|
|
||||||||
Departments of
1
Pediatrics and
2
Obstetrics & Gynecology, National University of Singapore, Singapore 119074, Singapore.
3 The Childrens Medical Institute and
4
Molecular Diagnosis Center, Department of Laboratory Medicine, National University Hospital, Singapore 119074, Singapore.
5 Departments of Pediatrics and of Gynecology &Obstetrics, The Johns Hopkins University School of Medicine, Baltimore, MD 21287.
aAddress correspondence to this author at: Department of Pediatrics, National University of Singapore, Level 4, National University Hospital, 5 Lower Kent Ridge Rd., Singapore 119074, Singapore. Fax 65-6779-7486; e-mail paecs{at}nus.edu.sg.
| Abstract |
|---|
|
|
|---|
90% of ß-thalassemia alleles in these places. We describe a multiplex minisequencing assay to detect these common mutations.
Methods: Gap-PCR was used to simultaneously amplify the ß-globin gene from genomic DNA and to detect the
619bp deletion mutation. Multiplex minisequencing was then performed on the amplified ß-globin fragment to detect an additional 15 common Southeast Asian and Indian ß-thalassemia mutations. Site-specific primers of different lengths were subjected to multiple rounds of annealing and single-nucleotide extension in the presence of thermostable DNA polymerase and the four dideoxynucleotides, each labeled with a different fluorophore. Minisequencing products were separated and detected by capillary electrophoresis, followed by automated genotyping. The optimized assay was subjected to a double-blind validation analysis of 89 ß-thalassemia and wild-type DNA samples of known genotype.
Results: Homozygous wild-type or mutant DNA samples produced electropherograms containing only a single colored peak for each mutation site, whereas samples heterozygous for a specific mutation displayed two different-colored peaks for that mutation site. Samples were automatically genotyped based on color and position of primer peaks in the electropherogram. In the double-blind validation analysis, all 89 DNA samples were genotyped correctly (100% assay specificity).
Conclusions: The described semiautomated multiplex minisequencing assay can detect the most common Southeast Asian and Indian ß-thalassemia mutations, is amenable to high-throughput scale up, and may bring population-based screening of ß-thalassemia in endemic regions a step closer to implementation.
| Introduction |
|---|
|
|
|---|
-thalassemias, which are predominantly produced by deletions in the
-globin gene cluster, most ß-thalassemias are caused by point mutations or small deletions within the ß-globin gene or its immediate flanking sequences (1). However, in each affected ethnic group, a few common mutations together with a variable number of rare mutations account for most of the cases (2).
In Southeast Asia and India, ß-thalassemia is a serious public health problem throughout the region. In these regions, <20 variations account for the overwhelming majority of ß-thalassemia alleles. These include -29 A
G; -28 A
G; codon 0 T
G; codon 8/9 +G; codon 17 A
T; codon 19 A
G; codon 26/HbE G
A; codon 27/28 +C; IVSI,1 G
T; IVSI,5 G
C; codon 35 -C; codon 41/42 -TTCT; codon 43 G
T; codon 71/72 +A; IVSII, 654 C
T; and the
619bp deletion (1)(3)(4)(5)(6)(7)(8). Three variants alone account for more than two-thirds of the ß-thalassemias in the combined region; IVSI,5 G
C (33%); codon 41/42 -TTCT (27%); and IVSI,1 G
T (9%) (1).
Various PCR-based strategies and technologies have been applied to the molecular analysis and prenatal diagnosis of ß-thalassemia. These include restriction fragment length polymorphism analysis (9), dot-blot hybridization with allele-specific oligonucleotides (10)(11)(12), denaturing gradient gel electrophoresis (13)(14), reverse dot-blot hybridization (15)(16)(17)(18), direct DNA sequencing (19), and amplification refractory mutation system (ARMS)1 (20)(21).
More recently, fluorescence-based multiplex minisequencing followed by gel electrophoretic size separation has been used to simultaneously detect multiple mutations and other nucleotide variants (22)(23)(24). In minisequencing, a primer is hybridized to DNA next to a variant nucleotide site and extended with DNA polymerase by a single appropriate dideoxyribonucleotide triphosphate (ddNTP) that matches the nucleotide at the target site. Minisequencing on solid-phase arrayed primers to detect 10 point mutations in the ß-globin gene has also been demonstrated (25). Arrayed minisequencing strategies are extremely attractive for multimutation screening applications, but they require robotics and other instrumentation that are beyond the reach of most diagnostic laboratories.
