforms.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
cDNA sequencing of the P53gene
cDNA sequencing was performed essentially as described (3) with some minor modifications of the RNA extraction procedure, PCR, primers, fluorescent label, and analytical platform.

Briefly, RNA was prepared from frozen tumor samples using the QuickPrepTM Total RNA Extraction Kit (Amersham Pharmacia Biotech) according to manufacturer’s instructions, under stringent conditions to avoid degradation and contamination. This procedure was followed by an enzymatic synthesis of cDNA, using RNA as template as described previously (3). Codons 188–393, extending into the flanking nontranslated 3' region, were amplified from the cDNA by PCR with two overlapping primer pairs (Table 1 ), to give fragments 3 (primers PNT3-1 and BK-1; codons 188–348; 482 bp) and 4 (primers PNT4-3 and PB4-8; codons 300–393; 445 bp). One of the primers in each primer pair was modified with biotin. Thus, biotin-labeled PCR products were generated to enable solid-phase sequencing (29). A manifold-based version of solid-phase sequencing was used (AutoLoadTM; Amersham Pharmacia Biotech) essentially as described elsewhere (30). The sequencing products generated using the primers PC3-6 (fragment 3) and PC4-S1 (fragment 4) were analyzed using an automated laser fluorescence sequencer (ALFexpress; Amersham Pharmacia Biotech). The sequence was finally compared with the P53 wild-type sequence by a prototype software program, p53 SBDecipher, Ver. 1.00 (Amersham Pharmacia Biotech). Nucleotide changes that had an impact on the protein were considered as mutations. Each mutation was verified by sequencing an entirely new PCR product from the corresponding cDNA. Additional confirmation was obtained by analysis of the neighboring PCR fragment when the mutation was located in an overlapping segment.


View this table:
[in this window]
[in a new window]
  		Table 1. Primers for cDNA sequencing and preparation of PCR products for PASSPORT.

selection of clinical specimens for comparative study using emd
The study included 178 tumor samples from patients with colorectal cancer (2) from which there were cDNA remaining from previous cDNA sequencing. The study was approved by the Ethical Committee, University of Uppsala. To maintain the blind nature of the comparative study, samples were given a new identity before the first round of PCR. A new unique identity was also given for each sample and each analytical platform before the second round of PCR to generate fragments intended for PASSPORT reactions. The integrity of the sample code remained unbroken until results on all samples for each platform were reported. This is referred to as the "blind phase of the study" as opposed to when the sample codes had been broken.

pcr
To save cDNA material, amplification was done in a nested fashion, generating the full-length open-reading frame of the P53 gene in the first PCR round. In the second round of amplification, the region covering codons 188–393 and the flanking nontranslated 3' region was amplified to give a PCR product of 703 bp, called fragment 34.

Amplification of "full-length" product from cDNA.
The PCR product covering the complete open-reading frame of P53 was generated using the following mixture in 100 µL: 1x GeneAmp PCR buffer II (PE Biosystems), 15 pmol each of primers PN8 and PN9, 1.5 mmol/L MgCl2, 200 µmol/L dNTPs (Amersham Pharmacia Biotech), 4 U of AmpliTaq Gold (PE Biosystems), and 2 µL of cDNA template. Thermocycling using touchdown was performed on a GeneAmp 9600 (PE Biosystems) as follows: 95 °C for 9 min; 10 cycles of 95 °C for 30 s, 72 °C for 30 s, and 72 °C for 45 s; 10 cycles of 95 °C for 30 s, cooling from 72 °C to 67 °C at 0.5 °C/cycle for 30 s, followed by 72 °C for 45 s; 20 cycles of 95 °C for 30 s, 66 °C for 30 s, and 72 °C for 45 s; 72 °C for 5 min; and cooling to 4 °C.

Amplification of fragment 34.
The PCR product was generated using the following mixture in 100 µL: 1x GeneAmp PCR buffer II, 15 pmol each of primers 34-1 and 34-2, 1.5 mmol/L MgCl2, 200 µmol/L dNTPs, 4 U of AmpliTaq DNA Polymerase, and 2 µL of the long PCR product obtained from the first round of amplification. The primers were modified for use on automated sequencers as follows: for the ABI 377, hexachloro-6-carboxyfluorescein (HEX)-labeled 34-1 and 6-carboxy-fluorescein (FAM)-labeled 34-2 were used; for the ALFexpress, both primers labeled with Cy5. AmpliWax Gem 100 (PE Biosystems) was used to separate reaction components, thereby enabling "hot start" PCR. Cycling conditions were as follows: 40 cycles of 95 °C for 15 s, 66 °C for 30s, 72 °C for 45 s; followed by 72 °C for 5 min, and cooling to 4 °C.

