Introduction
Rice is without a doubt the most significant grain on a worldwide scale. South Asia has been dubbed Asia's ‘meal cart’ due to its prominence in rice domestication (Pachauri et al., Reference Pachauri, Taneja, Vikram, Singh and Singh2013). Bangladesh is one of the leading rice-growing countries in Asia. Bangladesh ranks fourth in the world (Bandumula, Reference Bandumula2018). Apart from rice variations, Bangladesh cultivates a diverse range of cultivars and landraces. For several centuries, farmers have kept several thousands of distinct landraces (Gao, Reference Gao2003). These cultivars were selected for favourable characteristics, including crop yields, seed size and resistance to environmental stress. Rice landraces have been critical in ensuring regional sustainable development and agriculture's long-term viability, and their value as a genetic reservoir for selective breeding cannot be overstated (Ali et al., Reference Ali, Rasul, Islam, Ahmed, Abdullah, Jahan and Hasan2020).
However, rice landraces are being displaced by the widespread cultivation of modern varieties and numerous interventions in rice habitats throughout Bangladesh (Jahan et al., Reference Jahan, Ali, Raihan, Rahman, Abdullah, Huda and Hasan2020). For centuries, the majority of aboriginal crops have been referred to by their native creole names. As a result, records on the genetic nature and history of such genotypes are scarce or non-existent. As an outcome, their characterization could show their phylogeny, which would be extremely useful for the genetic improvement of existing rice varieties using this germplasm. The characterization of accessions enables the quantification and organization of genotypic variability and its conservation, preservation and use in crop improvement projects (Li et al., Reference Li, Yan, Agrama, Hu, Jia, Jia, Jackson, Moldenhauer, McClung and Wu2010; Aravind et al., Reference Aravind, Banumathy, Vanniarajan, Arunachalam, Ilamaran and Kalpana2019).
Additionally, the world's population is growing, which increases climate vulnerability due to various activities and poses a massive threat to global food security (Nelson et al., Reference Nelson, Rosegrant, Koo, Robertson, Sulser, Zhu, Ringler, Msangi, Palazzo, Batka and Magalhaes2009; Mekonnen et al., Reference Mekonnen, Tessema, Ganewo and Haile2021). Salinity already impacts more than 20% of the world's agricultural soils, and it is becoming an increasingly severe danger to food production due to climate change (Haque, Reference Haque2018; Mustafa et al., Reference Mustafa, Akhtar, Abdullah, Pitman and Läuchli2019; Thorslund et al., Reference Thorslund, Bierkens, Essink, Sutanudjaja and van Vliet2021). Around 2.8 million hectares of Bangladesh's south coastline region are afflicted by salt levels ranging from 3 to 23 ds/m. Traditional rainfed cultivars, which have a low production potential, are commonly grown in this area (Baten et al., Reference Baten, Seal and Lisa2015; Haque, Reference Haque2018).
Diversification is also critical for environmental adaptation (Mekonnen et al., Reference Mekonnen, Tessema, Ganewo and Haile2021). However, some rice varieties are becoming increasingly vulnerable to genetic erosion, or the loss of diversity, due to lower population sizes caused by land-use changes and degradation, as well as changes in local climate that may select against specific genotypes. As a result, there is an increasing urge to characterize and conserve the landraces of climate-vulnerable places. Numerous rice landraces are being planted throughout Bangladesh's saline-prone coastal regions (Lisa et al., Reference Lisa, Seraj, Elahi, Das, Biswas, Islam, Salam and Gomosta2004; Seraj et al., Reference Seraj, Lisa, Islam, Begum, Das, Ashwani and Takabe2006). Rice from these locations is in high demand by customers due to its unique cultural customs. Thus, the region's notable landraces are well-suited to the soil conditions, such as salinity. As a result, they are pretty salt tolerant. Rice cultivars that are already adapted to stressful situations have several unique genes that help them survive in the location. Variety is a significant part of agriculture due to the broad array of crops available, which results in genotypic variation in farming (Modi and Bornman, Reference Modi and Bornman2004). However, numerous commercially significant characteristics inherent in landraces are underutilized (Roy et al., Reference Roy, Marndi, Mawkhlieng, Banerjee, Yadav, Misra and Bansal2016; Umakanth et al., Reference Umakanth, Vishalakshi, Sathish Kumar, Rama Devi, Bhadana, Senguttuvel, Kumar, Sharma, Prasad and Madhav2017). As a result, the focus should be given to analysing rice landrace diversity, which will provide valuable information for breeding efforts aimed at increasing rice production.
