A Rapid Genotyping Assay for FTO and PPAR-γ Nutrigenetic Variants in Thai Adults with Overweight/Obesity

Abstract

Background

Nutrigenetic research linking fat mass and obesity-associated (FTO) and peroxisome proliferator-activated receptor gamma (PPAR-γ) variants to dietary response and obesity risk requires accessible genotyping tools, particularly for population studies in low- and middle-income countries. Current genotyping methods are typically limited to single or duplex formats, increasing cost and processing time for multivariant screening.

Objectives

The aim of this study is to develop, validate, and apply a cost-effective 4-plex high-resolution melting (HRM) assay for simultaneous detection of FTO (rs9939609 and rs1421085) and PPAR-γ (rs1801282 and rs3856806) variants using standard laboratory equipment.

Methods

Allele-specific primers with distinct melting temperatures were designed for multiplexed HRM detection of 4 obesity-associated single nucleotide polymorphisms. Analytical validation was performed against Sanger sequencing in 30 samples. Diagnostic accuracy was assessed by sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). The validated assay was applied to genotype 384 Thai adults (123 normal-weight, 261 with overweight/obesity) from Bangkok, Thailand. Genotype frequencies and Hardy–Weinberg equilibrium (HWE) were evaluated by the χ² test.

Results

The 4-plex HRM assay demonstrated 100% concordance with Sanger sequencing across all 4 variants (sensitivity: 100%; specificity: 100%; PPV: 100%; NPV: 100%). All triplicate measurements produced consistent genotype calls. Total assay time was 70 min. Based on the manufacturer’s catalog prices (QIAGEN, 2024), the estimated reagent cost per sample was ∼4- to 5-fold lower than 4 individual TaqMan assays and 6- to 8-fold lower than Sanger sequencing for the same 4 targets. Population screening revealed minor allele frequencies of 5.3%–8.2% for FTO and 0%–5.8% for PPAR-γ variants, with genotype distributions conforming to HWE for 3 of 4 variants.

Conclusions

This validated 4-plex HRM assay provides a rapid, accurate, and cost-effective tool for nutrigenetic screening of obesity-associated variants. The method is suitable for large-scale population studies where budget and infrastructure constraints limit access to high-throughput genotyping platforms.

Graphical abstract

Introduction

Obesity represents a major global public health concern driven by complex interactions between genetic susceptibility and environmental factors, particularly dietary intake patterns [1]. In Thailand, rapid urbanization and nutritional transition have contributed to rising obesity prevalence, with associated increases in type 2 diabetes, cardiovascular disease, and metabolic syndrome [2]. Understanding the genetic determinants of obesity is essential not only for risk stratification but also for developing gene-informed nutritional interventions, which is a central goal of the growing field of nutrigenetics [3,4]. However, the implementation of nutrigenetic approaches in population-level research requires accessible and affordable genotyping tools, particularly in low- and middle-income countries (LMICs) where the burden of nutrition-related chronic disease is increasing most rapidly.

Among obesity-associated genes, the fat mass and obesity-associated (FTO) gene and peroxisome proliferator-activated receptor gamma (PPAR-γ) gene are among the most extensively studied nutrigenetic targets. FTO variants rs9939609 and rs1421085, located on chromosome 16q12.2, influence energy intake regulation, appetite control, and macronutrient preference [[5], [6], [7], [8]]. Risk allele carriers demonstrate elevated consumption of sugar and dietary fat, and these gene–diet interactions contribute to obesity susceptibility across diverse populations [9,10]. With respect to predicted functional consequences, the FTO intronic variants rs9939609 and rs1421085 are regulatory rather than protein-coding: the obesity-risk allele is reported to disrupt a repressor-binding motif and de-repress downstream genes governing adipocyte energy balance, shifting cells toward lipid storage and increasing energy intake, which provides a plausible mechanism for their association with obesity susceptibility [7,8]. For PPAR-γ, the rs1801282 (Pro12Ala) missense variant modestly reduces receptor transcriptional activity, with the minor (Ala) allele generally associated with improved insulin sensitivity, whereas rs3856806 (C1431T) is a synonymous variant in linkage with functional haplotypes associated with altered body composition and metabolic phenotypes [11,12]. Through these mechanisms, both genes regulate adipocyte differentiation, lipid metabolism, and insulin sensitivity, and have been implicated in obesity, type 2 diabetes, and metabolic syndrome [5,6,12]. These 4 variants were chosen for their established associations with obesity-related phenotypes, their documented interactions with dietary factors [13,14], and their relatively high minor allele frequencies in Asian populations, making them suitable targets for a rapid nutrigenetic genotyping assay.

