Improved genetic screening with zygosity detection through multiplex high-resolution melting curve analysis and biochemical characterisation for G6PD deficiency

Abstract

Accurate diagnosis of glucose-6-phosphate dehydrogenase (G6PD) deficiency is crucial for relapse malaria treatment using 8-aminoquinolines (primaquine and tafenoquine), which can trigger haemolytic anaemia in G6PD-deficient individuals. This is particularly important in regions where the prevalence of G6PD deficiency exceeds 3%-5%, including Southeast Asia and Thailand. While quantitative phenotypic tests can identify women with intermediate activity who may be at risk, they cannot unambiguously identify heterozygous females who require appropriate counselling. This study aimed to develop a genetic test for G6PD deficiency using high-resolution melting curve analysis, which enables zygosity identification of 15 G6PD alleles. In 557 samples collected from four locations in Thailand, the prevalence of G6PD deficiency based on indirect enzyme assay was 6.10%, with 8.08% exhibiting intermediate deficiency. The developed high-resolution melting assays demonstrated excellent performance, achieving 100% sensitivity and specificity in detecting G6PD alleles compared with Sanger sequencing. Genotypic variations were observed across four geographic locations, with the combination of c.1311C>T and c.1365-13T>C being the most common genotype. Compound mutations, notably G6PD Viangchan (c.871G>A, c.1311C>T and c.1365-13T>C), accounted for 15.26% of detected mutations. The high-resolution melting assays also identified the double mutation G6PD Chinese-4 + Canton and G6PD Radlowo, a variant found for the first time in Thailand. Biochemical and structural characterisation revealed that these variants significantly reduced catalytic activity by destabilising protein structure, particularly in the case of the Radlowo mutation. The refinement of these high-resolution melting assays presents a highly accurate and high-throughput platform that can improve patient care by enabling precise diagnosis, supporting genetic counselling and guiding public health efforts to manage G6PD deficiency-especially crucial in malaria-endemic regions where 8-aminoquinoline therapies pose a risk to deficient individuals.

Keywords: G6PD deficiency; genotype; high‐resolution melting; mutations; structural stability.

FIGURE 1

Study workflow.

FIGURE 2

G6PD phenotype of samples across the different collection sites.

FIGURE 3

Distribution of G6PD activity among males and females, categorised into deficiency, intermediate deficiency and G6PD normal subjects.

FIGURE 4

HRM reactions for identification of G6PD mutations.

FIGURE 5

HRM reactions for identification of G6PD WT.

FIGURE 6

Representative HRM profiles for the identification of hemizygous/homozygous and heterozygous for G6PD Mahidol.

FIGURE 7

Distribution of G6PD activity by mutation types in (a) male and (b) female subjects.

FIGURE 8

Structural stability analyses of G6PD variants. The effect of mutations was assessed in the absence (solid line) and presence of NADP⁺ (10 μM: Dashed line and 100 μM: Dotted line). (a) Thermal stability assay. T _m was defined as the temperature at which half of the protein unfolded and presented as mean ± SD of triplicate measurements. (b) Thermal inactivation analysis. T _1/2 was defined as the temperature at which the enzyme lost 50% of its activity and presented as mean ± SE of triplicate measurements. (c) Susceptibility to Gdn‐HCl treatment. C _1/2 was defined as the concentration of Gdn‐HCl at which the enzyme lost 50% of its activity and presented as mean ± SE of triplicate measurements. (d) Susceptibility to trypsin digestion. NT: No trypsin treatment. Error bars represent the mean ± SE of triplicate measurements.

FIGURE 9

Cartoon view of G6PD dimer highlighting the mutation location in spheres (G6PD Radlowo; yellow, G6PD Chinese‐4; cyan, G6PD Canton; blue), important domains—dimer interface (green), tetramer interface (orange), catalytic βE—αe loop (red) and ligand binding sites.

FIGURE 10

Alterations in the intermolecular interactions due to mutation in the variant structures (green) in comparison to WT (cyan). (a) G6PD Radlowo, (b) G6PD Chinese‐4, (c) G6PD Canton and d) G6PD Chinese‐4 + Canton double variant showing changes at the R459L mutation site (left) and G131V mutation site (right).

FIGURE 11

Dimer interface of G6PD variants and WT at 100 ns showing (on the left) the distance of βN‐βN strands (res. 415–423) between chain A and chain B is higher in the variant (green) compared to the WT (cyan) and network interactions at the dimer interface (on the right) connecting residues in chain A (green) and chain B (grey) for (a) G6PD Radlowo, (b) G6PD Chinese‐4, (c) G6PD Canton and (d) G6PD Chinese‐4 + Canton.

FIGURE 12

Ligand binding pocket occupancy indicating the presence (orange) and absence (turquoise) of hydrogen bonds.