Submergence is one of the important abiotic stresses incurring substantial loss to rice production in Bangladesh. With a view to improving submergence tolerance in BR22 (popular name Kiron) is a promising antioxidant enriched, high yielding and strongly photosensitive transplanted Aman rice variety, BRRI dhan 51 was used as a donor parent to introgress the SUB1 QTL conferring submergence tolerance. Molecular marker within submergence tolerant QTL (SUB1), flanking markers and evenly distributed background markers were utilized in a marker aided selection program to develop advanced breeding lines with broad-spectrum tolerance to flash flooding submergence. Indel and simple sequence repeats (SSRs) markers were used in the marker assisted backcrossing scheme in BC4 F1 and BC5 F1 generation. Sub1C173 a gene specific indel marker SUB1 QTL was used to select plants possessing the tolerant genes (foreground selection). The result of recombinant selection also revealed that the size of the SUB1 QTL introgression from the donor parent was estimated as 2.5 Mb. The percentage of recipient genome recovery in the best plant viz. BR10190-3-1-20-3 was 95%. This research work illustrates the successful application of markerassisted breeding to introgress the submergence tolerant QTLs into the genetic background of BR22.

Amin A, Md. Iftekharuddaula K, Sarker A, Ghoshal S, Aditya TL, et al. (2018) Introgression of SUB1 QTL into BR22 Using Marker Assisted Backcrossing. Int J Plant Biol Res 6(5): 1103.

•    Submergence; Abiotic stress; SUB1 QTL; Molecular 
marker; Marker assisted selection; Inde

Rice is the staple food of about three billion people including eight countries in Africa, nine countries in North and South America and seventeen countries in Asia [1,2]. Approximately half the world’s population depends on rice for survival [3]. Although rice is a crop that requires irrigated condition for cultivation, most rice varieties cannot survive if the plants remain submerged for more than one week [4]. The major QTL responsible for submergence tolerance was mapped successfully to chromosome 9 named SUB1, stated to account for almost 70% of the phenotypic variation under submergence [5]. This SUB1 QTL has been introgressed through marker assisted selection programme into a few numbers of rice varieties.

Bangladesh Rice Research Institute has so far released two submergence tolerant rice varieties namely BRRI dhan 51 and BRRI dhan 52 and the Bangladesh Institute of Nuclear Agriculture (BINA) has also released two submergence-tolerant varieties namely BINA dhan 11 and BINA dhan 12 [6]. In many parts of submergence prone areas of Bangladesh including northern part the duration of flash flood submergence is more than three weeks; sometimes it remains within 3 to 4 weeks. The presently available flash flood tolerant varieties of Bangladesh cannot survive whenever the level of stress is more than three weeks.

BR22 (popular name Kiron) is a promising antioxidant enriched, high yielding, strongly photosensitive lowland rice variety having 150 days growth duration [7]. BRRI scientists reported that BR22 possessed the highest Hydroxyl Ion Scavenging Activity (HISA). BRRI scientists assumed that BR22 is potential to promote human health, due to its content of the Phenolic Compound (PC) which is able to inhibit the formation of reactive cell-damaging free radicals or reduce the concentrations of free radicals, thus reducing the risk of coronary heart disease, cancer and averting oxidative damage of lipid as well as lowdensity lipoprotein (LDL) [8-10]. The variety does not possess the submergence tolerant QTL-SUB1. Nonetheless submergence stress not only depends on duration of submergence but also on many other environment and agronomic factors like water temperature, turbidity, depth of flood water etc. With this context Mohanty et al. [11], stated that submergence is a polygenic trait and SUB1 QTL does not alone completely represent the trait. Even with two to three weeks of submergence tolerance, the present submergence tolerant varieties are often damaged whenever the environmental factors other than duration of submergence go beyond the capability of tolerance of the existing varieties. This is the reason why new submergence tolerant varieties with higher levels of tolerance are required. SUB1 QTL introgressed into different genetic background can result into producing a new submergence tolerant variety with more level of tolerance as the QTL works through interacting with other QTLs in the genetic background.

