Genetic fingerprinting for the protection of local rice (Oryza sativa L.) cultivars of Bangladesh

Rice (Oryza sativa L.) having diploid chromosome number 2n=2x=24 belongs to the significant grass family Gramineae and subfamily Oryzoidae is considered one of the most cultivated annual cereals because almost 1/2 of the world’s human population consume rice every day as their predominant staple food [1]. It is occupied in the central position other than rest of the agricultural component to contribute Bangladesh’s national economy [2]. It is believed that Asian farmers have been maintaining the selection and domestication process of a wide range of rice cultivars since ancient [3]. The Gene bank of Bangladesh Rice Research Institute has accumulated near about 8,500 rice germplasm from different ecological and indigenous sources of Bangladesh of which 8,044 genotypes have been registered so far [4]. Great variations in rice genotypes have been noticed in Asia, more particularly in China as well as Indian subcontinent regarding morphological, biochemical and molecular aspects [5],[6]. Zhao [7] has reported on the subject of the origin of rice which has been accomplished  close to the northeastern part of India (Assam) and southwestern  part of China (Yunan), both are truly recommended to the subtropical upland of Himalayas [8]. Oryza rufipogon and Oryza nivara are the two parts of Javanica rice which are assumed feasible immediate progenitors of Oryza sativa [9].Interestingly, the Assam center (nearby of Bangladesh) is also considered for the center of origin of such Javanica rice.
Day by day the number of rice genotypes become raise which ultimately harder the flexibility to differentiate of rice cultivars on the premise of morphological and biochemical attributes [4]. Therefore several molecular markers such as SSR (Simple Sequence Repeat), RAPD (Random Amplified Polymorphic DNA), RFLP (Restriction Fragment Length Polymorphism), and AFLP (Amplified Fragment Length Polymorphism) etc. are widely utilized to be identified of particular cultivars [10], or quality seed of hybrid varieties [11] and for documentation of the released varieties in seed grain trade as well [12]. Thus DNA fingerprinting data is additionally one of the example which is being globally practiced for the legal evidence of DUS (Distinctness, Uniformity, and Stability) [13]–[20].
PCR-based assays, co-dominant inheritance pattern, and elevated multi allelic variation/polymorphism are the major dynamic factors which steer the microsatellites/SSRs as the precious genetic markers of choice for the breeders [21]–[23]. The motif of the hyper variable microsatellites/SSRs are fairly located and abundantly well distributed all over the rice genome [24]–[28]. Hence, microsatellites/SSRs are also becoming efficient tools for the breeders as well as geneticists to incorporate genetic maps of rice having enormous wealth of diverge genetic variation [29]–[32]. Still, more than 50,000 microsatellites/SSRs have been designed in between Indica and Japonica rice accessions which are being contributed to construct genetic map for characterization and documentation of rice [30],[33]–[36]. Based on the above scientific reports, such characterization and documentation process have been recently utilized in several varieties/landraces/cultivars/wild types of Oryza sativa, Triticum aestivum, Zea mayes, Saccharum officinarum,  Brassica napus, Glycine max, Solanum tuberosum, Corchorus capsularis, and other crop species of Bangladesh [13]–[19],[37]. However, in this study, the genetic fingerprinting techniques were utilized through five SSR markers for the protection of 110 local rice genotypes grown in Bangladesh. Moreover, the genetic fingerprinting techniques of this research can be further exploited for the protection and establishment of Intellectual Property Rights (IPR) of other crop species of Bangladesh.

Raising of rice seedlings and extraction of DNA
Genetic Resource and Seed (GSD) division of Bangladesh Rice Research Institute (BRRI) was provider of 108 cultivar’s seeds and the seeds of rest two cultivars named ACI-1 and Alok-932024 were supplied by ACI Pvt. Ltd of Bangladesh. Genetic fingerprinting lab of department of Genetics and Plant Breeding (GPB), Bangladesh Agricultural University (BAU), Bangladesh was being concerned to grow these seedlings as well as this experiment to be conducted. Genomic DNA of each cultivar was extracted from 2-5 inner succulent shoots of two-weeks-old fresh seedlings germinated in sterile petridishes. In that case, Rahman et al., [13]–[19],[37],[38] illustrated modified CTAB (cetyl trimethyl ammonium bromide) DNA extraction method supported by Aljanabi and Martinez [39] was followed to isolate genomic DNA.