Despite the relevance of these variants to nutritional science, nutrigenetic research in LMICs remains constrained by the cost and infrastructure requirements of conventional genotyping methods. TaqMan allelic discrimination assays require proprietary fluorescent probes at ∼$5–$7 per single nucleotide polymorphism (SNP) per sample. Sanger sequencing, although accurate, involves multistep post-PCR processing and costs ∼$8–$10 per reaction. Higher-throughput platforms such as Sequenom MassARRAY or microarray-based systems require specialized instrumentation often unavailable in resource-limited research settings [2,15]. These barriers limit the scale at which gene–diet interaction studies can be conducted in populations where such data are most needed.

High-resolution melting (HRM) analysis offers a rapid, sensitive, and cost-effective alternative for SNP detection that requires only standard real-time PCR equipment and a single intercalating dye [15]. This technique enables identification of genetic variants through differential melting profiles without post-PCR processing, facilitating rapid, closed-tube genotyping. Compared with probe-based real-time PCR methods, HRM uses low-cost fluorescent dyes and requires minimal optimization [15]. Although HRM has been successfully applied for single-plex and duplex genotyping of individual obesity-associated variants [16], no multiplexed HRM assay currently enables simultaneous screening of multiple FTO and PPAR-γ nutrigenetic variants in a single reaction. We have previously demonstrated the utility of HRM-based FTO genotyping in characterizing gene–diet interactions and dietary behaviors in Thai adults [15,16]. Here, we report the development and validation of a 4-plex HRM assay for simultaneous detection of 4 FTO and PPAR-γ variants, and its application to screen 384 adults from Bangkok, Thailand. This method aims to provide a practical, accessible genotyping tool for nutrigenetic research and population screening in resource-limited settings.

Methods

Blood sampling and ethics

Whole blood samples were collected into EDTA anticoagulant tubes and transported to the laboratory under controlled conditions at 4°C. Upon receipt, samples were stored at −20°C for ∼1–3 mo before analysis. This protocol maintains sample integrity for genetic assessments, consistent with previous findings, indicating that blood specimens remain stable for ≤ 7–12 mo when preserved in EDTA tubes at −80°C.

For HRM assay validation, 30 blood samples were analyzed for FTO (rs9939609 and rs1421085) and PPAR-γ (rs1801282 and rs3856806) variants. For determining variant frequencies and demonstrating method applicability in population screening, 384 blood samples were collected from participants (83 males and 301 females) residing in Lak Si District, Bangkok, Thailand. The predominance of female participants reflects the community-based recruitment in this district and represents a limitation of the screening dataset. It reduces the precision of sex-stratified genotype estimates, particularly for the small male subgroups, and may contribute to the deviation from Hardy–Weinberg equilibrium (HWE) observed for 1 variant in females. Because the primary objective of this study was analytical validation of the genotyping method rather than epidemiological inference, this imbalance does not affect the assay-performance conclusions.

The study received ethical approval from the Ethics Committee of the Faculty of Tropical Medicine, Mahidol University, Thailand, on June 22, 2023 (MUTM 2023-042-01). All research procedures were carried out in accordance with the principles outlined in the Declaration of Helsinki. Written informed consent was obtained from all participants before enrollment.

The study cohort partially overlaps with populations described in our previous reports from the same community-based recruitment in Lak Si District: associations between FTO variants and macronutrient intake [17] and dietary lipid associations with inflammatory cytokines and leptin [18]. The present study addresses a distinct research objective: the development, analytical validation, and practical demonstration of a multiplexed HRM genotyping assay. Population screening data are presented to illustrate method applicability and are not intended to report novel genetic associations.

Anthropometric measurements

Anthropometric evaluations encompassed measurements of height, weight, and circumferences of the waist and hips. BMI was calculated as weight in kilograms divided by the square of height in meters (kg/m²). Waist circumference (WC) and hip circumference (HC) were measured in centimeters using a nonelastic plastic measuring tape with participants standing upright in light clothing, breathing normally, with the tape placed midway between the lower rib margin and the iliac crest. The waist-to-hip ratio was calculated as the ratio of WC to HC. All anthropometric data collection followed standardized procedures outlined by the International Biological Program [19] and used calibrated equipment. Body composition parameters, including fat mass percentage, fat-free mass, muscle mass, basal metabolic rate, metabolic age, and visceral fat rating, were assessed via bioelectrical impedance analysis using the HBF-375 device (Omron Healthcare).