evaluation of pcr products by silver staining
PCR products (4 µL) were separated on GeneGel 24/12.5 (Amersham Pharmacia Biotech) according to the manufacturer’s instructions and using GenePhor (Amersham Pharmacia Biotech). The gels were stained in a Hoefer Autostainer using DNA silver stain (Amersham Pharmacia Biotech).

emd analysis
PASSPORT reactions were prepared for analysis of fragment 34 (703 bp) of the P53 gene covering codons 188–393 on the ABI PRISM 377, using 32P and ALFexpress II, respectively, for the presence of mutations, according to manufacturer’s instructions. Codons 205–393 were considered informative in PASSPORT analysis.

mutation databases
BRCA1 gene-related data and the mutation spectrum were retrieved from the National Human Genome Research Institute, Division of Intramural Research, Breast Cancer Information at http://www.nhgri.nih.gov/intramural_research/lab_transfer/bic/member/brca1_mutation_database.html.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
mutation detection using cDNA sequencing
The spectrum of mutations detected using cDNA sequencing in the portion of the P53 gene that was relevant in this study (codons 205–393) is shown in Fig. 1 . In total, 56 different mutations were found in 178 samples. In five samples, two mutations were found. Thus, in total 51 samples were mutated.



View larger version (11K):
[in this window]
[in a new window]
  		Figure 1. Mutation spectrum in codons 205–393 of the P53 gene in 178 colorectal tumors as determined by cDNA sequencing.

{blacksquare}, missense mutations; {dotsquare}, nonsense mutations; {square}, insertions and deletions.

samples available for analysis by passport
Depending on the success rate in the PCR, various numbers of samples fulfilled the criteria set to qualify for PASSPORT analysis on individual platforms (Table 2 ). For assessment of the cDNA sequence data, the number of samples fulfilling inclusion criteria on all analytical platforms was further reduced to 98.


View this table:
[in this window]
[in a new window]
  		Table 2. Selection of samples containing various numbers of mutations in codons 205–393 of the P53 gene as determined by cDNA sequencing.

mutation detection using passport mutation scanning on three different analytical platforms
Preliminary analysis.
Initially, we attempted to amplify the desired PCR fragment on 178 samples. After PCR products not fulfilling quality and quantity requirements for further analysis were eliminated, 143 remained for analysis on the ABI PRISM 377, 155 remained for analysis using 32P, and 114 remained for analysis on the ALFexpress.

The results shown in Table 3 indicate that there is a good agreement between PASSPORT analysis on different analytical platforms and cDNA sequencing in terms of sensitivity and specificity for mutation detection. If it is assumed that cDNA sequencing data represent the true mutation spectrum, then the sensitivity for detection of mutations, i.e., the ability to accurately detect mutations, was 90.5–100%. Similarly, the specificity, i.e., the ability to accurately detect wild-type sequence, was 78.6–86.4%.


View this table:
[in this window]
[in a new window]
  		Table 3. Evaluation of sensitivity and specificity for detection of mutated samples by analysis of a 703-bp PCR fragment informative for codons 205–393 of the P53 gene.1

Four samples that were interpreted as wild-type in the PASSPORT assay for ABI PRISM 377 did in fact contain mutations according to sequencing. When reexamined, three of the samples did contain weak cleavage products corresponding to the expected fragment sizes when compared with cDNA sequence data but were not registered as mutated during the blind phase of the study. All samples containing mutations according to sequencing were correctly identified as mutated when PASSPORT for 32P was used.

Three samples that did contain mutations according to cDNA sequencing were interpreted as wild-type by PASSPORT assay for ALFexpress.

Typical results from PASSPORT analysis of the three analytical platforms are shown in Figs. 2–4 .



View larger version (41K):
[in this window]
[in a new window]
  		Figure 2. Analysis of a 703-bp fragment of the P53 gene from sample A344 containing a missense mutation (Arg273His) by PASSPORT for ABI PRISM 377.

Forward primer labeled with FAM (top) and reverse primer with HEX (bottom). Traces from wild-type and unknown sample were superimposed in each panel. Two complementary fragments of ~452 (FAM) and 258 (HEX) bp were generated. The internal size standards are not shown.



View larger version (118K):
[in this window]
[in a new window]
  		Figure 3. Analysis of a 703-bp fragment of the P53 gene from 15 unknown samples (lanes 1–10, 12–16) from colorectal tumors for the presence of mutations using PASSPORT for radioactive labels.