Major thrust is that BR22 is a strongly photosensitive variety. In case of multiple flash floods, the duration of photo insensitive or weakly sensitive variety is so delayed to be affected by cold at the later part of the growth phase. The duration of a strongly photosensitive Sub1-variety is expected not be delayed. With this point of view, efforts have been undertaken to introgress SUB1 QTL into BR22 utilizing marker assisted backcrossing technique. To convert the high yielding, slightly salinity tolerant BR22 into a submergence tolerant type by incorporating SUB1 QTL from BRRI dhan 51 through marker assisted backcrossing (MABC) technique. In this scenario, three levels of selection approaches viz. foreground selection, recombinant selection, and background selection have been performed. MABC has three major advantages over conventional backcrossing. For instance, trait specific target gene can be identified through foreground selection, the linkage drag can be minimized through recombinant selection, and the recovery of recurrent parent genome can be predicted through background selection [12].

It is expected that improved BR22-Sub1 lines with more submergence tolerance will be more adaptable under the submergence prone regions of Bangladesh, predominantly in the multiple flash flooding affected northern region of the country.

Plant materials

The experiment was designed and carried out in order to introgress SUB1 QTL into the genetic background of BR22 (recurrent parent). BR22 is a T. aman variety having 150 days growth duration, high yielding capability, strong photosensitivity, short and bold grain and does not possess SUB1 QTL. Need to refer to this; cite appropriate reference, carrying the submergence tolerant QTL served as the donor parent during the cross. The parental crosses were initiated in T. Aman, 2011 (Kharif, 2011) season and crossing as well as marker assisted selection was attempted in every T. Aman season. This research work was started from BC4 F1 to BC5 F1 generation as a part of development of breeding pipelines of plant breeding division, Bangladesh Rice Research Institute, Gazipur-1701, using marker assisted selection techniques.

Genotyping approach

Leaf samples were collected following ten stick method (leaf picking-up from every ten plants tagged by sticks). DNA was extracted following modified Mini scale method [13] as well as CTAB method [14]. The DNA concentration of samples was adjusted to 25 ng/μl. PCR for Simple Sequence Repeats (SSRs) was performed in 10 μl reactions containing around 25 ng/μl of DNA template (3 μl DNA), 1 μl 10X TB buffer (containing 200 mM Tris–HCl pH 8.3, 500 mM KCl), 0.2 μl of 10 mM dNTP, 1.35 μl 25 mM MgCl2 , 0.5 μl each of 10 μM forward and reverse primers and 0.2 μl of Taq DNA polymerase (5 U/μl) using G-STORM thermal cycler [15,16] and Gene Atlas thermal cycler.

CR for STS (Sequence Tagged Sites) Primer Sub1C173 was performed in 20 μl reactions containing around 25 ng/μl of DNA template (4 μl DNA with 10-20X dilution factor), 2 μl 10X TB, 2.7 μl 25mM MgCl2 , 0.4 μl of 10 mM dNTP,1 μl each of 10 μM forward and reverse primers, 1 μl DMSO (Dimethyl Sulphoxide) and 0.4 μl of Taq DNA polymerase (5 U/μl) using G-STORM and Gene Atlas thermal cycler. The PCR products of SSR and STS markers were resoluted using Polyacrylamide gel electrophoresis. The size (in nucleotide base pairs) of the most intensely amplified band for each microsatellite marker was measured based on its migration in relation to a molecular-weight size marker (1 Kb+ DNA Ladder) with Alpha Ease 4.0 software. The homozygous allele of the recipient parent for the particular SSR marker was scored as ‘A’. Again, the homozygous allele of the donor parent for the particular SSR marker was scored as ‘B’. However, heterozygous alleles were scored as ‘H’. The molecular data was analyzed with Graphical Genotyper (GGT 2.0) software [17]. The software produced graphical sketch of the percentage of recurrent parent chromosomal segments in the selected backcross population.

Selection approaches of backcross population

Simple sequence repeats (SSRs) and indel markers were used for selection [18,19]. The different selection levels (Figure 1) followed in this backcross breeding is discussed as follows:

Foreground selection

In BC3 F1 & BC4 F1 generations of T. Aman 2014 & 2015 season, the foreground selection was done over 77 and 175 plants respectively using Sub1C173, an indel marker specific to SUB1 QTL. After foreground selection, the heterozygous plants were marked by sticks in the field. Phenotypic selection was applied during active tillering stage based on resemblance to recipient parent BR22. Different characters considered during phenotypic selection were tiller numbers, tillering pattern, flag leaf angle, leaf shape, leaf length, leaf breadth, leaf color etc.