DNA extraction

Genomic DNA was isolated from blood samples using the QIAamp DNA Blood Mini Kit (250) (QIAGEN), following the manufacturer’s protocol. DNA concentration was quantified using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific).

Primer design for HRM assay

Primers were specifically designed to detect the FTO gene variants rs9939609 (A > T) and rs1421085 (C > T), as well as the PPAR-γ gene variants rs1801282 (C > G) and rs3856806 (C > T), using the National Center for Biotechnology Information (NCBI) Primer-BLAST tool. Each primer pair was constructed to generate PCR products with distinct melting temperatures (Tm), enabling discrimination of all 4 variants in a single multiplexed reaction. Two primer sets were employed per SNP: one specific to the mutant allele and another to the wild-type allele. Primer sequences and optimized concentrations are presented in Table 1.

TABLE 1.

HRM primer for each genetic variant

Gene	rs	Referenceaccessionnumber		Primer sequence	Tm (°C)	Ampliconsize (bp)	Concentration(nM)
FTO	9939609	NC_000016.10	(WT) F: T	5ʹ GCG ACT GCT GTG AAT TTT 3ʹ	76.0	100	600
			(MT) F: A	5ʹ GCG ACT GCT GTG AAT TTA 3ʹ			600
			R:	5ʹ TTT GCT TTT ATG CTC TCC CA 3ʹ
FTO	1421085	NC_000016.10	F: T	5ʹ CAG GTC CTA AGG CAT GAT 3ʹ	80.1	97	300
			F: C	5ʹ CAG GTC CTA AGG CAT GAC 3ʹ			600
			R:	5ʹ TGG CCC AGT GGG GAG AT 3ʹ
PPAR-γ	1801282	NC_000003.12	F: C	5ʹ GAG ATT CTC CTA TTG ACC 3ʹ	78.1	100	600
			F: G	5ʹ GAG ATT CTC CTA TTG ACG 3ʹ			600
			R:	5ʹ CGT CCC CAA TAG CCG TAT 3ʹ
PPAR-γ	3856806	NC_000003.12	F: C	5ʹ CAG ATT GTC ACG GAA CAC 3ʹ	82.2	198	1500
			F: T	5ʹ CAG ATT GTC ACG GAA CAT 3ʹ			600
			R:	5ʹ AAC AAT ATG CAT AAA ATA GAT CAT 3ʹ

Abbreviations: FTO, fat mass and obesity-associated; HRM, high-resolution melting; MT, minor allele; PPAR-γ, peroxisome proliferator-activated receptor gamma; Tm, melting temperatures; WT, wild-type; rs, Reference SNP cluster ID.

PCR amplification and melting curve analysis

A multiplex PCR was performed to simultaneously amplify target regions of the FTO and PPAR-γ genes using a single reaction mix. Optimization of assay parameters, including primer concentrations, reaction conditions, and detection settings, was performed to enhance both sensitivity and specificity. The HRM assay was carried out in a 12.5 μL reaction volume comprising 6.25 μL of 2× HRM Type-It mix (QIAGEN), molecular-grade water, varying concentrations of specific primers (Table 1), and 2.5 μL of genomic DNA template (3–10 ng/μL). PCR amplification and HRM analysis were conducted on the Rotor-Gene Q platform (QIAGEN), with melting analysis performed from 70°C to 90°C, acquiring fluorescence data at 0.1°C increments with a 2-s stabilization at each step. Each run included wild-type reference controls (wild-type gDNA confirmed by sequencing) and mutant positive controls (gDNA containing a confirmed minor allele). Data were analyzed using Rotor-Gene Q software, and all experiments were conducted in triplicate to ensure reproducibility. Assay reproducibility was confirmed by concordance of genotype calls across triplicate measurements for each sample and each SNP target [17,18].

Cost analysis

Reagent cost per sample for the 4-plex HRM assay was calculated based on the manufacturer’s catalog list prices accessed in 2024: Type-it HRM PCR Kit (Cat. No. 206544, QIAGEN; 400 × 25 μL reactions) and QIAamp DNA Blood Mini Kit (Cat. No. 51106, QIAGEN; 250 preparations). Custom oligonucleotide primers were synthesized at standard desalted purity (25 nmol scale). The HRM assay reaction volume (12.5 μL) is half the manufacturer’s standard protocol (25 μL), effectively doubling the number of reactions per kit. Cost per sample was calculated as: (master mix cost per 12.5 μL reaction) + (primer cost per reaction) + (DNA extraction cost per sample). Total per-sample cost for 4 SNPs was compared with the estimated cost of performing 4 individual reactions using alternative genotyping methods. For comparison purposes, the estimated costs of alternative genotyping methods were calculated using the same cost framework, including reagent and per-reaction consumable costs.