The following samples were also included: a wild-type sample containing the homoduplex control (lane 11); the labeled wild-type probe (lane 17); an unrelated control template of 546 bp containing a mismatch that generates two cleavage products of ~415 and 139 bp (lane 18); and a size marker to facilitate size estimation of cleavage products (lane 19). Cleavage products representing mutations in unknown samples can easily be distinguished from the background cleavage products representative for this particular fragment and are present in all samples.



View larger version (26K):
[in this window]
[in a new window]
  		Figure 4. Analysis of a 703-bp fragment of the P53 gene from sample X743 containing a missense mutation (Arg248Glu) by PASSPORT for ALFexpress.

Cy5-labeled primers (forward and reverse) were used. Traces from wild-type (thick line) and mutated (thin line) sample were superimposed. Two complementary fragments of ~188 and 532 bp were generated. A constitutive cleavage product at ~51 bp and the full-length product of 703 bp were used as internal size markers for evaluation.

Detailed comparison of cDNA sequence data.
In all comparisons, several samples were initially classified as mutant in PASSPORT but wild-type by cDNA sequencing. This could be attributable to either false-positive results in the PASSPORT analysis or false-negative results in the cDNA sequencing. To determine which was the case, the following analysis was undertaken.

In total, 98 samples generated results on all PASSPORT platforms. The mutational status of those samples found to be mutated with respect to cDNA sequencing and EMD is shown in Table 4 . Among these, samples fulfilling the following criteria were identified:


View this table:
[in this window]
[in a new window]
  		Table 4. Summary of mutations detected by sequencing and EMD in 98 samples.1

Six samples fulfilled these selection criteria on all three analytical platforms: samples 21C10, 17H7, 17J4, 17J5, 17E9, and 17I7. When the cDNA sequence data on the six samples were reassessed, two of the six samples were confirmed to have mutations close to the locations indicated by PASSPORT analysis. One sample was incorrectly reported as wild-type despite the presence of a clear mutation in the sequence data. The other sample contained a mutation at the borderline of detection for cDNA sequencing, and when reevaluated was found to be mutated. Considering these results, we decided that these two samples (21C10 and 17H7) could be reclassified as mutated. The resulting specificities and sensitivities are shown in Table 3Up (values in parentheses). The four remaining samples might represent false negatives in sequencing (see the Discussion), although this was not confirmed in the present study. This would give sensitivities and specificities, respectively, of 91.6% and 85.3% for ABI PRISM 377, 100% and 91.3% for 32P, and 91.7% and 84.6% for ALFexpress (detailed analysis not shown).

positional accuracy of passport mutation scanning
The positional accuracy of PASSPORT Mutation Scanning was evaluated by comparing mismatch locations generated from PASSPORT analysis with data from cDNA sequencing on samples displaying single point mutations; this was a subset of the "true positives" shown in Table 3Up . The approximate locations of detected mismatches were calculated from the sizes of the complementary cleavage fragments in PASSPORT and were expressed as the codon (ABI PRISM 377) and the alternative codons (ALFexpress and 32P) identified to contain a mismatch. Finally, the deviation in location between cDNA sequencing and PASSPORT analysis, respectively, was calculated. The results are shown in Fig. 5 .



View larger version (12K):
[in this window]
[in a new window]
  		Figure 5. Positional accuracy of missense and nonsense mutations of PASSPORT for ABI PRISM 377 {square}, 32P , and ALFexpress {blacksquare}, respectively, compared with cDNA sequencing data.

Deviations are expressed in codons.

potential of passport mutation scanning
To illustrate the potential advantages of using PASSPORT for detection of genetic variability, an experiment was performed on a relatively long PCR product containing several mismatches. Furthermore, in situations where few samples are to be analyzed for genetic variability, alternative labeling procedures may be used, e.g., by incorporating labeled dUTPs during PCR. To illustrate this, a 1362-bp fragment corresponding to a part of the IL4r gene from an individual heterozygous at six different positions was prepared by incorporation of R110-dUTP during PCR (31).

The fragment was analyzed in a blind fashion using PASSPORT for the presence of mismatches according to manufacturer’s instructions for heterozygous samples. The results are shown in Fig. 6 and Table 5 .



View larger version (18K):
[in this window]
[in a new window]
  		Figure 6. Analysis, using PASSPORT for ABI PRISM 377, of a 1362-bp fragment from the IL4r gene that was heterozygous at six locations according to sequencing data.

Arrows indicate cleavage products caused by heterozygosity. See Table 5Up for explanation.


