Making DNA melting useful.

Featured Article: Wittwer CT, Reed GH, Gundry CN, Vandersteen JG, Pryor RJ. High-resolution genotyping by amplicon melting analysis using LCGreen. Clin Chem 2003;49:853– 60. DNA melting (the denaturation of DNA by heat) is a fundamental property of DNA that can be used as an analytical tool. Classic studies of DNA melting monitored the absorbance near 260 nm as the temperature was increased and required large (microgram) amounts of DNA and analysis times of hours. Although DNA melting was an accurate research tool, these limitations relegated DNA melting to the specialized laboratory, away from most research laboratories and clinical testing. Fluorescence analysis of DNA melting as a companion to real-time PCR was introduced in 1996 with the LightCycler , the product of university academics and the small company Idaho Technology (1 ). In 1998, Roche Applied Science adopted the LightCycler worldwide. The PCR conveniently produces nanogram amounts of DNA that can easily be monitored with fluorescence. DNA fluorescence typically increases during real-time PCR and decreases during melting analysis. For the first time, no handling, processing, or separation of the sample after the PCR was required for analysis. Methods for both probe and PCR product (amplicon) melting were developed for the LightCycler and were later incorporated into other real-time instruments. DNA melting monitors stability across temperature as a “dynamic dot blot,” an approach that is inherently more robust than the prior singletemperature “static” analysis. Single-base variants, such as the F5 [coagulation factor V (proaccelerin, labile factor)] Leiden variant, were genotyped by melting hybridization probes (2 ), and in 2003 these tests became the first genetic tests cleared by the US Food and Drug Administration. Real-time PCR with SYBR Green I became more popular than with fluorescently labeled probes because of the cost and the added benefit of assessing PCR product purity by melting after amplification. The characteristic melting curves of amplicons after the PCR usually allowed the differentiation of unique products (3 ). Usually but not always. Small differences, such as a single-base change in 1 copy of diploid DNA, were difficult to detect. Real-time PCR instruments, including the carousel LightCycler, did not focus on melting quality. In 2000, we initiated a project with Idaho Technology to build a high-resolution melting instrument. Our goal was to see what additional information might be obtained if melting-curve quality were improved. The resulting instrument, the HR-1, used a LightCycler capillary surrounded by an aluminum ingot with a wound resistive coil. Melting rates of 0.1 °C/s to 0.3 °C/s produced 40 –120 points per degree Celsius, all of which could be plotted and analyzed without approximation or smoothing. We learned that “high resolution” did not necessarily mean “slow,” because melting curves of excellent quality were acquired over 30 °C in 2 min. As with the PCR, melting analysis does not have to be slow to be good. We also learned that new analysis methods, including normalization and difference plots, were required to display the small differences revealed by high-resolution melting. Heterozygotes were best detected by comparing melting-curve shapes after “curve overlay” or “temperature shifting,” rather than by their melting temperatures, or Tms. High-resolution amplicon melting was first demonstrated with PCR products amplified with a fluorescently labeled primer (4 ). Fluorescein was attached to the 5 end of one primer, and the fluorescence intensity was sensitive to whether the DNA was singleor double-stranded. After the PCR, any variation within the melting domain that included the labeled primer was detected, demonstrating that even single-base changes could be genotyped. A fluorescently labeled primer was required, however, and variants in other domains were not seen. If fluorescence could be generated with dyes rather than with labeled probes, both concerns might disappear. SYBR Green I, however, produced variable results. We investigated dyes other than SYBR Green I by looking for their ability to detect the heteroduplexes produced by the PCR after amplification of a heterozygote. In collaboration with Idaho Technology, we synthesized and characterized 30 asymmetric cyanine dyes. The dyes most sensitive to heteroduplexes be1 Department of Pathology, University of Utah Health Sciences Center, Salt Lake City, UT; 2 ARUP Laboratories and Institute for Clinical and Experimental Pathology, Salt Lake City, UT. Address correspondence to the author at: Department of Pathology, University of Utah Medical School, 50 N. Medical Dr., Salt Lake City, UT 84132. Fax 801-5816001; e-mail [email protected]. 3 In 2009 alone, this report was cited more than 75 times. Received; accepted April 13, 2010. Previously published online at DOI: 10.1373/clinchem.2010.146175 Clinical Chemistry 56:9 1500–1501 (2010) Citation Classic