Download Distinguishing Different DNA Heterozygotes by

Clinical Chemistry 51, No. 7, 2005
10. Stamer UM, Bayerer B, Wolf S, Hoeft A, Stüber F. Rapid and reliable method
for cytochrome P450 2D6 genotyping. Clin Chem 2002;48:1412–7.
11. Zackrisson AL, Lindblom B. Identification of CYP2D6 alleles by single
nucleotide polymorphism analysis using pyrosequencing. Eur J Clin Pharmacol 2003;59:521– 6.
12. Chou WH, Yan FX, Robbins-Weilert DK, Ryder TB, Liu WW, Perbost C, et al.
Comparison of two CYP2D6 genotyping methods and assessment of
genotype-phenotype relationships. Clin Chem 2003;49:542–51.
13. Gaikovitch EA, Cascorbi I, Mrozikiewicz PM, Brockmöller J, Frotschl R, Kopke
K, et al. Polymorphisms of drug-metabolizing enzymes CYP2C9, CYP2C19,
CYP2D6, CYP1A1, NAT2 and of P-glycoprotein in a Russian population. Eur
J Clin Pharmacol 2003;59:303–12.
14. Scordo MG, Caputi AP, D’Arrigo C, Fava G, Spina E. Allele and genotype
frequencies of CYP2C9, CYP2C19 and CYP2D6 in an Italian population.
Pharmacol Res 2004;50:195–200.
15. Syvänen AC. Accessing genetic variation: genotyping single nucleotide
polymorphisms [Review]. Nat Rev Genet 2001;2:930 – 42.
16. Bradford LD. CYP2D6 allele frequency in European Caucasians, Asians,
Africans and their descendants [Review]. Pharmacogenomics 2002;3:229 –
43.
17. Lundqvist E, Johansson I, Ingelman-Sundberg M. Genetic mechanisms for
duplication and multiduplication of the human CYP2D6 gene and methods
for detection of duplicated CYP2D6 genes. Gene 1999;226:327–38.
18. Lovlie R, Daly AK, Molven A, Idle JR, Steen VM. Ultrarapid metabolizers of
debrisoquine: characterization and PCR-based detection of alleles with
duplication of the CYP2D6 gene. FEBS Lett 1996;392:30 – 4.
19. Steen VM, Andreassen OA, Daly AK, Tefre T, Borresen AL, Idle JR, et al.
Detection of the poor metabolizer-associated CYP2D6(D) gene deletion
allele by long-PCR technology. Pharmacogenetics 1995;5:215–23.
20. Cann HM, de Toma C, Cazes L, Legrand MF, Morel V, Piouffre L, et al. A
human genome diversity cell line panel. Science 2002;296:261–2.
21. Fuselli S, Dupanloup I, Frigato E, Cruciani F, Scozzari R, Moral P, et al.
Molecular diversity at the CYP2D6 locus in the Mediterranean region. Eur J
Hum Genet 2004;12:916 –24.
22. Levo A, Koski A, Ojanperä I, Vuori E, Sajantila A. Post-mortem SNP analysis
of CYP2D6 gene reveals correlation between genotype and opioid drug
(tramadol) metabolite ratios in blood. Forensic Sci Int 2003;135:9 –15.
23. Johansson I, Lundqvist E, Dahl ML, Ingelman-Sundberg M. PCR-based
genotyping for duplicated and deleted CYP2D6 genes. Pharmacogenetics
1996;6:351–5.
24. Bender K. SNaPshot for pharmacogenetics by minisequencing. Methods
Mol Biol 2004;297:243–52.
25. Knaapen AM, Ketelslegers HB, Gottschalk RW, Janssen RG, Paulussen AD,
Smeets HJ, et al. Simultaneous genotyping of nine polymorphisms in
xenobiotic-metabolizing enzymes by multiplex PCR amplification and single
base extension [Technical Brief]. Clin Chem 2004;50:1664 – 8.
26. Schultz J. FDA guidelines on race and ethnicity: obstacle or remedy? J Natl
Cancer Inst 2003;95:425– 6.
27. Raimundo S, Toscano C, Klein K, Fischer J, Griese EU, Eichelbaum M, et al.
A novel intronic mutation, 2988G⬎A, with high predictivity for impaired
function of cytochrome P450 2D6 in white subjects. Clin Pharmacol Ther
2004;76:128 –38.