View this table:
[in this window]
[in a new window]
  		Table 5. PASSPORT analysis of part of the IL4r gene corresponding to a 1362-bp fragment from an individual heterozygous at six different locations as determined by DNA sequencing.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The primary purpose of this study was to evaluate the accuracy in terms of sensitivity, specificity, and positional accuracy of a new technology, PASSPORT Mutation Scanning, which is intended for the detection of genetic variation caused by unpredictable mutations or unknown SNPs. For the evaluation, we chose as a model system a 703-bp fragment of the P53 gene informative for codons 205–393 in which mutations at different locations may occur. The set of samples, originating from colorectal tumors, had been analyzed previously by cDNA sequencing for the presence of mutations (2). The selection of samples thus represented the relevant mutation spectrum for the P53 gene in tumors from a consecutive series of colorectal cancer patients.

To eliminate any possible operator influence, the study was performed in a blind fashion. Thus, samples were recoded before PCR amplification, and none of the operators and evaluators of EMD results knew the results from any other analytical platforms or previously performed cDNA sequencing.

The accuracy of PASSPORT for detection of samples harboring mutations was clearly acceptable. The sensitivity and specificity exceeded 90% and 80%, respectively, on all platforms. We addressed two essential questions when we assessed the accuracy of two methods: (a) Are the samples homogeneous regarding P53 mutation status? (b) If not, are the two methods equally sensitive in detecting low-abundance P53 mutations in tumors?

DNA sequencing is considered the gold standard for mutation analysis. At first sight it seems quite simple to implement large-scale DNA sequencing to identify genetic variation of any type and at any possible location. However, DNA sequencing is not free from weaknesses and limitations. For example, the analytical sensitivity is limited to 25–30% at best (32), which hampers its utility when somatic mutations are studied in cases where a mutation may be diluted below the threshold of detection of sequencing in heterogeneous tumors. Another limitation is the maximum length of template that can be reliably sequenced for the detection of heterozygotes, which seldom extends beyond 500 bases. Furthermore, the primary application of DNA sequencing technology is in de novo sequencing, with some 98–99% accuracy. Even in the very best case, i.e., when genomic DNA is used as sample source, this means that in a reasonably large gene of 1000 bases, a significant number (10–20) of false-positive base calls are generated, suggesting presence of mutations that require reanalysis to be resolved. When somatic mutations are analyzed in heterogeneous sample material, this problem will increase further.

One attempt to address the question of cDNA sequencing accuracy involved independent analysis of samples on three different PASSPORT platforms. An identical preparation of cDNA for each individual tumor sample was used for cDNA sequencing as well as for all PASSPORT platforms. The PCR products analyzed on PASSPORT were prepared by nested PCR. Any artifact that could occur during this procedure that introduced mutations in the full-length P53 product would affect subsequent analysis on the different PASSPORT platforms and the outcome of the study. Guided by the data on localization of mutations, however, we found very good agreement between PASSPORT on the three analytical platforms, respectively, and cDNA sequencing on samples carrying mutations. Furthermore, data generated on the three PASSPORT platforms suggested the presence of mutations in six samples, of which two were confirmed by reevaluating cDNA sequence data (Table 4Up ). Taken together, we found no evidence indicating that the PCR procedure in this set of samples contributed to the observed discrepancies between PASSPORT and cDNA sequence data.

In a detailed study, Finkelstein et al. (33) investigated the expression of p53 protein as a tool to characterize the potential heterogeneity of colorectal tumors. They found several typical expression patterns of p53 protein in a minor proportion of tumors, as determined by p53 immunohistochemical analysis and subsequent DNA sequencing of exons 5–8, suggesting in several cases that there were two expanding cell populations with different p53 mutational status. If tumors were heterogeneous, the detection of mutations in tumors would depend on the analytical sensitivity of the technologies used. Recent data suggest that the analytical sensitivity of EMD technology may extend below 5% of mutated vs wild-type PCR products (22), which is superior to the 25–30% that may be possible with DNA sequencing (32).

Given the experimental data available, we cannot exclude the possibility that mutations are present at low abundance in samples. These might be detectable in PASSPORT but not by cDNA sequencing, thus affecting the specificity slightly, as suggested by the results presented here. Thus, it is quite possible that four samples (17J4, 17J5, 17E9, and 17I7 in Table 4Up ), for which EMD on all three platforms had given comparable positive results, might be false negatives in sequencing because of low abundance of mutated cells in the sample. Confirmation of this interpretation would demand analysis by an independent detection method. We conclude, however, that the specificity of P53 mutations determined by PASSPORT on samples originating from colorectal tumors might be underestimated compared with a similar study performed on breast tumors (22).