In this report, we present a rapid screening procedure based on liquid-phase multiplex minisequencing and capillary electrophoresis to detect 16 of the most common Southeast Asian and Indian ß-thalassemia mutations.
| Materials and Methods |
|---|
|
|
|---|
|
ß-globin gene gap-pcr and
619bp mutation detection
A gap-PCR strategy was adopted to amplify an intact ß-globin gene fragment and/or a
619bp deletion junction fragment, if present, involving the use of forward primer ß-F (5'-ACGGCTGTCATCACTTAGAC-3'; GenBank HUMHBB sequence nucleotides 6201062029) and two reverse primers, ß-R1 (5'-AAGAGGTATGAACATGATTAGC-3'; HUMHBB sequence nucleotides 6346663445) and ß-R2 (5'-CAGATTCCGGGTCACTGTG-3'; sequence nucleotides 6429964281; Fig. 1A
). Genomic DNA template (100 ng) was amplified in a T3 thermal cycler (Biometra) in a total volume of 50 µL containing 0.2 µM each of the three primers, 200 µM each deoxynucleotide triphosphate, and 1 U of HotStarTaq polymerase in 1x supplied PCR buffer (Qiagen). An initial denaturation step at 95 °C for 15 min was followed by 35 cycles of incubation at 98 °C for 45 s, 55 °C for 45 s, and 72 °C for 90 s; final extension was at 72 °C for 5 min.
|
A 10-µL aliquot of each amplified product was resolved on a 1% agarose gel in 1x Tris-borate-EDTA at 15 V/cm for 40 min. The presence of a 1671-bp deletion junction fragment, in addition to a 1457-bp fragment, indicated heterozygosity for the
619bp allele (Fig. 1B
).
multiplex minisequencing
Excess PCR primers and unincorporated deoxynucleotide triphosphates in each PCR product were fragmented and functionally inactivated, respectively, in a one-step reaction by the addition of ExoI and shrimp alkaline phosphatase (SAP; United States Biochemical). We added 1 µL of SAP (1 U/µL) and 0.5 µL of ExoI (10 U/µL) to 2.5 µL of PCR product to produce a final volume of 4 µL. The mixture was incubated at 37 °C for 15 min, followed by enzyme deactivation at 80 °C for 15 min.
To each tube of purified PCR product, we added 1 µL of panel A or B mutation-detection primer mixture (see below), 2.5 µL of HPLC-grade water, and 2.5 µL of SNaPshotTM Multiplex Ready Reaction Mix (Applied Biosystems) containing AmpliTaq® DNA polymerase and fluorescently labeled ddNTPs. Each 10-µL multiplex minisequencing mixture was subjected to 25 single-base extension cycles consisting of denaturation at 96 °C for 10 s, primer annealing at 50 °C for 5 s, and extension at 60 °C for 30 s.
After cycle minisequencing, unincorporated fluorescent ddNTPs were enzymatically inactivated by incubation with 1 U of SAP at 37 °C for 1 h, followed by enzyme deactivation at 75 °C for 15 min.
ß-globin gene mutation-detection primers
We divided 15 ß-globin gene mutation-detection primers into two panels, A and B, which was necessitated by the close proximity and clustering of many of these mutations. Panel A consisted of eight primers, whereas panel B consisted of seven primers (Fig. 1C
). Within each panel, each mutation-detection primer differed in total length from the others. This was achieved through the addition of variable-length nonspecific polynucleotide tails to the 5' ends of the primers. The ability to differentiate between the mutation-detection primers based on length thus allowed the minisequencing reaction to be multiplexed. The use of nonspecific tails simplified standardization of annealing temperatures for all primers in both panels regardless of total primer length. All minisequencing primers were purified by polyacrylamide gel electrophoresis or HPLC. Table 2
provides details of the primers in each panel and their concentrations in the respective primer mixtures.
