Bakanae disease is one of the emerging diseases of rice incited by Fusarium fujikuroi (Nirenberg, 1976) (teleomorph: Gibberella fujikuroi Sawada (Ito and Kimura, 1931), a filamentous hyphomycete fungus. The disease is prevalent in temperate and tropical rice-growing regions of the world (Amoah et al., 1995; Desjardins et al., 1997; Amatulli et al., 2010; Eğerci et al., 2021). Reported earlier as a minor disease, bakanae has emerged during the last decade as a serious problem in Basmati-growing areas in India, causing yield losses of up to 70% as well as impairing grain quality (Bashyal, 2018). The symptoms of the disease include seedling elongation, seedling mortality, reduced tillers, and elongated non-productive tillers (Bashyal et al., 2016). The disease is primarily seed borne, and secondary spread is through spores on plant parts and soil [Sun, 1975; Anderson and Webster, 2005]. Hot water treatment, and chemical and biological control measures are the major interventions to manage the disease; however, genetic resistance is considered effective (Hayasaka et al., 2001; Hossain et al., 2015; Lee et al., 2018; Sarwar et al., 2018), as it is economical, sustainable, and eco-friendly.

Identification of resistant sources of bakanae disease has found little progress so far due to the complexity of disease responses, which is associated with weather conditions and the pathogen complex known as the G. fujikuroi species complex (GFSC) (Desjardins et al., 2000; Wulff et al., 2010; Bashyal and Aggarwal, 2013; Jeon et al., 2013). GFSC is a complex of Fusarium species such as Fusarium andiyazi, Fusarium proliferatum, and Fusarium verticilloides, which are also associated with the disease, together with F. fujikuroi. This complexity poses challenges in screening rice genotypes under natural conditions. The development of a robust screening technique under artificial inoculation conditions for rice genotypes (Fiyaz et al., 2014) has, however, eased the identification of resistant genotypes. Using this method, several genotypes resistant to bakanae have been identified and successfully utilized in mapping quantitative trait loci (QTLs) governing resistance (Fiyaz et al., 2016). To date, as many as 12 QTLs from various chromosomes have been mapped to impart resistance either through linkage mapping using bi-parental mapping populations or by association mapping using diversity panels comprising germplasm from diverse ecogeographic regions (Yang et al., 2006; Hur et al., 2015; Fiyaz et al., 2016; Volante et al., 2017; Ji et al., 2018; Lee et al., 2018; Chen et al., 2019; Cheon et al., 2019; Kang et al., 2019; Lee et al., 2021; Lee et al., 2022). A maximum number of QTLs have reported on chromosome 1, which includes qB1, qBK1, qBK1.1, qBK1.2, qBK1.3, qFfR1, qBK1 ^wd, and qBK1^z (Yang et al., 2006; Hur et al., 2015; Fiyaz et al., 2016; Ji et al., 2018; Lee et al., 2018; Lee et al., 2021). Other QTLs reported are qBK3.1 on chromosome 3 (Fiyaz et al., 2016), qBK4_31750955 and qBK4^T on chromosome 4 (Volante et al., 2017; Lee et al., 2022), qBK_628091 on chromosome 6 (Volante et al., 2017), qFfrR9 on chromosome 9 (Kang et al., 2019), and qB10 on chromosome 10 (Yang et al., 2006). However, only one QTL qBK1 has been fine mapped so far, confined to a 35-kb genomic region, and flanked by InDel markers InDel 18 and InDel 19-14 (Lee et al., 2019). Based on gene expression studies, four putative candidate genes (LOC_Os01g41770, LOC_Os01g41780, LOC_Os01g41790, and LOC_Os01g41800) have been identified. The molecular mechanism underlying resistance to bakanae remains largely unknown. However, a transcriptome study has indicated the involvement of jasmonic acid biosynthesis pathway-related genes in the resistant genotype Selenio (Matić et al., 2016).

