Ethical clearance to conduct this study was obtained from the Human Research Ethics Committee of Thammasat University, Thailand (No. COE No. 007/2566), according to the Declaration of Helsinki. Written informed consent was obtained from all individuals.

Leftover ethylenediaminetetraacetic acid-anticoagulated blood samples were collected from 805 unrelated healthy Thai blood donors at the Blood Bank, Thammasat University Hospital, Pathumthani, Thailand, from March 2023 to June 2023. All the donors who qualified for inclusion had already passed the questionnaire and physical screening procedures for blood donation provided by the National Blood Centre, Thai Red Cross Society. Blood samples positive for hepatitis B virus, hepatitis C virus, HIV, or syphilis were excluded from the study. In addition, the study included 1724 leftover DNA samples from the Blood Bank, Thammasat University Hospital, Pathumthani; 300 samples from the Blood Bank Section, Faculty of Medicine, Chiang Mai University, Chiang Mai; and 427 samples from the Regional Blood Centre, 12th Songkhla, Thai Red Cross Society, Songkhla, Thailand. DNA controls for the Kp(a+b−) and Kp(a+b+) phenotypes were obtained from Laboratori d’Immunohematologia Banc de Sang i Teixits, Barcelona, Spain.

Genomic DNA was extracted from the samples using the DNeasy Blood & Tissue Kit (QIAGEN, Valencia, California, United States) according to the manufacturer’s instructions and stored at −20 °C before further analysis.

Altogether, 805 blood samples were analysed to detect the Kp^a and Kp^b antigens, using the CAT method with an anti-Kp^a/Kp^b ID-Card (Bio-Rad Laboratories, Cressier, Switzerland) made up of six microtubes containing either anti-Kp^a or anti-Kp^b of human origin within the gel matrix. Briefly, 10 µL of 5% red cell suspension in ID-Diluent 1 (Bio-Rad Laboratories, Cressier, Switzerland) was added to the appropriate microtubes of the ID-Card. The ID Card was then centrifuged, and the results were recorded.

Genomic DNA from 400 of the samples tested using CAT were randomly selected and sequenced to identify KEL*03 and KEL*04 alleles. Primers were designed for each target gene (Table 1) using National Center for Biotechnology Information Primer-BLAST (https://www.ncbi.nlm.nih.gov/tools/primer-blast/).¹⁹ Polymerase chain reaction amplification for the KEL gene exon 8 containing single nucleotide variant c.841 C > T [rs8176059], p.Arg281Trp, was performed using SEQ-KP-F forward and SEQ-KP-R reverse primers, with an expected fragment size of 358 base pairs (bp). Each PCR reaction mix (total volume: 40 µL) was prepared using 3 µL of genomic DNA (50 ng/L), 3 µL each of the forward and reverse primers (10 µM), 20 µL of 2X PCR reaction mixture (Phusion High-Fidelity PCR Master Mix, New England BioLabs, Ipswich, Massachusetts, United States), and 11 µL of sterile distilled water. The PCR reactions were carried out in a T100 Thermal Cycler (Bio-Rad Laboratories, Inc., Hercules, California, United States). PCR reactions included 30 s of initial denaturation at 98 °C, followed by 10 cycles of 10 s at 98 °C and 60 s at 69 °C, then 20 cycles of 10 s at 98 °C, 50 s at 60 °C, and 30 s at 72 °C, and a final extension step for 5 min at 72 °C. The PCR products were separated on a 1.5% agarose gel with a 100 bp DNA molecular weight marker (Fermentas, Carlsbad, California, United States), stained using SYBR Safe DNA Gel Stain (Invitrogen, Paisley, United Kingdom), electrophoresed in 1X Tris-borate-EDTA solution at 100 Volts for 30 min, and then visualised with a blue-light transilluminator. The PCR products were then purified using a gel extraction kit (GeneJet Gel Extraction Kit, Fisher Scientific Baltics UAB, Vilnius, Lithuania), and the eluted fragments were sequenced directly using Sanger sequencing (U2Bio Sequencing Service, Bangkok, Thailand) with the aforementioned PCR primers.

KEL*03 and KEL*04 alleles were identified using an in-house PCR-SSP technique with custom primers and amplification conditions (Table 1). The KP-t-A-F and KP-REV primers were used for KEL*03 amplification, while the KP-c-B-F and KP-REV primers were used for KEL*04 amplification. Two sets of PCR reaction mixtures, set A for identifying KEL*03 alleles and set B for KEL*04 alleles, were performed for each sample. Each PCR reaction mixture (total volume: 10 µL) was prepared using 1 µL of template genomic DNA (50 ng/L), 1 µL each of the forward and reverse primers (10 µM), and 5 µL of 2 × PCR reaction mixture (Green Hot Start PCR Master Mix, Biotechrabbit GmbH, Hennigsdorf, Germany). Co-amplification of the human growth hormone (HGH) gene using 1 µL of HGH-F and HGH-R primers (6 µM) was conducted as the internal control in each set. The PCR reactions were carried out in a T100 Thermal Cycler (Bio-Rad Laboratories, Inc., Hercules, California, United States) and included 30 s of initial denaturation at 95 °C, followed by 10 cycles of 30 s at 95 °C and 60 s at 69 °C, then 20 cycles of 10 s at 95 °C, 50 s at 62 °C and 30 s at 72 °C, and a final extension step for 5 min at 72 °C. After electrophoresis, PCR products were examined with a blue-light transilluminator. The expected PCR product size for both the KEL*03 and KEL*04 amplification reactions was 158 bp, while that of the internal control, the HGH gene, was 434 bp. Known DNA controls of KEL*03/KEL*03, KEL*03/KEL*04 and KEL*04/KEL*04 genotypes were used as positive and negative controls in parallel with the test samples. The KEL*03/KEL*03 genotype was indicated if amplification was observed with only the KEL*03 reaction (set A), while the KEL*04/KEL*04 genotype was indicated if amplification was observed with only the KEL*04 reaction (set B). If amplification occurred in both sets of reactions, the KEL*03/KEL*04 genotype was indicated. The specificity of PCR-SSP was examined by cross-amplifying the KEL*03-specific primers with the KEL*04/KEL*04 genotype and vice versa. The PCR-SSP method was conducted on a total of 3256 DNA samples (including the 400 samples tested using DNA sequencing) to identify KEL*03 and KEL*04 alleles.

To evaluate the repeatability of the newly developed PCR-SSP, 100 randomly selected samples with known genotypes by DNA sequencing were retested by PCR-SSP. Upon completion of PCR-SSP genotyping in all 3256 samples, 100 randomly selected DNA samples from among the 400 blood donors with genotyping obtained by standard DNA sequencing were retested using PCR-SSP under the same conditions as previously described to test the repeatability of our in-house PCR-SSP technique for KEL*03 and KEL*04 genotyping. To improve the validity and reliability of the evaluation, technicians were blinded to the PCR-SSP results. DNA sequencing was performed in the event of any discrepant genotyping results obtained using the PCR-SSP and serological assays.

KEL*03 and KEL*04 allele frequencies among Thai blood donors were compared with those reported in Filipino, Southeast Asian, Chinese, Japanese, Korean, South Asian, Native American, Alaska Native/Aleut, Hawaiian/Pacific Islander, South African, Brazilian, Swiss, and German populations.^{9,11,15,16,17}

Gene and allele frequencies were simply estimated by counting the number of times each gene and allele was observed in samples from each population. The calculation of allele frequency for KEL*03 is based on the gene counting method; every KEL*03/KEL*03 individual has two KEL*03 alleles, and a KEL*03/KEL*04 individual has one. This is relative to all alleles in the population, that is, two times the total number of samples. This was also applied to the KEL*04 allele frequency calculation: KEL*03 allele frequency=(2×KEL*03/KEL*03+KEL*03/KEL*04)/(2×the total number of samples)[Eqn 1] KEL*04 allele frequency=(2×KEL*04/KEL*04+KEL*03/KEL*04)/(2×the total number of samples)[Eqn 2]

The Chi-square (χ²) test was used to evaluate whether the observed genotype frequencies agreed with the expected frequencies under the Hardy-Weinberg equilibrium. The two-tailed Fisher’s exact test with Bonferroni adjustment of homogeneity was used to determine the differences in allele frequencies between central, northern, and southern Thai populations and to compare these differences with those reported in other populations.^{9,11,15,16,17} An adjusted p-value of less than 0.004 was considered statistically significant. All statistical analyses were conducted using SPSS^® version 16.0 (SPSS Inc., Chicago, Illinois, United States).