Previously published online at DOI: 10.1373/clinchem.2004.046466
Distinguishing Different DNA Heterozygotes by HighResolution Melting, Robert Graham,1 Michael Liew,2 Cindy
Meadows,2 Elaine Lyon,2 and Carl T. Wittwer1,2* (1 Department of Pathology, University of Utah School of Medicine, Salt Lake City, UT; 2 Institute for Clinical and Experimental Pathology, ARUP, Salt Lake City UT; * address
correspondence to this author at: Department of Pathology, University of Utah School of Medicine, Salt Lake
City, UT 84132; fax 801-581-4517, e-mail carl.wittwer@
path.utah.edu)
High-resolution melting was recently introduced as a
technique to genotype single-nucleotide polymorphisms
(SNPs) within small amplicons (1 ). This closed-tube
1295
method (including rapid-cycle PCR) can be completed in
⬍15 min and does not require real-time PCR instruments
(2 ), allele-specific PCR (3 ), or fluorescently labeled oligonucleotides (4 – 6 ). The process is made possible by heteroduplex-detecting DNA dyes that can be used at saturating concentrations without inhibiting PCR (7 ). Wildtype and homozygous mutant samples are distinguished
by melting temperature (Tm) shifts. Heterozygous samples are best distinguished from homozygotes, not by Tm,
but by altered curve shape. Heterozygous samples produce heteroduplexes that melt at lower temperatures than
homoduplexes. Melting curves of amplified heterozygotes include 2 homoduplexes and 2 heteroduplexes,
giving a skewed composite melting curve easily distinguished from the curves for homozygotes by curve shape
(1 ).
However, it was not clear whether different heterozygotes within the same amplicon could be distinguished
from each other based on curve shape differences. Four
different classes of SNPs have been defined based on the
homo- and heteroduplexes that are produced after amplification (1 ). Because the heteroduplex mismatches of
SNPs in different classes are different, their melting
curves should be distinguishable if the resolution of the
instrumentation is sufficient. It is also possible that different heterozygotes in the same class may be distinguishable. Although the mismatches are the same, nearestneighbor stability parameters depend on the bases
adjacent to the mismatches; therefore, the predicted stabilities are often different.
We selected DNA samples retrospectively from those
submitted to ARUP Laboratories for standard clinical
genotyping of HFE, factor V Leiden, and factor II polymorphisms. DNA was extracted from blood samples
by use of the MagNa Pure instrument (Roche) and
genotyped by use of adjacent hybridization probe
(HybProbeTM) methods (4, 5 ) on a LightCycler®. One
advantage of probe melting analysis for genotyping is
that sequence alterations other than those expected can
often be identified under the probes (8 ). During routine
clinical analysis at ARUP, when aberrant melting peaks
are observed at unexpected temperatures, samples are
sequenced to identify the sequence alteration. Several
heterozygous sequence variants near the expected mutations were thus identified. Three factor II heterozygotes
(all SNPs), 4 factor V heterozygotes (2 SNPs, 1 single-base
deletion, and 1 compound heterozygote), and 6 HFE
heterozygotes (4 SNPs and 2 compound heterozygotes)
were available for comparison. After selection and deidentification of samples, DNA concentrations were
checked on an ND-1000 (NanoDrop Technologies, Inc.).
When available, 3 individuals of each heterozygous genotype were processed.
Previously described primers were used (1 ) unless it
was necessary to redesign them to flank all sequence
variants. SNPWizard (http://DNAWizards.path.utah.edu)
was used for primer design. Primers were synthesized by
Integrated DNA Technologies and used without further
purification. PCR was performed as described previously
1296
Technical Briefs
(1 ), except that 1X LCGreen ® PLUS (Idaho Technology)
replaced LCGreen I. LCGreen PLUS is brighter than
LCGreen I and can be used on a wider variety of melting
instruments. The amplicon sequences, primer regions,
and specific PCR conditions used for amplification are
listed in Table 1 of the Data Supplement that accompanies
the online version of this Technical Brief at http://
www.clinchem.org/content/vol51/issue7/.
