Ranbir Basmati possessing strong aroma and excellent cooking quality was used as a recurrent parent while a PAU148 possessing BB (xa13 and Xa21) resistant genes along with semi-dwarf (sd1) gene was used as a donor parent. The marker-assisted backcross breeding (MABB) scheme was used to pyramid semi-dwarf and BB resistance genes in Ranbir Basmati. The individual F₁ plants produced by crossing Ranbir Basmati/PAU148 and confirmed by gene-specific markers were backcrossed with recurrent parent to generate the backcross population. The backcross method was followed up to BC₂F₂ generation and at each generation, foreground selection as well as background selection was performed to select positive plants carrying desirable genes. The schematic diagram for pyramiding of semi-dwarf and BB resistance genes in the recurrent parent is depicted in Fig. 1. Further, before using the donor parent PAU148 for introgression of BB resistance and semi-dwarf genes in Ranbir Basmati, validation of the molecular markers linked to these genes was carried out in the selected donor parent. The details of the molecular markers (xa13, Xa21 and sd1) are presented in Table 1.

Total genomic DNA was extracted from rice leaves following the modified protocol of Sahu et al. [16] and quantified spectrophotometrically. The PCR reaction mixture for foreground selection of xa13 and Xa21 contained 50 ng genomic DNA template, 10 pmoL of each of the primers, 5 mM dNTP's, 10X PCR buffer, 1 U taq polymerase in a volume of 10 μL. For sd1 gene, the PCR reaction mixture contained 5X Q-solution, 10 mM dNTP's, 2.5 U Hot star taq polymerse. The PCR profile for all the genes studied was different with respect of temperature and timing and is provided in Supplementary Table 1. The PCR amplified products were separated by electrophoresis on a 2% agarose gel and visualized on gel documentation system (Bio-Rad Laboratories Inc., USA).

A parental polymorphism survey was conducted with the recipient parent (Ranbir Basmati) and donor parent (PAU148), and a total of 384 SSR markers selected from www.gramene.org were screened for their polymorphism. The SSR markers were selected from each short and long arm of rice chromosome 12. Out of 384 SSR markers, 51 polymorphic markers uniformly spanning across the genome were selected (Supplementary Table 2). These polymorphic SSR markers were used for background selection in order to select plants having maximum recovery of the recurrent parent genome. Graphical GenoTypes (GGT) version 2.0 [17] software was used for the assessment of the genomic contribution of the parent in the selected genotypes based on SSR marker data. Further, polymorphic SSR primers were resolved on 3.5% agarose gel and visualized.

The BC₂F₂ introgressed lines carrying xa13 and Xa21 genes both either single or in combination were selected and evaluated for BB resistance. Leaf clip inoculation method was used for artificial inoculation of pyramided lines with BB disease [18]. The inoculum was prepared by suspending the bacterial mass in distilled water to a concentration of 10⁶ cells/mL. The inoculation was carried out by clipping the tip of the leaf at the tillering stage with scissors that had been dipped into the inoculum. The symptoms become visible after five to six days after inoculation and scoring was done using IRRI-SES scale [19] after 15 days of inoculation (Supplementary Table 3).

The pyramided lines along with parents (Ranbir Basmati and PAU148) were evaluated at Experimental Research Farm, Sher-e-Kashmir University of Agricultural Sciences and Technology of Jammu (SKUAST-J) during kharif season in 2016. Twenty-five day-old seedlings of selected pyramided lines were transplanted with spacing of 15 × 20 cm in a randomized complete block design (RCBD) with two replications. Data were collected from five randomly selected plants from each entry in each replication for agro-morphological and grain quality characters including plant height, days to 50% flowering, days to maturity, effective tillers per plant, panicle length, 1000 grain weight, yield per plant, grain length, grain breadth, L/B ratio, kernel length after cooking (KLAC), kernel breadth after cooking (KLBC) and aroma. The scale used for aroma test was 0, 1, 2, 3 for no, mild, strong, and very strong aroma, respectively [20].

