α- and β-Globin Gene Mutations in Individuals with Hemoglobinopathies in the Chattogram and Sylhet Regions of Bangladesh

Abstract Hemoglobinopathies, including α- and β-thalassemias and sickle cell disease, are among the most widely disseminated hereditary blood disorders worldwide. Bangladesh is considered a hotspot for hemoglobinopathies, and these diseases cause a significant health concern in the country. However, the country has a dearth of knowledge on the molecular etiology and carrier frequency of thalassemias, primarily due to a lack of diagnostic facilities, limited access to information, and the absence of efficient screening programs. This study sought to investigate the spectrum of mutations underlying hemoglobinopathies in Bangladesh. We developed a set of polymerase chain reaction (PCR)-based techniques to detect mutations in α- and β-globin genes. We recruited 63 index subjects with previously diagnosed thalassemia. Along with age- and sex-matched control subjects, we assessed several hematological and serum indices and genotyped them using our PCR-based methods. We identified that parental consanguinity was associated with the occurrence of these hemoglobinopathies. Our PCR-based genotyping assays identified 23 HBB genotypes, with the codons 41/42 (–TTCT) (HBB: c.126_129delCTTT) mutation leading the spectrum. We also observed the presence of cooccurring HBA conditions, of which the participants were not aware. All index participants in this study were on iron chelation therapies, yet we found they had very high serum ferritin (SF) levels, indicating inefficient management of the individuals undergoing such treatments. Overall, this study provides essential information on the hemoglobinopathy mutation spectrum in Bangladesh and highlights the need for nationwide screening programs and an integrated policy for diagnosing and managing individuals with hemoglobinopathies.


Introduction
Hemoglobinopathies are autosomal recessive conditions affecting the quantity and quality of hemoglobin (Hb), molecules within red blood cells (RBCs) [1,2]. Broadly, these disorders fall into two groups: (i) the structural variants of Hb, best known of which is sickle cell anemia, and (ii) the thalassemias that refer to the reduced or no production of one or more globin genes, e.g. aand b-thalassemia (a-and b-thal) [2]. Mutations that result in the production of a structurally abnormal b-globin chain cause sickle hemoglobinopathies [3]. This abnormal Hb, Hb S (or HBB: c.20A>T) forms a gel polymer at conditions of low oxygen tension to give the red cell a sickled shape and eventually causing a wide variety of clinically severe complications [4]. Thalassemias result from unbalanced Hb synthesis caused by decreased production of at least one of the globin chains (a, b, c and d) [5,6].
Mutations, especially large deletions in the a-globin genes (HBA1 and HBA2), causing decreased or no production of the a polypeptide chains, results in a-thal [7,8]. A single gene mutation (-a/aa or two gene mutation (--/aa or -a/-a) causes a-thal trait with microcytosis, usually without anemia, while significant clinical phenotypes, e.g. Hb Bart's hydrops fetalis syndrome and Hb H diseases result from deletion of all four a-globin genes (--/--) and three a-globin genes (--/-a), respectively [7,9,10]. On the other hand, b-thal results from mutations that either abolish (b 0 ) or reduce (b þ ) synthesis of the b polypeptide chains [11,12]. Over 900 disease-causing variants have been reported on the b-globin gene, most of which are point mutations, e.g. small deletions or insertions [13]. In the absence of effective treatment that is inexpensive and easily accessible, hemoglobinopathies substantially impact the quality of life and health care in Bangladesh.
Hemoglobinopathies are the most prevalent and widely distributed monogenic inherited blood disorders globally, occurring more frequently in the Mediterranean, Middle Eastern, sub-Saharan, Southern Chinese, Indian subcontinent and Southeast Asian countries. Unsurprisingly, Bangladesh being in Southeast Asia, hemoglobinopathies are highly prevalent in the country [14,15]. However, in-depth knowledge of the prevalence of hemoglobinopathy patients and carriers is unavailable due to the lack of precise diagnostic approaches, inadequate infrastructure, and limited access to information [16,17]. The very limited epidemiological data on hemoglobinopathies in Bangladesh is available only from a few studies. However, more comprehensive data about the prevalence and distribution of mutations are of operational importance for the effective management of people with hemoglobinopathies in the country.
This study sought to describe the development and implementation of polymerase chain reaction (PCR)-based procedures for the simple, reliable, high throughput screening of the commonly occurring Hb mutations in a cohort obtained from Bangladesh. Based on previously published reports, we developed and used sets of the multiplex amplification-refractory mutation system (MARMS) and allele-specific PCR (ASPCR), to identify common b-globin mutations responsible for b-thal and sickle cell disease [18][19][20]. We also screened for seven commonly occurring a-thal deletions in the cohort using a multiplex gap-PCR strategy that we developed and reported previously [15].