Validation by Sanger sequencing and quality control

To validate the accuracy of the HRM assay, DNA sequencing was performed on a subset of 30 randomly selected samples. Genomic DNA previously extracted served as the template for PCR amplification. The PCR reaction was carried out in a final volume of 25 μL, consisting of 1.25 units of Taq DNA polymerase (Thermo Fisher Scientific), 200 μM of each deoxynucleotide triphosphate, 0.25 μM of each forward and reverse primer (Table 2), 50 ng of gDNA, and Taq buffer supplemented with (NH₄)₂SO₄. The thermal cycling conditions included an initial denaturation step at 95°C for 3 min, followed by 30 cycles of denaturation at 95°C for 30 s, annealing for 30 s at either 55°C (for PPAR-γ rs3856806) or 60°C (for FTO rs9939609, FTO rs1421085, and PPAR-γ rs1801282), and extension at 72°C for 1 min. This was followed by a final extension at 72°C for 10 min, and a hold at 4°C.

TABLE 2.

Primer for sequencing in this study

Gene	rs	Direction	Sequence	Tm (°C)
FTO	9939609	F:	5ʹ TAT CTT TTG GCA GAT CAG AAC 3ʹ	48.5
		R:	5ʹ GAA CAA ATG TTC AAG TCA CAC 3ʹ	48.5
FTO	1421085	F:	5ʹ GTC TCT AAG CCC AAC AAA CG 3ʹ	51.8
		R:	5ʹ GTT CCC ATC TTT AAG GTC AGA 3ʹ	50.5
PPAR-ץ	1801282	F:	5ʹ CCA ATT CAA GCC CAG TCC T 3ʹ	51.1
		R:	5ʹ TGA ACG CGA TAG CAA CGA G 3ʹ	51.1
PPAR-ץ	3856806	F:	5ʹ TTG TGT TTT CCA TAT GTG C 3ʹ	44.6
		R:	5ʹ TTT CAC AGT AAA TTT CTT AGG 3ʹ	44.6

Abbreviations: FTO, fat mass and obesity-associated; PPAR-γ, peroxisome proliferator-activated receptor gamma; Tm, melting temperatures; rs, Reference SNP cluster ID.

Statistical analysis

All data were expressed as mean ± SD. Diagnostic accuracy of the multiplex HRM assay was assessed by calculating sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) against Sanger sequencing as the reference standard. Assay reproducibility was confirmed by concordance of genotype calls across triplicate measurements. Genotypic and allelic frequencies were determined, and the HWE was assessed using the χ² test. To demonstrate method applicability, genotype frequencies were compared between normal-weight (BMI <23.0 kg/m²) and overweight/obesity groups (BMI ≥23.0 kg/m²), and anthropometric parameters were compared by weight status within genotype groups using Student’s t-test. Detailed analyses of gene–diet interactions and genotype–phenotype relationships in this cohort have been reported separately [15,16]. A P value of <0.05 was considered statistically significant. Statistical analyses were performed using SigmaPlot (Systat Software Inc.) and SPSS version 22 (IBM Corp.).

Results

Development and validation of 4-plex HRM assay

To evaluate the performance of the developed 4-plex HRM assay, 30 genomic DNA samples were genotyped for 4 SNPs: FTO rs9939609, FTO rs1421085, PPAR-γ rs1801282, and PPAR-γ rs3856806. All results obtained from the HRM assay were confirmed by direct DNA sequencing.

For the FTO rs9939609 variant, the HRM assay identified 13 samples with the homozygous TT genotype, 10 samples with the heterozygous TA genotype, and 5 samples with the homozygous AA genotype. These results were fully concordant with the sequencing data. Similarly, for FTO rs1421085, 14 samples were identified as homozygous TT, 11 as heterozygous TC, and 5 as homozygous CC by the HRM assay, matching exactly with DNA sequencing results.

Regarding the PPAR-γ rs1801282 polymorphism, all 30 samples were homozygous for the wild-type CC genotype. No heterozygous (CG) or homozygous mutant (GG) genotypes were detected by either HRM or sequencing. For PPAR-γ rs3856806, the HRM assay detected 19 samples with the CC genotype, 4 with the heterozygous CT genotype, and 5 with the TT genotype, consistent with sequencing analysis.

The overall genotyping concordance between the HRM assay and DNA sequencing was 100% for all 4 SNPs analyzed (Table 3). The HRM genotyping results showed 100% sensitivity, 100% specificity, 100% PPV, and 100% NPV for all analyzed SNPs. All experiments were conducted in triplicate with consistent genotype assignments across replicates, confirming assay reproducibility. These results demonstrate the reliability and accuracy of the developed 4-plex HRM assay for multiplex SNP detection (FIGURE 1, FIGURE 2).

TABLE 3.

Validation results of 30 samples detected by 4-plex HRM and DNA sequencing

Gene	HRM assay	DNA sequencing
FTO rs9939609 (TT)	13/30	13/30
FTO rs9939609 (T + A)	10/30	10/30
FTO rs9939609 (AA)	5/30	5/30
FTO rs1421085 (TT)	14/30	14/30
FTO rs1421085 (T + C)	11/30	11/30
FTO rs1421085 (CC)	5/30	5/30
PPAR-γ rs1801282 (CC)	30/30	30/30
PPAR-γ rs1801282 (C + G)	0/30	0/30
PPAR-γ rs1801282 (GG)	0/30	0/30
PPAR-γ rs3856806 (CC)	19/30	19/30
PPAR-γ rs3856806 (C + T)	4/30	4/30
PPAR-γ rs3856806 (TT)	5/30	5/30

Abbreviations: HRM, high-resolution melting; FTO, fat mass and obesity-associated; PPAR-γ, peroxisome proliferator-activated receptor gamma; rs, Reference SNP cluster ID.

FIGURE 1.

Identification of FTO and PPAR-γ gene minor allele by the multiplexed HRM assay. The assay is based on base complementarity between allele-specific primers and the DNA template. (A) Mutant samples produce a peak at the corresponding Tm, whereas WT samples do not produce PCR products, resulting in a flat line. (B) WT positive samples produce a peak at the corresponding Tm, whereas MT samples do not produce PCR products, resulting in a flat line. The differential melting profiles allow clear discrimination between WT and minor alleles for each target SNP. HRM, high-resolution melting; FTO, fat mass and obesity-associated; MT, minor allele; PPAR-γ, peroxisome proliferator-activated receptor gamma; SNP, single nucleotide polymorphism; Tm, melting temperature; WT, wild-type.

FIGURE 2.

Sanger sequencing validation results for FTO rs9939609. Representative electropherograms showing DNA sequencing confirmation of HRM genotyping results. (A) Wild-type homozygous genotype (TT) showing a single T peak. (B) Heterozygous genotype (TA) showing overlapping T and A peaks at the polymorphic site. (C) Mutant homozygous genotype (AA) showing a single A peak. These sequencing results demonstrate 100% concordance with HRM assay genotype calls, validating the accuracy of the developed method. HRM, high-resolution melting; FTO, fat mass and obesity-associated.

Cost analysis

Based on the manufacturer’s catalog list prices (QIAGEN, 2024), the estimated reagent cost per sample for the 4-plex HRM assay was ∼4- to 5-fold lower than performing 4 individual TaqMan allelic discrimination assays and ∼6- to 8-fold lower than Sanger sequencing of 4 targets. The primary cost advantage derives from the multiplexed single-reaction format: the 4-plex HRM assay genotypes 4 SNPs in one 12.5 μL reaction using a single intercalating dye, whereas TaqMan and Kompetitive Allele-Specific PCR (KASP) platforms require 4 separate probe-based reactions, and Sanger sequencing requires 4 PCR amplifications plus post-PCR cleanup and sequencing runs.

Application to population screening: variant frequencies

After the analytical validation, the 4-plex HRM assay was applied to screen 384 Thai adults to demonstrate its utility for population-level genotyping. A total of 384 participants were included, consisting of 123 individuals with normal-weight (19 males and 104 females) and 261 individuals with overweight or obesity (66 males and 195 females). The distribution of genotypes for the 4 SNPs was analyzed according to sex and weight status (Table 4).

TABLE 4.

