Abstract
The ascomycete fungus, Pyricularia oryzae or Magnaporthe oryzae, is known to cause blast disease in more than 80 host plants of the Gramineae family—cereals including rice and grasses. The improvement of the Supa234 rice line (IR97012-27-3-1-1-B, containing badh2 gene for aroma) developed at IRRI-ESA Burundi consisted of introgression of R genes (Pita and Pi9) for blast resistance. The F2 population obtained via the cross had been screened for blast resistance using inoculation with Pyricularia oryzae spore’s suspension. The objectives of this study were to assess the presence of Pita and Pi9 genes for blast resistance and to assess the presence of the badh2 gene for aroma in the screened F2 plants using molecular markers. Genotyping was carried out in 103 F2 plants which grew to maturity using the KASP genotyping method with SNP markers (snpOS0007, snpOS0006, and snpOS0022) targeting the Pita and Pi9 genes for blast resistance and the badh2 gene for aromatic fragrance. The genotyping results showed that 38 F2 plants had the Pita gene present in both alleles, 31 F2 plants with the Pita gene in one allele, and only one plant (3B1) was found with the Pi9 gene in one allele. The badh2 gene for aroma was detected in 27 F2 plants on both alleles and in 57 F2 plants on one allele. There were thirteen plants which had both the Pita gene and the badh2 gene for aroma, and only one plant (3B1) had a combination of the three genes (Pita, Pi9, and badh2). Seven plants resistant to blast disease (2H2, 2H4, 1G2, 1C12, 1E13, 1B12, and 1C5) with the Pita and badh2 genes were found, and only one resistant plant (3B1) had a combination of the three genes Pi9, Pita, and badh2 which is recommended to be bulked for the development of the Supa aromatic rice variety resistant to blast disease. The plants generated by the best line 3B1 should further be evaluated for grain quality (Supa type) after F5 generation in the field.
1. Introduction
Rice (Oryza sativa) is a staple food worldwide for more than half of the world population [1]. Consumers prefer rice varieties with good grain quality like aroma, long grain, and amylase content. The aromatic trait enhances the market value of rice [2]. Nonaromatic rice has badh2 gene in chromosome 8 encoding for betaine aldehyde dehydrogenase enzyme with 503 amino acids while in the aromatic one, the number of encoded amino acids is 251. The badh2 gene produces GABA, a four-carbon nonprotein amino acid acting as a natural pesticide playing several roles including detoxification of free radicals, plant development, and plant defense [3]. The aromatic trait is coded by the mutant form of badh2 gene with 8 bp deletion in exon 7 of the badh2 gene, encoding a chemical compound 2-acetyl-1-pyrroline (2AP) [4–6]. The presence of the badh2 mutant gene encoding for the pleasant aroma by producing 2-acetyl-1-pyrroline was established by Nadaf et al. [3] to be associated with some weakness like yield losses, sterility, and susceptibility to abiotic and biotic stresses including blast disease. The mutant form of the badh2 gene has been associated with plant susceptibility to diseases as it suppresses the expression of the badh2 gene [3]. Blast disease which is known to occur in 85 countries worldwide [7, 8] is manifested in temperate and humid regions as the main cause of reduction of rice production [9, 10]. The blast disease can cause high yield loss of 10 to 85% when factors or enhancers of epidemic development (high mean temperature, relative humidity higher than 85-89%, the presence of dew, excessive nitrogen fertilization, and drought stress) are present [11].
Methods used in controlling blast disease include adjustment of planting time, burning diseased tissues, use of healthy seeds, and cultural systems like fungicide and fertilizer management without ignoring the use of resistant plant varieties bearing genes for blast resistance [12, 13]. Molecular screening of major rice blast resistance genes has been determined using molecular markers, which showed close-set linkage to 11 major rice blast resistance genes (Pi-d2, Pi-z, Piz-t, Pi-9, Pi-36, Pi-37, Pi5, Pi-b, Pik-p, Pik-h, and Pi-ta2), in a collection of 32 accessions resistant to Magnaporthe oryzae [13]. Out of the 32 accessions, the Pi-d2 and Pi-z appeared to be omnipresent and gave positive expression. The analysis of QTLs links genetic markers with DNA base variations, like single-nucleotide polymorphisms (SNPs) and simple sequence repeats (SSRs) or microsatellite to the QTLs of interest [14]. The most popular markers used in QTL analysis are SSRs, also known as microsatellites. However, SSR markers have been replaced by SNPs as molecular markers of choice in plant genetic analysis due to their codominant inheritance, their biallelic nature, chromosome-specific location, and genome-wide distribution [15]. The objective of this study was to assess the presence of R genes (Pita and Pi9) for blast resistance and the badh2 gene for aroma in the F2 generation using rice genotyping methods. In this study, markers (SNP and InDels) linked to the Pita and Pi9 genes conferring blast resistance and the badh2 gene for aroma inherited in the improved parent (Supa aromatic 234) were genotyped using the KASP method in 103 F2 plants which had been phenotypically screened for blast resistance.
2. Materials and Methods
2.1. Rice Seeds
In this study, seeds of 208 individual rice plants of the F2 population and 72 seeds of parent or control lines (12 of Supa234, 12 of Vuninzara, 12 of Gigante, 12 of IRBL9-W, 12 of CO39, and 12 of BC3) were obtained from IRRI-ESA Burundi. The F2 generations had been developed by IRRI-ESA breeders in Burundi for the purpose of improvement of Supa234 rice (IR97012-27-3-1-1-B) aromatic line for resistance to blast disease. The IRBL9-W, a highly resistant parental monogenic line with the Pi9 gene, and BC3 bearing the Pita gene were used as resistant controls. The Supa234 and CO39 rice lines without genes for blast resistance were used as susceptible controls. Supa234 line (IR97012-27-3-1-1-B, aromatic) containing the badh2 gene for aroma was also used as a positive control for aromatic fragrance. The plants were grown in a randomized complete block design (RCBD) in trays. Rice seeds were sown in Minuro trays (plastic trays 36 cm wide and a depth of 56 cm) filled with soil collected from Gihanga rice-growing areas. The soil was dried under the sun for two weeks, to diminish plant contamination, and then ground. Each Minuro tray had 104 wells (each with a ) arranged in 8 rows and 13 columns. The seeds were sown at a rate of four per well which were later thinned to one after germination. One hundred and thirty-five plants selected from 208 F2 plants, four parents (Supa234, Vuninzara, Gigante, and IRBL9-W), and two controls (CO39 and BC3) were screened for blast resistance at vegetative and reproductive stages. The screening was carried out into petri dishes by inoculating detached leaves with blast spores using the spot inoculation method [16]. Five microliters of conidial/spore suspension were inoculated on both sides of each leaf segment. Each plant sample had its own negative control, which were inoculated with a mixture of Tween 20 and deionized water. After inoculation, the petri dishes containing the leaf segments were maintained at under continuous fluorescent light (11 to 12 μEm-2 s-1) for 24 hours. Excess moisture on the leaflets was removed after 24 h by blotting with sterile pieces of laboratory tissue paper [16]. The leaves were then incubated at in the dark room for 10 days. To maintain the moisture level, sterile distilled water was added once every 3 days to the petri dishes to avoid desiccation of the leaf segments during incubation.