The SSR–PCR bands were examined under ultra violet transilluminator and photographed under gel documentation unit (Bio-Rad Laboratories Inc., USA). The SSR bands were counted and scored manually as 1 for their presence and 0 for their absence to generate the binary data for diversity analysis. The molecular data was analysed using NTSYS-PC (Numerical Taxonomy and Multivariate Analysis System) computer package [21]. The genetic similarity between accessions was calculated by Dice (SSR) and Jaccard's similarity coefficient. The dendogram was constructed based on sequential UPGMA (unweighted pair group method with arithmetic mean) using software package NTSYS PC 2.11 to infer genetic relationships and phylogeny.

The donor parent (PAU148) was validated for the presence of xa13, Xa21 and sd1 genes using gene-specific markers xa13 promo, pTA248 and h, respectively. The amplified fragment with respect to xa13 promo in resistant and susceptible lines was 500 bp and 250 bp [7]. With primer pair pTA248, it was 1000 bp and 650 bp [8] and, with respect to primer pair ‘h’, it was 348 bp and 731 bp, respectively [22] (Fig. 2). The results revealed that all the markers used for xa13, Xa21 and sd1 genes were able to distinguish resistant lines from susceptible as well as from dwarf to tall ones.

Parental polymorphism survey was carried out using 384 SSR markers distributed across the 12 chromosomes of rice. Out of 384 SSR markers, 51 (13.28%) markers were found to be polymorphic between Ranbir Basmati and PAU148, while 333 (86.71%) markers were monomorphic. The 51 polymorphic SSR markers were used for background selection at each generation. The number of alleles varied from 1 to 2. The Polymorphism Information Content (PIC) value was calculated by using the formulae given by Roldan-Ruiz et al. [23]. The highest PIC value was shown in case of primer RM18 and RM22 (0.48), while the lowest PIC value was observed in RM317, RM499, RM3625, and RM527 (0.07) (Supplementary Table 2).

Marker-assisted backcross breeding (MABB) method was used to introgress semi-dwarf and BB resistance genes in Ranbir Basmati. During each generation from F₁ to BC₂F₂, foreground selection was carried out and plants having resistance alleles of all the three genes (xa13, Xa21 and sd1) were selected and advanced to the next generation. F₁ plants were also tested for the hybridity and true F_1S were backcrossed with recipient parent to get BC₁F₁ generation seeds. In BC₁F₁, a total of 325 plants were grown and screened for the presence of BB resistance and semi-dwarf genes using molecular marker xa13-promo, pTA248, and ‘h’. Out of 325 BC₁F₁ plants, 36 plants were found to be positive for xa13 + sd1 genes, 13 for xa13 + Xa21 and 23 for Xa21 + sd1 genes. Totals of 26, 60 and 82 plants showed positive for xa13, Xa21 and sd1 genes, respectively. A total of 17 plants were found to be positive for all three genes, viz. xa13, Xa21 and sd1.

In subsequent BC₂F₁ generation out of 150 plants, 5 plants showed positive for all three genes (xa13, Xa21 and sd1). Thirty-five plants were positive for xa13 + sd1, 17 for Xa21 + sd1 genes and 16 for xa13 + Xa21 genes. Similarly, 29 plants were positive for xa13, 14 for Xa21 and 37 for sd1.

Furthermore, in BC₂F₂ generation, out of 121 plants, 19 plants were found to be positive for all three genes (xa13, Xa21 and sd1), while 34 plants were found to be positive for xa13 + sd1 genes, 25 for Xa21 + sd1 genes and 23 for Xa21 + xa13 genes. For Xa21, xa13 and sd1 genes 34, 44 and 93 plants, respectively, were found to be positive (Fig. 3). Background selection was started from BC₁F₁ and continued up to BC₂F₂ generation. In each generation, plants showing maximum recovery of recurrent parent were advanced to the next generation. Further, the background selection of 19 introgressed line (Fig. 4) resulted in the highest recurrent parent genome (RPG) recovery in introgressed line SBTIL121 (86.9%) (Fig. 5). The superior BC₂F₂ plant was further advanced to the next generation.

The NILs along with parents were screened for BB resistance. The Xoo strain, which was most prevalent in the area, was used as an inoculum. The donor parent PAU148 showed lesion length between 1–5% with disease scoring scale value ‘1′, while the recurrent parent Ranbir Basmati showed average lesion length between 51–100% with disease score ‘7′. The pyramided lines showed very small lesion length (disease score 1) and were highly resistant to Xoo isolates. The introgressed line SBTIL121 having the maximum recovery of recurrent genome and possessing the superior grain quality displayed resistance to BB with less lesion length (Fig. 6). The plants with a score of 1 were considered as resistant, 3 as moderately resistant, while those with 7–9 were considered as susceptible. The IRRI standard evaluation system (IRRI–SES scale) was adopted for the screening of the introgressed lines [19]. Twenty-three introgressed lines carrying two genes together in homozygous condition (xa13xa13 Xa21Xa21) showed lesion lengths between 1 and 5% with disease score ‘1′, while it is significant to note that single BB-resistant genes in homozygous condition (i.e. either xa13xa13 or Xa21Xa21) showed lesion lengths between 6 and 12% with disease score ‘3′. The single BB-resistance gene showed partial level of BB resistance (Table 2). Thus, all the lines having xa13 + Xa21 gene combination showed a higher level of resistance, while lines showing either xa13 or Xa21 alone showed moderate levels of resistance against Xoo isolates.

The BC₂F₂ generation was grown in the field and evaluated for various agro-morphological and grain quality traits. The pyramided lines were analysed to estimate the magnitude of the genetic variability for different morpho-physiological and grain and cooking quality. Significant variation was observed for all the traits viz. plant height, days to 50% flowering, days to maturity, effective tillers per plant, 1000 grain weight, panicle length, yield per plant, grain length, grain breadth, and L/B ratio. All the traits were at par with the recurrent parent. The introgressed line SBTIL80 showed maximum yield per plant of 67.3 grams with a maximum number of effective tillers, panicle length, and 1000 grain weight, which are considered to be yield-contributing traits. Similarly, introgressed lines SBTIL39, SBTIL25, SBTIL121, and SBTIL31 also showed yield per plant more than recurrent parent ranging from 42.4 g to 58.3 g. These observations allow us to conclude that agro-morphological traits play an important role in the plant's yield. The plant height of all the introgressed lines ranged from 130.9 cm to 137.5 cm and the minimum plant height was recorded in introgressed line SBTIL37. Most of the pyramided lines have 1000 grain weight higher than the recurrent parent (Table 3).

The BC₂F₂ introgressed lines possessing all the target genes were also analysed for grain and cooking quality parameters. The KLBC for Ranbir Basmati and PAU148 was observed to be 7.13 mm and 8.16 mm respectively, while KBBC was 1.56 mm for Ranbir Basmati and 1.57 mm for PAU148. After cooking, KLAC values for Ranbir Basmati and PAU148 were 11.52 mm and 14.15 mm, respectively, while those for KBAC were 2.25 mm and 2.50 mm for Ranbir Basmati and PAU148, respectively. Interestingly, the KLAC and KBAC in introgressed line SBTIL121 were 13.98 mm and 2.48 mm, respectively, which is more than 2.46 mm in length and 0.23 mm in breadth than the recurrent parent (Fig. 7). The KLAC and KBAC were also measured in other introgressed lines that are depicted in Table 4.

The cluster analysis of BC₂F₂ population was also done using molecular data, and the dendogram was formed using software NTSYS version 2.10. Cluster I consists of Ranbir Basmati and all the 19 pyramided lines that were sub grouped into cluster I-A, which contains Ranbir Basmati and SBTIL21, while cluster I-B is further divided into two sub-groups, cluster I-Ba and cluster II-Bb. Cluster I-Ba contains introgressed lines SBTIL25, SBTIL30, SBTIL29, SBTIL27, SBTIL31, SBTIL36, SBTIL37, SBTIL38, SBTIL39, SBTIL42, SBTIL49, SBTIL80, SBTIL81, SBTIL82, SBTIL121, and SBTIL85, while cluster I-Bb consists of SBTIL33 and SBTIL34. Cluster II contains PAU148, which is a donor parent (Fig. 8). The results revealed that all the introgressed lines are similar to those of Ranbir Basmati, a recurrent parent, and are grouped in same cluster.