The prevalence of gene minor allele by weight status and sex

SNPs	Normal (n = 123)		Over and obese (n = 261)
	Male (n = 19)	Female (n = 104)	Male (n = 66)	Female (n = 195)
	n (%)		n (%)
FTO rs9939609
Genotype
TT	14 (73.7)	62 (59.6)	40 (60.6)	110 (56.4)
TA	4 (21.1)	35 (33.7)	23 (34.8)	70 (35.9)
AA	1 (5.3)	7 (6.7)	3 (4.5)	15 (7.7)
P-HWE	0.363	0.503	0.893	0.413
MAF	0.158	0.236	0.220	0.256
FTO rs1421085
Genotype
TT	14 (73.7)	61 (58.7)	38 (57.6)	108 (55.4)
TC	4 (21.1)	36 (34.6)	25 (37.9)	71 (36.4)
CC	1 (5.3)	7 (6.7)	3 (4.5)	16 (8.2)
P-HWE	0.363	0.594	0.661	0.376
MAF	0.158	0.240	0.235	0.264
PPAR-γ rs1801282
Genotype
CC	18 (94.7)	103 (99.0)	66 (100.00)	191 (97.9)
CG	1 (5.3)	1(1.0)	0	4 (2.1)
GG	0	0	0	0
P-HWE	0.906	0.960	0.950	0.884
MAF	0.026	0.005	0	0.010
PPAR-γ rs3856806
Genotype
CC	11 (57.9)	84 (80.8)	49 (74.2)	159 (81.5)
CT	7 (36.8)	14 (13.5)	16 (24.2)	27 (13.8)
TT	1 (5.3)	6 (5.8)	1 (1.6)	9 (4.6)
P-HWE	0.933	0.000	0.812	0.000
MAF	0.237	0.125	0.136	0.115

P-HWE was calculated from χ² test.

Abbreviations: HWE, Hardy–Weinberg equilibrium; FTO, fat mass and obesity-associated; MAF, minor allele frequency; PPAR-γ, peroxisome proliferator-activated receptor gamma; SNP, single nucleotide polymorphism; rs, Reference SNP cluster ID.

For FTO rs9939609, the TT genotype was the most common among both normal-weight (73.7% males and 59.6% females) and participants with overweight/obesity (60.6% males and 56.4% females). The TA genotype ranged from 21.1% to 35.9%, whereas the AA genotype was the least frequent across all groups. A similar pattern was observed for rs1421085, with TT being the predominant genotype in both weight groups. The genotypes of both FTO SNPs were in HWE (P > 0.05) in both normal and groups with overweight/obesity.

For the PPAR-γ gene, rs1801282 showed a high prevalence of the CC genotype across all subgroups, with the CG genotype being rare and the GG genotype absent in all groups. The rs1801282 genotypes were in HWE in all analyzed groups, including the overall cohort (P HWE = 0.884). For rs3856806, the CC genotype was most prevalent, ranging from 57.9% in normal males to 81.5% in females with overweight/obesity. However, the P value for HWE of PPAR-γ rs3856806 was <0.05 in female subgroups, indicating deviation from expected Mendelian proportions.

Genotype–phenotype associations

To demonstrate the assay's usefulness, we examined genotype distributions by weight status (Table 4) and compared BMI between weight groups within each genotype. For the FTO variants, the nonrisk genotype (TT) was more common in the normal-weight group than in the overweight/obesity group (rs9939609: 59.6%–73.7% compared with 56.4%–60.6%; rs1421085: 58.7%–73.7% compared with 55.4%–57.6%) (Table 4). Risk-allele genotypes were therefore more common in the overweight/obesity group. For the PPAR-γ variants, the risk allele was uncommon, and genotype distributions were similar between weight groups (rs1801282 CC: 94.7%–99.0% compared with 97.9%–100%; rs3856806 CC: 57.9%–80.8% compared with 74.2%–81.5%) (Table 4). Within every genotype, participants with overweight/obesity had significantly higher BMI than normal-weight participants for all 4 SNPs (P < 0.05; Figure 3). Among risk allele carriers of FTO rs9939609, FTO rs1421085, and PPAR-γ rs3856806, the overweight/obesity group also showed significantly higher adiposity and other anthropometric measurements (P < 0.05; Supplemental Table 1). rs1801282 was not included here because its minor allele was too rare in this cohort.

FIGURE 3.

Distribution of genotypes by BMI status. (A) BMI distribution comparing normal-weight and groups with overweight/obesity for homozygous dominant genotypes in each variant. (B) BMI distribution comparing normal-weight and groups with overweight/obesity for combined heterozygous and homozygous recessive genotypes. Significant differences (P < 0.05) were observed between groups for all analyzed SNPs. Data are presented as mean ± SD. Statistical significance was determined using Student’s t-test. FTO, fat mass and obesity-associated; PPAR-γ, peroxisome proliferator-activated receptor gamma; SNP, single nucleotide polymorphism.

