| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Editorials |
Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN 55905
The development of technically simple and reliable methods to detect sequence variations in specific genes is becoming more important as the number of genes associated with specific diseases grows. DNA sequencing is considered the "gold standard" for characterization of specific nucleotide alteration(s) that lead to genetic disease. Although sequencing was long considered too cumbersome, expensive, and operator-dependent for use in the clinical laboratory, a combination of clinical need and improved technology has brought automated DNA sequencing into routine clinical use. Sequencing technology is now firmly entrenched in the clinical molecular diagnostics laboratory, but it remains too expensive and time-consuming for all of the laboratory’s mutation-detection needs. Several PCR-based mutation-detection strategies can be used to identify both characterized and uncharacterized mutations and sequence variations.
The degree of allelic heterogeneity, or the number of different disease-causing mutations in a single gene, influences the method used for mutation detection. For diseases that exhibit no or limited heterogeneity (such as sickle cell anemia or factor V Leiden), assay systems designed to detect specific mutations are appropriate. These types of strategies are also appropriate for disorders in which allelic heterogeneity is high but only a limited set of mutations are typically analyzed, such as cystic fibrosis. For disorders in which the mutational spectrum is wide (e.g., mismatch-repair genes in hereditary nonpolyposis colon cancer), a scanning method is needed. A scanning method is also appropriate for analysis of newly identified disease genes for which there is little or no information regarding the number of disease-causing mutations.
A wide array of methods and technologies have been proposed for detection of specific mutations, and many of them are in use in clinical and research laboratories worldwide. Examples of the various technologies for specific-mutation analysis include restriction enzyme digestion of PCR-amplified DNA, with or without the introduction of differential restriction sites with mismatch-containing primers (1); ligase-mediated assays (2); and allele-specific primer extension on a variety of detection platforms, including gel electrophoresis, flow cytometry, and mass spectrometry (3)(4)(5). Non-PCR platforms, such as the InvaderTM chemistry, are growing in popularity (6). Several methods rely on fluorescence energy transfer (FRET) chemistry; these include TaqManTM, Molecular BeaconTM, and FRET probe assays (7)(8)(9)(10).
FRET probes became popular with the introduction of the LightCyclerTM, an adaptation of the air-driven RapidCyclerTM, which, legend has it, was prototyped by Dr. Wittwer in his garage with a hair dryer, to real-time PCR instrumentation or combination thermal cycler-fluorometers. The availability of a moderately priced instrument that offered rapid cycle times, an integrated fluorometer with melting curve analysis software, and FRET probe chemistry made the LightCycler the instrument of choice for many clinical laboratories. FRET probe methods have been developed for a wide variety of disorders (11)(12)(13). In addition to the ability of this chemistry to detect point mutations, it has also been extensively used for quantitative analysis in molecular microbiology and molecular hematopathology laboratories.
The choice of instrumentation and chemistry for clinical laboratories is complex and depends on several factors, including cost of instrumentation and reagents, labor requirements, test volume and menu, and workflow in the laboratory. A limiting factor of the LightCycler is that although the cycle time is rapid (30–45 min for a typical 35-cycle PCR reaction), it has limited throughput because of the labor requirements for preparing reagents and samples in small batches compatible with the 32-sample rotor capacity (of which several are typically dedicated to controls and blanks) and dispensing the reagents and samples into the LightCycler capillary cuvettes. This limitation has been addressed in several ways. First, Roche Diagnostics Corporation, which acquired the marketing rights to the LightCycler product line from Idaho Technologies, adapted an automated DNA extraction robot to dispense samples and reagents into the capillary cuvettes in the circular rotor. The introduction of robotics eliminated much of the tedium associated with preparing multiple LightCycler runs and has decreased the possibility of sample mix-up.
The Wittwer group took another approach to higher-throughput melting curve analysis. They realized that the most expensive components of real-time PCR instrumentation, the fluorometer optics, were idle for the overwhelming majority of the time in a melting curve analysis. They also realized that when the objective is detection of mutations, melting curve analysis is an endpoint technique, and little is gained by performing these assays on combination thermal cycler-fluorometer instruments. Thus, they developed a dedicated melting-curve-analysis instrument that had a simplified thermal control mechanism and was compatible with 96- or 384-well plates. Because the melting curve analysis itself requires only 10–15 min, a laboratory with higher throughput needs can set up multiple PCR reaction plates, carry out the PCR reaction in low-cost thermal cyclers, and perform the fluorometry on a dedicated instrument, thereby achieving a quantum jump in throughput while maintaining the advantage of a closed-tube, homogeneous system. The instrument, originally developed in Dr. Wittwer’s laboratory, is now commercially available as the LightTyperTM (14). In the 96-well plate format, the instrument is capable of analyzing several thousand samples per shift; with the 384-well format, that number rises to several tens of thousands.
