Maryam Neishabury PhD**, Farzin Pourfarzad MS* **, Sayeh Jalil-Nejad MD*, Maryam Azad BS*,

Christian Oberkanins PhD***, Walter Krugluger MD PhD****

Iran, with more than 25,000 affected individuals, represents one of the areas in the world with an unusually high prevalence of beta-thalassemia. The gene frequency of thalassemia shows great variation within Iran from one area to another. Provinces around the Persian Gulf and the

Table 1. Mutations investigated by ARMS and reverse hybridization.

Caspian Sea with a gene frequency of more than 10% constitutes the thalassemia major zones in Iran. Fars province in southern Iran shows a gene frequency of 8 to 10 percent, whereas the prevalence of the disease varies between 4% and 8% in other parts of the country.^5-7

The use of sensitive and reliable diagnostic methods plays a critical role in good screening and therefore, prevention of beta-thalassemia. Regarding the heavy emotional and economical burden imposed on the society, identification of beta-thalassemia carriers and prenatal diagnosis of the disease through an accurate and quick process has been a major goal for researchers in Iran during the past two decades.

During the last 8 to 10 years, amplification refractory mutation system (ARMS) and restriction fraction length polymorphism (RFLP) were the principal techniques for diagnosis. However, Iran with a multiethnic population of around 65 million, represents a vast spectrum of beta-thalassemia mutations, making these techniques too labor-intensive, slow and expensive. Therefore, in recent years modern molecular biology techniques, such as reverse hybridization and direct sequencing, have been implemented for the diagnosis of this disease. We report the application of reverse hybridization for detection of hemoglobin S and the 10 most common mutations (Table 1) and demonstrate its advantages versus ARMS in the detection and prenatal diagnosis of beta-thalassemia mutations as well.

ARMS is a PCR-based method, which uses allele-specific priming. In this method, an oligo-nucleotide primer with a triple end complementary to the sequence of a specific mutation, coupled with a common primer is used in one PCR reaction. In parallel, a corresponding normal primer coupled with a common primer is used in another PCR reaction. The presence of an amplified product in the first reaction indicates the presence of the mutation while its absence suggests presence of the normal DNA sequence at that specific site. In the second reaction the presence of an amplified product suggests presence of a normal DNA sequence at that specific site while its absence suggests presence of the mutation⁸ (Figure 1).

Figure 1 shows ARMS-PCR turn on a 2% agarose gel. Presence of an amplified mutant the ARMS primer product indicates presence of the mutant allele. All samples contain an internal control band. Sample 1 contains an amplified product in the normal primer but lacks it in the mutant primer; hence, implying a normal individual. Sample 2 contains an amplified product in both the normal and mutant primers; assigning the individual of minor type. Sample 3 contains an amplified product only in the mutant primer; hence, it is the sample of an individual who is of major type.

Figure 1. ARMS-PCR gel for beta-thalassemia diagnosis.

A total of 8 to 10 specimens of EDTA blood samples, amniotic fluid or chorionic villous sampling (CVS) were obtained from randomly selected clinically diagnosed beta-thalassemia minor and major patients. DNA was extracted from the samples using the saturated salt technique.⁷

In order to carry out ARMS, the DNA of interest was first extracted. After DNA extraction, PCR reactions were set up in two separate tubes for each sample. One test tube for the amplification of the normal ARMS primer and the second for the amplification of the mutant ARMS primer. The primers used for ARMS were kindly provided by JM Old.⁹

For reverse hybridization, a commercially available assay (ß-globin) strip assay (Vienna Lab, Austria)¹⁰ was used according to the instructions provided by the manufacture. The ß-globin regions of interest were amplified from isolated DNA in a

Figure 2. Reverse hybridization test strip.

Table 2. Number of specific mutations detected by ARMS and reverse hybridization (RH) in DNA samples extracted from blood, chorionic villi (CV) and amniotic fluid (AF) of beta-thalassemia patients.

single multiplex PCR reaction using three pairs of primers and hybridized to pre-made test strips containing the wild type and mutant-specific probes for fourteen mutations, including HbS and HbC. After several washes specifically bound sequences were detected by enzymatic color reaction (Figure 2).

Figure 2 shows examples of reverse hybridization strips. The strip contains probes for 14 mutations. The right-hand box represents the reference strip.¹⁰ On the left, the first case shows no bands at any of the mutation sites, hence, she is categorized as a normal individual. The second case shows a single band at the codon 5 mutation site. The third case shows two bands; one at codon 39 and one at HbS. The fourth case has a single band at codon 8 while the fifth case shows a double band, both at codon 8. This is the only individual in this group which is of major type. The sixth case displays a single band at codon 8 to 9 and the seventh case, a single band at intervening sequence (IVS) 1.1. The eighth case represents a single band at IVS 1.6, and the ninth case a band at IVS 1.110. Finally, the tenth case has a single band at IVS 2.1.

Of the 82 samples detectable by both ARMS and reverse hybridization, 80 gave identical results in both methods. Specifically, the mutations detected in both were identical. However, the remaining 2 samples showed IVS 2.1 mutations by reverse hybridization but resulted in a false- negative result with ARMS. ARMS was repeated several times for these two samples. In one case, after two rounds of repeating ARMS, we obtained a positive result. In the other case, the DNA was re-extracted to obtain a positive result. Of the ten mutations and hemoglobin S that were screened in this study, IVS2.1 was the most common mutation detected, which is in accordance with a previous study reported by Najmabadi, et al.⁷ A total of 78 mutant alleles have been detected. These included 29 IVS 2.1, 12 IVS 1.110, 7 codon 8, 7 codon 8/9, 6 IVS 1.6, 5 IVS 1.1, 3 codon 39, 3 IVS 2.745, and 1 codon 5 mutant allele (Table 2).

Currently, over 25,000 individuals with beta- thalassemia exist in Iran.⁵ Prenatal diagnosis is the key to prevention and control of this disease. In this study we introduced the application of reverse hybridization for the detection of beta-thalassemia mutations and compared it with ARMS, which is one of the most commonly used techniques for diagnosis of this disease in Iran. The most common mutation detected was IVS2.1. This is in accordance with our previous findings.⁷

Our results suggest that reverse hybridization is a more reliable technique that can also reduce false-negative results. Only one PCR reaction is required to screen for more than 20 mutations on a

Table 3. Comparison of different factors determining the efficiency of ARMS and reverse hybridization in beta thalassemia diagnosis.

single test strip by reverse hybridization, while in the ARMS technique, 8 PCR reactions must be performed for each sample to detect one single mutation. These include two PCR reactions (using mutant and normal primers) for the sample itself, the negative control, the heterozygote positive control and the homozygote positive control. It means that with using ARMS and assuming every PCR reaction works, we had to perform from 8 to as many as 88 PCR reactions to screen for the 11 mutations in our study. Repeating a reverse hybridization test under the same conditions present no difficalty, whereas reproduction of ARMS process is not as straight forward. Unlike ARMS which requires a documentation process after experimentation, reverse hybridization test strips are self documented and thus, can be stored and referred to with ease later, thereby reducing possible record errors. These factors, as well as increasing the accuracy of the tests, reduce our turnover time from 2 to 3 days for ARMS to 6 hours for in reverse hybridization. This is of vital importance in Iran, where the legislation allows therapeutic abortion of an affected fetus only within the first 16 weeks of gestation. In addition, as most families consult a genetic diagnostic center only after pregnancy, there is a limited window of time available to detect the possible mutation affecting the fetus and hence, a faster procedure would provide families with sufficient time to make appropriate decisions.

Reverse hybridization is a more environment-friendly technique as it avoids the use of carcinogens such as ethidium bromide and produces less toxic waste. From the economical point of view, reduced labor time and using simpler equipment decrease the cost of test per sample by reverse hybridization.

In Iran, with its vast spectrum of beta- thalassemia mutations, a procedure such as reverse hybridization is much more efficient and practical than ARMS because as many as 25 beta- thalassemia mutations can be simultaneously screened on a single strip in a reasonably short time. This number can be increased by designing a second strip to be run in parallel i.e. 50 mutations on 2 strips.

Nevertheless, in other countries, such as Cyprus, where only a few known mutations exist, a procedure such as ARMS could be sufficient for their needs. Table 3 compares the diagnostic characteristics of the two techniques.

The existing version of the reverse hybridization strip would cover 44% to 66% (depending on the region of the country) of the mutation spectrum in Iran. An extended version of the reverse hybridization assay has recently been released and covered 22 mutations, including codon 22 (7 bp del), IVS 1.25 (25 bp del), codon 30 (G->C), codon 36/37 (-T), codon 44 (-C), and IVS 1.116 (T->G), mutations that are frequently found in the Iranian population.

In conclusion, with the addition of probes for these new mutations to the reverse hybridization test strip, we believe that this technique is the best option that can be employed for carrier detection and prenatal diagnosis of beta- thalassemia in Iran.