2.2. DNA Isolation
Leaf samples from the 135 selected F2 individual plants, parents, and controls were collected 21 days after planting. The leaf samples were lyophilized to remove moisture and kept at -80°C. The DNA for genotyping was extracted from the leaf disks of 103 F2 plants (which grew to maturity) among the 135 plants screened for blast resistance, parents, and controls. Parents and the control lines CO39 and BC3 were also screened. IRBL9-W, Vuninzara, and Gigante parents and the BC3 rice line were used as positive controls for the Pi9 and Pita gene markers while Supa234 parents were used as the positive control for the badh2 gene maker. BC3 and Supa234 were also used as negative controls. Leaf tissues weighing 0.5 g from each sample were used to extract genomic DNA. Leaf samples were separately crushed using a mortar and pestle, and the powdered samples were collected in sterilized 1.5 ml Eppendorf tubes. In each tube, 400 μl of CTAB lysis buffer containing 6.25 mM of potassium ethyl xanthogenate, 0.5% CTAB, 700 mM NaCl, 10 mM EDTA, and 100 mM Tris, pH 7.5, was added and mixed by vortexing for 30 s. The tubes were then incubated for 1 hour in a water bath at 65°C. An equal volume of chilled chloroform isoamyl in the ratio of 24 : 1 was added to each tube and centrifuged for 10 min at 13,000 rpm at 4°C. The supernatant was then transferred to new sterilized Eppendorf tubes. In each tube, 400 μl of isopropanol was added and kept overnight at -20°C for nucleic acid precipitation. This was followed by centrifugation at 13,000 rpm at 4°C for 8 minutes. The liquid phase was then gently decanted off leaving the DNA pellets. The DNA pellets were washed by adding 400 μl of 70% ethanol followed by centrifugation at 13,000 rpm. The 70% ethanol was then decanted, and the samples air-dried by inverting the Eppendorf tubes on sterilized laboratory tissues. Finally, DNA was dissolved using 100 μl of TE buffer containing 10 mM Tris, pH 7.5, and 0.5 mM EDTA. The extracted DNA was stored at -21°C before genotyping. The quality and quantity of DNA was determined using agarose gel electrophoresis using 0.8% agarose.
2.3. Genotyping
The extracted genomic DNA samples were genotyped using the Kompetitive Allele Specific PCR (KASP) genotyping technique [17] in the Intertek laboratory in Sweden. The KASP markers for the Pi9 and badh2 genes used were designed by IRRI (Table 1). The SNP-specific KASP assay mix, the universal Master mix (genotyping mixture), and the DNA sample used for all PCR reactions had a total volume of 10 μl. In 96-well plates for the PCR, one well contained a mixture of 5 μl genotyping mixture (4.4 μl of 2x KASP Master mix and 0.6 μl of KASP assay mix) and 5 μl of 50 ng DNA from each sample. Each KASP assay mix comprised three assay-specific nonlabelled oligonucleotides specific to a SNP or InDel marker comprising two allele-specific forward primers and one common reverse primer. Each primer harbored a unique tail sequence corresponding with a universal fluorescence resonant energy transfer (FRET) cassette and a primer-tail was labelled with FAM dye while on the tail of the second primer was labelled with HEX dye. The KASP Master mix on the other hand contained two universal FRET cassettes (HEX and FAM), ROX (passive reference dye), free nucleotides, Taq polymerase, and MgCl2 in an optimized buffer solution.
The Kompetitive Allele Specific PCR genotyping was performed in the following conditions according to Devran et al. [18]: one cycle for hot activation at 94°C for 15 min and the DNA denaturation was performed in 10 cycles at 94°C for 20 sec. The primer annealing and elongation were performed in 10 cycles for 60 seconds by dropping the temperature from 61 to 55°C at a rate of 0.6°C per each cycle. After, the temperature was raised to 94°C for 20 secs in 26 cycles to allow new denaturation and then lowered to 55°C for 60 seconds during annealing and elongation. When the amplification reactions were completed, 5 μl of the amplified products was transferred into the 384-well plates and detected on a BMG PHERA Star plate reader with a fluorescent resonance energy transfer (FRET) using the genotype cluster analysis Kraken caller software from LGC Genomics assigning a genotype to each produced color. Geotypes were scored according to the guideline of Table 2.
2.4. Data Analysis
The traits associated with each genotype and the positions of each SNP marker were generated by R software [19]. Based on the genotypic traits, a numerical scoring method was used assigning 1 to a positive allele and 0 for a negative allele. The scores were used to calculate the genotypic relationship between the parents and the F2 populations and analysis of molecular variance (AMOVA) using GenAlex software version 6.5 [20]. Principal coordinate analysis (PCoA) showing the genetic differentiation between the rice plants was generated using GenAlex software version 6.5. A dendrogram showing the relationship between the plants was drawn based on the genetic dissimilarity using the neighbor-joining method using Darwin software version 6 [21].
3. Results
3.1. Molecular Marker Results
The genotyping results of the rice plants are presented in Table 3. There were 39 plants with the Pita gene for blast resistance represented on two alleles (score 1 : 1); among them, 23 plants including 1C2, 1G2, 1C5, 1H7, 1D5, and 1B12 were either resistant or highly resistant in both stages of development (vegetative and reproductive) (Table 3). The Pita gene was also present in allele 1 in 30 plants; among them, 10 were resistant or highly resistant in both stages including 3B1, 1D1, 3E5, and 1D8 (Table 3). The IRBL9-W parent had the Pi9 gene for blast resistance in both alleles (score 1 : 1) while only one F2 plant, 3B1, had the Pi9 gene in only one allele (score 1 : 0) (Table 3). The Supa234 (IR97012-27-3-1-1-B, the aromatic), Gigante parents, and 27 F2 plants had the badh2 gene for aroma present in both alleles (score 1 : 1); among them, 14 plants including 1G2, 1C5, 3E5, and 1B12 were resistant or highly resistant in both stages. There were fifty-seven F2 plants including 1A1, 3B1, 1F11, and 1A6 which had the badh2 gene present in one allele (score 1 : 0) (Table 3). There were also plants including 1C1, 3C6, and 3E8 that did not have any of the targeted genes Pita, Pi9, and badh2 (Table 3).
Among the resistant or highly resistant F2 plants, only one plant, 3B1, had a combination of three genes (Pita, Pi9, and badh2), each present in one allele and nine plants (1G2, 1C5, 1D5, 1B12, 1C12, 1E13, 2B2, 2H2, and 2H4) having a combination of Pita and badh2 genes in both alleles (Table 3). However, there were 4 plants, 1H3, 2H5, 1G11, and 4C6, which did not possess the targeted R genes (Pita and Pi9) for blast resistance but were resistant to blast disease in both stages.
3.2. Genetic Variation between Plants
For the analysis of genetic variation within the screened plants, GenAlEX Software version 6.5 [20] grouped the categories into populations where F2 plants were grouped in population 1, Gigante parent in population 2, Vuninzara in population 3, Supa234 (IR97012-27-3-1-1-B) in population 4, and IRBL9-W in population 5. The BC3 control plants were grouped in population 6 and CO39 in population 7. Based on the genotype scores, the genetic variation calculated between the populations indicated that the number of observed alleles per locus (Na) ranged between 0.00 and 2.00 while the number of effective alleles (Ne) per locus ranged from 1 to 1.49 (Table 4). The F2 plants (pop 1) had the highest number of effective alleles (Ne) (1.49) while all the other rice populations only had one (1) effective allele.
The F2 population had the maximal percentage of polymorphic loci (% P) (100%) while the parents and controls had no polymorphic loci (% P) (0%) (Table 4). In this study, high genetic diversity was observed in F2 rice plants (pop 1) with the mean Shannon’s Information Index while within the parents, there was no genetic diversity () (Table 4). The mean expected heterozygosity (He) ranged from 0 for populations 2, 3, 4, 5, 6, and 7 (parents and control’s lines) to 0.28 for pop 1 (F2 plants) (Table 4).
3.3. Analysis of Molecular Variance (AMOVA)
The analysis of molecular variance (AMOVA) for the seven populations showed that the genetic variation among populations (52%) was slightly higher compared to that within populations (48%). However, the variations were not significant () (Table 5).