Quantification and optimization of DNA concentration
UV-absorption spectrophotometer (Spectronic® Genesis™) was utilized at 260nm absorption to measure the purity and concentration of DNA which was subsequently estimated and converted into 25ng/µl with TE buffer, and finally stored at 4^oC before to amplify with SSR primers.

Selection of polymorphic SSR markers for rice genotypes
A total, 50 SSRs comprising on all the twelve chromosomes of rice were obtained. A set of seventeen SSR primers described previously [38] were selected from them by surveying allelic polymorphism data from the available rice genome database (http://www.gramene.org) as illustrated by Rahman et al.,  [13]–[19],[37]. At first, three to five of those primers were tested through ten randomly selected genotypes setting with the recommended PCR thermal profile [13]–[19]. The expected ranges (base pairs length) PCR product was then validated based on the most excellent response to amplify the target genomic region of the template DNA. In such a way five SSR primers viz., RM153, RM251, RM333, RM335 and RM475 which depict 5, 3, 10, 4, and 2 of rice chromosome [36] were preferred by visualizing clear and predictable amplified alleles, and finally employed for SSR analysis in this study (Figure 1). The selected primers were subsequently run with all 110 cultivars at once which displayed clear and repeatable polymorphic bands.

Polymerase chain reaction (PCR) amplification profile for SSRs
Eppendorf^™ oil-free thermal cycler gradient was utilized in this PCR reaction. Approximate 15 μl reaction volume consists of 50ng sample DNA (2.0 µl), 10x PCR Buffer (3.0 µl), 1 µM of each forward and reverse SSR primer (1.0 µl), 0.25 mM dNTPs (1.5 µl), 1 unit ampli Taq DNA polymerase (0.5 µl), and nuclease free double distilled deionized water (6.0 µl) were utilized to perform PCR reaction. PCR settings were carried out by the conditions described by Panaud et al., [34],[36] with minor modifications suggested by previous research [13]–[19],[37] as follows: 95°C for 5 min (an initial denaturation) followed by entire 35 cycles,  95°C for 40 sec (denaturation), 55°C for 30 sec (annealing) and 72°C for 1 min (elongation/extension), then a final elongation/extension cycle at 72°C for 7 min. Amplified PCR reactions were then stored at –20^oC for further utilization.

Electrophoresis and inspection of banding patterns
Top vision submarine horizontal electrophoresis system (BIORAD^™ Sequencing Cell) was utilized to electrophoresis the PCR reactions (Figure 1). Prior to electrophoresis, each 07 µL amplified PCR aliquot and 3 µL of loading buffer [38] were mixed gently. This mixture was then loaded on 2% agarose gel, and placed into the submarine horizontal gel chamber with 1x TBE running buffer (Trizma base, boric acid and EDTA; pH 8.0). A five microlitre (5 µL) 100 bp standard DNA (Gene ruler, Fermentas^®) ladder was added in both left and right side of the gel to compare the molecular weight of the amplified PCR products of each cultivar. PCR mixtures (10 μl) were subjected to electrophoresis at 100V and 50W for 2 hrs 40 mins. The electrophoresis sample was then kept as photographic image by the camera polaroid gel documentation system (UVP, BioDoc-It^™ System).

Genotyping of alleles and data analysis
DNA FRAG v3.03 computer software [39] was utilized to measure the most profoundly amplified alleles per loci (Figure 1) by using 100bp recognizing size standard DNA ladder [40]. The individual unambiguous DNA fragments were referred as alleles of the respective SSR markers. The allele frequency data (DNA fragment) was exported as diploid datasheet arrangement (AA, AB, CC and so on) on  POPGENE v1.31 computer program [42], and therefore, utilized for the several statistical analysis including “observed number of alleles-Na [42]”, “effective number of alleles-Ne” [43], “allelic diversity index (PIC=1-∑X_i², where X_i  indicates the  frequency  of  the i^th allele), Shannon’s Information index-I [44]”, “Hardy-Weinberg equilibrium (He and Ho of Levene [45] and Gene flow-N_m)”, “Nei’s gene diversity index (h) [46]”, “chi-square & probability index”, and “Wright’s fixation index-F_is [47]”. POPGENE v1.31 software was also applied to estimate genetic distance and similarity among the genotypes.  Finally, an UPGMA (Unweighted Pair Group Method of Arithmetic Means) phylogenetic tree (Figure 2) was assembled by means of Nei’s [48] genetic distance (D) which was visualized via Treeview computer software [49]. The generated cluster on UPGMA diagram (Figure 2) was then used to explain the relationships among the cultivars in this study.