There are several circumstances that decide the choice of technology for the detection of genetic variation. In the first situation, the analytical target has been predefined regarding location and type, which permits the design of tests to classify unknown samples in two or more categories.

The other situation is when genes or parts of genes contain variations at unpredictable locations or in a region that is poorly characterized (e.g., scanning for SNPs) or that contains sporadic mutations, such as the P53 gene. We advocate the use of PASSPORT in this situation. In the case of mutations, a single nucleotide base change at a frequency of 1:1000 or 1:10 000 and located in the coding region of a gene could be sufficient to alter, and possibly completely abolish, the normal function of a protein. It is important to identify any such mutation irrespective of its location and type. Omission of any part of a gene in the analysis can be justified only if the mutation spectrum has been well characterized after comprehensive analysis of many samples and the clinical implications have been evaluated thoroughly. Thus, in most cases, the complete coding region and possible splice sites of a gene should be analyzed to identify any type of aberrations located at any nucleotide position.

When new disease-causing genes are discovered, the initial mutation spectrum often is rather simple. That is sometimes true even after extensive studies, as in the case of the ras gene and colorectal cancer (34). However, it is common that the number of mutations increases as more affected patients are investigated, often producing a rather complex spectrum. The P53 (10 exons, 393 codons) and BRCA1 (22 exons, 1893 codons) genes, for example, harbor mutations at unpredictable locations. More than 700 sequence variants (as of September 1999) have been reported in the BRCA1 gene, of which most have been classified as mutations. Henceforth, all detected mutations in these genes should be regarded as potentially pathogenic because their biological implications are difficult to establish until large clinical studies have been conducted.

A commonly used technique for the detection of mutations is SSCP (15), which is based on aberrant electrophoretic behavior of single-stranded DNA when mutated is compared with wild-type. SSCP has attracted much attention because of its simplicity and low cost. SSCP has been claimed to reveal 70–95% of mutations in a fragment of 200 bp (35)(36). Beyond this size of fragment, the sensitivity of mutation detection decreases rapidly (37). Furthermore, SSCP does not reveal the location of a mutation; it only has the potential to identify aberrant fragments from normal. Very few studies have been conducted to independently compare the performance of SSCP with DNA sequencing. Recently, however, such a study revealed that only 62% of mutations were detected by SSCP when analyzed exon-wise compared with DNA sequencing. The authors concluded that SSCP analysis could not be used reliably to screen for P53 mutations (38). Another weakness of SSCP is its inability to correctly classify fragments harboring infrequent mutations and, concomitantly, an SNP at reasonable frequency. The presence of SNPs will confound correct identification of samples containing mutations.

A more recent technology for the detection of DNA variability is dHPLC (14), which utilizes the aberrant behavior of mismatch-containing heteroduplexes compared with wild-type homoduplexes when separated by ion-pair reversed-phase HPLC under partially denaturing conditions. dHPLC has been claimed to efficiently detect mutated samples, using a highly automated system with limited resource requirement (39). Most blind studies published to date have used moderately sized fragments (<600 bp), possibly because of the ready availability of existing material for evaluation. In one recent study, only 8 of 42 different amplicons exceeded 350 bp in size (40), and in another study the median size of amplicons tested was 280 bp (41). Although these studies indicate high accuracy, the performance of dHPLC should be interpreted relative to the method’s overall productivity, including the amplicon size.

In contrast, in another study PCR fragments averaging 300 bp covering 106 genes were analyzed for genetic variability by dHPLC (42). The technique gave ~40% false-positive reactions and identified only 87% of those SNPs previously detected by sequencing

PASSPORT, on the other hand, is capable of analyzing substantially larger fragments and has been shown to maintain high accuracy even when 2.2-kb amplicons are analyzed (43). This permits full exploitation of gene structure in terms of exon/intron composition. For example, some exons in BRCA1 have a size of 3–4 kb. If cDNA is prepared from samples, exons can be analyzed in a contiguous segment of several kilobases. PASSPORT has other advantages over dHPLC: it localizes the variation and can detect at least five mismatches in a fragment. This is particularly important when it is used for mutation analysis in fragments containing several SNPs, a situation that might be difficult to resolve accurately by dHPLC.