Although various researchers have characterized and analysed rice germplasm, research on the variability of common landraces and cultivars grown in Bangladesh's local regions is scarce. Given the widespread cultivation and importance of landraces, the research explored morpho-qualitative characteristics of 150 landraces collected from Bangladesh's coastal belt, including 50 advanced lines generated from coastal germplasm.
Materials and methods
Collection and characterization of rice germplasm
One hundred fifty rice cultivar samples were collected in Bangladesh's coastal zone, as indicated in online Supplementary Table S1. The samples were taken from local farmers in accordance with the ITPGRFA article no. 9. Farmers were recognized as the custodians and innovators of the material studied. Three replications of a randomized complete block design were used in the experiment. Data were collected following the recommendation of the Biodiversity International and the Africa Rice Center (AfricaRice), formerly known as the West Africa Rice Development Association (WARDA) (Imtiaz and Upadhyaya, Reference Imtiaz and Upadhyaya2010).
Data analysis
For three consecutive years, data were collected on ten qualitative morphological characters in 25 randomly picked plants (online Supplementary Table S2). Major gene contribution (MGC), character efficiency content (CEC) and gene diversity were calculated using PowerMarker 3.23 (Liu and Muse, Reference Liu and Muse2005). GenAlEx V6.5 (Peakall and Smouse, Reference Peakall and Smouse2006) was used to examine the analysis of morphological variance (AMOVA), F, and G statistic. STRUCTURE 2.3.3 was used to explore the population structure model-based application (Pritchard et al., Reference Pritchard, Stephens and Donnelly2000). For each K, there were three replications. For each run with 100,000 Monte Carlo Markov Chain repetitions, a burn-in duration of 100,000 iterations was used. For each genotype, a spectrum of genetic groups ranging from 1 to 10 was employed to assess its membership. The associated allele frequency and the admixture model were taken into consideration. To ascertain the threshold of K, the graph of LnPD computed for each K was employed. It was constructed using the Structure Harvester application.
Results
Importance of characters to the variability of the collected landraces
The prevalence of the MGC to every locus was determined using Fig. 1. The frequency ranged between 0.407 (leaf blade intensity of green colour, LBIGC) and 0.920 (LSAC, LBPA and LBAC). Six of the thirteen features had a significant gene contribution greater than 0.7, whereas the average was 0.694. LBIGC exhibited the most remarkable morpho-qualitative diversity (0.705), whereas LSAC, LBPA and LBAC exhibited the least (0.147). The overall diversity was determined to be 0.412. The amount of character efficiency in 150 rice germplasm was determined by calculating the CEC levels for each gene encoding a character. The character's genes were utilized as the ground for the computation. CEC values varied from 0.655 (LBIGC) to 0.136 (LSAC, LBPA and LBAC), with an average of 0.362 (Fig. 1).
Trait inputs to the diversity status of Bangladeshi coastal germplasm. The blue colour denotes significant gene contribution, while the maroon colour represents gene diversity. The colour green represents the Character efficiency content. BLSC: Basal Leaf Sheath Colour, LSAC: Leaf Sheath: Anthocyanin colouration, LBPA: Leaf Blade: P/A, LBAC: Leaf Blade: Anthocyanin. Colouration, LBIGC: Leaf Blade: Intensity of Green colour, LBA: Leaf Blade: Attitude, AC: auricle colour, CC: Collar colour, LigS: Ligule shape, FLA: Flag leaf: attitude, LeafS: Leaf: Senescence, StC: Stigma colour, SeedC: Seed colour.
Morpho-quantitative trait diversity and population structure of rice germplasm
The population formation and relationships of 150 rice germplasm samples from coastal Bangladesh were studied using STRUCTURE v2.3.4 (Pritchard et al., Reference Pritchard, Stephens and Donnelly2000). K-value was applied to identify the population structures of germplasm based on the qualitative data of 13 characteristics. A total of 41, 10, 33 and 36 pure genotypes were found in Pops 1, 2, 3 and 4 (Fig. 2). Subpopulations 1, 2, 3 and 4 had admixtures of 14, 1, 8 and 7 lines, respectively.
(a) K value for the purpose of the total of subpopulations. The K values ranged from 1 to 10. (b) Population differentiation pattern of coastal rice germplasm represented by a bar plot. Each vertical coloured column denotes a different genotype. The X-axis indicates the types of coastal rice, while the Y-axis represents the subpopulations.
Morpho-qualitative divergence and analysis of morphological variance
GenAlEx was used to determine the morphological variation in 150 rice genotypes' qualitative attributes. Seven per cent of the morphological variance was observed between populations, while ninety-three per cent was shown within individuals (Table 1).
Summary of analysis of morphological variance (AMOVA)
The F st is a method for determining population differences due to genetic structure. When comparing populations, a F st value greater than 0.15 can be considered significant (Frankham et al., Reference Frankham, Briscoe and Ballou2002). Thus, when the F st values from the STRUCTURE are investigated, a considerable divergence was identified within each of the four subpopulations (Table 2). However, a F st value of 0.111 was observed among the four subpopulations. This score shows a degree of phenotypic difference between the subpopulations as a whole (Hartl and Clark, Reference Hartl and Clark1997). The finding is consistent with the AMOVA result (Table 2), which indicated that 93% of the overall variations were contained within the individual variants. The remaining 7% was revealed as a result of demographic variation. A N m value less than one indicates that gene exchange between subpopulations is limited (Wright, Reference Wright1965). In this study, the N m value (2.007) indicates that gene flow between groups is continuous (Perdereau et al., Reference Perdereau, Kelleher, Douglas and Hodkinson2014). According to Eltaher et al. (Reference Eltaher, Sallam, Belamkar, Emara, Nower, Salem, Poland and Baenziger2018), this shows the possibility of adequate gene flow or genetic interchange. When this occurs, it is possible for considerable genetic differentiation to occur between subpopulations. The F st (0.317) value, which was less than 50%, offered information about homozygosity.
Wright's F statistics
We demonstrated that all 13 traits have an effect on population differentiation using G statistics and Wright's F statistics (Table 3).
F- and G-statistics for the studied traits
BLSC, Basal Leaf Sheath Colour; LSAC, Leaf Sheath: Anthocyanin colouration; LBPA, Leaf Blade: P/A; LBAC, Leaf Blade: Anthocyanin. Colouration; LBIGC, Leaf Blade: Intensity of Green colour; LBA, Leaf Blade: Attitude; AC, auricle colour; CC, Collar colour; LigS, Ligule shape; FLA, Flag leaf: attitude; LeafS, Leaf: Senescence; StC, Stigma colour; SeedC, Seed colour.
Between Pop 2 and Pop 3, the Shannon mutual information index was revealed to be the greatest when all subpopulations were included (1.252). While the lowest value (1.091) was found between Pops 3 and 4, as illustrated in Table 4. The highest significant gene exchange (1.552) occurred between Pop 3 and Pop 4. The smallest value (0.232) was observed between Pop 2 and Pop 3.
Pairwise population studies of Shannon mutual information (sHua) and degree of gene exchange (N m)
Discussion
Overemphasis cannot be placed on how important it is to examine genotypic variability in rice landraces in the saline region of Bangladesh. As a result of this examination, the abandoned heritage resource will be better conserved, and genes responsible for desirable features will be discovered. Achieving this will allow rice breeders to use these qualities for rice improvement. En outre, in the places afflicted by salt, there exist many rice landraces. A systematic characterization of the genotypes is therefore essential. There is now parental material for crop improvement from the qualitative data gained for the 150 germplasm taken from saline-affected coastal Bangladesh.
Impact of the traits on the rice diversity
Regardless of the type of crop, significant genes are vital to the level of diversity. An average value of 0.69 was found in this investigation. If the phenotypic expression of the evaluated traits was pronounced, then the value is high. It is known that significant genes play a vital role in determining the qualitative attributes of coastal germplasm in Bangladesh (Kinoshita et al., Reference Kinoshita and Kato2017; Bazrkar-Khatibani et al., Reference Bazrkar-Khatibani, Fakheri, Hosseini-Chaleshtori, Mahender, Mahdinejad and Ali2019; Rahman et al., Reference Rahman, Thomson, De Ocampo, Egdane, Salam, Shah-E-Alam and Ismail2019). Scientists have identified several genes that determine the quality of rice grain (Zhao et al., Reference Zhao, Zhou, Ponce and Ye2015; Wang et al., Reference Wang, Pang, Wang, Chen, Zhu, Shen, Ali, Xu and Li2017; Park et al., Reference Park, Park, Baek, Baek, Jeong, Cho, Lee, Lee, Suh, Kim and Kim2019).
According to a study conducted on the parameters of qualitative features, the rice germplasm sampled from coastal Bangladesh exhibited a low to moderate level of diversity. Diversity is used to measure the mean heterozygosity and genetic distances among individual plants (Shete et al., Reference Shete, Tiwari and Elston2000; Luo et al., Reference Luo, Brock, Dyer, Kutchan, Schachtman, Augustin, Ge, Fahlgren and Abdel-Haleem2019). According to Chen et al. (Reference Chen, He, Zou, Chen, Yu, Liu, Yang, Gao, Xu, Fan and Li2011), the average diversity of 300 rice samples gathered from various rice-growing regions worldwide was 0.358. The average diversity in our study was 0.238, which is equivalent to the 0.36 diversity reported for unique landraces gathered in India under farmer management (Kumar et al., Reference Kumar, Bisht and Bhat2010). However, it is smaller than the 0.43 value for diversity found in the US accession panel (Agrama et al., Reference Agrama, Eizenga and Yan2007). Similarly, the mean diversity of rice samples from China is 0.47 (Jin et al., Reference Jin, Lu, Xiao, Sun, Corke and Bao2010), which is more than the average diversity observed in our experiment. Additionally, it is less diverse than the diversity observed (0.68) by Ali et al. (Reference Ali, McClung, Jia, Kimball, McCouch and Georgia2011). Nachimuthu et al. (Reference Nachimuthu, Muthurajan, Duraialaguraja, Sivakami, Pandian, Ponniah, Gunasekaran, Swaminathan, Suji and Sabariappa2015) estimate that most diversity panels with worldwide accessions have a diversity of between 0.5 and 0.7.
The CEC metric quantifies the gene diversity associated with each character. CEC values were generated for each trait to conduct the diversity analysis. Three character variables, including LBIGC, auricle colour (AC) and flag leaf attitude, were used to summarize the diversity status of the local rice germplasm (FLA). Any of the three characters may be used to differentiate landraces and choose landraces for crop development initiatives (Oladosu et al., Reference Oladosu and Rafii2014; Sharifi and Ebadi, Reference Sharifi and Ebadi2018).
Population structure of rice germplasm
The population genetic parameters are critical when estimating the average genotypic diversity of a region or field. Additionally, they are valuable for determining the differences in allele frequencies between different populations and the disparities in polymorphism levels between populations (Kumar et al., Reference Kumar, Bisht and Bhat2010). Understanding population genetic structure will aid in tracking the loss of variety through time and space. Additionally, it will aid in the development of a rational conversation approach for managing farmer landraces on-farm.
Numerous scholars frequently employ STRUCTURE's model-based technique to examine the population structure of rice (Huda et al., Reference Huda, Hasan, Abdullah and Sarker2019; Luo et al., Reference Luo, Brock, Dyer, Kutchan, Schachtman, Augustin, Ge, Fahlgren and Abdel-Haleem2019; Mishra et al., Reference Mishra, Kumar, Shamim, Tiwari, Fatima, Srivastava, Singh and Yadav2019; Hasibuzzaman et al., Reference Hasibuzzaman, Islam, Miah and Hasan2020). Moderately constructed accessions revealed four subpopulations with F st values of 0.398, 0.364, 0.206 and 0.281 among the 91 rice landraces from eastern and northeastern India (Das et al., Reference Das, Sengupta, Parida, Roy, Ghosh, Prasad and Ghose2013). Islam et al. (Reference Islam and Khalequzzaman2018) used a model-based approach to analyse the population structure of red rice germplasm from Bangladesh, resulting in four subpopulations. Additionally, a pairwise F st analysis revealed significant divergence across all pairs of populations, ranging from 0.108 to 0.207. As a result of this finding, it is concluded that the four subpopulation structural groupings are highly distinct. The 150 germplasm samples in our study were divided into four subpopulations using structure-based population analysis. Each group had a small number of admixtures. Between Pop 3 and Pop 4, the highest number of gene exchange was discovered. Local farmers typically retain such germplasm in a traditional manner. For many years, local farmers have been preserving germplasm at the community level.
Our analysis of the germplasm's population structure indicates that farmers may sustain a major portion of its purity level satisfactorily. Outcrossing and natural mutations may play a significant role in the establishment of population admixtures. It is critical to define the rice landraces that indigenous consumers highly prize. According to our findings, natural selection, genetic migration, mutation and purity can explain four distinct subpopulations. Future research will find it beneficial to understand the rice landrace's diversity and population structure. With the understanding of genetic differentiation and key gene contributions, local rice germplasm can be preserved. Additionally, the data from this study will be beneficial to farmers in establishing intellectual property rights for the rice landraces from coastal Bangladesh that was studied, as some of the landraces demonstrated distinct traits. Certain landraces exhibited unique anthocyanin colouration on the leaf sheath and blade. Anthocyanins contribute to reproduction by enticing pollinators and seeds disperser and providing protection from abiotic and biotic stressors (Van den Ende and El-Esawe, Reference Van den Ende and El-Esawe2014). Anthocyanins have been linked in the responses of plants to a range of concerns, including drought (An et al., Reference An, Zhang, Bi, You, Wang and Hao2020), UV-B (Bai et al., Reference Bai, Saito, Honda, Hatsuyama, Ito and Moriguchi2014) and heavy metals (Gao et al., Reference Gao, Chen, Lin, Liu, Wei, Chen and Li2020) and also resilience to pests (Fan et al., Reference Fan, Fan, Wang and Yang2016) and diseases (Sivankalyani et al., Reference Sivankalyani, Feygenberg, Diskin, Wright and Alkan2016). There are indications that anthocyanins are beneficial to health (Wallace and Giusti, Reference Wallace and Giusti2019), rendering anthocyanin metabolism intriguing for plant breeders. Thus, the classification of the studied landraces will aid plant breeders and farmers identify and select favourable traits for crop development. However, little effort has been made to characterize Bangladeshi rice germplasm. As a result, deliberate efforts must be made to compile a comprehensive inventory of these important plant materials to estimate the presence of novel beneficial genes from this resource.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1479262122000090