Allele frequency and allelic diversity index (PIC)
Using 5 polymorphic loci in 110 cultivars of rice, a complete of 99 alleles were found in this quantitative investigation where RM335 revealed the foremost observed alleles (25) followed by RM333 (21), RM251 (20), RM475 (19) and RM153 (14) as shown in Table 1. The Highest effective number (Ne) of alleles (19.852) was also found in RM335 (Table 2). The diversity index of alleles or Polymorphism Information Content (PIC=1-∑Xi²) values usually reflect particular allele diversity into a species [50]. The average PIC value was 0.864 with the number ranging from 0.879 (RM153) to 0.949 (RM335) (Table 1).

Genetic variation statistics
Aspect of allele frequency of each cultivars and total cultivars were taken into consideration to calculate the comprehensive Shannon’s Information Index (I) which is fluctuated from 2.284 (RM153) to 3.080 (RM335). Fixation index (F_is) (a measure of genetic differentiation) was recorded from 0.687 to 0.958 having 0.838 average values (Table 2). Highest gene flow (N_m) was estimated through RM251 microsatellite loci (0.046) while RM153 showed the lowest gene flow (0.005) in this study (Table 2).

Across 110 rice varieties, RM251 (0.290) yielded the very best average heterozygosity (H_O) in current study followed by RM475 (0.172), RM335 (0.154), RM333 (0.090) and RM153 (0.036) (Table 2). Highest heterozygosity can be explained as a result of length and distance of RM251 marker on the genetic map relative to centromere [14].

Varietal identifications
Comparative SSR profiles and DNA molecular weight (band position) against five SSR primers (Figure 1), all the varieties were distinguished from each other with a minimum of single and/or arrangement of five primers (Table 3).

Analysis of genetic distance and phylogenetic tree
The summary of Nei’s genetic distance (D) from 5995 varietal pairs among 110 rice cultivars varied from zero to 2.832. Out of these varietal pairs, 58.87% (3529) showed no genetic distance [11]. Such genetic distance and similarity in this study separated all the 110 cultivars into several clusters (“a” to “x”) at once on the UPGMA diagram (Figure 2).

All the utilized polymorphic SSR markers recorded a complete of 99 unique alleles (Table 1) which was significantly higher than the total number of alleles reported by the several previous researches [13]–[19],[33]. On an average it yielded 19.80 alleles per primer with an effect of 38.81%. In agreement with earlier works [13]–[19],[33] reported a total of 18 alleles [16],[17] through analyzing with three primers (RM11; RM151 and RM153) and 78 [15] alleles with five primers (RM1; RM151; RM153; RM334 and RM335), respectively, while running on diverse ecotypes of rice genotypes of Bangladesh from the protected rice materials of the BRRI. In those investigations, PIC values were recorded 0.670; 0.707; 0.698 [16],[17] & 0.862; 0.923; 0.831; 0.865 and 0.910, respectively [15].  In another study,  a total of 238 rice accessions (Indica and Japonica) by using entire ten microsatellite markers were investigated by Yang and his associates [51] where they observed maximum 25 exclusive alleles. Genotypes under the low PIC value study represent closely related variants, while superior PIC indicates considerable enormous diversity, which is ideal for the development of new variants as well [52]. The frequency of short tandem repeats of microsatellites as well as their repeat sequences have a command on the quantity of amplified alleles and their resultant PIC values  of the experimental genotypes [35],[36],[53],[54]. In addition, Ni et al., revealed from his investigation that more extensive repeats including GA- sequenced repeats acquiesce more quantity of distinctive alleles with superior PIC standards [54]. In contrast, it has been suggested that the motif of (CTT)_n and amplified AT-affluent trinucleotide repeats also exhibits adequate and greater polymorphism of alleles [35]. RM333 primer containing (CTT)_n motif was one of the most instructional SSR marker because it gave 6-7 distinctive alleles and standard PIC range in Temnykh’s experiment [36]. In support of RM335 [(CTT)₂₀] SSR primers, 25 unique alleles with 0.910 PIC numeral were recorded which were the foremost alleles and the maximum PIC numeral in this investigation. The PIC valuation is considered as the discriminating strength of a promising marker to the genetic diversity study of the breeding materials selection program for the breeders because it regulates the frequency of observed and effective alleles of a particular DNA marker [55]–[57]. However, remarkably elevated PIC (0.879 to 0.949) values in this investigation indicated that the chosen markers have the required properties to be used in this DNA fingerprinting research among the 110 rice cultivars grown in Bangladesh [58]. However, the observation of this study was partially supported by the points of accuracy and usefulness from the above discussions.
Location of particular DNA genetic markers on the precise chromosomes, frequencies and size of alleles through their PIC numeral are given in Table 1. Mutation and chromosomal crossover are the two general events of heredity which are usually occupied at distal proximity from the centromere of the chromosome. They usually effort the formation of abundant alleles and eventually diversity of a specific locus [13]–[16],[18],[19],[37]. 24.7cM (chromosome 5); 79.1 cM (chromosome 3); 110.4cM (chromosome 10); 21.5cM (chromosome 4) and 92.5cM (chromosome 2) are the located positions of RM153; RM251; RM333; RM335 and RM475 primers on rice chromosome [36]. These primers were applied in the current observation of 110 rice cultivars including 2 hybrids and one variety of Jhum cultivation system. Overall gene flow values and genetic diversity were  observed in several populations of Oryza officinalis and computed 0.316 and 0.442, respectively, by examining of entire 14 microsatellite markers [59]. A wide range of allele frequency along with PIC value, major genetic variation in sense of observed and expected heterozygosity (Ho and He) were detected in this investigation. The observed and expected heterozygosity (Ho and He) values were estimated from 0.036 (RM155) to 0.290 (RM 251), and from 0.883 (RM155) to 0.954 (RM335), respectively. Superior expected heterozygosity (He) content than the observed heterozygosity in this study indicated that the selected SSR markers were remarkably abundant informative for the DNA fingerprinting among the rice cultivars [60],[61]. Partial consistent in theses’ observations were computed in several Bangladeshi local rice cultivars by a number of previous research groups [4],[17],[38]. Greater genetic variation and a lower gene flow value in 110 rice varieties argued that the most studied varieties in this experiment were landraces [38].
In these research materials, a total of 92, 15, 02 and 01 cultivars are recommended for transplant aman (T. Aman), broadcast aman (B. Aman), boro and jhum ecotype by BRRI [14]. In diallel fashion, a total of 5995 varietal pairs were possible among 110 cultivars where 2466 (41.13%) varietal pairs were computed as to be prominent genetic distance (Nei’s genetic distance-D) with each other [14]. In a previous study, while analyzing 94 varieties of six different ecotypes of rice grown in Bangladesh, a total of 4371 varietal pairs were computed of which 37% appeared nil genetic distance, and merely 1% showed highest genetic distance (2.583) [15],[62]. This closeness may be possible due to the genetic make-up of the locus for which the primers were responsible to distinguish along with low ecotype variation. Among the 5995 varietal pairs, only 0.917% appeared highest (2.832) genetic distance in this research. However, the superior genetic distance (D) is often observed while the cultivars or any genotypes were occupied from the landraces or wild relatives, in one side, and the high-yielding varieties (HYVs) on the opposite side in their crossing events [38].  Thus, the variation between highest and lowest genetic distance among the 110 cultivars proved their existence of variability. The resulting such genetic variability of the cultivars can be applied as a parent material in the future variety improvement programs to seek out the most efficient cultivars for further crossing or breeding.
However, Nei’s genetic distance (D) while analyzed on the UPGMA dendrogram considering 110 cultivars at a time, the dendogram separated the varieties, Dhul Abiz, Biropa and Bhor Gelam (Cluster “a”) from other 107 cultivars (Cluster “b”). Cluster “b” subsequently separated into sub-cluster “c” (Dumai Sail and Jol kumari) and sub-cluster “d” containing other 105 rice varieties. Sub-cluster subsequently formed other sub-clusters namely, “e”, “f”, “g”, “h” and so on (Figure 2). The varieties, as for example, Jol Kumari, Sandik sail, Jhul Kata, Thakor, Aalok 932024 and Tulu Sail were found in different sub-clusters “c”, “g”, “i”, “o”, “w” and “x”, respectively, due to their genetic distance. The major sub-clusters (“u”-“x”) were found to cover 39 of the 110 cultivars starting from ACI 1 to Lal Amon, all of which are traditional rice varieties of Bangladesh except ACI 1 and Aalok 932024.  UPGMA dendrogram within the groups; Transplant Aman (A), Broadcast Aman (B), Boro and Jhum (C) are given in Figure 2. As two Boro varieties (ACI 1 and Aalok 932024) and one Jhum variety (Ful Badam) were used in this experiment it was not possible to analyze them individually: they were therefore combined for a single analysis. An attempt was made to distinguish the varieties as their ecotype situations. The UPGMA dendrogram was constructed for this purpose (Figure 2). The results showed that the ecotypes have distinct clusters to represent Jhum and BRRI Accessions have formed the unique cluster different from all others as expected. The groups Transplant Aman (T. Aman) and Broadcast Aman (B. Aman) formed two closely linked sub-sub-clusters under one sub-cluster, while showing distinct difference from these sub-groups. Jhum formed a unique cluster and Boro a sub-cluster.
In these 110 cultivars, all the cultivars were distinguished from one another with either through 1^st, 2^nd, 3^rd, 4^th, and 5^th SSR primers (Table 3) and also through qualitative and quantitative traits of Breeders [18]. Many varieties had similar names, which had created a number of problems related to final characterization. The varieties Moisha Mida (T. Aman), Moisha Mira (T. Aman), Moisa Mira (B. Aman); Kala Gura (B. Aman), Kala Gora (T. Aman) and Thakor (T. Aman), Thakur Dhan (T. Aman) have similar names but when studied by both qualitative and genetic fingerprinting those showed distinct differences [14].

In this study, the registered local rice cultivars grown in Bangladesh were exploited to distinct each and every rice cultivar based on identification of the DNA band patterns by means of specific primers, generally termed as genetic fingerprinting. All the utilized cultivars in this research were distinguished from one another with either 1^st, 2^nd, 3^rd, 4^th, or 5^th SSRs. It is important to note that some of the varieties of traditional types had very similar names, which usually gave an understanding of repeats, but it was interestingly found to be distinctly different from one another due to molecular data. An example is the varieties named Moisa Mira, Moisha Mira, Moisha Mida, Thakor, Thakur Dhan, both pairs of which were distinct from one another on molecular traits. Therefore, these were not repeated but individually distinct cultivars or land races of rice collected from different source areas of Bangladesh by BRRI scientists at different times. However, Intellectual Property Rights (IPR) and Plant Variety Protection (PVP) of wild relatives, landraces, cultivars and commercial varieties of Bangladeshi rice will be guided from the outcomes of this research.   Such series of works on more rice genetic materials as well as other crop species should be done as a regular study by the genetic resource centers of different institutes in collaboration with universities, where government should give adequate financial and special manpower support with appropriate incentives for those who will lead the program.

Authors express thanks to the DANIDA for supporting this research work through the Agriculture Extension Component/Seed Wing of the Ministry of Agriculture, Government of the People’s Republic of Bangladesh. In the same line, they extend thanks to Mr. Anwar Faruqe, Former Joint Secretary MoA and Former Director General of the Seed Wing, for his support and keen interest in the project activities.

LR, MSEA and MKB were involved in conception and design of the experiments. MSR and MKHS contributed to perform the experiments. LR, UKN, MSR and MKHS contributed to drafting the article. LR and MSR contributed to revising it critically for important intellectual content. MSR made the final approval of the version to be published.

Authors declared that they have no conflict of interest.