|
capillary electrophoresis and genotype analysis
Multiplex minisequencing products were resolved by automated capillary electrophoresis on an ABI PRISM® 3100 Genetic Analyzer (Applied Biosystems). Briefly, 9 µL of HiDiTM formamide and 0.5 µL of GeneScan-120 LIZ size calibrator (Applied Biosystems) were added to 0.5 µL of multiplex minisequencing product; the fluorescently tagged extended primers in the mixture were electrophoretically separated across a 36-cm capillary containing POP-4 polymer (Applied Biosystems) for 25 min and analyzed using GeneScanTM application software (Applied Biosystems). Automated allele calling and tabula-tion of results were accomplished with the aid of GenotyperTM 3.7 application software (Applied Biosystems).
| Results |
|---|
|
|
|---|
619bp deletion mutation directly. Under the recommended PCR conditions, only a 1457-bp fragment will be amplified from a wild-type allele by primers ß-F and ß-R1 (Fig. 1
619bp allele, the annealing sequence for primer ß-R1 is deleted; thus 1671-bp fragment will be amplified from primers ß-F and ß-R2. In individuals heterozygous for the
619bp allele, therefore, both fragments are amplified and easily discriminated by simple agarose gel electrophoresis.
To detect the point mutations in the ß-globin gene, we performed multiplex minisequencing on the amplified PCR products. Given the close proximity of the majority of ß-globin gene mutations, the 15-mutation assay was divided into two multiplex panels and mutation-detection primers were designed to either the sense or complementary strand of the gene to eliminate primer-primer overlap and primer-dimer interactions in the reaction (Fig. 1C
and Table 2
).
In minisequencing, a mutation-detection primer anneals such that its 3' nucleotide ends just before the mutation site and a cycle sequencing reaction is performed in the presence of Taq DNA polymerase and a mixture of four terminator nucleotides (ddNTPs), each labeled with a different fluorescent molecule. As a result, each primer molecule is extended by one of the four dye-terminators, and the fluorescent tag(s) on the extended primer serves as a reporter of the wild-type and/or mutant genotype of the template DNA. In wild-type or homozygous mutant samples, only a wild-type or mutant dye-terminator is attached to the primer; therefore, only one primer peak is detected on an electropherogram. With a heterozygous sample, however, either the wild-type or mutant dye-terminator will attach to the mutation-detection primer; hence two different fluorescent signals/peaks will be detected. The same primer tagged with two different dye-terminators may or may not migrate identically, depending on molecular weight differences between the nucleotide-fluorophore combinations. In addition, wild-type and mutant allele peak heights may differ significantly because of differences in fluorescence emission of the different fluorophores (Fig. 2
).
|
In each multiplex minisequencing panel, several mutation-specific primers were designed for simultaneous annealing and single-base extension. All primers except one were designed with 5' nonspecific poly(dA) or poly(dGACT) tails of different lengths to enable easy differentiation of terminator-incorporated primers based on size. Table 2
summarizes the primers and their concentrations in the respective primer mixtures as well as the expected results after incorporation of dye-terminators. Thus, the position of the extended primer peak in the electropherogram specifies the mutation locus, whereas the peak color(s)/fluorescence specifies the genotype (Figs. 2
and 3
).
|
To demonstrate that all the mutations can be detected in both the homozygous and heterozygous state, we analyzed genomic DNA samples carrying these mutations. Where patient samples homozygous for particular mutations were unavailable, we cloned the mutant ß-globin genes from heterozygous DNA samples and performed the multiplex minisequencing directly on the purified plasmid constructs. Such tests were performed to confirm the absence of spurious "wild-type" peaks that could arise from nonspecific incorporation of other dye-terminators in the homozygous mutant state and that would lead to misdiagnosis of homozygotes as heterozygotes. In all instances, expected wild-type and/or mutant peaks were observed in the presence of the wild-type and/or mutant alleles, thus confirming the specificity of the assay (data not shown).
GeneScan 3.7 application software was used to automatically analyze the results after capillary electrophoresis. The GeneScan electropherograms of several patient samples after multiplex minisequencing with primer panels A and B are shown in panels A and B of Fig. 2
, respectively. To further automate the allele-calling process, the results were reanalyzed with Genotyper 3.7 application software. A "macro" file was created for both multiplex minisequencing panels, such that each peak in the Genotyper-generated electropherogram could be automatically identified and labeled by launching the macro application (Fig. 3
). The labels were designed to provide information on the mutation site, the nucleotide incorporated, and whether it was a wild-type or mutant allele. Additionally, a template was created to automatically generate a tabulated report of the Genotyper electropherogram results (Table 3
). Automated allele-calling and report generation serve not only to simplify the diagnostic process but, more importantly, to minimize human errors in data transcription.
|
As a final validation of the multiplex minisequencing assay, we performed a double-blind analysis of 81 ß-thalassemia patient samples and 8 wild-type controls. The genotypes of these samples were determined previously either by ARMS-PCR analysis or by direct DNA sequencing (Table 1
). Samples were coded and assayed by different individuals, and results were scored independently by the person performing the assay and by a third individual. The genotypes scored by both individuals were completely concordant, and all 89 samples were correctly genotyped.
| Discussion |
|---|
|
|
|---|
T are the most common mutations in Southern Chinese. These two mutations together with codon 17 A
T; -28 A
G; IVSI,5 G
C; IVSI,1 G
T; and codon 71/72 +A account for
90% of all ß-thalassemia alleles in this population (3)(4)(5)(8). In contrast, codon 26/HbE G
A is very common in a large part of mainland Southeast Asia, from Burma in the west to Vietnam in the east, as well as in aboriginal Malays (1). In India, codon 8/9 +G; IVSI,1 G
T; IVSI,5 G
C; codon 41/42 -TTCT; and
619bp together account for >90% of ß-thalassemia mutations (1)(6). The cosmopolitan population of Singapore consists mainly of Southern Chinese (76.8%), Malays (13.9%), and Indians (7.9%), with Eurasians making up the largest proportion of the remaining ethnic groups (26). With a 34% ß-thalassemia carrier frequency in Singapore (27) and much higher frequencies in many parts of Southeast Asia and India (1), prenatal diagnosis and screening of ß-thalassemia are currently the best solutions to minimizing the incidence of thalassemia major and alleviating its considerable healthcare burden.
Many molecular methods have been used for the screening and diagnosis of ß-thalassemia. Among them, hybridization with allele-specific probes is the most widely used method (4)(15)(16)(17)(18). A major advantage of the minisequencing reaction principle over hybridization with allele-specific oligonucleotide probes is that the distinction between the sequence variants is based on the high accuracy of the nucleotide incorporation reaction catalyzed by a DNA polymerase rather than on differences in thermal stability between mismatched and perfectly matched hybrids formed with the allele-specific oligonucleotide probes (28). Current versions of thermostable enzymes used in minisequencing have extremely low error rates and have considerably enhanced efficiency and specificity for ddNTPs (29)(30)(31)(32). These properties provide negligible primer misincorporation and excellent discrimination between homozygous and heterozygous genotypes. Moreover, the same reaction conditions can be used for detecting any variable nucleotide irrespective of the nucleotide sequence flanking the variable site (28).
Minisequencing is a direct method compared with denaturing gradient gel electrophoresis because it determines the exact nucleotide at the mutation site and does not need other methods to confirm results. Minisequencing also accurately differentiates between heterozygosity and homozygosity within the same reaction, based on color and number of the peaks at each mutation site in the electropherogram.
Another advantage of minisequencing is its multiplexing capability, with several mutations being screened in the same reaction tube. Although several mutations can be simultaneously detected by multiplex ARMS (21) or combined ARMS (33), only two to five mutations can be detected at the same time. Our multiplex minisequencing assay detects a total of 15 mutations in two parallel reactions. The use of two reactions was necessitated by the fact that a few mutation sites in exon 1 and intron 1 of the ß-globin gene are very close together, such as codon 17; codon 19; codon 26; codon 27/28; IVSI,1; and IVSI,5. Placing all 15 mutation-specific minisequencing primers in the same reaction would have produced inefficient primer annealing because of overlap of primer-binding sites and/or formation of primer-dimers from overlapping sense and antisense primers. Had the mutations been located sufficiently distant from each other, simultaneous detection of all 15 mutations in the same reaction would have been theoretically possible.
Compared with direct DNA sequencing, minisequencing assays are more rapid because automated fluorescent capillary electrophoresis of minisequencing products requires only 25 min compared with 2.5 h for capillary electrophoresis of standard sequencing products. Additionally, data analysis of minisequencing electropherograms is comparatively simple and easily automatable with the aid of Genotyper 3.7 application software, which not only provides substantial savings in time and manpower but, more importantly, minimizes human error during data analysis. Furthermore, Genotyper can be programmed to generate a tabulated report of the electropherogram results, further reducing human errors in data transcription.
The major advantage of DNA sequencing is its ability to interrogate every nucleotide in the ß-globin gene. There is application software available for alignment of sample DNA sequences against a wild-type sequence template. However, once any variant nucleotide has been identified, the mutation site it represents must be manually determined. Additionally, several sequencing reactions are required to achieve complete coverage of the gene, and a single standard sequencing reaction may not be able to detect all the common mutations. For example, Cd41/42 and IVSII,654 are
840 bp apart and thus require high-quality very long-read sequencing to accurately genotype both mutations in the same sequencing reaction.
Our assay contains two main modifications to the manufacturer-recommended SNaPshot minisequencing protocol. The first is that we discovered that minisequencing primers with very long homopolymeric dA tails produces a high background of nonspecific peaks. Conversion of the tails to poly(dGACT) resolved the nonspecific peaks. The second modification was that we successfully decreased the SNaPshot reagents used in the minisequencing reaction to 50% of the recommended amount without any adverse effects on the specificity or reproducibility of the assay.
We have calculated the test costs of this assay at
S$7.50 (US $4.20) per patient sample. This cost includes PCR amplification reagents and primers, purification enzymes, minisequencing reagents and primers, and GeneScan electrophoresis costs (polymer, capillary array, and other reagents) but excludes manpower. Because of the automated electrophoresis and data analysis capabilities of the Genetic Analyzer instrument and Genotyper software, respectively, this assay is most cost-effective in diagnostic laboratories with moderate to high patient sample volumes. This is because up to 96 DNA samples can be analyzed within 12 h by a single technologist on 50% effort (4h of an 8-h day), with fewer samples leading to shorter turnaround times.
A double-blind validation analysis of this assay on 89 genomic DNA samples of known genotype produced 100% assay specificity. These samples were from wild-type individuals, heterozygous carriers, or ß-thalassemia patients homozygous or compound heterozygous for the 16 mutations in the two panels. It should be noted that, like all other previously described assays, our assay does not detect all ß-thalassemia mutations. Therefore, patients who are compound heterozygous for a common point mutation and a rarer large deletion, such as in
ß-thalassemia, will appear as "homozygous" for the common point mutation. This is because the deleted allele is not amplified by the ß-globin gene primers, and thus, only the allele carrying the common point mutation will be amplified and detected. For affected patients who are genotyped as homozygous for a mutation, it might be prudent to genotype the parents to rule out the rare possibility of compound heterozygosity with a large deletional allele.
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
- and ß-thalassemias in Chinese. J Formos Med Assoc 1998;97:5-15.[Web of Science][Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
![]() |
W. Thongnoppakhun, S. Jiemsup, S. Yongkiettrakul, C. Kanjanakorn, C. Limwongse, P. Wilairat, A. Vanasant, N. Rungroj, and P.-t. Yenchitsomanus Simple, Efficient, and Cost-Effective Multiplex Genotyping with Matrix Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry of Hemoglobin Beta Gene Mutations J. Mol. Diagn., July 1, 2009; 11(4): 334 - 346. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Glynou, P. Kastanis, S. Boukouvala, V. Tsaoussis, P. C. Ioannou, T. K. Christopoulos, J. Traeger-Synodinos, and E. Kanavakis High-Throughput Microtiter Well-Based Chemiluminometric Genotyping of 15 HBB Gene Mutations in a Dry-Reagent Format Clin. Chem., March 1, 2007; 53(3): 384 - 391. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. P. Vichinsky, E. A. MacKlin, J. S. Waye, F. Lorey, and N. F. Olivieri Changes in the Epidemiology of Thalassemia in North America: A New Minority Disease Pediatrics, December 1, 2005; 116(6): e818 - e825. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. P. Yip, S. F. Pun, K. H. Leung, and S. Y. Lee Rapid, Simultaneous Genotyping of Five Common Southeast Asian {beta}-Thalassemia Mutations by Multiplex Minisequencing and Denaturing HPLC Clin. Chem., October 1, 2003; 49(10): 1656 - 1659. [Full Text] [PDF] |
||||
![]() |
W. Wang, E. S.K. Ma, A. Y.Y. Chan, D. H.K. Chui, and S. S. Chong Multiple Minisequencing Screen for Seven Southeast Asian Nondeletional {alpha}-Thalassemia Mutations Clin. Chem., May 1, 2003; 49(5): 800 - 803. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |