This study was reviewed and approved by the Institutional Review Board of the University of Phayao, Phayao, Thailand (1.2/026/67) and was conducted according to the tenets of the Declaration of Helsinki. Written informed consent was obtained from the patient.

The patient was a 44-year-old Thai male who presented with a three-week history of tachypnea, progressive dyspnea on exertion, non-productive cough, and low-grade fever. His past medical history was notable for chronic anemia of unknown etiology, for which he had received intermittent blood transfusions, the last being administered when the patient was 39 years old. Six months after this last transfusion, the patient underwent a complete blood count examination at a local hospital, which revealed: red blood count (RBC), 3.99 × 10¹²/L; Hb, 9.0 g/dL; hematocrit (Hct), 31.9%; mean corpuscular volume (MCV), 79.9 fL; mean corpuscular hemoglobin (MCH), 22.6 pg; mean corpuscular hemoglobin concentration (MCHC), 28.2 g/dL, and red cell distribution width (RDW), 29.9%. These values, obtained from the medical records, represent the patient's steady-state hematological profile in the absence of acute illness and at least six months after any transfusion. Physical examination and abdominal ultrasonography indicated hepatosplenomegaly. He was diagnosed with congestive heart failure secondary to chronic anemia of unknown etiology, complicated by pneumonia. A peripheral blood sample anticoagulated with ethylenediaminetetraacetic acid was collected prior to blood transfusion for further hematological and molecular analysis. Blood specimens and initial hematological data from the hospital were sent to the School of Allied Health Sciences, University of Phayao. Hb analysis was performed using two automated cation-exchange high-performance liquid chromatography (HPLC) systems: VARIANT II (Bio-Rad, Hercules, CA, USA) and β-Thalassemia Short Program and Premier Resolution (Trinity Biotech, in Bray, County Wicklow, Ireland). The latter is a novel cation exchange chromatography quantifying Hb A2 even in the presence of Hb E in high-resolution mode. Capillary electrophoresis (CE) was performed using the MINICAP Flex Piercing system (Sebia, Lisses, France).

Genomic DNA was extracted from peripheral blood leukocytes using the Total Blood DNA Isolation Kit (Vivantis Technologies, Selangor, Malaysia). Common α-thalassemia mutations (–SEA, –THAI, -α3.7, -α4.2, Hb Constant Spring (HBA2:c.427T>C) [αCS], and Hb Paksé (HBA2:c.429A>T) [αPaksé]) were identified using gap-polymerase chain reaction (PCR) and allele-specific PCR (AS-PCR) [11–13]. AS-PCR was also used to identify Hb variants with the same retention time (RT) on Hb-HPLC and migration zones on the electropherogram as those occasionally documented in the Thai population. These variants included Hb Hekinan II (HBA1:c.84G>T) and Hb Wiangpapao (HBA1:c.133C>T) [14]. Mutational analyses of α1-, α2-, and β-globin genes were performed by amplifying their sequences using a previously described protocol [15,16]. Direct DNA sequencing of amplified globin genes was performed using an ABI PRISM™ 3130 XL analyzer (Applied Biosystems, Foster City, CA, USA).

Mutation positions were identified and analyzed using the human abnormal Hb thalassemia libraries HbVar (http://globin.bx.psu.edu/hbvar) and IthaGenes (https://www.ithanet.eu/db/ithagenes). Variants were named according to HbVar (http://globin.bx.psu.edu/cgi-bin/hbvar/query_vars3). The α-globin gene haplotype was assessed using PCR-based restriction enzyme digestion at one broadly triallelic inter-ζ-globin hypervariable region (HVR) and at six polymorphic restriction sites: (1) the XbaI site of the 5′ ζ2-globin gene, (2) SacI site of the 3′ ζ2-globin gene, (3) AccI site of the 3′ ψα2-globin gene, (4) RsaI site of the 5′ α2-globin gene, and (5) PstI sites of the 5′ α1-globin gene and (6) 5′ θ1-globin gene. The PCR products containing each of these polymorphic sites were amplified via PCR and digested with appropriate restriction enzymes [17]. Haplotypes were determined by analyzing the presence or absence of cleavage at each site and combining these outcomes into a unified pattern.

The GTC–GAC mutation in codon 10 of the α2-globin gene, causing Hb Chumphae, abolishes a BstPAI recognition site (5′-GACNN▼NNGTC-3′). PCR-restriction fragment length polymorphism (RFLP) assays were developed to detect this mutation. A 1085-base pair (bp) α-globin gene fragment was amplified using primers C1 (5′-TGGAGGGTGGAGACGTCCTG-3′) and C3 (5′-CCATTGTTGGCACATTCCGG-3′) under specific PCR conditions [15]. PCR amplification was performed in a total volume of 50 µL containing 100–200 ng genomic DNA, 60 pmol of each primer, and 1.5 units of Taq DNA polymerase (Vivantis) in 10 mM Tris–HCl (pH 9.1), 50 mM KCl, and 0.1% Triton™ X-100. For restriction digestion, the 10 µL of amplified fragment was completely digested by 1 unit of BstPAI (New England Biolabs, Beverly, MA, USA) at 37 °C for six hours. Digested fragments were then analyzed using 1.5% agarose gel electrophoresis and visualized under ultraviolet (UV) light after ethidium bromide staining. The 1085-bp amplified fragment from the normal allele (αA) was digested into 818-bp and 267-bp fragments, whereas the Hb Chumphae allele (αChumphae) remained undigested.

A simple rapid DNA assay using AS-PCR was developed for the GTC–GAC mutation in Hb Chumphae. The forward primer SP60 (5′-CCTGCCGACAAGACCAACGA-3′) and common primer C3 produced an 834-bp fragment specific to the αChumphae allele. Two additional primers, γ4 (5′-GGCCTAAAACCACAGAGAGT-3′) and γ5 (5′-CCAGAAGCGAGTGTGTGGAA-3′), generated a 578-bp internal control fragment. The total PCR mixture (50 μL) contained 100–200 ng genomic DNA, 24 pmol SP60 and C3, 60 pmol γ4 and γ5, 200 μM dNTPs, and 2.0 units Taq DNA polymerase (Vivantis) in 10 mM Tris–HCl (pH 9.1), 50 mM KCl, and 0.1% Triton™X-100. Amplification was performed using a thermal cycler (Cycler Personalis, Bio-Rad, USA). After an initial denaturation step at 94 °C for 3 min, the reaction was followed by 30 cycles at 94 °C for 30 s, 64 °C for 30 s, 72 °C for 1 min, and a final extension at 72 °C for 10 min. The amplified products were analyzed via 1.5% agarose gel electrophoresis and visualized under UV light after ethidium bromide staining.

Variant pathogenicity was evaluated using the American College of Medical Genetics and Genomics and Association for Molecular Pathology (ACMG/AMP) best-practice guidelines, as specified by the ClinGen Hemoglobinopathy Variant Curation Expert Panel [18,19]. Additionally, the impact of the amino acid substitution on the α-globin chain and the structure and function of the α-globin chain were predicted using network-based HumDiv-trained Polymorphic Phenotype V.2 (PolyPhen-2) (http://genetics.bwh.harvard.edu/pph2/) and the Sorting Intolerant from Tolerant (SIFT) web server (https://sift.bii.a-star.edu.sg/) [20]. Additionally, the PredictSNP program, freely available at https://loschmidt.chemi.muni.cz/predictsnp1/, was used to estimate whether the effect of amino acid substitution was neutral or deleterious [21]. In addition, homology models of three-dimensional (3D) structures of wild-type and variant proteins were generated from amino acid sequences using the SWISS-MODEL website [22,23]. The crystal structure of deoxygenated Hb complexed with 5HMF-NO (PDB code: 7ud7.1) was selected based on sequence identity. The resulting 3D protein structures were visualized using Discovery Studio Visualizer and PyMOL (Biovia, D. S., Discovery Studio Visualizer, v24.1.0.23298; Dassault Systems, San Diego, CA, 2024, and Schrodinger, LLC, The PyMOL Molecular Graphics System, Version 3). Swiss-PdbViewer version 4.1 was used to predict contact sites of residues of interest [24]. The crystal structure of alpha Hb-stabilizing protein (α-HSP; PDB code 1Y01) was used as a template for α-HSP and α-globin chain binding simulation. To investigate the interaction between α-HSP and α-globin chain in either native or mutant HBA2 proteins, protein–protein docking simulation was performed using ZDOCK (ZDOCK 3.0.2, https://zdock.umassmed.edu). Contracting residues for docking were selected based on α-HSP interaction sites with the α-globin chain.

The patient’s hematological findings indicated severe anemia: (RBC count: 1.74 × 1012/L, Hb: 3.7 g/dL; Hct: 17.0%; MCV: 84.7 pg; MCH: 21.3 pg; MCHC: 25.1 g/dL; and coefficient of variation of RDW: 40.5%; reticulocyte count: 7.7%). The serum ferritin level during the acute crisis was elevated at 930.3 ng/mL (reference range: 23.9–336.2 ng/mL). A blood smear showed hypochromia and anisocytosis. The VARIANT II HPLC revealed abnormal Hb peaks eluting faster than Hb A peaks. This abnormal Hb peak was incompletely eluted from Hb A at a specific RT of 2.06 min, accounting for 18.0% of the total Hb (Figure 1A). Additionally, Hb Bart’s (γ4) was distinctly observed on the chromatogram. A comparable analysis using the Premier Resolution-HPLC system detected an abnormal Hb peak completely eluting from Hb A at RT of 3.882 min (21.7% of the total Hb) and Hb H and Hb Bart's at RT of 0.330 min (2.1% of the total Hb) (Figure 1B). Similarly, CE Hb analysis showed an abnormal Hb peak in migrating zone 12, resembling Hb Bart’s, and accounting for 15.2% of the total Hb (Figure 1C). A comprehensive overview of Hb analysis profiles and levels is presented in Table 1.

Figure 1.

Hemoglobin (Hb) analysis results obtained using three different automated Hb analyzers. (A) Hb-chromatogram profile on the HPLC-Variant II system. (B) Hb-chromatogram profile on the HPLC-Premier Resolution system. (C) Hb-electropherogram pattern by capillary electrophoresis.

Sequence analysis of the selectively amplified α- and β-globin genes revealed a pattern consistent with homozygosity for a missense mutation at codon 10 (GTC>GAC) in the α2-globin gene. This mutation caused the valine-to-aspartic acid substitution [α10(A8)Val>Asp; HBA2:c.32T>A] (Figure 2A) responsible for Hb Chumphae. Notably, only A nucleotide was identified at this position, indicating that one chromosome carried the α0-thalassemia gene with two deleted α-globin genes. Consequently, only one α2-globin gene was amplified for DNA sequencing. Further DNA analysis confirmed SEA deletion α0-thalassemia, indicating that the patient coinherited Hb Chumphae with α0-thalassemia, corresponding to the genotype (--SEA/α10T>Aα). No mutations were detected in the β-globin gene.

Figure 2.

Mutation identification using DNA sequencing, polymerase chain reaction (PCR)-restriction fragment length polymorphism, and allele-specific PCR. (A) The sequencing profile of the α2-globin gene shows antisense strand sequences in codon 10, where the GTC–GAC mutation causes hemoglobin (Hb) Chumphae. (B) GTC-GAC mutation identification using BstPAI digestion of the PCR product. M represents the 100-bp DNA ladder markers. Lane 1 displays undigested amplified DNA for gel electrophoresis. Lanes 2–5 show digested amplified DNA samples. Lanes 2–4 contain BstPAI-digested amplified DNA from non-mutated samples, showing 818- and 267-base pair (bp) digested fragments. Specifically, lane 4 also has a 1085-bp undigested fragment, indicating the presence of Hb Chumphae. (C) Hb Chumphae mutation is identified using an allele-specific PCR assay. Primers SP60 and C3 amplify an 834-bp fragment specific to Hb Chumphae, whereas primers γ4 and γ5 amplify a 578-bp internal control band. M represents the 100-bp DNA ladder markers. Lane 1: negative DNA control; lanes 2 and 3: normal DNA samples; lane 4: DNA of the patient with Hb Chumphae co-inherited with α0-thalassemia.

The GTC to GAC mutation at codon 10 abolished the BstPAI restriction site (5′-GACNN▼NNGTC-3′) on the α- gene. This mutation was confirmed using PCR–RFLP analysis (Figure 2B). In the normal allele, the 1085-bp fragment was digested into 267-bp and 818-bp fragments, whereas the fragment in the Hb Chumphae allele remained undigested. The developed PCR–RFLP method successfully confirmed the presence of Hb Chumphae in the patient.

A new AS-PCR technique was developed for rapid DNA diagnosis of Hb Chumphae. The amplified 834-bp fragment from the Hb Chumphae allele was clearly observed in patients harboring Hb Chumphae, whereas the 578-bp fragment specific for the normal allele was observed in normal individuals (Figure 2). This result indicated that the newly developed AS-PCR technique successfully enabled rapid diagnosis of Hb Chumphae.

Further α-globin gene haplotype analysis revealed a haplotype (±, −, S/M, −, −, −, −) associated with Hb Chumphae. Two polymorphic sites showed heterozygous patterns: the XbaI site of the 5′ ζ2-globin gene and the broadly triallelic inter-ζ-globin HVR. These two polymorphic sites are commonly found in the αA allele [25,26]. Accurate segregation analysis was not possible owing to a lack of specimens from family members. However, the αChumphae allele was possibly associated with haplotype (−, −, M, −, −, −, −), whereas the αA allele was associated with haplotype (+, −, S, −, −, −, −).

The mutation site was located outside of the α1β1 or α1β2 interface and crucial functional binding sites for oxygen, proteins, and heme (Figure 3A). Interaction predictions showed that the residue at position 11 (valine in the normal α-chain on the left and aspartic acid in the mutated α-chain on the right) was involved in intra-chain interactions within the monomer (Figure 3B). The replacement of valine with aspartic acid resulted in significant changes in the interactions; aspartic acid carries a negative charge and likely disrupts the hydrophobic interactions originally stabilized by valine. The introduction of aspartic acid created new interactions, specifically with residues Val74, Asp127, and Lys128. These additional interactions stabilized or destabilized the Hb structure.

Figure 3.

Homology modeling of HBA2 proteins demonstrating the α-globin chain in a ribbon structure. Methionine is numbered as 1 in this format, regardless of processing. (A) In the three-dimensional (3D) structure of HBA2 proteins, residues are labeled according to their binding sites, and mutation position is labeled in violet. (B) In the 3D structure of hemoglobin, the α-globin chain is gray, and the β-globin chain is pink. The normal α-chain is on the left, and the mutated α-chain is on the right. Side chains of residue position 11 are red, side chains of residues that interact with residue position 11 are brown, and side chains of additional interacting residues with mutant position 11 are blue. (C) The 3D structures of α-HSP (purple) and α-globin chain (green) are shown in ribbon representation. Contracting areas of the protein complex are shown in surface representation. Residue position 11 is shown as pink.

In protein–protein docking simulations, the top-ranked poses for each complex yielded ZDOCK scores of 1227.522 and 1253.109, respectively, suggesting favorable binding configurations. Valine to aspartic acid substitution at this position did not significantly affect intermolecular interactions (Figure 3C). Furthermore, pathogenicity predictions for Hb Chumphae showed an SIFT score of 0.00 and a PolyPhen-2 score of 0.999, classifying it as probably damaging. Furthermore, PredictSNP predicted the mutation as deleterious. Additionally, the HBA2:c.32T>A mutation met the ACMG/AMP criteria to be classified as ‘pathogenic’ for α-thalassemia and related disorders (Table 2).