Article Text

Download PDFPDF

Rapid screening for the most common β thalassaemia mutations in south east Asia by PCR based restriction fragment length polymorphism analysis (PCR-RFLP)
  1. Eijkman Institute for Molecular Biology, Jl Diponegoro 69, Jakarta 10430, Indonesia
  2. Department of Clinical Pathology, University of Indonesia, Salemba 6, Jakarta 10430, Indonesia
    1. Eijkman Institute for Molecular Biology, Jl Diponegoro 69, Jakarta 10430, Indonesia
    2. Department of Clinical Pathology, University of Indonesia, Salemba 6, Jakarta 10430, Indonesia

      Statistics from

      Request Permissions

      If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

      Editor—The heterogeneity of the molecular lesions which underlie the failure of erythropoietic cells to synthesise normal haemoglobin in β thalassaemia1 is a complicating factor in its molecular diagnosis. However, although more than 100 different mutations have been identified, mostly single base substitutions or small deletions and insertions in the β globin gene, in many populations the bulk of β thalassaemia is caused by a population specific spectrum of only a small number of mutations.1The strategy for the detection of mutations in patients, therefore, normally involves screening in the first instance for a small number of mutations that are the most common for the population concerned.

      Two of the most commonly used screening procedures are dot blot or reverse dot blot hybridisation2 and the amplification refractory mutation system (ARMS).3 While these procedures are satisfactory in the hands of the more experienced specialised laboratories, the exacting conditions required for the performance of allele specific hybridisation or PCR amplification steps have made them less reproducible in less advanced laboratories.4 5 In south east Asia, where in some regions the frequency of β thalassaemia can be as high as 10%,6 7 many laboratories in medium sized provincial hospitals and universities are suitably equipped for routine molecular biology laboratory testing, but do not have the skill to comply with the strict conditions demanded by the above procedures, a situation which is perhaps shared by many other countries in which β thalassaemia is common. We have endeavoured, therefore, to develop a more robust and accurate alternative method for the detection of β thalassaemia mutations in south east Asia, based on restriction endonucleases that recognise naturally occurring or PCR generated restriction sites associated with the β thalassaemia mutations. The approach is widely used in the detection of pathological mutations in mitochondrial DNA8 and has also been applied in the molecular diagnosis of β thalassaemia.9

      A strategy has been devised to allow rapid detection of nine of the most common mutations in south east Asia which requires the amplification of two segments only of the β globin gene. The mutations are at positions IVS-1 nt5, IVS-1 nt1, codon 26, codon 15, codon 17, codon 19, codon 30, IVS-1 nt2, and codon 41-42 of the β globin gene (fig 1 (top), table 1), which together account for around 70-90% of β thalassaemia mutations in most populations of south east Asia.6 7 10 The PCR amplifications use primer sets TLF62028-TLR62320 and TLF62392-TLR62703. Primer TLR62320 includes a G at the position equivalent to nt8 of intron 1 instead of the normal A to create a GCTAGC site for Cac8I in the presence of the IVS-1 nt5 G>C mutation. ACac8I site is also created in the presence of the IVS-1 nt2 T>C mutation which represents less than 1% of the β thalassaemia alleles in Indonesia. The mutations G to T at IVS-1 nt1, G to A at codon 26, and A to G at codon 19 abolish the natural occurring sites for BslI,MnlI, and MaeII, respectively. The mutations G to A at codon 15, A to T at codon 17, and G to C at codon 30 create sites for SfcI,BfaI, andBsp1286I, respectively. The detection of the 4 bp deletion of codon 41-42 is essentially as described by Changet al 9; a C has been introduced at the second position of codon 41 in the sequence of primer TLF62392, which together with the 4 base deletion creates a TCGA site forTaqI.

      Figure 1

      Detection of the common β thalassaemia mutations of south east Asia by PCR-RFLP. (Top) Detection strategy; see table 1 for details. (Bottom) Results for the detection of (A) IVS-1 nt5, (B) codon 26, (C) IVS-1 nt1, (D) codon 41/42, (E) codon 15, (F) codon 19, and (G) IVS-1 nt2 mutations; lane a, uncut PCR product; lane b, patient homozygous for the β thalassaemia mutation; lane c, heterozygote for the β thalassaemia mutation; and lane d, normal subject.

      Table 1

      PCR primers and restriction endonucleases used in the detection procedure

      As shown in fig 1(A-D) the method produces unambiguous results. Thus, for the detection of the IVS-1 nt5 mutation, the normal allele (indicated by a 293 bp undigested product) can be distinguished readily from the mutant allele (a 257 bp digested fragment). Heterozygosity could be easily detected (fig 1A). Similarly, the 122 bp fragment of the mutant allele could be distinguished from the 60 and 62 bp fragments of the normal allele among the digestion fragments ofMnlI in the detection of the HbE codon 26 mutation (fig 1B). Definitive electrophoretic patterns were also obtained in the detection of IVS-1 nt1 (fig 1C), codon 41-42 (fig 1D), codon 15 (fig 1E), codon 19 (fig 1F), and IVS-1 nt2 (fig 1G) mutations. Similar results were obtained in the detection of codon 17 and codon 30 mutations (data not shown). In all the above cases, the PCR-RFLP results agreed with those of ARMS and were confirmed by DNA sequencing.

      The detection procedure described here does not involve an allele specific hybridisation or an allele specific PCR amplification step and is thus less prone to non-specific reactions which could lead to false positive/negative results. We have applied our new procedure to the detection of the underlying mutations in a number of difficult samples in which ARMS gave ambiguous results. In all cases the PCR-RFLP method proved to be more reliable and gave definitive identification of the underlying mutation, as confirmed by DNA sequencing data (data not shown). The nine mutations included here account for 70% of β thalassaemia alleles in Thailand,6 90% in Malaysia,6 53% in India,6 and 68-90% in Indonesia7 10 depending on the ethnic population. The procedure, therefore, is suitable as the front line screening for the molecular diagnosis of β thalassaemia in south east and perhaps also in eastern and southern Asia.


      This work was supported by the RUT grant No IIIo/13/I/-/IPD from the National Research Council (Indonesia) to Iswari Setianingsih, and grant in aids from PT Krakatau Steel and PT Inti through the Agency for Strategic Industries (Indonesia).