Ethics statement
This study was carried out according to the code of ethics of the World Medical Association (Declaration of Helsinki) and adhering to the Revised Biosafety Guidelines of Bangladesh, 2018. Informed written consent was obtained from the participants of this study before sample collection. All experimental protocols and ethics were approved by the Institutional Ethics Committee of Shahjalal University of Science and Technology, Sylhet, Bangladesh [IEC-101(1)/005].
The participants, primary screening, and hematological analysis A total of 140 unrelated Bangali index subjects born and currently living in Chattogram and Sylhet regions of Bangladesh were primarily recruited in this study from the Thalassemia Welfare Center Bangladesh (TWCBD), Chattogram (n ¼ 63) and the Department of Pediatrics, Sylhet MAG Osmani Medical College and Hospital (SOMCH), Sylhet (n ¼ 77). All individuals had been previously diagnosed to carry hemoglobinopathies by either microscopic analysis (n ¼ 91), high performance liquid chromatography (HPLC) (n ¼ 46), or Hb electrophoresis (n ¼ 95). We interviewed all the participants and/or their legal guardians for background data and clinical history using a structured questionnaire. Following the primary interview, hepatitis B virus surface antigen (HBsAg) and hepatitis C virus (HCV) antibody tests for all subjects were ordered from a commercial diagnostic laboratory. Out of the 140 samples, 51 for HBsAg, 18 for HCV, and eight for both HBsAg and HCV, were positive. We excluded these subjects from any further analysis on the ground of safety concerns. We collected $5 ml of peripheral blood from the remaining index subjects (n ¼ 63) in two separate vials by venipuncturing. We also recruited 63 age-and sex-matched healthy individuals as control subjects. The same background data and an equal amount of peripheral blood were collected from the control subjects.

Hematological and serum analysis
Peripheral blood counts and red blood indices were measured using an automated blood cell analyzer (KX-21N TM ; Sysmex Corporation, Kobe, Japan). In addition, we evaluated the serum ferritin (SF) levels using an automatic chemiluminescence immunoassay analyzer (Abbott Architect i2000SR; Abbott Laboratories, Abbott Park, IL, USA).

DNA isolation and genotyping
We extracted genomic DNA from peripheral blood samples using ReliaPrep TM Blood gDNA Miniprep System (Promega Corporation, Madison, WI, USA) following the manufacturer's instruction. A total of 12 sets of multiplex reactions were carried out for the detection of 31 b-globin (17 b 0 , eight b þ , three b þþ , two b 0 /b þ , and one sickle cell anemia mutation) and seven a-globin gene mutations (Supplementary Tables 1-13). Twenty-four of the primer sequences used for the b-thal mutation detection were obtained from earlier publications, while the rest were designed during the course of this project, and then, depending on the amplicon sizes, assigned to different MARMS reaction sets (Supplementary Tables 3-11) [18,19,[21][22][23]. For design of the new mutant primers, we designed them with a mismatch at the three terminal bases, and a second mismatch engineered at one of the nucleotides between the third and fifth nucleotides from the 3 0 end to ensure it would not amplify the normal (healthy) DNA [23,24]. Potential off-targets of the primers were searched using GGGenome (https://gggenome.dbcls.jp) that works similarly as the BLAST tool, except it is more optimized for searching genome databases with short sequence queries. Homozygosity and heterozygosity were determined using the corresponding normal primers in a separate reaction [23]. Primer sequences for the detection of the codon 6 (A>T) transversion, i.e. the Hb S or sickle cell anemia mutation, were adopted from a previous report (Supplementary Table  13) [25]. Following a previous report [15], two single-tube multiplex gap-PCRs were applied to investigate the presence of seven commonly occurring a-thal deletions (Supplementary Table 12 Tables  2-11). For the multiplex gap-PCR and ASPCR reactions, we used GoTaq Hot Start Green Master Mix (Cat. #M5122; Promega Corporation). All oligonucleotide primers were synthesized commercially by Integrated DNA Technologies (Singapore).

Statistical analysis
We tabulated and analyzed all data using Microsoft Excel and Statistical Package for Social Sciences (SPSS), IBM Inc., Armonk, NY, USA) version 27.0 tools. We expressed the quantitative data as mean ± SD, or median interquartile range (IQR), analyzed by t-test for two groups and the analysis of variance (ANOVA) for comparison of more than two groups. Qualitative data were reported by frequency and percentages analyzed by the v 2 -test or Fischer's exact test where suitable. We also analyzed the categorical and continuous data using a two-tailed v 2 test and Mann-Whitney U tests, respectively.

Background characteristics of the index and control subjects
The cohort of the index subjects was composed of 63 samples, composed of 36 females and 27 males ( Table 1). The mean age of this cohort was 14.54 ± 7.38 years, with a range between 3 to 19 years. Neither the age nor the male:female ratio was observed to be significantly dissimilar in the control subjects. Nineteen of the index subjects had parental consanguinity, the majority (n ¼ 12) of whom were first cousins.
Fifty-six (88.89%) of the index cases were on regular blood transfusion (92.86% monthly, 5.36 bimonthly and 1.79% every 2 weeks) ( Table 2). All patients reported receiving deferoxamine (DFO) as supplementary medication except one who was administered a combination of DFO and deferiprone (DFP). The mean treatment or management cost for the thalassemia patients was US$83.66 ± 27.45 per month. Table 3 shows the hematological indices of the index and control subjects. Our analysis showed significantly decreased values for Hb concentrations in the index subjects than the control subjects (p<.0001). Red blood cell counts were also significantly reduced in the index subjects (p<.0001). We observed significantly reduced values for mean corpuscular volume (MCV), mean corpuscular Hb (MCH), and mean corpuscular Hb concentration (MCHC) in the index subjects (p<.0001). The RBC count and RBC distribution width (RDW) were elevated in the index subjects as compared to the control subjects (p<.0001). Interestingly, a significant difference in the total white blood cell (WBC) counts was observed between the groups (p<.0001). However, we did not observe any significant difference in the platelets and packed cell volume (PCV) between the index and control subjects. We also observed significantly elevated SF levels in the index subjects compared to the control subjects (p<.0001) [ Figure 1(A)]. Over one-third of the index subjects had SF levels within the range of 1000.0-2500.0 ng/mL, while nearly 60.0% of the subjects had the levels above 2500.0 ng/mL [ Figure 1(B)]. Serum ferritin levels also

HBB genotyping for b-thalassemia and sickle cell anemia mutations
The MARMS-PCR analysis of the index samples revealed that 35 (55.6%) subjects were homozygous for a single mutation, 13 (20.6%) were compound heterozygotes, while a single mutation was detected in nine (14.3%) subjects (Table 4).
Also, there were four (6.3%) subjects, including two males and two females, with compound homozygous mutations, i.e. they harbored two homozygous HBB mutations. The most frequent mutant allele was the codons 8/9 (+G) (HBB: c.27_ 28insG, 21.64%). We did not detect a mutation in two subjects (3.18%). In addition, none of the samples were positive for the codon 6 (A>T) transversion, i.e. sickle cell anemia mutation.

HBA genotyping for a-thalassemia mutations
Our analyses using gap-PCR identified the presence of six a-thal deletions in the index subjects (Table 5). Heterozygous -a 3.7 (rightward) and --SEA (Southeast Asian) deletions were found in three and two subjects, respectively, and a heterozygous -a 4.2 (leftward) deletion was present in one subject. All the identified HBA mutations occurred in subjects carrying heterozygous b-thal mutations. We did not record any striking difference in phenotypic presentation due to the co-occurrence of HBB and HBA mutations.

Discussion
Hemoglobinopathies are among the most common heritable disorders in Bangladesh. It causes reduced or absence of   b-globin chain synthesis of Hb [2]. Identification of Hb variants and carrier identification, and prenatal analysis are necessary for the effective management of these diseases [26,27]. Premarital and prenatal genetic counseling for the prevention of hemoglobinopathies is also essential to avoid marriages between carriers of mutated gene. This study did a comprehensive investigation to detect commonly occurring aand b-globin mutations using inexpensive and straightforward techniques. We applied MARMS, gap-PCR, and ASPCR to detect 38 aand b-globin mutations. The PCR-based detection procedure developed in this study would help diagnose the Hb mutation more efficiently in the context of the lower-middle-income economy and underdeveloped health facilities of Bangladesh. The molecular epidemiologic insights into the Hb mutations obtained from this study would also help design hemoglobinopathy management strategies in the country.
All the index subjects in this study were on iron chelation therapy, yet they had very high SF levels, indicating inadequate chelation and vulnerability to developing iron overload-related complications in these patients (see Figure 1). It has been reported that ineffective erythropoiesis, increased gastrointestinal iron absorption, and multiple blood transfusions may lead to iron overload in patients with thalassemia [28][29][30][31]. Iron overload can result in serious health complications, e.g. cardiac and hepatic dysfunction and an impaired immune system. To avoid the complications of iron overload, patients with thalassemia are often recommended iron chelation therapies. The SF levels we observed were comparable with similar studies reported from India; however, they were much higher than those reported from other regions, e.g. North America [31][32][33]. To ensure efficient management of the individuals with thalassemic conditions, developing a national policy with a thorough delineation of how individuals undergoing iron chelation treatment should be handled pre-and post-therapy is required.
Of the 63 index subjects whose Hb status we studied, we recorded 23 HBB genotypes and three HBA genotypes. The most frequent mutant allele was the codons 8/9 (+G) deletion, with an allele frequency of 0.22 (Table 4). The combined allele frequency of the five most common genotypes, e.g. codons 41/42, codon 43 (G>T) (HBB: c.130G>T), 619 bp deletion (NG_000007.3: g.71609_72227del619), codon 15 (G>A) (HBB: c.47G>A) and IVS-I-5 (G>C) (HBB: c.92 þ 5G>C), accounted for over half (61.2%) of all the mutant alleles. In terms of ethnic backgrounds, these five mutations have origins in the Southeast Asian or Asian region (Supplementary Table 1) [13,34,35]. Interestingly, none of the previous studies on the Bangladeshi population reported these alleles as the most common HBB mutations in the Bangladeshi population [14,[36][37][38]. This difference can be related to inequitable sample size and regional distribution based on the ethnic background of patients [39]. Although Bangladesh is largely considered an ethnically homogeneous population, the country's Northern, Northeast, Southeast and Southern Hill regions are home to diverse Sino-Tibetan and Meitei tribal groups. A previous study [40] revealed a high carrier frequency of thalassemias in the tribal population of the country. The Sylhet and Chattogram areas fall in the country's Northeast and Southeast regions.
The population in these regions has also observed frequent admixture with Assamese and Manipuri people. In addition, historically, the preachers of Islam came to Bangladesh from the Arab and other Mediterranean areas, mainly through the Chattogram region and Sylhet was the nerve center of their preaching activities [41][42][43][44][45]. Moreover, people from the Mediterranean region have a long history of visiting this country for religious and trading purposes, and Chattogram and Sylhet regions were a hub for these peoples [43,44]. Many of these people came to these areas for good and, over the course of history, became part of the mainstream Bangali population. As a result, the population of Sylhet and Chattogram is, therefore, more diverse origin than it is often thought and may have more complexity in terms of ethnic and genetic origin compared to other parts of the country. These differences in ethnic and genetic origin may have contributed to the difference we observed in the HBB allele distribution compared to previous studies. Of note, there are only a couple of small-scale studies on HBB allele frequency that specifically surveyed the Bangali population in the Chattogram region, while no such studies have been reported from the Sylhet region. The high level of consanguinity in Sylhet and Chattogram regions, particularly in Sylhet, may have contributed to the observed differences. It is a recognized risk factor in the prevalence of thalassemia, which is among the most common autosomal recessive conditions among children from consanguineous parentage in Bangladesh [46]. This study also observed the presence of mutations commonly considered to have origins in the Mediterranean regions, potentially indicating the ethnic heterogeneity within the mutation spectrum in the population (Table 4 and Supplementary Table 1) [13,34,35].
Along with the HBB mutations, we found six subjects with heterozygous mutations in HBA (Table 5). Although all the index subjects enrolled in this study had a previous diagnosis of their HBB conditions, none reported a known HBA condition. No significant difference in terms of blood indices and SF levels were recorded between index subjects carrying only HBB mutations, e.g. homozygous 619 bp deletion, codon 15 (G>A), and both HBB and HBA mutations, e.g. homozygous 619 bp deletion with heterozygous -a 3.7 , homozygous codon 15 (G>A) with heterozygous -a 4.2 . However, these data should be interpreted carefully as the number of samples in each of these groups was not enough for any robust statistical analysis. Nevertheless, our study encompasses a nationwide comprehensive screening program for HBB mutations, along with HBA mutations [15]. It also provides evidence that PCR-based techniques could be efficiently used for such screening programs. Being highly selective, easy-to-use, and low-cost, such detection techniques can be integrated with the national management policy for hemoglobinopathies.
This study identified two index subjects with compound homozygous mutations, both carrying IVS I-1 (G>T)/codons 41/42 (-TTCT) genotype. Multiple mutations on the same chromosome were reported previously in Bangladesh and other parts of the Indian subcontinent but with entirely different genotypes [14,[47][48][49]. This indicates the unknown molecular intricacy of b-globin gene mutation conditions in the country. The heterogeneity of mutations also has a partial impact on the clinical variability of the disease. However, this impact becomes unpredictable if all the mutations are of the same type, e.g. b 0 /b þ . In the presence of several ameliorating factors such as a-thal deletions, other mutations, and the XmnI-G c polymorphism, the phenotype of the disease becomes altered [49][50][51].
One of the significant strengths of this study was that it could characterize (or at least partially characterize) all the index subjects using the designed PCR-based assays designed (Tables 4 and 5). However, to be fully confirmed, the PCR assays we designed can detect all the major or commonly occurring Hb mutations in the population, further studies using advanced techniques, e.g. direct sequencing, comparative genomic hybridization (CGH), or multiplex ligationdependent probe amplification (MLPA), are needed. Studies involving siblings and parental samples could also be advantageous [23]. Outcomes of this study and similar molecular epidemiological studies can be extended to families at elevated risk for genetic counseling, and families with rare and new mutations to better understand the diseases in the country.
This study highlights that adopting a molecular screening method for detecting mutations on the Hb genes could overcome the lapses of the resource-intensive techniques, especially in the context of lower-middle-income economies and underdeveloped health facilities. The MARMS, gap-PCR, and ASPCR-based cost-effective molecular methods can be applied in thalassemia-prone regions and help fight these incurable and life-threatening disorders. Disclosure statement