Rice (Oryza sativa) is one of the most important staple crops in the world (Priyadarshi et al. 2018). Understanding the genetic basis of agronomic traits important for rice production, such as yield, stress tolerance, and grain quality, is crucial for improving rice varieties and increasing global food security (Xing et al. 2010). Rice production is ham-pered by various abiotic stresses, because rice cultivating fields are located largely in stress prone areas. These abiotic stresses result in a yield loss of approximately 30% in the production (Lafitte et al. 2004). Thus, it is necessary to develop elite varieties tolerant to various abiotic stresses.

Many genes related to abiotic stress tolerance have been identified in rice (Yang et al. 2020). Some of them provide significantly enhanced tolerance to the crop, however one major gene is sometimes not enough to give enough tole-rance to the crop plant. Pyramiding major genes associated with abiotic stress tolerance is a promising strategy to enhance tolerance phenotypes, not only for specific stresses but also for multi-stress tolerance (Joshi et al. 2010). Gene pyramiding has been performed for various abiotic stress tolerances such as saline stress, drought stress and heat stress etc (Shailani et al. 2021; Withanawasam et al. 2022). Application of DNA markers targeting major genes un-derlying abiotic stress tolerance can facilitate gene pyramid-ing process.

With the advent of biotechnology, DNA markers have emerged as powerful tools revolutionizing the breeding process. DNA markers offer a cost-effective, efficient, and accurate means of identifying desirable traits in crops, allowing breeders to make informed selections and accele-rate the development of superior cultivars. Various DNA markers have been developed, starting from the early techn-iques like restriction fragment length polymorphism (RFLP), which use restriction enzymes and hybridization systems to detect genetic variation (Madhumati et al. 2014). Develop-ment of polymerase chain reaction (PCR) techniques enabled the development of PCR-based markers such as AFLP and RAPD markers which are relatively simple and cost- effective compared to RFLP (Kubelik et al. 1990). Ad-vancements in genome sequencing technologies have paved the way for the utilization of sequence-based DNA markers: simple sequence repeats (SSRs) (Wu and Tanksley 1993), sequence-tagged sites (STSs) (Inoue et al. 1994), and single nucleotide polymorphisms (SNPs) (Feltus et al. 2004).

SNPs are the most prevalent form of genetic variation, characterized by single-base differences within a DNA sequence, making them abundant across the genomes of individuals within a species. This abundance enables researchers to target specific genomic regions with ease, facilitating rapid marker development (Rafalski, 2002). Kompetitive Allele-Specific PCR (KASP) as a SNP marker genotyping method, has gained significant popularity due to its cost-effectiveness and scalability (Jagtap et al. 2020). KASP markers utilize a competitive PCR reaction, allow-ing bi-allelic scoring of SNPs as well as specific loci of insertions and deletions (Alvarez-Fernandez et al. 2021).

High-throughput SNP genotyping is particularly useful in crop breeding, enabling the simultaneous analysis of multiple SNPs in a high-throughput manner, drastically reducing genotyping time and labor costs (Thomson, 2014). Several high-throughput SNP genotyping platforms have been reported, including: Illumina Infinium iSelect HD array, Affymetrix Axiom array, Douglas Array Tape, Fluidigm dynamic arrays (Thomson, 2014). Among them, Fluidigm dynamic arrays adopt a flexible, PCR-based SNP platform using a nanofluidic integrated fluid circuit (IFC; Wang et al. 2009), which provides automated processing and multiplexing capabilities, enabling simultaneous analysis of multiple samples through the digital PCR principle. These features minimize manual labor and significantly decrease the consumption of reagents, leading to efficient and cost-effective genetic analysis with accelerated turn-around times. Converting KASP markers to Fluidigm- compatible assays offers streamlined experimental processes, reduced reagent consumption, and improved accuracy in genotyping results, enabling high-throughput analysis. This transition empowers researchers to accelerate their genetic studies, screen larger populations, and efficiently select in-dividuals carrying desirable genetic traits, ultimately contri-buting to the rapid development of improved crop varieties.

In this study, we developed SNP-based KASP markers for the major genes/QTL related to abiotic stresses and Fluidigm markers converted from the KASP markers. The Fluidigm markers were applied to analysis of genetic diversity through genotypic profiling of 172 domestic and abroad varieties.

