The upland BL nurseries were set up in the wet season at three rice research centers: Ubon Ratchathani Rice Research Center (UBN-RRC) in 2016 and 2018, Sakon Nakhon Rice Research Center (SKN-RRC) in 2016 and 2018, and Phrae Rice Research Center (PRE-RRC) in 2016 (Table 2, Figure 1). Seeds for 286 and 288 RILs from KD-IR71033 and KD-IR57514, respectively, were planted in single-row plots, 60 cm in length, in BL nursery beds. Every five entries were separated by a susceptible check (KDML105 or KTH17) and resistance check (SPR1 or HY71). KDML105 was planted alongside the nursery bed in spreader rows to enhance the spore transmission. Plots and spreader rows were inoculated with a local natural mix of strains. Standard Evaluation System (SES) ratings for BL [24] were used when all of the susceptible control rows died.

DS based on damage score, LL based on lesion leaf length.

Xoo isolates collected from the Ubon Ratchathani (UBN), Sakon Nakhon (SKN), Udon Thani (UDN), and Phrae (PRE) provinces from 2016 to 2018 were used for BB resistance evaluation at the UBN-RRC, SKN-RRC, and RRE-RRC in the wet season (Figure 1). The isolate was grown in nutrient agar media for 72 hours at 28˚C. The bacterial cells were suspended in sterile water and adjusted to ~108 CFU/ml. Xoo isolates were assayed for a resistance reaction in RIL populations, susceptible (RD10, PSL2) and resistance (RD7, IRBB5, IRBB7) controls. BB inoculation was performed in a greenhouse using the leaf-clipping method [25] . Forty-five days after sowing, two to three fully expanded leaves of each RIL plant were inoculated. Resistance reactions were recorded based on the mean of the percent of the leaf area that appeared to be diseased [24] and the lesion length [26] of an individual plant 12 - 14 days after inoculation. The percentage of affected leaf area was used for SKN2016, UDN2016, and UBN2017, while leaf lesion length was used for UBN2016, SKN2018, and UDN2018 (Table 2). A damage score between 0 and 3 and a lesion length shorter than 5 cm were regarded as resistant to BB.

RILs from the KD-IR71033 population were evaluated for WBPH resistance at the seedling stage and BPH resistance at the tilling stage. A WBPH population was collected from a rice field in the Phitsanulok (PSL) province in 2017 (Figure 1). The WBPH population was maintained on TN1 in a greenhouse at the Phitsanulok Rice Research Center (PSL-RRC). The seedbox screening test [27] was conducted to evaluate WBPH resistance in the test lines. Ten pre-germinated seeds of each test entry, susceptible (TN1) and resistance (PTB33) controls were sown in seedboxes (60 cm × 40 cm × 10 cm) containing well-puddled soil in the 12-cm rows. Seven days after sowing, the seedlings were infested with second and third instar nymphs of WBPH at ten nymphs per seedling. When all seedlings of TN1 were almost dead, the SES scoring [24] was used to assign a resistance score to each genotype. A BPH population from a single colony was collected in the outbreak field from the UBN province in 2015 (Figure 1) and was grown on a susceptible variety of TN1 in a temperature-controlled rearing room (25˚C ± 2˚C) at the UBN-RRC. The screening method, which was modified from Jairin et al. [28] , was conducted in a greenhouse at the UBN-RRC. The seeds of each RIL progeny, a susceptible cultivar (TN1), and a resistant cultivar (Rathu Heenati) were separately sown (10 × 20 cm) in 7 × 24 m² seedling plots. Twenty days after sowing, the seedlings were infested with 3^rd-4^th instar nymphs of BPH at ten nymphs per seedling. Then, the insects fed, mated, laid eggs, and hatched freely. We evaluated the severity scores of the test lines according to SES [24] until TN1 and the susceptible recurrent parents died.

Samples included the parental and the two mapping populations, 286 RILs from KD-IR71033 and 288 RILs from KD-IR57514. Genomic DNA was extracted from the young leaves of each sample using the standard cetyltrimethylammonium bromide (CTAB) method [29] and quantified with a Nanodrop-1000 spectrophotometer (Thermo Fisher Scientific).

A genotyping by sequencing (GBS) protocol using the ApeKI enzyme was applied to prepare the reduced representation libraries for sequencing. The forward adapters contained 9-bp unique barcodes in addition to the 21-bp Ion Forward adapter. The ApeKI restriction site was used to enable multiplex sequencing of the libraries. Genomic DNA digestion and adapter ligation were performed as described in Mascher et al. [30] . DNA fragments of 250 - 300 bp were selected using E-Gel™ SizeSelect™ Agarose Gels (Invitrogen). The libraries were sequenced on the Ion S5™ XL Sequencer (Thermo Fisher Scientific) with a loaded Ion 540™ Chip according to the manufacturer’s protocol. We multiplexed between 24 samples per run. All sequencing data generated in the present study were analyzed using Ion Torrent™ Suite Software Alignment Plugin V5.2.2 (Thermo Fisher Scientific) with the Nipponbare genome [31] as the alignment reference. The variants were called using the Torrent Variant Caller (Thermo Fisher Scientific). Filtered vcf files were imported into TASSEL 5 [32] . The default parameters of TASSEL were used to detect informative SNPs. SNPs were named according to their scaffold and base pair position within the Nipponbare genome.

Overall, 286 and 288 RILs from the KD-IR71033 and KD-IR57514 populations, respectively, were screened for BL resistance in the uniform BL nursery with natural infection at UBN-RRC, SKN-RRC, and PRE-RRC. IR71033 and IR57514 were resistant to BL, while KDML105 was highly susceptible in all of the nurseries. This indicated that IR71033 and IR57514 might be carrying broad field resistance against BL. The frequency distributions of the BL damage score for both RIL populations according to location and year are shown in Figure 2. The segregation of resistant and susceptible individuals in the KD-IR71033 population fitted a ratio of 1:3 (UBN2016 χ² = 1.45, P = 0.23; UBN2018 χ² = 3.53, P = 0.06; SKN2016 χ² = 3.08, P = 0.08; PRE2016 χ² = 1.29, P = 0.26), while in the KD-IR57514 population fitted a ratio of 3:1 (SKN2018 χ² = 3.17, P = 0.07; UBN2018 χ² = 3.68, P = 0.06).

Two scoring methods were used to quantify BB resistance. In 2017 at UBN-RRC, an SES-based damage score was used to evaluate the resistance of the rice lines, while in 2016 and 2018 at UBN-RRC and SKN-RRC, the lesion length on the fully expanded leaves was used (Table 2). IR71033 and IR57514 were resistant to BB, while KDML105 was highly susceptible to all strains in all locations. The frequency distributions of BB damage scores and the lesion lengths for both RIL populations according to the location are presented in Figure 2. The segregation of resistant and susceptible individuals in the RIL populations showed a good fit to the expected ratio of 1:3 (KD-IR71033 population: UBN2016 χ² = 0.27, P = 0.60; UBN2017 χ² = 0.04, P = 0.84; SKN2018 χ² = 0.72, P = 0.39; UDN2016 χ² = 2.14, P = 0.14; UDN2018 χ² = 0.44, P = 0.50, KD-IR57514 population: UDN2018 χ² = 0.17, P = 0.68; SKN2018 χ² < 0.01, P = 0.97). The segregation pattern indicated that BB resistance derived from IR71033 and IR57514 was controlled by a single recessive gene.

The donor parent plant IR71033 displayed moderate resistance to WBPH and BPH, while KDML105 was completely susceptible to both insect pests. The level of resistance against BPH of RIL plants could not detected at seedling stage using standard seedbox screening test. Therefore, we decided to screen all RILs at the tillering stage for BPH resistance. The distribution of the damage scores for the progeny of KD-IR71033 was skewed towards susceptibility (Figure 2). The segregation of resistant to susceptible plants was in agreement to 1:3 segregation for WBPH at the seedling stage (χ² = 0.56, P = 0.45) and BPH at the vegetative stage (χ² < 0.01, P = 0.97). The segregation pattern indicated that planthopper resistance derived from the donor IR71033 was controlled by a single recessive gene.

A total of 1897 polymorphic SNPs were used to construct the linkage map for KD-IR71033 and spanned a total genetic distance of 1899.7 cM with linkage groups ranging from 69.9 cM (chromosome 12) to 204.2 cM (chromosome 11). The number of markers mapped to each chromosome varied from 108 (chromosome 9) to 223 (chromosome 1). An average of one SNP per cM region was detected across the genome with three large gaps on chromosomes 9, 10, and 11 (Figure 3).

We used QTL IciMapping 4.1 to construct the linkage map for KD-IR57514 utilizing a total of 2237 polymorphic SNPs that spanned a total genetic distance of 2370.4 cM on all of the 12 chromosomes. The number of markers mapped to each chromosome varied from 64 (chromosome 7) to 322 (chromosome 4). The genetic length of the smallest linkage group was 120.3 cM on chromosome 9, and the largest was 311.3 cM on chromosome 2 (Figure 3).

The linkage maps for KD-IR71033 and KD-IR57514 and the phenotypic trait evaluation data were used to map the QTLs associated with BL resistance. MQM mapping analysis detected QTLs for BL resistance in the same genomic regions for all of the three environments (UBN-RRC, SKN-RRC, and PRE-RRC) from 2016 to 2018. At UBN-RRC, QTL-mapping analysis revealed three QTLs associated with BL resistance on chromosomes 9, 11, and 12 of KD-IR71033 (in 2016 and 2018) and KD-IR57514 (in 2018) mapping populations (Table 3). The percent of PVE for BL from these QTL ranged from 7.4% to 59.3% (Table 3). The major QTL from KD-IR71033 was located on chromosome 12 and flanked by SNP markers S12_8036165 and S12_10973592 in 2016 and S12_10254600 and S12_10973592 in 2018 with PVEs of 31.9% and 46.3%, respectively. At SKN-RRC, the only major QTL was detected on chromosome 12 between S12_10060136 and S12_14061647 in 2016 for KD-IR71033 with a PVE of 19.7%. In 2018, three QTLs were detected on chromosomes 9, 11, and 12 from KD-IR57514 at SKN-RRC with 6.1%, 7.3%, and 36.7% PVE, respectively. From PRE-RRC in 2016, we identified three QTLs on chromosomes 9, 11, and 12 in KD-IR71033 with 19.3%, 7.3%, and 47.4% PVE, respectively. Fewer QTLs were observed for BL at SKN-RRC than at other nurseries. All QTLs showed positive additive effect values, indicating that the alleles that confer resistance to BL are those from IR71033 and IR57514 (Table 3) and the major QTLs (qBL1-12^UBN16, qBL1-12^UBN18, qBL1-12^SKN16, qBL1-12^PRE16, qBL2-12^UBN18, qBL2-12^SKN18) were mapped in the same Pi-ta region on chromosome 12.

The MQM mapping analysis detected major QTLs for BB resistance on chromosome 11 of both KD-IR71033 (qBB1-11^UDN16, qBB1-11^UBN16, qBB1-11^UBN17, qBB1-11^SKN16) and KD-IR57514 (qBB1-11^UDN18, qBB1-11^SKN18) with LODs ranging from 8.5 to 52.2 that explained 13.5% to 57.3% of the trait variations (Table 3). The QTL detected from KD-IR71033 was flanked with SNP markers S11_28162887 and S11_28977529 and from KD-IR57514 with S11_27446665 and S11_28704532. The QTLs detected from KD-IR71033 explained 43.9%, 54.3%, 40.7%, and 13.5% of the PVE of qBB1-11^UDN16, qBB1-11^UBN16, qBB1-11^UBN17, and qBB1-11^SKN16, respectively. The QTL detected from KD-IR57514 in 2018 explained 28.2% and 41.7% of the PVE of qBB1-11^UDN18 and qBB1-11^SKN18, respectively (Table 3). The alleles that confer resistance to BB were from IR71033 and IR57514. The major QTL region was shared with the cluster region of Xa4 and Xa26.

WBPH and BPH resistance evaluations were conducted in a greenhouse at PSL-RRC and UBN-RRC in 2018 and 2016, respectively. IR71033 was moderately resistant to the planthoppers, while IR57514 and KDML105 were completely susceptible. Therefore, only the KD-IR71033 mapping population was used to screen and map WBPH and BPH resistance loci. From QTL mapping, two QTLs (qWBPH1-6^PSL18 and qBPH1-6^UBN16) associated with WBPH and BPH resistance, respectively, were identified. The major QTL for WBPH and BPH were located between the SNP markers S6_841740 and S6_1394987 on the short arm of chromosome 6 and explained 19.7% and 61.8% of the phenotypic variation, respectively (Table 3 and Figure 3). The major QTL region was found to contain the published BPH-resistance gene Bph32.

For BL resistance, the QTLs qBL1-9 and qBL2-9 on chromosome 9 from IR71033 and IR57514 were detected in the regions between SNP markers S9_8172820 and S9_9911937. There are 93 annotated genes within this region in the Nipponbare genome including three BL resistance genes (Pi3, Pi5, and Pi15). The qBL1-11 and qBL2-11 regions between the S11_25260851 and S11_27584310 markers on chromosome 11 contain 85 predicted genes, which include the Pik and Pi44 genes. The major QTLs (qBL1-12^UBN16, qBL1-12^UBN18, qBL1-12^SKN16, qBL1-12^PRE16, qBL2-12^UBN18, qBL2-12^SKN18) were in the Pi-ta region. The 718.9-kb QTL region contains 44 predicted genes.

For BB resistance, the qBB1-11 and qBB2-11 regions contain 85 predicted genes including the Xa3/Xa26, Xa4, Xa40, and xa44 genes. The QTLs from IR71033 qWBPH1-6 and qBPH1-6 associated with WBPH and BPH resistance, respectively, were detected in the Bph32 genomic region between S6_1045301 and S6_1394987 on chromosome 6. This 349.7-kb region of the Nipponbare genome contains 39 predicted genes.