These patterns agree with previous reports for these variants in Thai and other Southeast Asian populations [[20], [21], [22], [23]], confirming that the 4-plex HRM assay produces genotype data suitable for detecting established genotype–phenotype associations.

Discussion

The development of cost-effective, multiplexed genotyping methods is essential for advancing nutrigenetic research, particularly in resource-limited settings where conventional genotyping platforms are inaccessible. Most existing HRM methods for FTO and PPAR-γ genotyping are limited to single-plex or duplex formats, allowing identification of only 1 or 2 variants at a time [24]. Multiplexing beyond 2 targets typically requires multiple fluorescent dyes, which increases assay cost and complexity [25]. The 4-plex HRM method reported here addresses these limitations by enabling simultaneous detection of 4 common FTO and PPAR-γ nutrigenetic variants using a single fluorescent dye and standard laboratory equipment in a 70-min closed-tube reaction.

To the best of our knowledge, this is the first report of a 4-plex HRM assay for simultaneous genotyping of FTO rs9939609, FTO rs1421085, PPAR-γ rs1801282, and PPAR-γ rs3856806 in a single reaction. The assay design exploits allele-specific primers generating PCR products with distinct Tm, enabling unambiguous discrimination of all 4 variants from a single melting curve profile. Validation against Sanger sequencing in 30 samples demonstrated 100% concordance, with perfect sensitivity, specificity, PPV, and NPV across all 4 SNPs. All triplicate measurements produced consistent genotype calls, confirming high analytical reproducibility suitable for large-scale genotyping applications.

A key advantage of the 4-plex HRM assay is its cost-effectiveness for multivariant nutrigenetic screening. Based on the manufacturer’s catalog list prices (Type-it HRM PCR Kit, Cat. No. 206544, QIAGEN, 2024), the reagent cost per sample for simultaneous detection of 4 SNPs was estimated to be ∼4–5-fold lower than TaqMan-based genotyping, which requires 4 separate probe-based reactions with proprietary dual-labeled probes, and ∼6–8-fold lower than Sanger sequencing, which requires 4 PCR amplifications plus post-PCR gel electrophoresis, cleanup, and sequencing runs. The HRM assay eliminates the need for fluorescent probes, reducing per-reaction consumable costs, and the 70-min runtime enables same-day genotyping of study samples. These characteristics make the method particularly suitable for nutrigenetic screening studies in resource-limited settings where access to probe-based platforms or sequencing facilities may be restricted (Table 5).

TABLE 5.

Comparison of genotyping methods for simultaneous detection of 4 obesity-associated SNPs

Parameter	4-Plex HRM (this study)	TaqMan assay	Sanger sequencing	KASP assay
SNPs per reaction	4	1	1	1
Reactions needed (4 SNPs)	1	4	4	4
Estimated reagent cost per sample (4 SNPs)1	∼US$5	∼US$20–25	∼US$30–40	∼US$15–20
Relative reagent cost (4 SNPs)	1× (reference)	∼4–5× higher	∼6–8× higher	∼3–4× higher
Total run time (4 SNPs)	70 min	∼280 min (4 × 70 min)	∼8 h (incl. post-PCR)	∼360 min (4 × 90 min)
Post-PCR processing	None (closed-tube)	None	Gel + cleanup + sequencing	None
Fluorescent probes	No (intercalating dye only)	Yes (dual- labeled probes)	No	Yes (FRET cassette)
Specialized equipment	Standard qPCR with HRM	Standard qPCR	PCR + capillary sequencer	Standard qPCR
Hands-on time per sample	∼5 min	∼20 min (4 setups)	∼45 min (4 setups + cleanup)	∼20 min (4 setups)

Abbreviations: FRET, Förster resonance energy transfer; HRM, high-resolution melting; KASP, Kompetitive Allele-Specific PCR; SNP, single nucleotide polymorphism.

¹Costs are estimated based on catalog reagent prices (2024). TaqMan, KASP, and Sanger costs represent 4 individual reactions required to genotype the same 4 SNPs.