After PCR, capillaries were heated to 90 °C at 20 °C/s in
the LightCycler and then cooled to 40 °C at 20 °C/s to
favor heteroduplex formation (9 ). Samples were removed
from the LightCycler and analyzed on a high-resolution
melting instrument (HR-1; Idaho Technology). Each capillary was inserted into the instrument at 59 °C, the
temperature was increased at 0.3 °C/s, and fluorescence
was acquired from 65 to 95 °C, which required only 1–2
min. Heteroduplex detection in small amplicons is favored by rapid cooling before melting, rapid heating
during melting, and low Mg2⫹ concentrations (9 ). Melting
curves were analyzed by custom software with normalized, temperature-shifted curves displayed as difference
plots (1, 7 ). Difference plots display the difference between each sample and a “standard” (often the average of
multiple wild-type curves) and should not be confused
with derivative plots (⫺dF/dT), which are often used to
analyze low-resolution melting data. Difference plots are
a convenient way of viewing high-resolution melting data
because slight differences in curve shape and Tm become
obvious.
Predicted duplex Tms were calculated with previously
described nearest-neighbor thermodynamic models
(1, 10 ) as implemented in MeltingWizard (available at
http://DNAWizards.path.utah.edu). No corrections for
dUTP incorporation instead of dTTP (tends to lower Tm)
or influence of LCGreen PLUS (tends to increase Tm) were
used. To maximize differences in Tm, amplicon lengths
were kept short (1 ), although SNP typing has been
demonstrated in products more than 500 bp long (7 ).
Calculated Tm differences between homozygous wildtype and homozygous mutant amplicons varied from 0.0
to 1.3 °C for the different SNPs (Table 2 of the online Data
Supplement). For single SNPs, heteroduplex destabilization varied from 1.1 to 3.0 °C compared with the most
stable homoduplex, or 0.7–2.2 °C compared with the least
stable homoduplex. For compound heterozygotes, the
respective destabilizations were 2.8 – 4.0 °C and 2.0 –3.5 °C.
High-resolution melting techniques are necessary to
differentiate different heterozygotes within the same amplicon. Melting curves generated on the LightCycler did
not reliably separate heterozygotes (data not shown), in
contrast to high-resolution HR-1 analysis. As seen in Fig.
1, each heterozygote traces a unique melting curve path
according to its 4 duplex Tms. These unique melting curve
shapes are best seen with normalized, temperatureshifted data, shown in the form of difference plots. In Fig.
1A, factor II and 2 rare heterozygotes (all class 1 SNPs) are
clearly separated from the wild type and from each other.
In Fig. 1B, factor V Leiden and 3 rare heterozygotes (one
SNP, one 1-bp deletion, and one compound heterozygote)
Fig. 1. Melting analysis of several heterozygous sequence variants
near common mutations in factor II (20210G⬎A), factor V (Lieden;
1691G⬎A), and HFE (187C⬎G).
The melting curves were normalized and temperature-shifted, and the difference
between each sample and the average of the wild types is displayed at each
temperature (1, 7 ). Such “difference” plots should not be confused with the
commonly used derivative plots. Each mutation is indicated by its location
(20210) followed by the base pair change (G⬎A). All samples are heterozygous
or compound heterozygous, unless followed by a WT, which signifies the wild
type. Three individuals of each genotype are shown, except for factor II
20218G⬎A and factor V compound heterozygote 1696A⬎G/1690G⬎A, which
are in duplicate, and the HFE compound heterozygote 187C⬎G/189T⬎C, for
which only 1 sample was available.
Clinical Chemistry 51, No. 7, 2005
also followed distinct paths. As expected, single SNPs and
the 1-bp deletion (single unpaired loci) were more similar
to each other than the compound heterozygote (2 unpaired loci). In Fig. 1C, all heterozygotes tested were
clearly separated from each other and the wild type,
including the targeted HFE mutation, 3 rare SNPs, and 2
rare compound heterozygotes. In all cases studied, different heterozygotes could be distinguished, including those
within the same SNP class (1 ).
At the factor V and HFE loci, compound heterozygotes
were easily distinguished from “single” heterozygotes, as
might be expected because of 2, rather than 1, region of
destabilization. It is interesting that the predicted Tms of
the homoduplexes and/or heteroduplexes are different
depending on the cis/trans orientation of the sequence
variants (Table 2 of the online Data Supplement). This
suggests that melting analysis may allow haplotyping, at
least when the variations to be haplotyped are in the same
melting domain.
The presence of unexpected polymorphisms close to a
SNP of interest may or may not interfere with genotyping,
depending on the analysis method (10 ). All PCR-based
methods can be compromised if polymorphisms occur
under the primers and lead to undesired allele-specific
PCR. The same concern applies to internal primers used
for sequencing or extension reactions. These caveats aside,
some PCR-based methods, including sequencing, will
identify and distinguish polymorphisms that are missed
by restriction fragment length polymorphism assays and
isothermal probe-based assays. Hybridization probe assays that interrogate over a range of temperatures (melting assays) often detect the presence of unexpected polymorphisms but may require further studies to identify
them. In our study, high-resolution melting of small
amplicons distinguished all heterozygotes studied. This
included 21 pairwise comparisons, suggesting that most
randomly selected heterozygotes within small amplicons
can be distinguished.
The power of melting analysis to distinguish multiple
sequence variants has been controversial (11–14 ). When
wild-type probes are used on standard low-resolution
systems, 19%–52% of possible SNP heterozygotes may be
confused with the expected heterozygous variant (11 ).
Although better temperature resolution has been reported
on these instruments (12, 13 ), it is clear that rare sequence
variants under a probe may not be distinguished from the
common targeted variant. Similar concerns for genotyping by amplicon melting temperature have been described and focus on the limitations of Tm to distinguish
variants (14 ).
High-resolution melting analysis introduces the use of
melting curve shape, as well as Tm, to distinguish different variants. Because the composite melting curves of
heterozygotes include 2 homoduplexes and 2 heteroduplexes, the meaning of Tm is ambiguous and less useful as
a metric than is curve shape. Is the Tm of a composite
melting curve the temperature at which one-half of the
duplexes have melted (conventional definition), or is it
the peak of the derivative plot (common derivation)?
1297
These are not identical because such curves are skewed at
low temperatures from heteroduplex contributions. In
either case, the Tm is only one point on the melting curve.
Use of the complete melting curve, conveniently displayed as difference plots, allows differentiation of most
heterozygotes (21 of 21 random pairwise comparisons in
this study).
High-resolution melting analysis is convenient because
no processing or separation steps are required (7 ). In
cases in which the complexity of the target exceeds the
genotyping capabilities of amplicon melting, unlabeled
oligonucleotide probes can be added (15 ). In addition to
genotyping, high-resolution melting analysis is an accurate mutation scanning tool (16 ) that has been applied to
the medium chain acyl-CoA dehydrogenase (17 ), c-kit
(18 ), and SLC22A5 (19 ) genes, as well as HLA matching
(20 ). The sensitivity and specificity are greater than denaturing HPLC (21 ), and no electrophoresis is required, as
in temperature gradient (22 ), denaturing gradient (23 ), or
conformation-sensitive (24 ) gel electrophoresis.
References
1. Liew M, Pryor R, Palais R, Meadows C, Erali M, Lyon E, et al. Genotyping of
single-nucleotide polymorphisms by high-resolution melting of small amplicons. Clin Chem 2004;50:1156 – 64.
2. Wittwer CT, Ririe KM, Andrew RV, David DA, Gundry RA, Balis UJ. The
LightCycler: a microvolume multisample fluorimeter with rapid temperature
control. Biotechniques 1997;22:176 – 81.
3. Germer S, Higuchi R. Single-tube genotyping without oligonucleotide probes.
Genome Res 1999;9:72– 8.
4. Lay MJ, Wittwer CT. Real-time fluorescence genotyping of factor V Leiden
during rapid-cycle PCR. Clin Chem 1997;43:2262–7.
5. Bernard PS, Ajioka RS, Kushner JP, Wittwer CT. Homogeneous multiplex
genotyping of hemochromatosis mutations with fluorescent hybridization
probes. Am J Pathol 1998;153:1055– 61.
6. Crockett AO, Wittwer CT. Fluorescein-labeled oligonucleotides for real-time
pcr: using the inherent quenching of deoxyguanosine nucleotides. Anal
Biochem 2001;290:89 –97.
7. 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.
8. Gundry CN, Bernard PS, Herrmann MG, Reed GH, Wittwer CT. Rapid F508del
and F508C assay using fluorescent hybridization probes. Genet Test 1999;
3:365–70.
9. Gundry CN, Vandersteen JG, Reed GH, Pryor RJ, Chen J, Wittwer CT.
Amplicon melting analysis with labeled primers: a closed-tube method for
differentiating homozygotes and heterozygotes. Clin Chem 2003;49:396 –
406.
10. Peyret N, Seneviratne PA, Allawi HT, SantaLucia J Jr. Nearest-neighbor
thermodynamics and NMR of DNA sequences with internal A:A, C:C, G:G,
and T:T mismatches. Biochemistry 1999;38:3468 –77.
11. Schutz E, von Ahsen N, Oellerich M. Genotyping of eight thiopurine methyltransferase mutations: three-color multiplexing, “two-color/shared” anchor,
and fluorescence-quenching hybridization probe assays based on thermodynamic nearest-neighbor probe design. Clin Chem 2000;46:1728 –37.
12. Berrnard PS, Wittwer CT. Homogeneous amplification and variant detection
by fluorescent hybridization probes. Clin Chem 2000;46:147– 8.
13. Lyon E, Millson A, Phan T, Wittwer CT. Detection and identification of base
alterations within the region of factor V Leiden by fluorescent melting curves.
Mol Diagn 1998;3:203–10.
14. von Ahsen N, Oellerich M, Schutz E. Limitations of genotyping based on
amplicon melting temperature. Clin Chem 2001;47:1331–2.
15. Zhou L, Myers AN, Vandersteen JG, Wang L, Wittwer CT. Closed-tube
genotyping with unlabeled oligonucleotide probes and a saturating DNA dye.
Clin Chem 2004;50:1328 –35.
16. Reed GH, Wittwer CT. Sensitivity and specificity of single-nucleotide polymorphism scanning by high-resolution melting analysis. Clin Chem 2004;
50:1748 –54.
17. McKinney JT, Longo N, Hahn SH, Matern D, Rinaldo P, Strauss AW, et al.
Rapid, comprehensive screening of the human medium chain acyl-CoA
dehydrogenase gene. Mol Genet Metab 2004;82:112–20.
18. Willmore C, Holden JA, Zhou L, Tripp S, Wittwer CT, Layfield LJ. Detection of
1298
Technical Briefs
c-kit-activating mutations in gastrointestinal stromal tumors by high-resolution amplicon melting analysis. Am J Clin Pathol 2004;122:206 –16.
19. Dobrowolski SF, McKinney JT, Amat di San Filippo C, Giak Sim K, Wilcken B,
Longo N. Validation of dye-binding/high-resolution thermal denaturation for
the identification of mutations in the SLC22A5 gene. Hum Mutat 2005;25:
306 –13.
20. Zhou L, Vandersteen J, Wang L, Fuller T, Taylor M, Palais B, et al.
High-resolution DNA melting curve analysis to establish HLA genotypic
identity. Tissue Antigens 2004;64:156 – 64.
21. Chou LS, Lyon E, Wittwer CT. A comparison of high-resolution melting
analysis to denaturing high performance liquid chromatography for mutation
scanning: cystic fibrosis transmembrane conductance regulator gene as a
model. Am J Clin Pathol 2005;in press.
22. Wartell RM, Hosseini S, Powell S, Zhu J. Detecting single base substitutions, mismatches and bulges in DNA by temperature gradient gel electrophoresis and related methods. J Chromatogr A 1998;806:169 – 85.
23. Abrams ES, Murdaugh SE, Lerman LS. Comprehensive detection of single
base changes in human genomic DNA using denaturing gradient gel
electrophoresis and a GC clamp. Genomics 1990;7:463–75.
24. Nataraj AJ, Olivos-Glander I, Kusukawa N, Highsmith WE Jr. Single-strand
conformation polymorphism and heteroduplex analysis for gel-based mutation detection. Electrophoresis 1999;20:1177– 85.
Previously published online at DOI: 10.1373/clinchem.2005.051516