3.4. Principal Coordinate Analysis
The principal coordinate analysis (PCoA) of 103 F2 plants, 23 plants from 6 parents, and 19 plants from 2 controls populations clustered differently in the PCoA. The IRBL9-W parent (pop 5) with the Pi9 gene was in its own cluster but in the same quadrant I with the F2 plant 3B1 (Figure 1). However, the parents Vuninzara (pop 3) and Gigante (pop 2) bearing the Pita gene clustered together with the BC3 control (pop 6) in quadrant II while Supa234 (IR97012-27-3-1-1, aromatic) parent (pop 4) clustered with CO39 control (pop 7) without any R gene in quadrant IV. Other F2 plants clustered with Supa234 parent and CO39 and the remaining part of F2 population clustered alone in the PCoA (quadrant III) (Figure 1). There was no genetic differentiation between F2 plant 3B1 and IRBL9-W (Pi9) plants. There was also no genetic differentiation between Supa234 parent (IR97012-27-3-1-1) and CO39 control and some of the F2 plants (Figure 1). There was no genetic differentiation between parents Vuninzara and Gigante and the control BC3 as they clustered in the same quadrat (Figure 1). The F2 plants were distributed in three clusters (quadrants I, III, and IV) while the other populations were found in only one quadrat (Figure 1).
3.5. Phylogenetic Analysis
The neighbor-joining phylogenetic tree based on the genetic dissimilarity grouped the 103 F2 plants into three main clusters (clusters A, B, and C) (Figure 2). Cluster A containing Supa234 parents with the badh2 gene for aroma consisted of 35 F2 plants including the resistant plants like 1A11, 2C3, 3G12, 3E5, 4C6, and 1G11. Cluster A had three subclusters (1, 2, and 3) in which subcluster 2 contained Supa234 clustering with 4 F2 plants with bootstrap support of 43% (Figure 2). The smallest cluster B contained the negative control CO39, IRBL9-W (Pi9 gene donor parent) both without the Pita gene nor the badh2 gene and 9 F2 plants which are supported by 40% bootstrap except plant 3B1 (Figure 2). The third cluster C composed of mainly parents and controls with the Pita gene (BC3 control in subcluster 3, Vuninzara parent in subcluster 4, and Gigante parent in subcluster 5) and 58 F2 plants. The cluster C had 5 subclusters in which subcluster 3 supported by 41% bootstrap contained BC3 which was the positive control with the Pita gene.
The subcluster 4 contained the Vuninzara parent containing the Pita gene for blast resistance and subcluster 5 supported by 41% bootstrap value contained Gigante (Pita gene donor parent). Twenty resistant F2 plants clustered together with Vuninzara and Gigante parents in subclusters 4 and 5. In subcluster 4 containing Vuninzara which has the Pita gene, clustered resistant 13 plants including 1A10, 2H1, 1B10, 2E4, 1H10, 2G3, 1B13, 1G13, 1C2, 1F13, 2B4, 4B3, and 3G10. Subcluster 5, in which Gigante is clustered (parent with Pita gene), contained 7 blast resistant plants including 2H2, 2H4, 1G2, 1C12, 1E13, 1B12, and 1C5 (Figure 2).
4. Discussions
Marker-assisted selection is a breeding technique in which selection is done base on genotype of a marker of dominant or recessive alleles within a population [22]. In this study, marker-assisted selection was used in order to identify F2 plants of Supa aromatic rice line which may contain Pita and Pi9 genes for blast resistance and badh2 gene for aroma. KASP genotyping showed that Pita gene was predominant in the F2 population (except one plant 3B1 with Pi9 gene) even though not all plants had Pita gene and not all the positive plants for Pita gene were homozygous in both alleles (Table 3). The homozygous genotypes (resistant: resistant or resistant/homozygous) contained the Pita gene represented in both alleles and heterozygous (resistant: susceptible or resistant heterozygous) genotypes were characterized by the presence of the Pita gene on one allele while in the homozygous genotypes (susceptible: susceptible) Pita gene was absent in both alleles. This shows the state of segregation within the F2 population. This finding concurs with those reported by Jia et al. [23], in which there was a segregation in the F2 population for the Pita gene (resistant/heterozygous and resistant/homozygous). The badh2 gene for aroma was detected in 84 F2 plants in both allele or on one allele which demonstrate the inheritance of aroma from parent and segregation. In the present study, there were plants which had both the aroma and blast resistance genes, similar to the findings by Luo et al. [24] who reported a successful development of WH6725 resistant line to blast disease which possessed both genes. In the present study, the presence of Pita gene in resistant or moderately resistant plants may be attributed to the fact that Pita gene has been found to confer a medium-spectrum resistance [25].
The analysis carried out on genetic diversity and gene frequencies in the seven populations of plants used in this study (F2 plants, Gigante plants, Vuninzara plants, Supa234 plants, IRBL9-W plants, CO39 plants, and BC3 plants) showed genetic diversity (Shannon’s Information Index, ) within population 1 (F2 plants) while in parent’s populations, the Shannon’s Information Index was zero (Table 4). That genetic diversity ranging from 0 to 0.410 show a moderate diversity in the screened plants compared to the moderate genetic diversity ranging from 0.05 to 0.78 observed in fifty SSR markers used in germplasm of fifty red rice by Islam et al. [26]. The heterozygosis of zero found in parents and controls lines demonstrated that parents used in the cross carried out at IRRI-ESA were 100% homozygous or true breeding [27]. However, heterozygosis of 0.28 observed in F2 plants is associated to the state of segregation of F2 generation (called segregating population by Mendel).
Analysis of molecular variance (AMOVA) in the rice plant populations showed that there was a slightly higher variation among populations (52%) although variation was also observed within populations (48%) however the variations were not significant (). The low genetic variation among the population and within population, respectively, indicate that the plants under this study were closely related. This variation found is different from the reports on variations among population (34%) and within population (66%) [28], and among groups (35.28%) and within groups (64.72%) reported by Chakhonkaen et al. [29].
The F2 plants clustered in dendrogram and in PCoA according to the presence of R genes and aroma in alleles. According to the principal coordinate analysis (PCoA), there were no genetic differentiations between 3B1 F2 plant and IRBL9-W plants gene due to the presence of Pi9 gene in the 3B1 and IRBL9-W plants. This was also observed in the dendrogram where the two plants, 3B1 and IRBL9-W clustered in the vicinity. However, they were separated due to the presence of a single copy of Pita and badh2 in 3B1. The fact that Supa234 (IR70212-27-3-1-1-B) parent and CO39 control clustered together is due to the absence of the target R genes; hence, the F2 plants clustering together do not contain the R genes. The F2 plants were distributed in three plot areas as they were distributed in three cluster of dendrogram due to the fact their genetic characteristics differ where some offspring carried genes from one parent while others had genes from both parents and others did not have any of the targeted three genes (Pita, Pi9, and badh2).
5. Conclusion and Recommendation
The molecular marker genotyping of the rice plants for R genes for blast resistance shows the presence of Pita gene conferring resistance to blast disease in many F2 plants, represented in either both alleles or on one allele. The Pi9 gene was recovered in only one F2 plant 3B1 (represented on one allele). Resistant rice plants including F2 population of Supa234 (IR702-23-3-1-1-B, aromatic line) bearing badh2 gene for aroma and Pita gene for blast resistance were identified in this study. The resistant F2 line 3B1 obtained in this study combining Pi9, Pita, and badh2 genes can be used for development of an aromatic rice variety (Supa type) with resistance to blast disease.
Based on the above findings, there is a need to for further study on the resistant 3B1 F2 plant identified with the 3 targeted genes (Pita, Pi9, and badh2), by testing for blast resistance in field conditions to assess the stability of the resistance. The resistant 3B1 plant, found with aroma gene and resistance genes can be studied further for grain quality (Supa type) in final stage of variety fixation. Further research is necessary to check for other R genes which can be the source of resistance in the five resistant plants with badh2 gene for aroma (1G11, 3E5, 1A11, 3G12, and 2C3) which did not contain the targeted R genes (Pita and Pi9) or contained a single copy of the Pita gene.
Data Availability
The data used to support the findings of this study are available from the corresponding authors upon request.
Conflicts of Interest
The authors declare that they have no conflicts of interest.