The mutations usually include missense or nonsense types, but other types, such as deletions of the beta-globin gene and surrounding regions, also have been identified in thalassemia patients. According to the Human Gene Mutation Database (HGMD), currently, 835 disease-causing mutations have been found in HBB, including 404 missense/nonsense, 118 small deletions, 97 gross deletions, 73 regulatory, 53 splicing, 44 small insertions, 21 complex rearrangements, 19 small indels, and six gross insertions (http://www.hgmd.cf.ac.uk/ac/gene.php?gene=HBB).

In addition, different studies have reported variants with unknown significance in the 5′ region, near the splice sites, and in the 3′ area of HBB, including c.-273T>C (upstream of the gene) [9]; until date, there have been no comprehensive data regarding its role in the phenotype of thalassemia. The goal of this study was to clarify the significance of this variant using segregation and bioinformatics analysis. Thus, from our large number of samples recruited from cases of minor thalassemia, we collected data regarding their laboratories and genetic studies. All patients provided informed consent before undergoing molecular testing for HBB and HBA mutation analysis. This study was approved by the institutional review board of the Comprehensive Medical Genetics Center, Shiraz University of Medical Sciences (Approval No. 95.113.). Genomic DNA was extracted from the peripheral blood lymphocytes of these samples using DNA Extraction Kits (Yekta Tajhiz, Iran) according to the manufacturer's instructions, and the DNA concentration was measured by NanoDrop (ND1000, USA) and stored at −20℃ until use.

Sequences covering all coding and important non-coding regions of HBB and HBA 1 and 2 genes were amplified by PCR. The total volume used for the PCR was 50 µL including 1 µL of each primer (20 pmol/µL), 2 µL DNA template (50??00 ng), 25 µL TEMPase Hot Start 2x Master Mix Blue (Ampicon, A290806), and 21 µL dH2O. The PCR reactions were carried out according to Amplicon TEMPase Hot Start protocol and programs. Ten microliters of the PCR products were visualized on 2% agarose gel containing SYBR Safe. For mutation analysis for HBA 1 and 2 genes, ViennaLab StripAssays was used to locate all important deletions that were not detected by standard PCR.

From 200 samples of different types of suspected minor thalassemia, we found 12 cases with c.-273T>C. It is worth noting that one of the cases had this variant in a homozygous pattern and another of the cases with c.-273T>C had a pathogenic mutation. All heterozygous cases for c.-273T>C (cases 1??0) and also that of the homozygous form, including case 12 without any pathogenic mutations in HBB, had HBA2 levels in the normal range (Table 1). However, the case with c.-273T>C (case 11) that also had a pathogenic c.92+6T>C mutation showed an HBA2 level of 4.0%, indicating that c.-273T>C has no effect on HBB protein. In HBA 1 and 2, we found the different mutations listed in Table 1, which shows why some cases without any mutations in HBB had positive laboratory findings.

We also used bioinformatics software, such as FATHMM, CADD, and PhastCons programs, to predict the effects of this variant in the non-coding region of HBB. The FATHMM program (http://fathmm.biocompute.org.uk/disease.html) uses information concerning sequence homology and its score ranges between −16.13 and 10.64. If a score is lower than −1.5, then the corresponding nonsynonymous SNP (NS) is predicted as DAMAGING. The non-coding score of FATHMM for this variant was 0.042, which predicts that this variant does not have functional effects on the HBB protein. For CADD program, mutations with scores ≥10 are predicted to be the 10% most deleterious substitutions, whereas scores ≥20 signify the 1% most deleterious effects. Using SeattleSeq Annotation (http://snp.gs.washington.edu/SeattleSeqAnnotation138/), we found that the CADD score for this variant was 0.146, which predicts that c.-273T>C is a benign variant.

The phastCons program was also used to identify conserved genomic regions. Higher scores indicate a higher probability that the base is in a conserved element. The PhastCons score of the c.-273T>C variant was 0.250, suggesting that the region of the variant is less conserved among vertebrates.

In conclusion, our data confirmed that c.-273T>C does not impose any effect on the function of HBB protein and should be considered as a benign variant.