In conclusion, PASSPORT has properties that fulfill several requirements for methods to detect genetic variation. The more prominent features are (a) it has high accuracy for detection of genetic variation; (b) it can analyze large DNA fragments (<=2.2 kb); (c) it gives the approximate location of the genetic variation (90% within 2 codons); (d) it resolves genetic variation at several locations in a fragment; (e) it has high analytical sensitivity, detecting mutations in low abundance; and (f) one protocol is used for all fragments. PASSPORT also displays some weaknesses : (a) it requires PCR products of reasonable quality and quantity; (b) the cleavage efficiency varies with mismatch type and sequence context and requires some experience to interpret easily; and (c) the enzyme generates constitutive cleavage background patterns specific for each DNA fragment that may obscure weak, specific cleavage products. Used correctly, however, PASSPORT has the potential to substantially reduce costs for the analysis of mutations and SNPs while maintaining a high degree of accuracy.


   Footnotes
 
1 Nonstandard abbreviations: SNP, single nucleotide polymorphism; SSCP, single strand conformation polymorphism; dHPLC denaturing HPLC; EMD, enzymatic mutation detection; HEX, hexachloro-6-carboxyfluorescein; and FAM, 6-carboxy-fluorescein.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Berg C, Hedrum A, Holmberg A, Pontén F, Uhlén M, Lundeberg J. Direct solid-phase sequence analysis of the human p53 gene by use of multiplex polymerase chain reaction and {alpha}-thiotriphosphate nucleotides. Clin Chem 1995;41:1461-1466.[Abstract/Free Full Text]
  2. Kressner U, Inganäs M, Byding S, Blikstad I, Påhlman L, Glimelius B, Lindmark G. Prognostic value of p53 genetic changes in colorectal cancer. J Clin Oncol 1999;17:593-599.[Abstract/Free Full Text]
  3. Sjögren S, Inganäs M, Norberg T, Lindgren A, Nordgren H, Holmberg L, Bergh J. The p53 gene in breast cancer: prognostic value of complementary DNA sequencing versus immunohistochemistry. J Natl Cancer Inst 1996;88:173-182.
  4. Bharaj B, Angelopoulou K, Diamandis E. Rapid sequencing of the p53 gene with a new automated DNA sequencer. Clin Chem 1998;44:1397-1403.[Abstract/Free Full Text]
  5. Angelopoulou K, Diamandis E. Identification of deletions and insertions in the p53 gene using multiplex PCR and high-resolution fragment analysis. J Clin Lab Anal 1998;12:250-256.[Web of Science][Medline] [Order article via Infotrieve]
  6. Shuber AP, Skoletsky J, Stern R, Handelin BL. Efficient 12-mutation testing in the CFTR gene: a general model for complex mutation analysis. Hum Mol Genet 1993;2:153-158.[Abstract/Free Full Text]
  7. Delahunty C, Ankener W, Deng Q, Eng J, Nickerson DA. Testing the feasibility of DNA typing for human identification by PCR and an oligonucleotide ligation assay. Am J Hum Genet 1996;58:1239-1246.[Web of Science][Medline] [Order article via Infotrieve]
  8. Syvänen A-C, Ikonen E, Manninen T, Bengtström M, Söderlund H, Aula P, Peltonen L. Convenient and quantitative determination of the frequency of a mutant allele using solid-phase minisequencing: application to aspartylglucosaminuria in Finland. Genomics 1992;12:590-595.[Web of Science][Medline] [Order article via Infotrieve]
  9. Nyrén P, Pettersson B, Uhlen M. Solid phase DNA minisequencing by an enzymatic luminometric inorganic pyrophosphate detection assay. Anal Biochem 1993;208:171-175.[Web of Science][Medline] [Order article via Infotrieve]
  10. Holland PM, Abramson RD, Watson R, Gefland DH. Detection of specific polymerase chain reaction products by utilizing the 5' to 3' exonuclease activity of Thermus aquaticus DNA polymerase. Proc Natl Acad Sci U S A 1991;88:7276-7280.[Abstract/Free Full Text]
  11. Ryan D, Nuccie B, Arvan D. Non-PCR-dependent detection of the factor V Leiden mutation from genomic DNA using a homogeneous invader microtiter plate assay. Mol Diagn 1999;2:135-144.
  12. Ronaghi M, Karamohamed S, Pettersson B, Uhlén M, Nyrén P. Real-time DNA sequencing using detection of pyrophosphate releases. Anal Biochem 1996;242:84-89.[Web of Science][Medline] [Order article via Infotrieve]