The genotype distributions of FTO rs9939609, rs1421085, and PPAR-γ rs1801282 were in HWE (P > 0.05), suggesting that the study sample is genetically representative and free from major confounding factors such as population stratification, inbreeding, or genotyping error [26]. These findings are consistent with HWE-conforming distributions reported for FTO and PPAR-γ SNPs in East Asian and Southeast Asian populations [27,28]. However, the P value for HWE of PPAR-γ rs3856806 was <0.05 in female subgroups, indicating deviation from expected Mendelian proportions. Because the assay showed 100% concordance with Sanger sequencing, this deviation is unlikely to be technical in origin and more probably reflects population stratification, nonrandom mating, recruitment-related selection bias, or the limited size of the female subgroups, consistent with deviations reported in previous studies of this variant [29,30]. Analysis of these samples, together with additional samples, is needed to validate this finding.

The practical utility of affordable FTO and PPAR-γ genotyping for nutrigenetic research has been demonstrated through studies using genotype data generated by this HRM approach. In the same study population, we previously reported that FTO rs9939609 risk allele carriers exhibited significantly higher consumption of total fat, saturated fat, and sugar [17], and that FTO variant status was associated with altered dietary lipid profiles influencing inflammatory cytokine and leptin levels [18]. These findings illustrate that a low-cost genotyping method can generate actionable nutrigenetic data suitable for investigating gene–diet interactions in LMIC populations. The 4-plex format described here further enhances this capability by enabling simultaneous assessment of multiple gene–diet interaction targets in a single assay.

The validated method opens several avenues for future research. The multiplexing strategy could be expanded to include additional nutrigenetic variants associated with macronutrient metabolism, micronutrient status, or dietary response phenotypes. The assay’s compatibility with standard real-time PCR instruments that are increasingly available in clinical and research laboratories across Southeast Asia supports its potential for broader adoption. In addition, the method could be adapted for point-of-care nutrigenetic screening in community health settings, supporting personalized dietary recommendations based on individual genetic profiles.

Several limitations should be acknowledged. First, analytical validation was performed using 30 samples, which, although demonstrating 100% concordance with Sanger sequencing, represents a modest validation set; larger multicenter validation studies would strengthen confidence in assay transferability across laboratories and instruments. Second, the HWE deviation observed for PPAR-γ rs3856806 in female subgroups requires confirmation in these and additional samples; given the 100% concordance with Sanger sequencing, this deviation is more likely attributable to population or sampling factors than to technical error. Third, the cross-sectional study design and recruitment from a single Bangkok district limit the generalizability of the population screening data, though the primary objective was method validation rather than epidemiological inference. Fourth, the cost analysis is based on catalog reagent prices and does not account for equipment depreciation, personnel time, or institutional overhead, which may vary across settings. Finally, the assay was optimized on the Rotor-Gene Q platform; performance on other real-time PCR instruments with HRM capability requires separate validation.

In conclusion, a cost-effective 4-plex HRM assay was successfully developed and validated for simultaneous genotyping of FTO (rs9939609, rs1421085) and PPAR-γ (rs1801282, rs3856806) nutrigenetic variants. The method demonstrated 100% concordance with Sanger sequencing, requires only standard laboratory equipment and a single fluorescent dye, and completes in 70 min at substantially lower reagent cost than probe-based or sequencing-based alternatives. Application to 384 Thai adults confirmed that the assay generates reliable genotype data consistent with established population frequencies and genotype–phenotype associations. The same allele-specific, distinct-melting-temperature design could be extended to develop rapid HRM assays for other nutrigenetic variants. Examples include variants involved in macronutrient and micronutrient metabolism, lipid response, and appetite regulation, which would enable broader, low-cost screening panels for nutrigenetic research. This accessible genotyping tool lowers technical and financial barriers to nutrigenetic research, supporting the implementation of gene–diet interaction studies and population-level genetic screening in resource-limited settings.

Author contributions

The authors’ responsibilities were as follows – SP, PP: designed research and wrote the draft; SP: conducted research; UB: provided essential reagents and materials; SP, UB: analyzed data; and all authors: read and approved the final manuscript.

Data availability

Data described in the manuscript will be made available upon request pending application to and approval by the corresponding author.

Declaration of Generative AI and AI-assisted technologies in the writing process

During the preparation of this work, the authors used AI-assisted technologies in order to improve the grammar of the manuscript. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

Funding

This research was supported by Mahidol University. The funding source had no involvement in study design; collection, analysis, and interpretation of data; writing of the report; or the decision to submit the article for publication.

Conflict of interest

The authors report no conflicts of interest.

Appendix A. Supplementary data

The following is the Supplementary data to this article: