Rapid [beta]-globin genotyping by multiplexing probe melting temperature and color.
The human [beta]-globin gene sequence (GenBank accession no. U01317) was used to design primers and probes for the amplification of a 214-bp segment containing exon 1 (Fig. 1a). Because of the high homology between the [beta]-globin and [delta]-globin sequences, the primers (sense, 5'-GTCAGGGCAGAGCCATCTA-3'; antisense, 5'-GTTCTATTGGTCTCCTTAAAGGTG-3') were designed with 3' and additional mismatches to [delta]-globin. In addition, because of the close proximity of the hemoglobin mutations, a unique combination of probes was designed to detect Hb S, C, and E alleles. Two probes were labeled with acceptor fluorophores, LightCycler Red 640 (LC Red 640) and LightCycler Red 705 (LC Red 705; Roche Molecular Biochemicals), as mutation detection probes. The third probe was a dual-labeled fluorescein donor probe that spanned the distance between the mutation detection probes. When annealed, resonance energy was transferred from each fluorescein label to either the LC Red 640- or the LC Red 705-labeled probes. The codon 6 detection probe (5'-CTCCTGTGGAGAAGTCTGC-LC Red 640) completely matched the Hb S allele antisense strand. The codon 26 probe (LC Red 705-GTTGGTGGTAAGGCCCTGGT-phosphate) completely matched the Hb E allele antisense strand. Both the LC Red 640 and LC Red 705 probes were obtained from Idaho Technology Biochem. The fluorescein-labeled probe was labeled with two fluorescein (F) molecules attached to the 5' and 3' ends (F-GTTACTGCCCTGTGGGGCAAGGTGAACGTGGATGA-F; Operon). Fifty-five blinded samples of human genomic DNA were randomly selected from samples submitted to Neo Gen Screening for sickle cell hemoglobinopathy screening. The DNA (80-130 ng) was prepared from blots on filter paper (10) and had been genotyped previously by allele-specific cleavage and gel electrophoresis (2).
[FIGURE 1 OMITTED]
PCR and melting curve analysis were preformed on the LightCycler[TM] (Roche Molecular Biochemicals). Asymmetric amplification of the antisense strand occurred in 10 [micro]L of 50 mmol/L Tris, pH 8.5 (25[degrees]C), 3 mmol/L Mg[Cl.sub.2], 500 mg/L bovine serum albumin, 0.2 mmol/L each deoxyribonucleoside triphosphate, 0.5 [micro]mol/L sense primer and 1.0 [micro]mol/L antisense primer, 1 U of KlenTaq DNA polymerase (AB Peptides), 0.2 [micro]mol/L LC Red 640 probe, 0.2 [micro]mol/L LC Red 705 probe, 0.4 [micro]mol/L dual fluorescein-labeled probe, and 2.0 [micro]L of human genomic DNA. The samples were thermally cycled 40 times with three temperature segments. The first segment was 94[degrees]C for 0 s at 20[degrees]C/s for denaturation. A second segment of 63[degrees]C for 30 s at 20[degrees]C/s allowed for both primer and probe annealing. A third temperature segment for extension consisted of a temperature ramp at 1[degrees]C/s from 63 to 75[degrees]C. After amplification, the temperature was raised to 94[degrees]C for 5 s, lowered to 35[degrees]C at 1[degrees]C/s, and held at 35[degrees]C for 20 s. Melting curve profiles where obtained by raising the temperature to 80[degrees]C at 0.1[degrees]C/s while collecting fluorescence data continuously. Genotyping of the samples by [T.sub.m] was accomplished after color compensation of each channel. [T.sub.m]s were determined by converting melting curves to -dF/dT derivative peaks and fitting the peaks to gaussian curves (LightCycler software; Roche Molecular Biochemicals).
Melting curve analysis of codon 6 (Hb S and Hb C), and codon 26 (Hb E) point mutations is shown in Fig. 1b. Panels A-C of Fig. 1b show derivative plots that display "melting peaks" for the LC Red 640-labeled probe, and panel D displays results from the LC Red 705 probe. Panel A of Fig. 1b shows single peaks for both the homozygous Hb S ([T.sub.m] 5 63.7[degrees]C; SD 5 0.18[degrees]C; n = 25) and the wild-type ([T.sub.m] 5 56.4[degrees]C; SD 5 0.26[degrees]C; n = 49) genotypes. Two melting peaks are present for the S-trait corresponding to each allele. Panel B of Fig. 1b shows the melting peaks for homozygous Hb C ([T.sub.m] 5 50.9[degrees]C; SD 5 0.48[degrees]C; n = 8). The C-trait has both the Hb C and wild-type peaks. The genotype of a compound heterozygote, Hb S/C, is shown in Fig. 1b, panel C, along with the corresponding Hb C and Hb S genotypes. Genotyping of the Hb E allele is shown in fig. 1b, panel D. Single melting peaks are shown for homozygous Hb E ([T.sub.m] 5 66.1[degrees]C; SD 5 0.13[degrees]C; n = 3) and wild-type ([T.sub.m] 5 57.1[degrees]C; SD 5 0.20[degrees]C; n = 52) genotypes, whereas the E-trait has two melting peaks corresponding to each allele. The codon 6 and codon 26 probes discriminated the 10 most common hemoglobin genotypes in exon 1. Homozygous samples for the wild type (codon 6, n = 21; codon 26, n = 50), Hb S (n = 1), Hb C (n = 2), and Hb E (n = 1) as well as S-trait (n = 23), C-trait (n = 5), E-trait (n = 2), and the compound heterozygotes Hb S/E (n = 1), C/E (n = 1), and S/C (n = 1) were correctly genotyped. The codon 6 probe sequence that showed the best resolution between the wild-type, Hb S, and Hb C alleles was completely matched to the Hb S allele. Probes that matched either the wild-type or Hb C alleles did not resolve all genotypes as effectively. The total time for amplification and genotyping of 32 samples was <1 h.