A constant theme that runs through automated specific-mutation-detection strategies, with the exception of primer extension and detection by mass spectrometry, is the reliance on fluorescent oligonucleotides. Because the majority of current methods also depend on FRET, these assays (such as TaqMan or Molecular Beacon assays) require either single oligonucleotide probes labeled with two different fluorophores or two separate, singly labeled probes (FRET probes). Although in recent years the manufacturing of custom oligonucleotides has become almost a commodity business, labeled oligonucleotides are still expensive. For example, a recent set of 20mer PCR primers ordered by the Mayo Molecular Genetics Laboratory was obtained for under $15.00 per primer. In contrast, a recently purchased set of sensor and anchor probes for melting curve analysis cost more than $500.00 each for the same synthesis scale. In addition to the purchase price, fluorescently labeled oligonucleotides require special care in handling (to avoid photobleaching in room light) and are not (in my experience) as stable as unlabeled oligonucleotides and therefore have to be replaced or resynthesized much more often. Dr. Wittwer’s group has recently developed a second chemistry for melting curve analysis, the Simple ProbeTM, which requires only one fluorescently labeled oligonucleotide probe (15). This was a major advance in decreasing the cost and complexity of specific-mutation analysis.
In addition to advancing specific-mutation-detection techniques, Dr. Wittwer and his team have expanded melting curve analysis to mutation scanning. One key to this advance was the development of a DNA binding fluor, LightCycler GreenTM, which can be added to a PCR mixture at saturating concentration without inhibiting the thermostable polymerase. The second key to using melting curve analysis for mutation scanning was the development of a device with sufficient resolution to detect minute changes in LightCycler Green fluorescence attributable to differences in the melting characteristics of unmodified PCR products. This instrument, the HR-1TM, and the use of the novel dye were reported in this journal in March and June of 2003, respectively (16)(17).
In this issue of the journal, Dr. Wittwer and his colleagues have returned to specific-mutation analysis and have taken the next step in decreasing the cost of these assays by eliminating the use of fluorescent probes altogether (18). Following the development of a DNA-binding dye that was compatible with PCR at saturating concentrations and the demonstration that small changes in melting temperatures in whole PCR products were reproducible and detectable, this group reasoned that they should be able to detect the changes in melting of matched and mismatched oligonucleotides by use of dye alone. Here they report that such is indeed the case. Using a single reporter oligonucleotide probe that is perfectly complementary to one allele and contains a destabilizing mismatch to the other allele, the investigators show that the fluorescence from a small (20–24 bp) duplex is detectable and that the change in fluorescence as a function of melting temperature can be monitored.
The use of unbalanced PCR primer ratios, which leads to asymmetric PCR or the overproduction of one strand, the target strand of the reporter probe, was shown to be key for the generation of sufficient signal. This maneuver should come as no surprise to seasoned LightCycler users because it is typical in standard FRET probe analysis as well. What is surprising is the observation that such a small stretch of double-stranded DNA would be observable against the much larger signal arising from the PCR product. Of course, it is not fluorescence signal from the probe–target duplex per se that is of interest, it is the change in signal as a function of temperature. Here the low melting temperature of the small duplex, or the change in fluorescence as the oligonucleotide melts from the single-stranded PCR product strand, is sufficiently resolved from the derivative peak produced by the melting of both the intact PCR product and primer-dimer as to be readily analyzable.
Extensive validation of the concept is provided in the current report. Zhou et al. (18) have investigated the ability of the unlabeled probes to detect all possible base mismatches and have optimized the procedure with respect to probe length and PCR primer ratio for asymmetric PCR. They have also provided data showing that this method is compatible with three different platforms: the LightCycler, the LightTyper, and the HR-1. The method described in this report represents a major advance in the field because it is one of the very few that can deliver lower cost assays.
Will widespread use of this method decrease the cost to patients and third-party payers of molecular genetic tests? Alas, the answer is probably not. The reason is that reagent costs make up a rather small fraction of the final price of a clinical test, with royalties, technologist time, and overhead typically being larger contributors to the final price. However, although the cost of fluorescent probes per test is already low, if this technology allows laboratories to spend less of their reagent budget on fluorescent oligonucleotides and less on quality control of reagents (by decreasing the number of oligonucleotides being managed), then there will be modest savings, which may ultimately lead to lower prices to the consumer. The area in which this technology is likely to have the largest financial effect is genomics research. Obvious applications include scenarios in which target areas of the genome have been identified in a limited set of samples by expensive genome-scanning methods [e.g., single-nucleotide polymorphism (SNP) chips, genome-wide short tandem repeat (STR) sets, and haplotype mapping] and fine mapping with a selected set of SNPs over thousands of samples is required.
In conclusion, Dr. Wittwer and his colleagues have pioneered rapid, affordable mutation-detection technology and have moved these developments out of the engineering laboratory and into the clinical laboratory. Looking to the future, I can hardly wait to see what the Wizard of Salt Lake will come up with next.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