Since the reported QTLs have been proven effective only against specific isolates of F. fujikuroi prevalent in different regions, the wide utilization of the reported QTLs in breeding is limited. However, Lee et al. (2018) found that the pyramiding of the two QTLs qBK1 and qBK1^wd provided a better level of resistance as compared to individual QTL introgressed lines. Identification of novel QTLs and their deployment into elite cultivars is a proven strategy to broaden the spectrum of resistance and restrict the evolution of pathogens. Furthermore, validating the identified QTLs is of utmost importance prior to their utilization in breeding programs. One of the approaches for validation of a QTL is the development of QTL introgressed near-isogenic lines (QTL-NILs), which can also be used for fine mapping and identification of underlying candidate genes. Also, developing functional markers for the candidate gene bears significance in improving precision in marker-assisted breeding.

In a population generated from the cross between PB1121 and Pusa1342, a major QTL qBK1.2 was mapped on chromosome 1, which explained 24.74% phenotypic variance (Fiyaz et al., 2016), but remains to be fine mapped. This QTL is effective against the F. fujikuroi isolates prevalent in Basmati-growing regions of India. Therefore, this study reports the validation and fine mapping of qBK1.2 while developing tightly linked markers for their use in marker-assisted breeding.

Among the RILs generated from the cross between PB1121 and Pusa1342, RIL28 was chosen for the study because it carried qBK1.2 and showed a percent seedling mortality of 32% as compared to 95.33% in the susceptible genotype PB1121 under bakanae infection in the artificial screening conditions ( Figures 2A, B ). Furthermore, RIL28 exhibited 76.47% similarity to PB1121 based on the 119 SSR markers of which 28 were polymorphic ( Figure 3 ).

The F₁s from the cross made between PB1121 and RIL28 when subjected to the test of hybridity using the qBK1.2 linked markers, RM10153 and RM5336, showed one true hybrid out of 12 plants tested. A true F₁ was backcrossed with PB1121, 35 BC₁F₁s were generated with qBK1.2, and RPG recovery ranged from 78.57% and 82.14%. Out of these 35 BC₁F₁s, an individual plant with RPG recovery of 82.15% was backcrossed with PB1121 to generate 90 BC₂F₁s. The RPG recovery among the BC₂F₁ plants carrying qBK1.2 ranged from 85.71% to 96.42%. A BC₂F₁ plant with 96.42% RPG recovery was selfed to generate a BC₂F₂ population comprising 1,100 individuals. Genotyping with foreground markers followed by phenotyping using F. fujikuroi F250 isolate ( Figure 4 ) resulted in the identification of 24 BC₂F_2:3 recombinants falling in both resistant and susceptible classes.

The severity of bakanae disease has been increasing over the years in the Indo-Gangetic plains of India wherein Basmati rice is one of the major crops. The majority of the Basmati rice varieties show susceptibility to bakanae. Although genetic resistance to bakanae is reported by several studies, the sources of resistance lie outside the Basmati gene pool. Naturally, Basmati rice has a narrow genetic base, and hence, the deployment of resistance genes is sought from non-Basmati sources. This necessitates a long breeding process to safeguard the specific Basmati traits along with the recruited trait. Recently, it was established that marker-assisted backcross breeding can ease this cumbersome breeding process and accelerate varietal development with enormous success (Lee et al., 2018).

The QTL qBK1.2 identified on chromosome 1 lies over a span of 260 kb on the short arm flanked by the SSR markers, i.e., RM10153 and RM5336 (Fiyaz et al., 2016). In the current study, we used the approach of fine mapping qBK1.2 using QTL-NILs. NILs provide several advantages such as low background noise and genetic similarity to the sensitive parent. Because of these benefits, NILs are considered ideal for QTL validation, fine mapping, comparative genomics, marker development, gene expression studies, and even deployment as new cultivars. The BC₂F₂ population generated in the current study had an RPG recovery of 97.91%. Moreover, the F_2:3 population provided an opportunity for evaluating resistance reactions under the artificial screening system. The variation observed among the NILs was quantitative and could be due to environmental influence. The environmental influence of disease response makes the management of phenotyping cumbersome, expensive, and time-consuming.

The SNP genotyping among the recombinants delimited qBK1.2 to a region of 130 kb between the SNPs rs3164311 and rs3295562. It was found that when this region was absent in the S-NIL1 and S-NIL2, the disease intensity reached above 94% ( Figure 4 ). Since the other regions between the flanking markers were similar to the donor, the likelihood of the region being the candidate locus was very high, making us conclude that the QTL resided in this region. A similar strategy was used to fine map several QTLs in rice including qBK1 (Lee et al., 2019). The fine-mapped region comprised 11 annotated genes with putative disease resistance function. The expression pattern of a gene under mock and challenged inoculation conditions provides significant contrast in its involvement in response under infection (Matić et al., 2016; Cheng et al., 2020). Genes LOC_Os01g41770 and LOC_Os01g41780 were annotated to encode leucine-rich repeat motifs, which were identified to negatively regulate bakanae disease resistance (Lee et al., 2019). Eight of 11 genes that were subjected to expression profiling between the parental lines PB1121 and Pusa1342 showed significant differential expression. The observed differences may be attributed to the response to infection dynamics and/or due to inherent differences in the genomes of these genotypes. Therefore, comparing the expression between the NILs could help identify the genes responding to the infection. Accordingly, the R-NIL4 carrying qBK1.2 in the genetic background of PB1121 was profiled along with the parental lines. The expression pattern ruled out the involvement of five loci, i.e., LOC_Os01g06720, LOC_Os01g06730, LOC_Os01g06760, LOC_Os01g06876, LOC_Os01g06900, and LOC_Os01g06790, in the manifestation of resistance to bakanae because they failed to show a differential response between qBK1.2 carriers (R-NIL4 and Pusa1342) and the non-carrier (PB1121). Two genes, that is, LOC_Os01g06750 and LOC_Os01g06870, have shown significant differential expression between susceptible (PB1121) and resistant individuals (Pusa1342 and R-NIL4). A comparative analysis of encoded protein using BLASTp showed that these genes are orthologues of Verticillium wilt (Verticillium dahliae) resistance gene of tomato, i.e., Ve1, which is rich in leucine-rich repeats. Ve1 gene of tomato was identified to provide resistance in tomato against race 1 strains of V. dahliae and Verticillium albo-atrum (De Jonge et al., 2012). It was also shown that Ve1 requires the cascade of signaling genes, such as Enhanced Disease Susceptibility 1 (EDS1), Non-race-specific Disease Resistance 1 (NDR1), NB-LRR Protein Required for HR-Associated Cell Death 1 (NRC1), ACIF, MAPK/ERK kinase 2 (MEK2), and SERK3/BAK1 (Fradin et al., 2009). The other gene, i.e., Ve2, has been shown to act antagonistically to Ve1 using RNAi, and it was found that Ve1 regulates the expression of defense-related genes by minimizing the effects of Ve2 on these genes (Nazar et al., 2018).

LOC_Os01g06750 is an interrupted gene with four introns, while LOC_Os01g06870 is an uninterrupted gene. The gene sequence variation among the genotypes determines the target phenotype. Although SNPs were identified in LOC_Os01g06750 between PB1121 and Pusa1342, the translated product revealed minor substitution of amino acids with no major effect on functional motifs, i.e., leucine-rich repeat having a role in the resistance-related functions (McHale et al., 2006). This indicates its function in providing tolerance to bakanae at the transcriptional or posttranscriptional level or by generating protein products through alternate splicing of the exons. However, the translated product of LOC_Os01g06870 had 21 amino acid substitutions, which led to the creation of the LRR domain in the resistant individual Pusa1342. The differential expression of LOC_Os01g06870 and functional difference in gene product make it the potential candidate gene in providing tolerance against bakanae disease and can be conclusively proved with functional validation by generating transgenic events in the susceptible genotype.

We conclude that LOC_Os01g06870 is a potential putative candidate gene conferring tolerance to bakanae disease, underlying the major QTL qBK1.2 between the SNPs rs3164311 and rs3295562. These markers are resources of immense importance for a rice breeding program targeting to introgress qBK1.2 for developing bakanae resistance in rice.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

AK: Data curation, Formal Analysis, Investigation, Validation, Visualization, Writing – original draft. RE: Formal Analysis, Methodology, Writing – review & editing. GK: Conceptualization, Funding acquisition, Project administration, Supervision, Writing – review & editing. SM: Data curation, Writing – review & editing. BB: Resources, Writing – review & editing. PB: Writing – review & editing. HB: Writing – review & editing. KV: Writing – review & editing. NS: Resources, Writing – review & editing. AS: Conceptualization, Project administration, Supervision, Writing – review & editing.