|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Molecular Diagnostics and Genetics |
1 Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN.
aAddress correspondence to this author at: Laboratory Genetics, 920 Hilton Bldg., Mayo Clinic, 200 First St. SW, Rochester, MN 55905. Fax 507-284-0670; e-mail sthibodeau{at}mayo.edu.
| Abstract |
|---|
|
|
|---|
15 codons of the RET gene have been linked to disease phenotypes in the vast majority of cases. These missense mutations range from very simple single nucleotide base changes to more numerous changes at a given codon; they therefore are often tested for by more than one DNA-based diagnostic method. We developed and evaluated a PyrosequencingTM technology-based approach for MEN2 mutation testing that allows both simple and complex mutations to be analyzed on one platform. Methods: Archived DNA from peripheral blood of patients referred to the Mayo Clinic Molecular Genetics laboratory for MEN2 testing was selected. One to all of codons 609, 611, 618, 620, 630, 634, 768, 804, and 918 were analyzed by Pyrosequencing technology to match the original analysis of each patient. Template PCRs were set up using an automated liquid handler; the subsequent post-PCR preparation step was performed manually, and the sequencing was performed by a PSQ 96 instrument. Samples were tested in batch sizes expected to occur routinely.
Results: We analyzed samples from 217 patients who previously tested negative for MEN2 and 230 patients who previously tested positive, for a total of 1449 sequencing reactions. One discrepant result was found (100% concordant for negatives and 99.6% concordant for positives). A total of 37 unique mutations or alterations of unknown significance were analyzed.
Conclusion: Pyrosequencing technology offers an accurate, nonisotopic, simple, and rapid method for the analysis of DNA from patients suspected of having MEN2.
| Introduction |
|---|
|
|
|---|
Missense mutations in the RET protooncogene have been shown to cause MEN2, and these occur primarily in exons 10, 11, 13, 14, 15, and 16 (1). Within these exons, 13 codons (609, 611, 618, 620, 630, 634, 768, 790, 791, 804, 883, 891, and 918) are altered in the vast majority of cases. Most of the reported mutations are clustered in exons 10 and 11, with the remaining scattered among exons 13, 14, 15, and 16.
An association between alterations at specific codons and the MEN2 subtypes has been shown (2)(3). For MEN2A, a single missense mutation in any one of six cysteine residues in exon 10 (codons 609, 611, 618, and 620) or 11 (codons 630 and 634) is the primary causative factor in
98% of cases (1). Infrequently, exon 13 (codons 790 and 791) (4)(5)(6) and 14 (codon 804) (7) mutations are associated with MEN2A. For MEN2B, missense mutations in codon 883 of exon 15 or codon 918 of exon 16 account for
5% and 95% of cases, respectively (1)(8)(9). For familial MTC, mutations in the same codons associated with MEN2A plus two additional codons in exons 13 (codon 768) and 15 (codon 891) are found in
93% of cases (1). Because of these findings, genetic testing has now become an important clinical tool and is frequently used to help establish the diagnosis of MEN2 and to predict the occurrence of disease in at-risk individuals.
For mutations found in codons 609634, mutations can occur at any one of the three bases within the codon. In addition, for any one of the three bases within the codon, the number of base substitutions can vary from several to all of the other three possible bases. In short, any base substitution that leads to a loss of the cysteine residue generally produces abnormal protein function and must be tested for. In contrast, the missense mutations that occur in the remaining affected codons (768918) typically occur only at one of the bases and are substituted by only one or two of the other possible bases. Because of the overall complexity, various methods have historically been used for gene testing. The most efficient testing strategy for some of the less complex MEN2 mutations (codons 768918) have involved techniques such as restriction fragment length polymorphism analysis or a fluorescence resonance energy transfer/hybridization probe-based method (10). However, because of the more complex nature of the remaining alterations (codons 609634), single-strand conformation polymorphism analysis (11), conformation-sensitive gel electrophoresis (12), or manual or automated sequencing (13) have frequently been used. In either case, these methods all involve a PCR step followed by several other post-PCR techniques, such as restriction enzyme digestions, additional thermocycler-based sequencing reactions, and/or the preparation, loading, and running of gels. Depending on the overall testing strategy, several of these methods may be used to perform MEN2 testing. As a result, skills in many different methods may be required.
Because of the various types of mutations observed with MEN2, a single DNA diagnostic method that is simple, inexpensive, and efficient for the detection of all predominant mutations would be of great use. The goal of this study was to develop and evaluate such a method for the detection of mutations within the RET protooncogene using the PyrosequencingTM platform.
| Materials and Methods |
|---|
|
|
|---|
dna extraction
Archived DNA (stored at -70 °C) was originally obtained from peripheral whole blood samples by phenolchloroform extraction, extraction with the Puregene DNA Purification Kit (Gentra Systems), or extraction with the QIAamp 96 DNA Kit (Qiagen Inc.). Phenolchloroform- and Puregene-extracted DNA was suspended in Tris-EDTA buffer (10 mmol/L Tris, 1 mmol/L EDTA, pH 8.0), and the concentration was adjusted to 250 ng/µL. Qiagen-extracted DNA was resuspended in the manufacturer-supplied buffer and volume.
template generation
Five pairs of primers (IDT Technologies) were designed to amplify five regions of the RET gene. One primer of each pair has a 5' biotin-triethylene glycol label necessary for post-PCR processing. The PCR template primers and the amounts of MgCl2 and AmpliTaq® Gold Polymerase (Applied Biosystems) used in this study are listed in Table 1
. Pyrosequencing template was made by PCR amplification 2.0 µL (50500 ng) of genomic DNA in a 25-µL reaction in the presence of 5 pmol of each forward and reverse primer, 200 µM each deoxynucleotide triphosphate, 50 mM KCl, 10 mM Tris (pH 8.3), and 0.01 µg/µL gelatin. Template PCRs were set up on a MultiPROBE II HT liquid handler (PerkinElmer, Inc.) in batches of
15 (patients + controls) to mimic routine conditions and to test the day-to-day performance of the entire system (template generation, post-PCR processing, and operation and analysis on the Pyrosequencing instrument). Thermocycling conditions for all PCRs were 35 cycles of 30 s at 95 °C, 30 s at the temperature listed in Table 1
, and 30 s at 72 °C. A final 72 °C hold for 10 min ended the reaction.
|
sequencing primers
Sequencing primers were designed to sequence six areas within the five regions amplified. Primer sequences, the template targeted, and the codons analyzed are shown in Table 2
. Three of the sequencing primers each interrogated two codons, for a total of nine codons analyzed.
|
template preparation
From the PCR-generated template, the desired strand to be sequenced was separated from its complement, and an appropriate sequencing primer was annealed to it according to the manufacturers instructions. Briefly, 25 µL of PCR-generated template was incubated with 90 µg of streptavidin-coated Dynabead M-280 magnetic beads (Dynal AS) and a volume-to-volume quantity of 2x BW-buffer II (10 mmol/L Tris-HCl, 2 mol/L NaCl, 1 mmol/L EDTA, 1 mL/L Tween 20, pH 7.6). This mixture was incubated at 65 °C for 15 min, with shaking, to bind the template to the beads. A 96-pin magnetic tool (Pyrosequencing AB) was used to transfer up to 96 samples at a time to solutions as follows. The beads with bound template were first transferred to 50 µL of denaturing solution (0.5 mol/L NaOH), then to 100 µL of 1x Annealing buffer (20 mmol/L Tris-acetate, 5 mmol/L magnesium acetate, pH 7.6), and finally into a solution of 1x Annealing buffer containing 5 pmol of the appropriate sequencing primer. Lastly, this mixture was heated to 95 °C for 2 min and then cooled to and incubated at room temperature for at least 5 min to bind the sequencing primer to the template.
pyrosequencing
After template preparation, a 96-well plate containing the samples was loaded into the instrument (PSQ 96 System; Pyrosequencing AB) along with the appropriate reagents (PSQ 96 SNP Reagent Kit; Pyrosequencing AB). This instrument sequences the templates by dispensing reagents and deoxynucleotide triphosphates in a user-defined order, achieving real-time sequencing by synthesis in an automated fashion. This is achieved by creating and monitoring an enzyme cascade initiated by nucleotide incorporation that produces light emission (14).
The PSQ 96 system records a DNA sequence analysis in the form of a pyrogram. Pyrograms are traces of light emission plotted as a function of a user-defined order of nucleotides that are dispensed into a reaction well. At each nucleotide dispensation, the presence or absence of a light peak represents the successful or unsuccessful incorporation of this nucleotide at the 3' end of the sequencing primer. Any base change in a template being analyzed is recognized by loss or gain of a specific peak at one or more dispensation positions, one or more changes in peak height at a dispensation position, or some combination of both. At each base position in the template being sequenced, dispensation of a nucleotide may produce a normal amount of signal when both alleles are wild type, one-half the signal if that position is heterozygous, and no signal if that position is homozygous mutant (not typically seen in MEN2). In the case of a heterozygote, the sequencing of one allele may lag or advance, producing peaks at positions not seen in a typical analysis. Thus, the genotype is identified by comparing the patient pyrogram with a wild-type control pyrogram or a positive control pyrogram. In addition, a patients pyrogram can be compared with theoretical wild-type or mutant patterns that can be produced by the system software. See Ahmadian et al. (15) for further details.
| Results |
|---|
|
|
|---|
An example of a group 1 analysis, in this case for codon 918, is shown in Fig. 1
. In this case, the analysis is performed such that the complement to the wild-type base is dispensed, followed by the one expected mutant base. This general strategy was also used to obtain results for the other codons in this group: 768 and 804 (data not shown). For these two codons, however, an extra dispensation was added to determine whether the wild-type base was substituted with a second possible base. The third mutant codon (TTG) for codon 804 is conveniently detected by the nucleotide dispensation directed at the next base of that codon.
|
However, unlike group 1 analyses, in which only one base is typically interrogated per sequencing reaction, interrogation of up to six bases per reaction was typically performed for the group 2 analyses. Fig. 2
depicts a theoretical wild-type, an actual wild-type, a theoretical mutant, and an actual mutant pattern for a member of this group (codons 618/620). Although the mutation example shown here occurs only at codon 618, until and unless the position of the polymerase synchronizes on both alleles, the pyrosequencing will remain asynchronous and produce an abnormal pyrogram, even subsequent to the site of the mutation. This yields differences in the pyrogram well beyond codon 618. With this type of analysis, any changes to any of the three bases of codons 618 or 620 can be detected in a single pyrosequencing reaction, with each mutation producing a unique pyrogram. This general strategy was also used to obtain the genotypes for the 609/611 and the 630/634 pairs of codons (data not shown).
|
Using the above methods, we analyzed DNA from 217 patients who had previously tested negative and 230 patients who had previously tested positive for changes in up to nine codons each, for a total of 1449 sequencing reactions. The results were compared with the original results that were obtained by manual sequencing, restriction fragment length polymorphism analysis, or LightCycler analysis. Among the positive patients, a total of 37 unique mutations were analyzed: 33 known mutations and 4 alterations of unknown significance. Of these, 9 were from group 1 (total of 26 specimens analyzed) and 28 were from group 2 (total of 204 specimens analyzed). All of the negative samples were scored correctly (100% concordance). Among the 230 positive samples, all previously detected mutations were correctly identified. However, one patient appeared to have an additional mutation not previously observed (99.6% concordance). In this one discrepant case, the pyrogram suggested the presence of an additional mutation (630 TGC>TGT) along with the previously identified mutation (634 TGC>CGC). On repeat analysis (PCR and pyrosequencing), however, only the expected 634 TGC>CGC mutation was detected. Ideally, resequencing a portion of the template used for the first analysis would have provided additional information regarding the source of error, but unfortunately, this was not possible because the entire amount of template was consumed in the first reaction. To determine whether this artifact was reproducible, the PCR and pyrosequencing were performed 20 additional times. Each of these gave consistent results, all showing the 634 mutation only. The discordant result and the pattern from the original run are illustrated in Fig. 3
.
|
For those patients with a gene alteration, the codons analyzed, the wild-type sequence of each codon, and the distribution of the gene alterations (33 unique mutations among the 9 codons tested) are summarized in Fig. 4
. Also shown are the disease-causing alterations at each codon that have been reported previously (1)(16)(17)(18)(19) and, for codons 609620 and 634, the other possible alterations that produce an amino acid other than a cysteine. Note that several novel alterations not previously reported in the germ line were included in this study. Fig. 4B
shows several alterations of unknown significance in codon 631. These alterations have also not been reported previously in the germ line. In all, 37 unique alterations were analyzed.
|
| Discussion |
|---|
|
|
|---|
To date, many of the assays utilizing pyrosequencing have been directed at the analysis of single-nucleotide polymorphisms (20)(21). Assays for these single-base-pair changes are robust and relatively simple to design. Pyrosequencing, however, also has the ability to interrogate more complex changes in DNA sequences (22)(23). In this study, we evaluated several pyrosequencing-based assays designed to interrogate a series of both simple and complex common mutations associated with MEN2 and evaluated their performance characteristics. Using this method, we studied 217 patients who had previously tested negative and 230 who had previously tested positive for MEN2. The samples were analyzed over multiple runs to mimic those conditions expected in a clinical laboratory. In all, 1449 sequencing reactions were performed, and overall, the quality of results obtained with this technique was excellent. However, although there were no false-negative calls, there was one false positive.
In this one discrepant case, the pyrogram suggested the presence of an additional mutation (630 TGC>TGT) along with the correctly identified mutation (634 TGC>CGC). After a repeat of the entire analysis (PCR and pyrosequencing), only the expected 634 TGC>CGC mutation was detected. To further explore the potential sources of error, we considered all steps of the assay, including PCR contamination, PCR-related errors, post-PCR preparation, and instrument- or pyrosequencing reagent-related issues. This investigation provided no obvious sources of error because (a) a sample with the 630 TGC>TGT genotype (potential PCR contaminant) has never been analyzed in this laboratory; (b) misincorporation by AmpliTaq Gold polymerase early in the PCR would have to be significant relative to the presumed much higher concentration of the correct template; (c) post-PCR preparation of the sample involved only an amplicon that would be unlikely to produce false priming (609/611 target); (d) other pyrograms produced by the instrument for samples ran at the same time as the discordant sample showed nothing unusual; and finally (e) reagent errors would have had to occur twice and be of opposite nature to produce the observed one reduced peak followed by the one increased peak with no changes seen thereafter. Therefore, at this point there is no obvious source of error in any of the steps leading to or during the sequencing reaction that provides a reasonable explanation for the discordant result.
Of note, there are two additional considerations with regard to this technology. The first is that because pyrosequencing involves manipulation of post-PCR product, the possibility of future PCR contamination is increased relative to "closed tube" systems. Although we did not experience PCR contamination, proper precautions must be followed to prevent this from occurring. The second consideration is that the PCR template requirement for the sequencing reaction is fairly high. To provide adequate signal, a fairly robust PCR is therefore required. The recommended amount of template for the instrument used in this study, after post-PCR processing, is 2 pmol. However, there is an additional instrument platform with an
10-fold increase in sensitivity, the PSQ 96 HS. In addition to the higher sensitivity charge-coupled device camera, the platform accommodates reduced reaction volumes.
One additional feature of this technology is its ability to determine whether two closely positioned mutations are in cis or trans (data not shown). This can be done by taking advantage of a characteristic of pyrosequencing whereby a mutation causes a lag or advance of the polymerase on the mutated strand vs the wild-type strand. When this occurs the sequencing of the two strands beyond the mutation is asynchronous. Because the sequencing of these two strands is now out of phase, the two alleles can be distinguished from each other. As a result, when the second mutation is encountered, it will appear on the pyrogram differently if it is in cis vs trans. As long as the two alleles are not allowed to synchronize, the upper limit on the distance between the two mutations (i.e., closely spaced) is limited only by the ability to obtain adequate signal. Signal decreases as the reagents are catalyzing substrates on and after each dispensation. For the reagent set used in our study, up to 3540 dispensations could be done routinely without significant loss of signal; this therefore may be considered a general limit for mutation separation with these reagents.
In spite of the considerations described above, however, pyrosequencing provides many advantages because it seems not susceptible to many of the problems inherent in other methodologies. It is fairly rapid: generally results can be obtained within hours of completion of the template PCR. The microtiter plate format makes the assays amenable to automation; as more mutations are found or as the test volume increases, the same platform can be used. This method is simple to use because both the group 1 and group 2 types of mutations described in this study require the same approach, a standard template PCR is followed by a post-PCR preparation step and then by automated sequencing on the instrument. Finally, this sequencing-based technique offers high accuracy, as is the case with conventional sequencing, because any base change in the region being interrogated can be identified, both known and unknown. Overall, the Pyrosequencing platform has been successfully applied to the analysis of both simple and complex mutations within the RET protooncogene for the diagnosis of MEN2.
| Acknowledgments |
|---|
| References |
|---|
|
|
|---|
T, in codon 768 of the RET proto-oncogene in a patient with medullary thyroid carcinoma. Am J Med Genet 2002;110:85-87.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
![]() |
R. L. Margraf, R. Mao, W. E. Highsmith, L. M. Holtegaard, and C. T. Wittwer RET Proto-Oncogene Genotyping Using Unlabeled Probes, the Masking Technique, and Amplicon High-Resolution Melting Analysis J. Mol. Diagn., April 1, 2007; 9(2): 184 - 196. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. L. Margraf, R. Mao, W. E. Highsmith, L. M. Holtegaard, and C. T. Wittwer Mutation Scanning of the RET Protooncogene Using High-Resolution Melting Analysis Clin. Chem., January 1, 2006; 52(1): 138 - 141. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |