| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
+-Thalassemias by Determining the Ratio of the Two -Globin Gene Copies by Oligonucleotide Hybridization and Melting Curve Analysis
1 Department of Molecular Medicine, Bernhard Nocht Institute for Tropical Medicine, Hamburg, Germany, and Institute of Medical Biometry and Statistics, University Hospital Schleswig-Holstein—Campus Lübeck, Lübeck, Germany;2 Kumasi Centre for Collaborative Research, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana;3 Department of Paediatrics, University of Ulm, Ulm, Germany;
aaddress correspondence to this author at: Bernhard Nocht Institute for Tropical Medicine, Bernhard-Nocht-Strasse 74, 20359 Hamburg, Germany; fax 49-40-42818-512, e-mail timmann{at}bni.uni-hamburg.de
Most cases of 
-thalassemia result from large deletions at the -globin locus (1). The -globin gene cluster contains a tandem array of 2 nearly identical -globin genes (HBA; Fig. 1A
) (2). The 0-thalassemias are characterized by deletions that inactivate both -globin genes of a given chromosome, whereas in +-thalassemias, one gene remains functional. The most widespread +-thalassemias are those designated –3.7 and –4.2, according to the lengths of the deleted fragments (3).
|
The +-thalassemias cannot be diagnosed on clinical grounds, and routine laboratory tests fail, except in the perinatal period, when transient expression of hemoglobin (Hb) Barts may be found (4). Diagnostic assays are also important for genetic counseling because, when occurring in compound heterozygosity with 0-thalassemia, +-thalassemia causes Hb H disease with a thalassemia intermedia phenotype (5).
The classic method to diagnose -thalassemias is the laborious and expensive Southern blot hybridization. Alternative assays include qualitative gap-PCRs (3)(6), which depend on the generation of long PCR products, which is not consistently reproducible (3) in the GC-rich sequences of the -globin gene cluster. In addition, gap-PCRs require the use of deletion-type–specific primers, which limits their sensitivities to distinct variants. Moreover, as for any PCR-based, allele-specific method, they carry the risk that alleles may be amplified from minimal contaminants, such as maternal DNA in prenatal diagnostics or fetal DNA circulating in maternal blood (7)(8)(9).
Quantitative PCR has been proposed to avoid these problems (9). Because of limited sensitivity, however, real-time PCR approaches have been successful only for the diagnosis of 0-thalassemias (9)(10)(11).
We studied a group of 122 individuals from the Ashanti Region of Ghana, West Africa, which has a moderate prevalence of –3.7-thalassemia. Ethics approval was obtained from the Committee for Research, Publications and Ethics of the School of Medical Sciences, Kwame Nkrumah University of Science and Technology, Kumasi. Informed consent was obtained from all participants, and the procedure was explained in the local language. We obtained 5 mL of peripheral blood by venipuncture and isolated genomic DNA by standard procedures. For reference, we used gap-PCR (6) to classify all samples, including additional diagnostic samples with -thalassemias other than the –3.7 deletion (–4.2, –20.5, – –SEA, – –MED, and – –FIL).
The genotypes of one reference sample each of –3.7/–3.7, /–3.7, and / DNA were confirmed by Southern blot hybridization (data not shown). The sequence of the wild-type HBA cluster was obtained from the National Center for Biotechnology Information (GenBank accession no. J00153). The sequences of the HBA2 amplification product used for Southern blot hybridization and of an amplification product obtained by LightCycler PCR were confirmed by bidirectional sequencing on an automated DNA sequencer.
In agreement with previous observations (11), we found that real-time PCR can detect only a minimal 2-fold gene copy reduction and, accordingly, did not allow us to distinguish between the 3 HBA copies of heterozygous +-thalassemia (–/) and the 4 copies in / DNA. We therefore developed an assay that quantifies HBA2 and HBA1 individually. Segments (257 bp) containing intron 2, exon 3, and the flanking 3' untranslated region (3'UTR) of both HBA1 and HBA2 were amplified simultaneously by use of a single pair of primers annealing to sequences identical in both genes. The amplification products were subjected to melting curve analysis (MCA) (12) by use of anchor- and sensor-oligonucleotide hybridization on a LightCycler system (Roche Applied Science). Sequence differences between HBA1 and HBA2 were used to design a sensor probe completely matching the 3'UTR of HBA2 and containing 3 mismatches to the 3'UTR of HBA1 (Fig. 1A
). A fluorescence resonance energy transfer technique (12)(13) with fluorescein (FL)- and LightCycler Red 640 (LC-Red640)-labeled and phosphorylated (p) anchor/sensor oligonucleotides (TIB® Molbiol) was used for detection.
Each 10-µL PCR contained 0.33 µM forward primer (5'-CCTCTTCTCTGCACAGCTCCTAA-3'), 1.0 µM reverse primer (5'-CTGCCGCCCACTCAGACT-3'), 0.1 µM anchor probe (5'-TGAGCACCGTGCTGACCTCCAAATACCGT-FL-3'), 0.1 µM sensor probe (5'-LC-Red640-AAGCTGGAGCCTCGGTAGCCGTTC-p-3'), and 
10 ng of template DNA in LightCycler FastStart DNA Master Hybridization Probes (Roche) mixture for hot-start PCR, with MgCl2 added to a final concentration of 4 mM. PCR conditions were initial denaturation at 95 °C for 480 s; 52 cycles of 95 °C for 4 s, 60 °C for 10 s (acquisition of fluorescence at the end of the annealing time in "single" mode), and 72 °C for 10 s; heating to 95 °C for 5 s; and cooling to 50 °C for 120 s. Melting curves were then obtained by heating the sample from 50 to 82 °C with a ramp rate of 0.1 °C/s (acquisition of fluorescence in step mode). Estimates of the numbers of template HBA copies were derived from real-time PCR data obtained during the amplification phase of the test reaction. The samples had crossing points of 36 (range, 32–40) cycles. Each LightCycler run contained a blank and reference samples of –3.7/–3.7, –3.7/, and / DNA. The first derivatives of the melting curves were generated with the LightCycler algorithm, using a linear mode without background suppression and with a temperature range of 7.5 °C to average.
The MCA discriminated –/–, –/, and / DNA (Fig. 1, B and C
). We believe that alternative reporting systems, such as quenching of fluorochromes, or the use of other PCR methods, such as the TaqMan techniques, may give results of similar sensitivity.
We refined the method by determining the area under the melting curve (AUC), using the built-in LightCycler algorithm without normalization. The intraassay variation of AUC2/AUC1 ratios in a simultaneous 30-fold replication of a single sample gave ratios <0.05 for –/–, 0.80 (SD = 0.029; range, 0.74–0.87) for –/, and 1.32 (0.057; range, 1.21–1.43) for / (see Fig. 1A
of the Data Supplement that accompanies the online version of this Technical Brief at http://www.clinchem.org/content/vol51/issue9/). The corresponding interassay variations in 30 successive assays were <0.05, 0.77 (SD = 0.081; range, 0.56–0.91), and 1.26 (0.089; range, 1.12–1.45), respectively (see Fig. 1B
of the online Data Supplement). Applying the assay to DNA from 122 individuals of a population with a moderate prevalence of the –3.7 deletion produced 3 groups of AUC2/AUC1 ratios (Table 1
; see also Fig. 1C
of the online Data Supplement). MCA gave reproducible and clearly discernible results and yielded the same results as the allele-specific method. For serial testing, it may be an advantage that PCR amplification and MCA can be performed in a single test tube and in 
80 min. MCA does not require the analysis of a single-copy gene for reference because HBA1 and HBA2 serve as reference to each other. Because MCA does not use allele-specific PCR, it will not be affected by small amounts of contaminating DNA (7)(8)(9).
|
Nearly all of the deletional +-thalassemias, including the deletion types –2.7, –3.5, –3.7I, –3.7II, –3.7III, and –4.2, leave intact one 3'UTR of either HBA2 or HBA1 (3); therefore, all of these can most likely be diagnosed by MCA. We confirmed the applicability of the MCA test in an individual homozygous for the –4.2 deletion (Table 1
). The majority of 0-thalassemias will not be identified. Complete deletion of both -globin genes on a given chromosome in the homozygous state will cause PCR amplification and probe hybridization to fail, thereby indicating a need for further analysis. This applies only to prenatal diagnostics because of perinatal fatality of homozygotes. 0 heterozygosity will yield an HBA2/HBA1 ratio similar to that of wild-type DNA. Accordingly, wild-type MCA ratios were found in heterozygous individuals with variants of 0-thalassemias referred to as – –SEA, – –MED, and – –FIL (1) (Table 1
). An exception is the –20.5 deletion, which affects both globin genes but spares the 3'UTR of HBA1 and therefore should give MCA results similar to those for +-thalassemias. This was confirmed in 1 individual carrying the –20.5 deletion (Table 1
). Compound heterozygous +-thalassemias and 0-thalassemias (– –/–) producing Hb H disease are expected to give MCA results similar to those of homozygous +-thalassemias.
Hematologic indices can aid interpretation of results. An AUC2/AUC1 ratio of 1.2–1.5 would indicate / and not – –/ if the erythrocyte mean cell volume is within reference values. Likewise, a ratio <0.05 would indicate – –/– and not –/– if brilliant cresyl blue staining reveals an increase in Hb H inclusion bodies.
Several rare +-thalassemias (1) are expected to escape detection, such as deletions not involving a 3'UTR in ()5.3, nondeletional variants of the -globins, variants affecting HBA regulatory elements, or certain multiplications of HBA genes, such as, for example, in –/ genotypes.
MCA offers new perspectives in the diagnosis of +-thalassemias because it is rapid, requires little blood, can be performed in a single tube, and is expected to be resistant to fetal/maternal contamination. It remains to be determined whether MCA can be combined with the previously described real-time PCR (9)(10)(11), which would allow diagnosis of the majority of individuals affected by +- and/or 0-thalassemias.
Acknowledgments
This study was supported by the National Genome Research Network (NGFN) of the German Ministry of Education and Research (BMBF) and by the Volkswagen Foundation, Germany. This work is part of the doctoral thesis of F.M. at the Faculty of Medicine, University of Hamburg, Germany.
References
-thalassemia. Steinberg MH Forget BG Higgs DR Nagel RL eds. Disorders of hemoglobin: genetics, pathophysiology, and clinical Management 2001:405-430 Cambridge University Press Cambridge, UK. .
globin genes lie close together in cellular DNA. Proc Natl Acad Sci U S A 1978;75:5950-5954.-thalassemia. Blood 2000;95:360-362.-thalassemia-1 of Southeast Asian type deletion in Taiwan. Clin Genet 2001;60:305-309.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
-thalassaemia (Hb Barts hydrops fetalis). Br J Haematol 2001;115:341-346.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
-thalassemia model. Diagn Mol Pathol 1994;3:246-254.[Web of Science][Medline]
[Order article via Infotrieve]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