Plant materials

Tolerant and susceptible rice varieties for specific abiotic stresses were used as control panels to validate newly designed SNP markers. For the validation of Saltol QTL markers, the susceptible variety IR64 was compared with the tolerant varieties IR64-Saltol (harboring FL478 allele) and Pokkali. Similarly, to validate the markers for qDTY2.2 and qDTY4.1, the susceptible variety IR64 was compared with the tolerant varieties IR64-DTY2.2+4.1 and Addeysel. The 172 domestic and foreign rice varieties used for profiling and screening are the same variety collection employed in the previous study (Lee et al. 2021).

SNP selection

SNPs linked to the QTLs for these abiotic stresses were selected based on data previously reported, and the posi-tions of selected SNP were confirmed by Rice SNP-Seek Database (https://snp-seek.irri.org/) and Rice Genome Annotation Project (http://rice.uga.edu) (Chin et al. 2011; Cotsaftis et al. 2011; Platten et al. 2019).

Genotyping using KASP marker assays

The KASP marker assays for the SNPs were designed based on the manual of LGC Genomics (http://www. lgcgenomics.com). The 50 bp sequences flanking selected SNPs on either side were used to design the KASP markers. The information of KASP marker assays and primer sequences was listed in Tables 1 and 2, respectively. The KASP markers assays were performed in a 96-well format and set up as 10 mL reactions with 4.86 mL of DNA template, 0.14 mL of assay mix (containing a common primer and two different allele specific primers), and 5 mL of master mix (containing fluorescence resonant energy transfer and Taq polymerase). Amplification was started at 94℃ for 15 minutes, followed by 10 cycles at 94℃ for 20 seconds and at 61℃ for 60 seconds (touchdown to 55℃, ‒1℃ per cycle), followed by 26 cycles at 94℃ for 20 seconds, 55℃ for 60 seconds, and a read step at 37℃ for 1 minute (Moon et al. 2019; Lee et al. 2021). After the amplification, the fluorescence signals of PCR products were read with the CFX Maestro software (Bio-Rad, U.S.A).

Table 1 . List of KASP markers developed in this study and relevant information

KASP marker assays ID	Trait	Chromosome	Position (bp)	Target gene/QTL	SNP (phenotypea))	SNP (color of dyeb))
Saltol-SNP-1	Salinity	1	10,217,743	Saltol	C(T):T(S)	C(F):T(H)
Saltol-SNP-2	Tolerance	1	11,430,304	-	A(T):G(S)	A(F):G(H)
Saltol-SNP-3		1	11,460,305	-	G(T):C(S)	G(F):C(H)
Saltol-SNP-4		1	11,463,299	-	A(T):T(S)	A(F):T(H)
Saltol-SNP-5		1	11,463,595	-	A(T):C(S)	A(F):C(H)
DTY2-2-SNP-1	Drought	2	2,085,377	qDTY2.2	G(T):C(S)	G(F):C(H)
DTY2-2-SNP-2	Tolerance	2	2,792,380	-	A(T):G(S)	A(F):G(H)
DTY2-2-SNP-3		2	3,200,039	-	A(T):G(S)	A(F):G(H)
DTY2-2-SNP-4		2	4,044,509	-	A(T):C(S)	A(F):C(H)
DTY4.1-SNP1	Drought	4	222,034	qDTY4.1	C(T):T(S)	C(F):T(H)
DTY4.1-SNP2	Tolerance	4	330,380	-	C(T):T(S)	C(F):T(H)
DTY4.1-SNP3		4	2,281,969	-	T(T):C(S)	T(F):C(H)
AG1-KP-SNP-1	Anaerobic	9	12,333,777	OsTPP7	C(T):T(S)	C(F):T(H)
AG1-KP-SNP-2	germination	9	12,326,525	-	G(T):A(S)	G(F):A(H)
AG1-KP-SNP-3	Tolerance	9	12,215,550	-	A(T):G(S)	A(F):G(H)
Sub1A-SNP1	Submergence	9	Sub1A gene	Sub1A	C(T):T(S)	C(F):T(H)
Sub1A-SNP2	Tolerance	9	Sub1A gene	-	A(T):G(S)	A(H):G(F)

^a)T: tolerant, S: susceptible.