A total of 805 blood samples were tested for Kp^a and Kp^b antigens using CAT. All donors (100%) had the Kp(a−b+) phenotype. Kp(a+b−) and Kp(a+b+) phenotypes were not found in this donor population. The sequencing results of the 400 randomly selected samples revealed that these donors were homozygous at the KEL gene locus, having the KEL*04/KEL*04 genotype (Figure 1). These results agreed with the serological testing results using CAT.

According to the PCR-SSP results, among 3256 examined samples, a 158 bp product was only detected using the particular primer sets for KEL*04 allele detection, thus indicative of the KEL*04/KEL*04 genotype. The HGH internal control was also amplified in all reactions (Figure 2). Three samples with the Kp(a−b+), Kp(a+b−), and Kp(a+b+) phenotypes as determined using DNA sequencing were used as known DNA controls to validate the genotyping results for both alleles. For all three control samples, there was no cross-amplification detected between KEL*03 and KEL*04 primer sets.

In total, to detect KEL*03 and KEL*04 alleles, we genotyped 2529 samples obtained from central Thai blood donors, 300 samples from northern Thai donors, and 427 samples from southern Thai donors using the newly developed PCR-SSP. All 3256 Thai blood donors were homozygous at the KEL gene locus, having the KEL*04/KEL*04 genotype. The KEL*03/KEL*03 and KEL*03/KEL*04 genotypes were not detected (Table 2). The retesting results for the sub-group of 100 blood sample donors with known genotype agreed with the first round of PCR-SSP testing for all samples. The frequencies of the three genotypes, including KEL*03/KEL*03, KEL*03/KEL*04, and KEL*04/KEL*04 genotypes, were consistent between the central Thai, northern Thai, and southern Thai populations according to the Hardy-Weinberg equilibrium (χ² = 0.00, degree of freedom = 2, p = 1.00).

The KEL*03 and KEL*04 frequencies among the Thai donors in this study were similar to those reported in Filipino (p = 0.29),¹¹ Southeast Asian (p = 1.00),¹¹ Chinese (p = 0.12),¹¹ Korean (p = 1.00),¹¹ South Asian (p = 0.05),¹¹ Alaska Native/Aleut (p = 0.008),¹¹ and Hawaiian/Pacific Islander (p = 1.00) populations¹¹ (Table 3). There were, however, significant differences in KEL*03 and KEL*04 allele frequencies between the Thai donors in this study and individuals in a South African population¹⁵ (KEL*03: 0.0000 vs 0.0067; and KEL*04: 1.0000 vs 0.9933; p = 0.002). In addition, there were significant differences in allele frequencies between Thai donors and individuals in Japanese (p < 0.001),¹¹ Native American (p < 0.001),¹¹ Brazilian (p < 0.001),⁹ Swiss (p < 0.001),¹⁶ and German populations (p < 0.001).¹⁷

One limiting factor of the PCR-SSP assay stems from the need for samples with known and relatively rare phenotypes, particularly Kp(a+b−) and Kp(a+b+), to ensure the test’s validity and reliability. In addition, an erroneous prediction of the K₀, Kmod, KEL: 1 weak, and other variants of the KEL gene – which are exceedingly rare in Thai and other populations – may occur when using PCR-SSP to determine KEL*03 and KEL*04 alleles. The confirmation of these phenotypes using serological techniques and KEL gene sequencing is thus suggested.^2,22 To obtain a more representative picture of the prevalent KEL gene variants among Thai blood donors, the genotypic results reported in this study need to be verified in large-scale studies involving all regions in Thailand.

This study found a 100% KEL*04 allele frequency in three Thai populations. This data could provide information on KEL*03 and KEL*04 allele frequencies to estimate the risk of alloimmunisation in Thai populations.