  13. Pushnova E, Akhmedova SN, Shevtsov SP, Schwarts EI. A rapid and simple DNA fingerprinting method using RFLP and SSCP analysis of the hypervariable noncoding region of human mitochondrial DNA. Hum Mutat 1994;3:292-296.[Web of Science][Medline] [Order article via Infotrieve]
  14. White MB, Carvalho M, Derse D, O’Brien SJ, Dean M. Detecting single base substitutions as heteroduplex polymorphisms. Genomics 1992;12:301-306.[Web of Science][Medline] [Order article via Infotrieve]
  15. Orita M, Iwahana H, Kanazawa H, Hayashi K, Sekiya T. Detection of polymorphisms of human DNA by gel electrophoresis as single-strand conformation polymorphisms. Proc Natl Acad Sci U S A 1989;86:2766-2770.[Abstract/Free Full Text]
  16. Fisher SG, Lerman LS. DNA fragments differing by single base-pair substitutions are separated in denaturing gradient gels: correspondence with melting theory. Proc Natl Acad Sci U S A 1983;80:1579-1583.[Abstract/Free Full Text]
  17. Cotton RGG, Rodrigues NR, Campbell RD. Reactivity of cytosine and thymine in single-base-pair mismatches with hydroxylamine and osmium tetroxide and its application to the study of mutations. Proc Natl Acad Sci U S A 1988;85:4397-4401.[Abstract/Free Full Text]
  18. Sarkar G, Yoon H, Sommer S. Dideoxy fingerprinting (ddE): a rapid and efficient screen for the presence of mutations. Genomics 1992;13:441-443.[Web of Science][Medline] [Order article via Infotrieve]
  19. Oefner PJ, Underhill PA. Comparative DNA sequencing by denaturing high-performance liquid chromatography (DHPLC) [Abstract]. Am J Hum Genet 1995;57(Suppl):A266.
  20. Taylor GR. Enzymatic and chemical cleavage methods [Review]. Electrophoresis 1999;20:1125-1130.[Web of Science][Medline] [Order article via Infotrieve]
  21. Pottmeyer S, Kemper B. T4 endonuclease VII resolves cruciform DNA with nick and counter-nick and its activity is directed by local nucleotide sequence. J Mol Biol 1992;223:607-615.[Web of Science][Medline] [Order article via Infotrieve]
  22. Del Tito BJ, Jr, Poff HE, III, Novotny MA, Cartledge DM, Walker RI, II, Earl CD, Bailey AL. Automated fluorescent analysis procedure for enzymatic mutation detection. Clin Chem 1998;44:731-739.[Abstract/Free Full Text]
  23. Kemper B. Branched DNA resolving enzymes (X-solvases). In: Nickoloff JA, Hoekstra MF, eds. DNA damage and repair, Vol. 1. Totowa, NJ: Humana Press, 1998:179–204..
  24. White MB, Giraud-Panis M-JE, Pöhler JRG, Lilley DM. Recognition and manipulation of branched DNA structure by junction-resolving enzymes. J Mol Biol 1997;269:647-664.[Web of Science][Medline] [Order article via Infotrieve]
  25. Raaijmakers N, Vix O, Töro I, Golz S, Kemper B, Suck D. X-ray structure of T4 endonuclease VII: a DNA junction resolvase with a novel fold and unusual domain-swapped dimer architecture. EMBO J 1999;18:1447-1458.[Web of Science][Medline] [Order article via Infotrieve]
  26. Youil R, Kemper B, Cotton RGH. Screening for mutations by enzyme mismatch cleavage with T4 endonuclease VII. Proc Natl Acad Sci U S A 1995;92:87-91.[Abstract/Free Full Text]
  27. Mashal RD, Koontz J, Sklar J. Detection of mutations by cleavage of DNA heteroduplexes with bacteriophage resolvases. Nat Genet 1995;9:177-183.[Web of Science][Medline] [Order article via Infotrieve]
  28. Youil R, Kemper B, Cotton RGH. Detection of 81 of 81 known mouse ß-globin promotor mutations with T4 Endonuclease VII—The EMC method. Genomics 1996;32:431-435.[Web of Science][Medline] [Order article via Infotrieve]
  29. Hultman T, Ståhl S, Hornes E, Uhlén M. Direct solid phase sequencing of genomic and plasmid DNA using magnetic beads as solid support. Nucleic Acids Res 1989;17:4937-4946.[Abstract/Free Full Text]
  30. Lagerkvist A, Stewart J, Lagerström-Fermér M, Landegren U. Manifold sequencing: efficient processing of large sets of sequencing reactions. Proc Natl Acad Sci U S A 1994;91:2245-2249.[Abstract/Free Full Text]