^b)F: FAM fluorescence, H: HEX fluorescence.

Table 2 . List of KASP marker assay primers used in this study.

KASP marker assays ID	FAM-labelled primer (5’-3’)	HEX-labelled primer (5’-3’)1)	Reverse primer (5’-3’)2)
Saltol-SNP-1	TTTGGTAAGACATGAATGAACATCAG	CTTTTGGTAAGACATGAATGAACATCAA	CCACTGCACTACATTTTCTTTCAACTGAA
Saltol-SNP-2	TTCCAGGAACATCATAAATTTTGCTATT	TCCAGGAACATCATAAATTTTGCTATC	GGGCATCCTGAGACAAATGTTTTCCAT
Saltol-SNP-3	GACACGGTGGAGAGGTCGAC	GACACGGTGGAGAGGTCGAG	ACTCGAGGCACTCCGGTGAGAT
Saltol-SNP-4	ATAAACAACAAGTGTGAGAAAACCCT	CTATAAACAACAAGTGTGAGAAAACCCA	GAACTCATTTCTCTCCTATGTGTTTCGTT
Saltol-SNP-5	GTTCAACAAATAGGACTTCATATTTATGGAA	CAACAAATAGGACTTCATATTTATGGAC	CGCAGACACTTGTAGCATTAGATTTCATT
DTY2-2-SNP-1	GTTCTGTTTGACTTTGACTTCG	GCTGTTCTGTTTGACTTTGACTTCC	CCAATGTATTTTTTTGCCTTTTTGTTTCAA
DTY2-2-SNP-2	GACACTAGAATCAACAGTAATCCTGAT	ACACTAGAATCAACAGTAATCCTGAC	GCATATTTGTTTCACTGCTGATGACTCTA
DTY2-2-SNP-3	GATTTTCTTAGTTGCAGTATTTTCTGATTCA	TTCTTAGTTGCAGTATTTTCTGATTCG	GATGCCAAAACACAATGCATATAAAGGAAA
DTY2-2-SNP-4	GAGTTTTTGAGCTGGTAAATTTTGGAATTATT	GTTTTTGAGCTGGTAAATTTTGGAATTATG	CACTTGAAATTTGAATGACGTGGAGTACTA
DTY4.1-SNP1	CAATCGCTGATGAAGCTGTCGAC	CAATCGCTGATGAAGCTGTCGAT	CTGCCGACTGCATTATCCCTACTTT
DTY4.1-SNP2	GCTGATTTCATTGAATTTTGTTGC	GGCTGCTGATTTCATTGAATTTTGTTGT	GTTCATACTACGTGAAAACGAACTTCACAA
DTY4.1-SNP3	GCCACCGTGTGGGAGATGGAA	CCACCGTGTGGGAGATGGAG	TGGCCATGGCGGCGGCGAA
AG1-KP-SNP-1	ACGTTACAATGGTTTGAGTATATGG	GTCTACGTTACAATGGTTTGAGTATATGA	TTGTAAGGGTAAGGGTGGGACCTAT
AG1-KP-SNP-2	GAGACGGAGAAGACGGAGAAG	GGAGACGGAGAAGACGGAGAAA	CCATCACCGCCAAGAAGTCATCTTA
AG1-KP-SNP-3	GGATTGCAATAAATTATTGTCAAGTTTTACAA	GCAATAAATTATTGTCAAGTTTTACAG	GGATACAGTGTTTGAAATATGGAAATGTTA
Sub1A-SNP1	AGAAGCAGAGCGGCTGCGG	CAAGAAGCAGAGCGGCTGCGA	CTTCCCACCCGCCGATCTTTCTT
Sub1A-SNP2	GTCCGAGCAGCACTCCAGC	CGTCCGAGCAGCACTCCAGT	GTCCGGCGACGCGCGCATA

¹⁾FAM/HEX-labeled primers are allele-specific primers.

²⁾Reverse primers are common primers.

Genotyping using Fluidigm marker assays

The 150 bp sequences flanking the selected SNPs on either side were used to design the Fluidigm markers. The flanking sequences with selected SNPs were uploaded on the D3 Assay Design (https://d3.fluidigm.com/) website, and the designed assays were confirmed. One Fluidigm SNP assay contains allele-specific primer 1 (ASP1), ASP2, locus-specific primer (LSP), and specific target amplification (STA) primer. The primer information of Fluidigm marker assays was listed in Supplementary Table S1. For geno-typing of Fluidigm markers, we followed the genotyping procedure described in the study by Seo et al. 2020. The BioMark^TM HD system (Fluidigm, San Francisco, CA, USA) and 96.96 Dynamic Array IFCs (Fluidigm) were used for genotyping. STA was performed prior to SNP genotyping analysis to enhance the success rate of SNP type assay. A 10 × STA primer pool was prepared by combining 2 mL of STA primer, 2 mL of LSP and 16 mL of DNA suspension buffer for each marker. For each of the sample, the STA was conducted in a total volume of 5 mL per reaction, which con-tained 2.5 mL of 2 × multiplex PCR master mix (Qiagen, Hilden, Germany), 0.5 mL of the 10 × STA primer pool, 0.75 mL of PCR-certified water, and 1.25 mL of genomic DNA with the following thermal cycling conditions: 15 minutes at 95℃, followed by 14 cycles of a 2-step ampli-fication including 15 seconds at 95℃ and 2 minutes at 60℃. The STA products were diluted 1:50 in DNA suspension for the SNP genotyping. To perform the SNP assays using the 96.96 IFC, we prepared the assay mix and sample mix. The assay mix consisted of 1.5 mL of PCR-certified water, 2.5 mL of 2 × assay loading reagent, and 1.0 mL of the SNP assay mix. The SNP assay mix was obtained by combining 3 mL of each ASP, 8 mL of each LSP, and 29 mL of DNA suspension buffer. For the sample pre-mix, we mixed 360 mL of 2 × Fast Probe Master Mix (Biotium, Fremont, CA, USA), 36 mL of SNP type 20 × sample loading reagent, 12 mL of SNP type 60 × reagent, 4.3 mL of 50 × ROX dye (Invitrogen, Waltham, MA, USA), and 7.7 mL of PCR certi-fied water. Subsequently, the sample mix was prepared by mixing 2.5 mL of each STA product and 3.5 mL of the sam-ple pre-mix in each well of 96-well plates. Finally, 5 mL of each sample mix and 4 mL of each assay mix were loaded into 96 sample inlets and 96 assay inlets of the 96.96 IFC, respectively. The SNP assays were conducted using two machines: the IFC controller HX (Fluidigm) and the BioMark^TM HD system (Fluidigm). The thermal cycling conditions con-sisted of an initial step at 95℃ for 15 seconds, 64℃ for 45 seconds, and 72℃ for 15 seconds, with a touchdown of ‒1℃ per cycle from 64 to 61℃, followed by 34 cycles of 95℃ for 15 seconds, 60℃ for 45 seconds, and 72℃ for 15 seconds. The genotyping results were obtained using the Fluidigm SNP Genotyping Analysis software.

Analysis of genetic diversity

Genetic diversity was analyzed based on the genotype profiling of 172 varieties with the developed Fluidigm mar-kers. MEGA (Molecular Evolutionary Genetics Analysis) X software was used to construct UPGMA (unweighted pair group method with arithmetic mean) tree for the genetic diversity. Each SNP type for genotype of Fluidigm markers was converted to independent code value, and the arranged code values were used to construct UPGMA tree.

The selection of major QTLs for the development of markers associated with key environmental stress tolerance

Gene pyramiding enhances multi-stress tolerance in crops. The studies of gene pyramiding have been reported for various abiotic stresses such as drought, salinity, cold, submergence and anaerobic germination (Das et al. 2018; Yang et al. 2020; Shailani et al. 2021; Shin et al. 2022). Application of DNA markers targeting major gene and/or QTLs remarkably increase the efficiency of pyramiding the genes for abiotic stresses. In this study, three major QTLs were selected for the development of SNP-based KASP markers associated with key environmental stress tolerance in rice. Specifically, we focused on the Saltol QTL for saline stress (Bonilla et al. 2002), and DTY2.2 and DTY4.1 QTLs for drought stress (Bernier et al. 2007; Venuprasad et al. 2009). These QTLs have been extensively characterized and well-documented in the literature, with several studies reporting their significant effects on rice tolerance to their respective environmental stresses. Therefore, we believe that these QTLs are appropriate targets for the development of KASP markers to aid in the breeding of rice varieties with enhanced tolerance to environmental stresses.