Probe melting temperature is dependent on external factors (salt concentration and pH) and intrinsic factors (concentration, duplex length, GC content, and nearest neighbor interactions) (11-14). Mismatches between the oligonucleotide probe and DNA target cause a decrease in the [T.sub.m] of the duplex (15, 16). By designing a probe specific to a mutation locus, genotyping by melting curve analysis in real time is possible. Each allele has a characteristic [T.sub.m], and multiple alleles can be identified by a single probe. Depending on the type of mismatch, varying degrees of destabilization occur. Mismatches that are relatively stable, e.g., G-T mismatches, have [T.sub.m] shifts of 2-3[degrees]C (7), whereas less stable C-A mismatches have 8-10[degrees]C [T.sub.m] shifts (6, 8). The mismatch position and the number of mismatches also affect [T.sub.m] and can be used to enhance polymorphism discrimination (16).
Single-color hybridization probes can identify multiple mutations under a probe and detect unexpected polymorphisms. The cystic fibrosis F508del locus was genotyped recently by melting curve analysis with hybridization probes (17). A 10[degrees]C [T.sub.m] shift was observed between the wild-type and the F508del alleles. Unexpectedly, some samples showed aberrant 5[degrees]C [T.sub.m] shifts. This unique [T.sub.m] was correlated to the F508C allele with a single T-C mismatch to the probe. Subtle changes in [T.sub.m] can also indicate unexpected polymorphisms. A probe to the factor V locus revealed two additional mutations sites characterized by differences in [T.sub.m] as small as 0.8[degrees]C (18). Because a single probe can characterize multiple alleles, we designed a single hybridization probe to cover the adjacent single base changes in exon 1 of the hemoglobin gene that produce Hb C and Hb S. Additional alterations in the sequence under the probes, such as Hb Leiden (deletion of codon 6 and/or 7) or Hb Chesterfield (codon 28 CTG [right arrow] CGG), should also be detectable (9).
Multiple hybridization probes can also be used for multiplex genotyping of loci that are not in close proximity. Multiplexed hybridization probes of the same color have identified four different hemochromatosis alleles and a fifth unexpected variant (8). Two probe pairs each identified two expected alleles with characteristic [T.sub.m]s. The [T.sub.m] of the unexpected variant was within 1.5[degrees]C of a mutant allele. Although the [T.sub.m]s of the five alleles were discernible in this case, unexpected mutations can cause genotyping errors if allele [T.sub.m]s overlap. Although the number of distinct [T.sub.m]s that can be differentiated is limited, different colors can be used to increase the extent of multiplexing. For example, hybridization probes have been used to detect apolipoprotein E mutations at the [epsilon]2/[epsilon]3 and [epsilon]3/[epsilon]4 loci with LC Red 640 and LC Red 705 dyes, respectively (19).
Other methods have been developed to detect polymorphisms with oligonucleotide probes. TaqMan[TM] probes (20, 21) and molecular beacons (22) can be used to correlate a single fluorescent probe to a specific allele. In these cases, the number of different detectable mutations is limited to the number of colors available. Although not performed in real time (21), the TaqMan system has been used to genotype three different point mutations simultaneously by using seven different colored probes (one for background correction). By multiplexing the probe [T.sub.m]s and only two colors, we were similarly able to genotype three different point mutations. An additional advantage of using hybridization probes is that unexpected sequence alterations can be detected.
The ability to multiplex PCR analysis by color and [T.sub.m] has many uses in addition to multiplex genotyping. For example, internal amplification controls often are needed for infectious disease and translocation testing to verify that amplifiable DNA or cDNA is present even if the target amplification is negative. Another common need is for multiplexing a competitor as an internal standard for PCR quantification.
This work was financially supported by an STTR grant from NIH (GM-58983). We thank Alex Chagovetz, Christine Litwin, and Elaine Lyon for insightful conversation, and Doug Searles for probe synthesis.
(1.) Saiki RK, Scharf S, Faloona F, Mullis KB, Horn GT, Erlich HA, Arnheim N. Enzymatic amplification of [beta]-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science 1985;230:1350-4.
(2.) Zhang YH, McCabe LL, Wilborn M, Therrell BL Jr, McCabe ER. Application of molecular genetics in public health: improved follow-up in a neonatal hemoglobinopathy screening program. Biochem Med Metab Biol 1994;52: 27-35.
(3.) Girodon E, Ghanem N, Vidaud M, Riou J, Martin J, Galacteros F, Goossens M. Rapid molecular characterization of mutations leading to unstable hemoglobin beta-chain variants. Ann Hematol 1992;65:188-92.
(4.) Barany F. Genetic disease detection and DNA amplification using cloned thermostable ligase. Proc Natl Acad Sci U S A 1991;88:189-93.
(5.) Wittwer CT, Herrmann MG, Moss AA, Rasmussen RP. Continuous fluorescence monitoring of the rapid cycle DNA amplification. Biotechniques 1997;22:130-7.
(6.) Lay MJ, Wittwer CT. Real-time fluorescence genotyping of factor V Leiden during rapid-cycle PCR. Clin Chem 1997;43:2262-7.
(7.) Bernard PS, Lay MJ, Wittwer CT. Integrated amplification and detection of the C677T point mutation in the methylenetetrahydrofolate reductase gene by fluorescence resonance energy transfer and probe melting curves. Anal Biochem 1998;255:101-7.
(8.) 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.
(9.) Hardison R, Chui DHK, Riemer C, Miller W, Carver M, Molchanova T, et al. Access to "A Syllabus of Human Hemoglobin Variants (1996)" via the World Wide Web. Hemoglobin 1998;22:113-27.
(10.) Heath EM, O'Brien DP, Banas R, Naylor EW, Dobrowolski SF. Optimization of an automated DNA purification protocol for neonatal screening. Arch Pathol Lab Med 2000;123:1154-60.
(11.) Wetmur JG. DNA probes: applications of the principles of nucleic acid hybridization. Crit Rev Biochem Mol Biol 1991;26:227-59.
(12.) Wetmur JG. Formation and structure of nucleic acid hybrids. In: Meyers RA, ed. Molecular biology and biotechnology. New York: VCH, 1995:605-8.
(13.) Brown T, Leonard GA, Booth ED, Kneale G. Influence of pH on the conformation and stability of mismatch base-pairs in DNA. J Mol Biol 1990;212:437-40.
(14.) Gaffney BL, Jones RA. Thermodynamic comparison of the base pairs formed by the carcinogenic lesion [O.sup.6]-methylguanine with reference both to Watson-Crick pairs and to mismatched pairs. Biochemistry 1989;28:5881-9.
(15.) Guo Z, Liu Q, Smith LM. Enhanced discrimination of single nucleotide polymorphisms by artificial mismatch hybridization. Nat Biotechnol 1997; 15:331-5.
(16.) Wallace RB, Shaffer J, Murphy RF, Bonner J, Hirose T, Itakura K. Hybridization of synthetic oligodeoxyribonucleotides to [phi][chi] 174 DNA: the effect of single base pair mismatch. Nucleic Acids Res 1979;6:3543-57.
(17.) Gundry CN, Bernard PS, Herrmann MG, Reed GH, Wittwer CT. Rapid F508del, F508C assay using fluorescent hybridization probes. Genet Testing 1999;3:365-70.
(18.) Lyon E, Millson A, Phan T, Wittwer CT. Detection and identification of base alterations with in the region of factor V Leiden by fluorescent melting curves. Mol Diagn 1998;3:203-10.
(19.) Bernard PS, Pritham GH, Wittwer CT. Color multiplexing hybridization probes using the apolipoprotein E locus as a model system for genotyping. Anal Biochem 1999;273:221-8.
(20.) Livak KJ, Marmaro J, Todd JA. Towards fully automated genome-wide polymorphism screening. Nat Genet 1995;9:341-2.
(21.) Lee LG, Livak KJ, Mullah B, Graham RJ, Vinayak RS, Woudenberg TM. Seven-color, homogeneous detection of six PCR products. Biotechniques 1999;27:342-9.
(22.) Marras SA, Kramer FR, Tyagi S. Multiplex detection of single-nucleotide variations using molecular beacons. Genet Anal 1999;14:151-6.
Mark G. Herrmann,  Steven F. Dobrowolski,  and Carl T. Wittwer  * ( Department of Pathology, University of Utah, 50 N. Medical Dr., Salt Lake City, UT 84132;  Neo Gen Screening, Pittsburgh, PA 15220; * author for correspondence: fax 801-581-4517, e-mail Carl_Wittwer@hlthsci.med.utah.edu)
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Technical Briefs|
|Author:||Herrmann, Mark G.; Dobrowolski, Steven F.; Wittwer, Carl T.|
|Date:||Mar 1, 2000|
|Previous Article:||Melting temperature assay for a UGT1A gene variant in Gilbert syndrome.|
|Next Article:||Procalcitonin is not a reliable marker for the assessment of severity in acute pancreatitis without infectious complications.|