In this study, gene-based indel marker Sub1C173 specific to the putative gene of Sub1C of SUB1 QTL was used in the foreground selection (Table 1). The individuals that were heterozygous for the foreground marker were selected in this selection step.

Recombinant selection

In this selection level, individual plants that were homozygous for the recipient allele at the marker loci distally or proximally flanking the Sub1 locus (i.e. recombinants) were identified. This was titled as “recombinant selection” [3]. The objective of this selection was to minimize the linkage drag. The flanking markers selected were as close as possible. Population sizes depend on distance of flanking markers from target locus. Flanking markers used for recombinant selection were selected from the tip of chromosome 9 around the Sub1 region was targeted. In this study, four markers (Table 2) were selected as flanking markers like RM5799, RM23843, RM8300 and RM23915.

Background selection

Microsatellite or Simple Sequence Repeat (SSR) markers covering all the chromosomes that were polymorphic between the donor and recurrent parent were used for background selection to recover the recipient genome (Figure 5). Out of 403 SSR primers surveyed, a total of 97 microsatellite markers were used for background selection initially. Maximum number of background markers used was 10 for chromosome 1 and chromosome 9. The minimum number of background markers used was 5 for chromosome 3. The average number of background markers over all the 12 chromosomes was 8.50 (Table 3). The microsatellite markers that revealed fixed (homozygous) alleles at non-target loci at one generation are not screened at the next BC generation. Only those markers that were not fixed for the recurrent parent allele are analyzed in the following generations. The segregates with fixed donor alleles were discarded from the selection in backcross-F1 generation.

Foreground selection

Foreground selection was performed among 77 BC4 F1 plants and 175 BC5 F1 plants in order to detect SUB1 QTL. Foreground selection was carried out using Sub1C173. The size of the resistant allele of this marker, which was obtained from the donor of SUB1 QTL viz. BRRI dhan 51, was 180 bp and that of susceptible allele obtained from BR22 was 196 bp. This marker produced very clear bands and it was possible to identify the genetic constitution of the Sub1-locus very easily using polyacrylamide gel electrophoresis. In BC4 F1 plants, out of 77 plants, 35 plants were found showing the locus for the gene-based marker as heterozygous state (Score H), 43 plants were found with the locus fixed for recipient allele (susceptible allele) (Score A) and only 3 plants were found with the locus fixed for donor allele (resistant allele) (Score B) (Table 4). In BC5 F1 plants, out of 175 plants, 87 plants had the heterozygous genotype (Score H), 74 plants were found with the susceptible allele (Score A) and only 14 plant were found with the locus fixed for donor allele (resistant allele) (Score B) (Table 4). In this marker assisted backcross breeding program, the allele fixed for recipient parent (BR22) for any marker has been scored as ‘A’, the allele fixed for donor allele as ‘B’ and that of with both the alleles of two parents, i.e. the heterozygous alleles as ‘H’. The plants with B score were produced due to accidental failure of backcrossing.

Here, in BC4 F1 generation, 31 plants showed ‘H’ score and 43 plants showed ‘A’ score which indicated that the results fitted to the expected 1:1 ratio of this generation with a non-significant chi square value of 0.82 at a probability level of 0.05. In BC5 F1 generation, 87 plants showed ‘H’ score and 75 plants showed ‘A’ score which indicated that the results fitted to the expected 1:1 ratio of these generation with a non-significant chi square value of 0.90 at a probability level of 0.05. Iftekharuddaula [20] also obtained the similar results in foreground selection with SUB1 QTL while introgressing this QTL into BR11 utilizing Marker Assisted Backcrossing technique.

Out of 87 plants, 10 plants were selected during maximum tillering stage based on resemblance to recipient parent BR22. It was expected that those 73 individuals possessed the submergence resistant allele, so these segregants were selected and promoted to further selection steps. Ten plants selected phenotypically were subjected for recombinant selection. The different characters considered during phenotypic selection were plant height, tillering pattern, leaf length, leaf width, leaf angle, etc. (Figure 2) shows the partial view of gel picture of the foreground selection with the gene-based marker Sub1C173 in BC5 F1 generation.

Recombinant selection

The foremost objective of recombinant selection in BC5 F1 generation was to obtain single recombinant type segregants at one side of the SUB1 QTL. Recombinant selection was carried out among 10 plants having heterozygous alleles for the gene-based marker in foreground selection. Recombinant type segregates were those types of segregants which minimized linkage drag at both sides of the QTL. In other sense, single recombinant type segregants possessed the recipient allele for the flanking markers used at any sides of the QTL (Figure 3). These types of segregants are very rare in frequency. A total of four flanking markers from the sub-telocentric Sub1-region of chromosome 9 were used in recombinant selection in BC5 F1 generation. Out of four flanking markers, two markers were taken from proximal end and other two markers were taken from distal end. The description of these flanking markers is presented in Table 5.

Frisch et al., stated that moderately small population sizes could be used for recombinant selection. For recombinant selection four flanking markers were used as RM5799, RM23843, RM8300 and RM23915. RM5799 and RM23843 were used for proximal end and RM8300 and RM23915 were used for distal end. The distance between the closest flanking marker and SUB1 QTL was 0.6 Mb at proximal end and 0.4 Mb at distal end (Table 6). The scores for the flanking markers viz. RM23843 and RM23915 were ‘A’ (homozygous recipient allele) whereas these were ‘H’ (heterozygous alleles) for the other two flanking markers RM5799 and RM8300. The results indicated that it was possible to minimize linkage drag or to obtain recipient alleles for the two flanking markers with ‘A’ scores. However, it was not possible to avoid linkage drag for the other two flanking markers, which will be achieved in the next backcross generation. The results of recombinant selection also revealed that the size of the SUB1 QTL introgression from the donor parent was estimated as 2.5 Mb. It was assumed that, the crossing over was taken place in the middle place of the markers.

Background selection

A total of 97 microsatellite markers were used for background selection over 10 BC4 F1 plants resulting from foreground, phenotypic and recombinant selection. Maximum number of background markers was used 11 for chromosome 9. Minimum number of background markers was used 5 for chromosome 3 (Table 3). Hospital et al., and Visscher [21,22] reported that, as a general rule, two to four markers per 100 cM could be efficiently used to accelerate the recovery of the recurrent parent genome in the early generations such as BC1 F1 or BC2 F1 . The reason for this was that in early generations, few recombinant events occurred so that donor chromosome segments were represented by a few long segments on each chromosome. Takeuchi et al. [23], used 116 RFLP markers covering 12 chromosomes in background selection in rice. Neeraja et al. [16], used 56 SSR markers as initial background markers for the development of Swarna-Sub1. However, the average number of background markers over all the 12 chromosomes was 7.83. The percent markers homozygous for the recipient parent ranged from 62.11% to 92.63% in those 10 plants (Table 7).

In the plant number BR10190-3-1-20-3 (Table 8) of BC5 F1 generation, 88 markers out of 97 markers (92.63%) were like recipient parent type. But if the alleles of the homologous chromosomes were considered, the percentage of recipient alleles in plant number BR10190-3-1-20-3 was 95.26%. So, this plant had been considered as the first best plant of that population. Figure 4 shows the graphical genotype of the best plant (plant no. BR10190-3-1-20-3). The contribution of recipient genome among the 10 selected individuals varied widely. The percentage of recipient alleles recovered were 68.95%, 69.95%, 83.68%, 84.21%, 85.26%, 88.42%, 88.42%, 93.95%, 94.26% and 95.26% in the selected plants viz. plant number BR10190- 3-1-16-7, BR10190-3-7-3-1, BR10190-3-1-19-5, BR10190-3-1- 16-2, BR10190-3-1-16-8, BR10190-3-7-14-7, BR10190-3-1-20- 3, BR10190-3-1-20-1, BR10190-3-7-14-6, BR10190-3-1-20-3 respectively (Table 8).

For available background markers, in the selected best plant (Plant No. BR10190-3-1-20-3), chromosome 1, 2, 3, 4, 5, 6, 7, 8, 11 & 12 were completely like the recipient type, whereas the recipient genome ranged from 81.8% (chromosome 9) to 93.8% (chromosome 10) in the remaining chromosomes (Table 8).

Seed production in backcross population

Totally 527 BC4 F1 seeds were produced from 10 selected plants of BC3 F1 generation (Table 9) and totally 632 BC5 F1 seeds were produced from 10 selected plants of BC4 F1 generation (Table 10).

In BC5 F1 generation, foreground and background selection will again be applied to identify the best plant which will be possible to self-pollinate in order to obtain homozygosity in BC5 F2 generation of T. Aman, 2016 season. In this way, BR22-Sub1 will be developed which will be reported elsewhere.

The first target of this experimental work was to introgress the SUB1 QTL into the genetic background of BR22 using markerassisted backcrossing in order to maximize the recovery of the desirable traits of the recurrent parent. During this process, three levels of selection approaches of marker assisted backcross breeding viz. foreground, recombinant and background selections were followed. In this study foreground selection was done in BC3 F1 generation to produce BC4 F1 seed and foreground, phenotypic, recombinant and background selections were done in BC5 F1 generation in order to introgress SUB1 QTL into BR22. After performing all the levels of selections, backcrosses were done with the selected plants of BC4 F1 generation to develop BC5 F1 generation.

In Marker Assisted Backcrossing of the present study, foreground, phenotypic, recombinant and background selection steps were applied in BC4 F1 generation in order to develop BR22- Sub1 variety. In foreground selection, Sub1C173, gene-based marker specific to putative Sub1C gene was used to select the backcross progenies with SUB1 QTL. In this selection step, BC4 F1 progenies segregated following 1:1 ratio which indicated that SUB1 QTL followed Mendelian segregation.

It was possible to minimize linkage drag or to obtain recipient alleles for the two flanking markers with ‘A’ scores. However, it was not possible to avoid linkage drag for the other two flanking markers, which may be achieved in the next backcross generation. The results of recombinant selection also revealed that the size of the SUB1 QTL introgression from the donor parent was estimated to be 2.5 Mb. The agronomically suitable genetic background of BRRI dhan 51 (proved earlier) nearby SUB1 QTL will hopefully not affect the phenotype of the newly developed breeding lines.

In the present study 3 best plants were identified possessing 95.26%, 94.26% and 93.95% recipient parent genetic background along with target SUB1 QTL in the heterozygous state. The rest of the genome was found segregating for the remaining heterozygous background markers. The recovery of recipient parent genome in the best plant of BC4 F1 generation indicated that one more backcrossing and a final self-pollination in the BC5 F1 generation would be required in order to develop BR22-Sub1 homozygous line. The results of the Marker Assisted Backcrossing works in BC5 F1 generation indicated that the introgression work will be completed in BC5 F2 generation.

The current output of Marker Assisted Backcrossing activities of this research work reflected that the approaches could routinely be used in order to introgress tolerant or resistant genes or QTLs, conferring tolerances or resistances to abiotic stresses, into the genetic background of high yielding varieties.

The newly developed BR22-Sub1 lines could be useful as a strongly photosensitive submergence tolerant rice line for the multiple flash flood affected greater Northern region of Bangladesh. However, the submergence tolerance, yield and other attributes of the newly developed BR22 line will be evaluated later.

The author is grateful to BRRI authorities for providing supports in this research work. This research activity was financially supported the Integrated Agricultural Productivity Project (IAPP) funded by the Global Agriculture and Food Security Program (GAFSP).

1. Schatz MC, Maron LG, Stein JC, Wences AH, Gurtowski J, Biggers E, et al. Whole genome de novo assemblies of three divergent strains of rice, Oryza sativa, document novel gene space of aus and indica, Genome Biol. 2014; 15: 506.

2. Elzupir AO, Alamer AS, Dutton F. Aflatoxin in Rice Crop: Prevalence and Assessment of Daily Exposure. Fungal Met. 2016; 1-19.

3. Jirawut WA. Increasing Rice Production: Solution to the Global Food Crisis. International students summit tokyo university of agriculture. Kasetsart University. 2001.

4. Goswami S, Labar R, Paul A, Adak MK, Dey N. Physio-Biochemical and Genetic Exploration for Submergence Tolerance in Rice (Oryza sativa L.) Landraces with Special References to Sub1 Loci. Am J Plant Sci. 2015; 6: 1893-1904.

5. Xu K, Xia X, Fukao T, Canlas P, Maghirang-Rodriguez R, Heuer S, et al. Sub1A Is an Ethylene Response Factor-Like Gene That Confers Submergence Tolerance to Rice. Nature. 2006; 442: 705-708.

6. Iftekharuddaula KM, Ahmed HU, Ghosal S, Moni ZR, Amin A, Shamsher Ali. Development of New Submergence Tolerant Rice Variety for Bangladesh Using Marker-Assisted Backcrossing. Rice Sci. 2015; 221: 16-26.

7. Kabir MA, Miah NM. Two modern photoperiod-sensitive rice varieties for Bangladesh. University Library, University of the Philippines Los Banos. 1989.

8. Dutta AK, Gope PS, Banik S, Makhnoon S, Siddiquee MA, Kabir Y. Antioxidant properties of ten high yielding rice varieties of Bangladesh. Asian Pac J Trop Biomed. 2012; 2: 99-103.

9. Rashid HM , Khan AH , Quddus MA , Elahi NE, Saha A. Performance of Photosensitive Modern Rice Varieties in the Medium Highland Phase-2 After Boro Rice. J Agronomy. 2002; 1: 71-73.

10. Dhan Gobeshona Somacar. Bangladesh Rice Research Institute, Gazipur. 2015.

11. Mohanty HK, Mallik S, Grover A. Prospects of improving flooding tolerance in lowland rice varieties by conventional breeding and genetic engineering. Curr Sci. 2000; 78: 132-137.

12. Collard BC, Mackill DJ. Marker-assisted selection: an approach for precision plant breeding in the twenty-first century. Philos Trans R Soc Lond B Biol Sci. 2008; 363: 557-572.

13. Zheng KL, Huang N, Bennett J, Khush GS. PCR-based marker-assisted selection in rice breeding. Philippines: International Rice Research Institute: 1995.

14. Allen GC, Flores-Vergara MA, Krasnyanksi S, Kumar S, Thompson WF. A modified protocol for rapid DNA isolation form plant tissues using cetyltrimethylammonium bromide. Nat Protoc. 2006; 1: 2320-2325.

15. Chen X, Temnykh S, Xu Y, Cho Y G, McCouch SR. Development of a microsatellite framework map providing genome-wide coverage in rice (Oryza sativa L.). Theor Appl Genet. 1997; 95: 553-567.

16. Neeraja CN, Rodriguez RM, Pamplona A, Heuer S, Collard BC, Septiningsih EM, et al. A marker-assisted backcross approach for developing submergence-tolerant rice cultivars. Theor Appl Genet. 2007; 115: 767-776.

17. Van Berloo R. GGT 2.0: versatile software for visualization and analysis of genetic data. J Hered. 2008; 99: 232-236.

18. McCouch SR, Teytelman L, Xu YB, Lobos KB, Clare K, Walton M, et al. Development and mapping of 2,240 new SSR markers for rice (Oryza sativa L.). DNA Res. 2002; 9: 199-207.

19. Temnykh S, Clerck GD, Lukashova A, Lipovich L, Carthinour S, McCouch SR. Computational and experimental analysis of microsatellites in rice (Oryza sativa L.): frequency, length variation, transposon associations, and genetic marker potential. Genome Res. 2001; 11: 1441-1452.

20. Iftekharuddaula KM, Newaz MA, Salam MA, Ahmed HU, Mahbub MA, Septiningsih EM, et al. Rapid and High-Precision Marker Assisted Backcrossing to Introgress the SUB1 QTL into BR11, the Rainfed Lowland Rice Mega Variety of Bangladesh. Euphytica. 2011; 178: 83- 97.

21. Hospital F. Selection in backcross programmes. Philos Trans R Soc Lond B Biol Sci. 2005; 360: 1503-1511.

22. Visscher PM, Haley CS, Thompson R. Marker-assisted introgression in backcross breeding programs. Genetics. 1996; 144: 1923-1932.

23. Takeuchi Y, Ebitani T, Yamamoto T, Sato H, Ohta H, Hirabayashi H, et al. Development of Isogenic Lines of rice cultivar Koshihikari with early and late heading by marker-assisted selection. Breeding Sci. 2006; 56: 405-413.