KASP marker development and validation

KASP markers were designed based on the selected SNP positions for each QTL. Marker validation was performed by genotyping of the control panels consisting of tolerant and susceptible varieties for target traits. Saltol locus, representing the 5 Mb region (10.4-15.6 Mb) in chromo-some 1, is one of the major genomic loci conferring see-dling stage salt tolerance in rice (Krishnamurthy et al. 2015). As SSR markers, RM3412 and RM8094, efficiently discriminate the salt tolerant rice genotypes from suscep-tible genotypes, we designed KASP markers in the region of 10.2 Mb-11.5 Mb where RM3412 and RM8094 are located (Islam et al. 2011; Naresh Babu et al. 2014). Five KASP markers were developed and validated. The result of validation for designed KASP markers was shown in Table 3. The validation for the five KASP markers (Saltol- SNP-1, Saltol-SNP-2, Saltol-SNP-3, Saltol-SNP-4, and Saltol-SNP-5) showed that the IR64-Saltol (tolerant variety) showed a tolerant genotype (Allele1) for all designed KASP markers, while Pokkali (the other tolerant variety) showed tolerant genotype for three designed KASP mar-kers (Saltol-SNP-2, Saltol-SNP-3, and Saltol-SNP-4), but showed susceptible genotype (Allele2) for the other two designed KASP markers (Saltol-SNP-1 and Saltol-SNP-5). This indicates that the genomic constitution of IR64-Saltol is different from Pokkali, possibly due to the recombination occurring during line development. The IR64 (susceptible variety) had a susceptible genotype for all designed KASP markers.

Table 3 . Validation of KASP markers with the control varieties.

Sample	Saltol-SNP-1			Saltol-SNP-2			Saltol-SNP-3			Saltol-SNP-4			Saltol-SNP-5
	Allele	RFU_1	RFU_2	Allele	RFU_1	RFU_2	Allele	RFU_1	RFU_2	Allele	RFU_1	RFU_2	Allele	RFU_1	RFU_2
IR64	Allele2	3765.08	4968.89	Allele2	2612.31	2545.97	Allele2	3812.58	6702.01	Allele2	2646.61	2499.42	Allele2	2669.36	2660.83
IR64-Saltol	Allele1	6514.09	2882.8	Allele1	3041.22	2049.5	Allele1	7919.24	3105.35	Allele1	3097.16	2023.94	Allele1	3408.14	2111.12
Pokkali	Allele2	3813.43	6124.81	Allele1	2831.07	1998.89	Allele1	7277.74	2992.35	Allele1	3145.07	2028.88	Allele2	2700.2	2704.92
NTC	No Call	3708.49	1896.96	No Call	2576.03	1874.63	No Call	3555.29	1889.66	No Call	2557.01	1854.77	No Call	2655.05	1856.12
Sample	DTY2.2-SNP-1			DTY2.2-SNP-2			DTY2.2-SNP-3			DTY2.2-SNP-4
	Allele	RFU_1	RFU_2	Allele	RFU_1	RFU_2	Allele	RFU_1	RFU_2	Allele	RFU_1	RFU_2
IR64	Allele2	2629.21	2630.05	Allele2	3726.04	5402.83	Allele2	3765.29	5668.48	Allele2	3908.31	6165.91
IR64-DTY2.2+4.1	Allele1	3257.18	2053.37	Allele1	7525.46	3127.95	Allele1	6976.09	2990.59	Allele1	7652.61	3160.55
Addey sel	Allele1	3265.54	2011.19	Allele1	7301.05	3169.9	Allele1	7151.86	3058.99	Allele1	7323.76	3030.61
NTC	No Call	2581.52	1854.69	No Call	3375.49	1890.55	No Call	3473.84	1928.7	No Call	3621.66	1897.18
Sample	DTY4.1-SNP1			DTY4.1-SNP2			DTY4.1-SNP3
	Allele	RFU_1	RFU_2	Allele	RFU_1	RFU_2	Allele	RFU_1	RFU_2
IR64	Allele2	2648.43	2507.94	Allele2	2630.02	2586.76	Allele2	2630.83	2636.64
IR64-DTY2.2+4.1	Allele1	3161.85	2040.5	Allele1	3171.15	2062.87	Allele1	2771.05	2604.9
Addey sel	Allele1	3199.2	2051.37	Allele1	3232.52	2074.23	Allele1	3236.89	2063.25
NTC	No Call	2560.05	1859.41	No Call	2560.05	1859.41	No Call	2560.05	1859.41

RFU_1: FAM fluorescent values, RFU_2: HEX fluorescent values, NTC: non-template control.

Several qDTYs QTLs for drought stress tolerance have been identified for growth and yield performance under drought stress conditions and successfully were introgressed into high yielding rice varieties (Yadaw et al. 2013; Shamsudin et al. 2016). Among them, the integration of two specific QTLs, namely qDTY2.2 and qDTY4.1, into the widely used high-yield rice variety IR64 resulted in a yield increase of 0.8-1.0 t/ha under conditions of severe drought stress (Swamy et al. 2013). Thus, we developed seven KASP markers for qDTY2.2 (DTY2.2-SNP-1, DTY2.2- SNP-2, DTY2.2-SNP-3, DTY2.2-SNP-4) and qDTY4.1 (DTY4.1-SNP1, DTY4.1-SNP2, and DTY4.1-SNP3) and validated them with control panels. The validation of the KASP markers targeting qDTY2.2 and qDTY4.1 QTL revealed that the two tolerant varieties (IR64-DTY2.2+4.1 and Addeysel) showed a tolerant genotype and the sus-ceptible variety IR64 showed susceptible genotype for all seven KASP markers.

KASP markers were converted to Fluidigm markers

In order to establish a high-throughput genotypic screen-ing marker set for assessing the response to multiple abiotic stresses, we next proceeded with the conversion of the developed KASP markers into Fluidigm-compatible mar-kers. In addition to the KASP markers developed in this study, we included five more KASP markers associated with the Sub1 (two markers) and AG1 (three markers) QTLs which are major loci for submergence and anaerobic germination in rice, respectively. These KASP markers had been previously developed in our earlier studies (Moon et al. 2019; Lee et al. 2021). Therefore, a total of 17 KASP markers associated with five abiotic stresses were converted into Fluidigm markers as part of the process to establish a Fluidigm marker set optimized for multi-abiotic stress screening.

The converted Fludium markers were used to analyze genetic diversity in order to perform genotype profiling of 172 varieties consisting of 78 domestic and 94 abroad varieties. Most of the 17 Fluidigm markers were clearly separated into tolerant and susceptible alleles (Fig. 1), and the genotype profiling and genetic diversity of the 172 varieties was performed based on the genotypes determined by the Fluidigm markers (Supplementary Table S2, Fig. 2). The phylogenetic tree analysis of 172 varieties for the five abiotic stress traits revealed that these varieties clustered into four distinct groups when classified based on domestic and foreign origins (Fig. 2A). However, in Fig. 2B, the phylogenetic tree analysis for the Saltol QTL only indicated that the majority of varieties distinctly clustered into two groups corresponding to domestic and foreign categories. Further analysis of phylogenetic trees for single abiotic stress trait revealed that for DTYs and AG1, the majority of varieties were divided into two distinct clusters (Supplementary Fig. S1). However, in the case of Sub1, the varieties were divided into four groups (Supplementary Fig. S1C). These suggest that most of the abiotic stress related QTLs were differentiated between domestic and foreign regions.

Figure 1. Representative clustering patterns of genotyping. Colored plots represent the genotyping results from the Fludium markers for 172 domestic and abroad varieties. Each color code of the plots has the meaning of genotypes: homozygotes of Allele X (red; FAM fluorescence) and homozygotes of Allele Y (green; HEX fluorescence). The plots of gray color have meaning of no call genotype or NTC (non-template control). (A-Q) Genotyping result for the marker of Saltol-SNP-1 (A), Saltol-SNP-2 (B), Saltol-SNP-3 (C), Saltol-SNP-4 (D), Saltol-SNP-5 (E), DTY2-2-SNP-1 (F), DTY2-2-SNP-2 (G), DTY2-2-SNP-3 (H), DTY2-2-SNP-4 (I), DTY4-1-SNP-1 (J), DTY4- 1-SNP-2 (K), DTY4-1-SNP-3 (L), AG1-KP-SNP-1 (M), AG1-KP-SNP-2 (N), AG1-KP-SNP-3 (O), Sub1A-SNP1 (P), Sub1A-SNP2 (Q).

Figure 2. Phylogenetic tree constructed using the UPGMA method. (A, B) The phylogenetic tree of 172 domestic and abroad varieties using the developed Fludium markers for four abiotic stresses (A) and saline stress (B). Red solid circles represent foreign varieties and the others are domestic varieties.

Due to climate change, various abiotic stresses are occurring more frequently, which necessitates the develop-ment of rice varieties with multiple stress tolerance. For gene pyramiding, the application of SNP-based high- throughput DNA markers is more essential than traditional gel-based markers. Furthermore, rather than using indivi-dual markers for diverse genes, a high-throughput marker set capable of simultaneous screening is required to in-crease the efficiency. In this study, SNP-based KASP markers were developed for QTLs associated with tolerance to key abiotic stresses. These KASP markers were then converted into Fluidigm markers to create a high-throughput Fluidigm marker set for efficient screening. Additionally, the developed marker set was employed to profile the genotypes of abiotic stress QTLs in 172 domestic and inter-national varieties.

Gene pyramiding refers to the process of stacking various genes into a single genotype to combine desirable traits through conventional breeding or recombinant DNA technology (Lombardo et al. 2016). Gene pyramiding can be accomplished by multiple rounds of marker-assisted backcross (MAB). MAB is an approach for integrating two or more specific genes into elite cultivars, thereby im-proving the deficient trait. The application of both fore-ground and background selection in MAB ensures the reliability of selecting the plants with target allele, along with maximum recovery of recurrent parent genome (Singh et al. 2013). In the validation of the developed KASP markers with control varieties, most of the tolerance varie-ties harbored tolerant alleles while sensitive varieties had sensitive alleles of each gene. However, in the test of the KASP markers targeting Saltol QTL, the genotype of Pokkali, a salt stress-tolerant variety, was consistent with the genotype of IR64-Saltol for the only three markers (Saltol-SNP-2, Saltol-SNP-3, and Saltol-SNP-4), but not for the other two markers (Saltol-SNP-1 and Saltol-SNP-5). It is speculated that recombination occurred in this region during the IR64-Saltol line development process. Indeed, Saltol-QTL was selected from recombinant inbred lines derived from the cross of Pokkali and IR29 (Gregori et al. 1997). As both IR64-Saltol and Pokkali are tolerant to saline stress, the two markers (Saltol-SNP-1 and Saltol-SNP-5), which showed genotypic variation in Saltol region between IR64-Saltol and Pokkali, can be used as recombination markers. In our previous study, we developed five KASP markers for the AG1 QTL and screened a control variety panel with the markers. As a result, two markers exhibited genotype variation among tolerant varieties (Lee et al. 2021). These markers could also serve as recombination markers. Recombination markers can be utilized to mini-mize the donor genome segments and increase the recur-rent parent genome during gene introgression through backcrossing, thereby enhancing selection efficiency and reducing linkage drag (Frisch et al. 1999).

The phylogenetic analysis revealed that the majority of varieties distinctly grouped into two clusters, corresponding to domestic and foreign categories. This observation might be attributed to the fact that most tolerance genes for abiotic stresses were inherited from indica varieties (Bonilla et al. 2002, Venuprasad et al. 2009). In the genotype profile results, the allele composition of the 172 varieties exhibited distinct differences between indica and japonica (Supple-mentary Table S2). Interestingly, in the case of Sub1, the varieties were divided into four groups instead of two groups, indicating that the genotypes were not clearly divided according to the origin. Notably, there are many instances where the PCR reaction was labeled as ‘no call’. No call indicates a failure in PCR amplification, often due to the absence of the genomic segment in the specimen. These markers could be utilized as dominant markers capable of determining presence and absence of genomic segments. However, for them to be used as co-dominant markers, the development of the additional markers needs to be necessary. For AG1, as previously mentioned in other studies, it seems that the AG1 genomic segment appears to be present in most domestic varieties (Lee et al. 2021). The high-throughput Fluidigm marker set developed in this study would be used as a valuable tool for screening the presence of major abiotic stress genes in the future.

This work was supported by a research grant from Hankyong National University for an academic exchange program in 2020.