  31. Olavesen M, Hampe J, Mirza MM, Saiz R, Lewis CM, Bridger S, et al. Analysis of single-nucleotide polymorphisms in the interleukin 4 receptor gene for association with inflammatory bowel disease. Immunogenetics 2000;51:1-7.[Web of Science][Medline] [Order article via Infotrieve]
  32. Leitner T, Halapi E, Scarlatti G, Rossi P, Albert J, Fenyö EM, Uhlén M. Analysis of heterogeneous viral populations by direct DNA sequencing. Biotechniques 1993;15:120-126.[Web of Science][Medline] [Order article via Infotrieve]
  33. Finkelstein SD, Przygodzki R, Pricolo VE, Sakallah SA, Swalsky PA, Bakker A, et al. Prediction of biological aggressiveness in colorectal cancer by p53/K-ras-2 topographic typing. Mol Diagn 1996;1:5-28.[Medline] [Order article via Infotrieve]
  34. Kressner U, Björheim J, Westring S, Wahlberg SS, Påhlman L, Glimelius B, et al. Ki-ras mutations and prognosis in colorectal cancer. Eur J Cancer 1998;34:518-521.
  35. Grompe M. The rapid detection of unknown mutations in nucleic acids. Nat Genet 1993;5:111-117.[Web of Science][Medline] [Order article via Infotrieve]
  36. Curtis D, Durkie M, Balac (Morris) P, Sheard D, Goodeve A, Peake I, et al. A study of Wilson disease mutations in Britain. Hum Mutat 1999;14:304-311.[Web of Science][Medline] [Order article via Infotrieve]
  37. Sheffield VC, Beck JS, Kwitek AE, Sandstrom DW, Stone EA. The sensitivity of single-strand conformation polymorphism analysis for the detection of single base substitutions. Genomics 1993;16:325-332.[Web of Science][Medline] [Order article via Infotrieve]
  38. Tolbert DM, Noffsinger AE, Miller MA, DeVoe GW, Stemmermann GN, Macdonald JS, Fenoglio-Preiser CM. p53 immunoreactivity and single-strand conformational polymorphism analysis often fail to predict p53 mutational status. Mod Pathol 1999;12:54-60.[Web of Science][Medline] [Order article via Infotrieve]
  39. O’Donovan MC, Oefner PJ, Roberts SC, Austin J, Hoogendoorn B, Guy C, et al. Blind analysis of denaturing high-performance-liquid chromatography as a tool for mutation detection. Genomics 1998;52:44-49.[Web of Science][Medline] [Order article via Infotrieve]
  40. Giordano M, Oefner PJ, Underhill PA, Cavalli Sforza LL, Tosi R, Momigliano Richiardi P. Identification by denaturing high-performance liquid chromatography of numerous polymorphisms and a candidate region for multiple sclerosis susceptibility. Genomics 1999;56:247-253.[Web of Science][Medline] [Order article via Infotrieve]
  41. Jones AC, Austin J, Hansen N, Hoogendoorn B, Oefner P, Cheadle JP, O’Donovan MC. Optimal temperature selection for mutation detection by denaturing HPLC and comparison to single-stranded conformation polymorphism and heteroduplex analysis. Clin Chem 1999;45:1133-1140.[Abstract/Free Full Text]
  42. Cargill M, Altshuler D, Ireland J, Sklar P, Ardlie K, Patil N, et al. Characterization of single-nucleotide polymorphisms in coding regions of human genes. Nat Genet 1999;22:231-238.[Web of Science][Medline] [Order article via Infotrieve]
  43. Ford ME, Whitcomb DC. Analysis of the hereditary pancreatitis-associated cationic trypsinogen gene mutations in exons 2 and 3 by enzymatic mutation detection from a single 2.2 kb polymerase chain reaction product. Mol Diagn 1999;4:1-7.[Web of Science][Medline] [Order article via Infotrieve]



The following articles in journals at HighWire Press have cited this article:


Home page
 Clin. Chem.Home page
T. Norberg, S. Klaar, L. Lindqvist, T. Lindahl, J. Ahlgren, and J. Bergh
Enzymatic Mutation Detection Method Evaluated for Detection of p53 Mutations in cDNA from Breast Cancers
Clin. Chem., May 1, 2001; 47(5): 821 - 828.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Chem.Home page
E. P. Diamandis
Sequencing with Microarray Technology--A Powerful New Tool For Molecular Diagnostics
Clin. Chem., October 1, 2000; 46(10): 1523 - 1525.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Inganäs, M.
Right arrow Articles by Tooke, N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Inganäs, M.
Right arrow Articles by Tooke, N.
Related Collections
Right arrow Molecular Diagnostics and Genetics

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS