Identication and Introgression of Bacterial Blight Resistance From African Rice (Oryza Glaberrima Steud.)

Bacterial blight (BB) caused by Xanthomonas oryzae pv. oryzae (Xoo) is one of the major diseases affecting rice production. In the present study, 31 accessions of O. glaberrima were screened for BB resistance consecutively for four seasons and identied 28 stable BB resistant accessions with a mean lesion length of < 3cm. Molecular characterization revealed absence of xa5, xa13, Xa21, Xa38, xa41(t) and xa45, indicating the possibility of presence of novel gene(s) conferring BB resistance. In order to transfer BB resistance from O. glaberrima, resistant accession EC861812 was crossed to the susceptible cultivar, IR64 and BC 2 F 2 population developed. Genetic studies in BC 2 F 2 population showed transgressive segregation which indicates that O. glaberrima also harbor benecial alleles for yield related traits. Six resistant BC 2 F 2 plants with mean lesion length of 0.1-0.8 cm were genotyped with 70 parental polymorphic SSR markers which revealed recurrent parent genome 65.1 to 84.5% with average of 67.5%. The BB resistant introgression lines identied in the present study will serve as pre-breeding lines for mapping of BB resistance derived from O. glaberrima. x 2 m 2 plots) in the experimental plots of ICAR-IIRR, Hyderabad following a spacing of 15 x 20 cm. Data were recorded for eight yield attributing traits viz., days to 50% owering, plant height, number of tillers per plant, number of productive tillers per plant, panicle length, number of spikelets per plant and spikelet fertility.


Introduction
Rice (Oryza sativa L) (2n = 2x = 24), is the staple food crop that belongs to grass family, Poaceae and is grown in many countries in Asia, Africa and Latin America. It is a major source of calories and is the principal food source for more than half of the world's population (Khush, 2005). The global demand for rice is expected to be 650 million tonnes by 2050, principally due to the rise in demand in Asian countries (Chukwu et al., 2019). Rice production is adversely affected due to several biotic stresses, including bacterial blight (BB) disease caused by Xanthomonas oryzae pv. oryzae (Xoo). Yield loss due to BB is generally in the range of 20-30% across the years. However, under favorable environmental conditions, it can reach up to 80% (Kim, 2018). Developing resistant cultivar is generally regarded as the most effective and economical means of controlling this dreaded disease. Presently at least, 46 genes conferring resistance against various races of Xoo have been identi ed in both cultivated rice and wild relatives of rice (Chukwu et al., 2019;Neelam et al., 2019;Chen et al., 2020). Because of the rapid changes in the virulence pattern of Xoo population resulting in emergence of new races, there are reports of resistance breakdown conferred by major resistance genes (Shanti et al. 2001). Hence, it is imperative to identify novel genes, which are effective against virulent isolates of the pathogen in order to cope with the dynamic and rapidly evolving pathogen population.
Genus Oryza consists of two cultivated species (O. sativa L. and O. glaberrima steud.) and 22 wild species, which are known to harbor several BB resistance genes. There are at least 12 BB resistance genes have been identi ed in different species of Oryza, namely Xa21 from O. longistaminata (Khush, 1989; Ronald et al., 1992) (Neelam et al., 2019). Based on the above information, it is clear that different species of Oryza are useful resources that can be explored for identi cation of new sources of BB resistance in rice. . The test entries along with recurrent parent and resistant check were raised in mud pots following the required care and fertilizer applications to raise healthy plants. The Xoo strain, IX-020 was grown in modi ed Wakimoto's culture medium (Laha et al., 2009) and using a 3-day old culture, bacterial suspension (10 8 cfu/ml) was prepared and used for inoculation. The plants at maximum tillering stage were inoculated by clipping top 2-3 cm of completely developed leaves with sterilized scissors dipped in the bacterial suspension (Kauffman et al., 1973). Disease reactions were recorded following Standard Evaluation System for Rice (IRRI, 2014) and by measuring the lesions length caused by BB on each inoculated leaf, 14 days after inoculation. The mean lesion length (cm) for each plant was calculated based on lesion lengths of ve inoculated leaves and used for assessing resistance or susceptibility of individual plants. Based on mean lesion lengths, accessions were classi ed as resistant (≤3 cm), moderate resistant (3-6 cm), moderate susceptible (6-9 cm) and susceptible (≥9 cm) as per Chen Table 1). Plants were inoculated with bacterial blight pathogen culture (strain IX-020) and were scored at 14 days after inoculation (DAI) Fig 1(A). In the rst experiment, the reactions of the genotypes were scored based on SES scale and in the subsequent experiments, the reactions were recorded by measuring the lesion length. Based on observations of multiple screenings, out of 31 accessions, 28 accessions showed high level of resistance with lesion length less than 3 cm. The average lesion length among the resistant entries ranged from 0.3 cm (EC # 861795, 861810) to 1.8 cm (EC861811). Two accessions, viz., EC861819 and EC861820 showed a moderate resistance reaction with average lesion length ranging from 3.9-4 cm while EC861804 showed extreme susceptibility with an average lesion length > 30 cm. The cultivar, IR64, exhibited a score of 7 with a lesion length ranging from  Figure 2).

Development and identi cation of BB resistant introgression lines
Among the 31 resistant accessions of O. glaberrima, EC861812 consistently showed high level of BB resistance with lesion length of 0.4-1.0 cm across different seasons; hence it was selected as the donor parent ( Table 2). The bacterial blight susceptible cultivar, IR64 (used as the female parent and recurrent parent) was crossed to O. glaberrima accession (EC861812) during wet season, 2016 to produce interspeci c F 1 seeds. The seed set was very low (less than 20%), as the F 1 s were products of inter-speci c hybridization. The F 1 plants (nine F 1 plants) were grown and inoculated with the Xoo strain IX-020 during wet season of 2017. The mean lesion length of the F 1 plants was below 4 cm (Table 3 and Figure 1B).
The F 1 s plants were found to be partially/completely pollen sterile and there was very low seed set on sel ng. Hence, the F 1 plants were backcrossed with recurrent parent, IR64 to generate BC 1 F 1 seeds. A total of 10 BC 1 F 1 seeds were obtained as the seed set percentage was very low. The BC 1 F 1 seeds were grown during dry season of 2017-18 and selfed to produce 206 BC 1 F 2 seeds. High level of spikelet sterility was also observed among plants of BC 1 F 1 population.
During wet season of 2018, BC 1 F 2 population was grown under eld conditions and the plants were inoculated with Xoo strain, IX-020 for their reaction to BB. The resistant plants identi ed in BC 1 F 2 population were further backcrossed to recurrent parent IR64 and 16 BC 2 F 1 seeds were obtained. The seeds of BC 2 F 1 were sown during dry season 2018-19 and inoculated with Xoo isolate (IX-020). Out of 16 BC 2 F 1 plants grown and evaluated for BB resistance, two plants were found to be highly resistant (lesion length ranging from 2.5-3cm), ve were moderately resistant (lesion length ranging from 3.3-5.5 cm) and nine plants were moderately susceptible (lesion length ranging from 6.2-8.8cm) ( Figure   1D, and Supplementary

Estimation of recovery of RP genome in BB resistant introgression lines
Six introgression lines (at BC 2 F 2 generation) showing high level of BB resistance (with lesion length 0.1-0.8 cm) were selected for estimation of recovery of recurrent parent genome. A total of 70 parental polymorphic SSR markers identi ed (out of 428 screened) across all the 12 chromosomes were used for genotyping of BB resistant introgression lines. The total number of markers per chromosome ranged from four (Chromosome 9 and 10) to seven (Chromosome 1, 6 and 12) with average of six markers per chromosome ( Table 5). The recurrent parent genome recovery percentage among the six selected plants was observed to range from 52.4 (Chromosome 2) to 86.8 (Chromosome 1) with an average of 72.8% ( Table 5). The heterozygous component of the genome ranged from 0% (chromosome 4 and 10) to 25.3% (chromosome 6) with an average of 7.98% (Table 5; Plate 4.7). Among the six introgression lines, the recurrent genome per individual line ranged from 65.1% (Plant # 58*-4-14*-3-6) to 84.5% (Plant # 58*-4-14*-3-5) with an average of 67.5% (Table 6; Fig 4 (A)).

Development of BB resistant introgression lines
Cross incompatibility and spikelet sterility are the two major bottlenecks affecting utilization of O. glaberrima for genetic improvement of O. sativa cultivars (Hiroko et al., 1962). During inter-speci c hybridization between two cultivated species, we have encountered very high level of spikelet sterility in F 1 s and subsequent selfed and backcross generations resulting in segregation distortion, which ultimately affected studies on trait segregation pattern and molecular mapping. Some studies have reported 100% spikelet sterility and partial female fertility in inter-speci c cross between two cultivated species of rice due to presence of gamete eliminator and pollen killer (Heuer and Miezan, 2003;Bhatia et al., 2017). There are 10 loci causing hybrid sterility loci (pollen killer or gamete eliminator) between O. sativa and O. glaberrima (Li et al., 2018). Repeated backcrossing of inter-speci c hybrids with the recurrent parent would help in restoring the fertility in the progenies (Sano 1986 andSano, 1990). In our study, we employed two rounds of backcrossing followed by sel ng (BC 1 F 1 -BC 2 F 2 ). During each generation of backcrossing and sel ng, BB resistant introgression lines were identi ed by screening against Xoo strain IX-020. Six BB resistant introgression lines with recovery of 65.1-84.5% recurrent parent genome identi ed in BC 2 F 2 will serve as important novel donor sources for BB resistance to further utilize in mapping and transfer to cultivated varieties for development of durable BB resistance rice cultivars.

Performance of introgression lines for yield attributing traits
We also made an attempt to understand the performance of O. glaberrima derived introgression lines for yield attributing traits and we found signi cant favorable variations for days to 50% owering, maturity, plant height, productive tillers in the BC 2 F 2 population. We also observed transgressive segregation for yield attributing traits namely plant height, productive tillers, panicle length indicating the bene cial alleles from O. glaberrima. However, number of fertile spikelets per panicle was considerably low among the introgression this could be due to high level of sterility observed in the BC 2 F 2 population. By selecting plants with high spikelet fertility and advancing through sel ng, we will be able to able x the problem of spikelet fertility as we select against spikelet. It is pertinent to note that the linkage drag from O. glaberrima is known to affect the performance of introgression lines for yield attributing traits. O. sativa cultivars have been successfully utilized for genetic improvement of O. glaberrima cultivars in Africa which has resulted in development of NERICA lines, which have shown promising performance for yield attributing traits (Jones et al., 1997). However, O. glaberrima genetic resources have not been extensively utilized for genetic improvement of indica cultivars (Sarla and Swamy, 2005). The transgressive segregation observed for yield attributing traits is an indication that O. glaberrima also harbor bene cial alleles for yield attributing traits. The utilization of high-through put genomic approaches is indeed essential for gainfully exploring O. glaberrima for yield attributing traits in order to achieve genetic gain in indica rice cultivars.

Conclusion
Oryza glaberrima is important genetic resource for identi cation of novel bacterial blight resistance.