Unlocking Genetic Variability in Rice: Enhancing Rice Bran Oil and Quality Traits

Rice (Oryza sativa L.) is a crucial cereal crop that is a primary food source for half of the global population. India, one of the world's largest producers and consumers of rice, has made significant progress in rice cultivation. Rice serves as a staple food for most of the Indian population, especially in the eastern and southern regions.  Due to the rapid increase in the global population, it is expected to feed 9 billion people by 2050 (Arunkumar et al., 2022). Rice grain quality has emerged as a primary concern for rice breeders and consumers. The milling yield of the grain plays a crucial role in determining both the yield of head rice and the rate of broken kernels in milled rice.

Rice bran, which comprises the outer brown layer of the rice kernel consisting of the pericarp, aleurone, seed coat, and germ, is a valuable by-product obtained after the rice polishing process (Sohail et al., 2017). Due to its high nutrient content, Unpolished rice containing bran is more nutritious than milled or polished white rice. Nevertheless, rice consumers prefer to consume polished white rice, even though brown rice contains valuable nutrient content. Rice bran containing 12 to 23 percent lipid can be extracted to obtain rice-bran oil (RBO). Typically, the yield of rice bran is around 8-10% of paddy rice (Punia et al., 2021). The amount of RBO extracted depends on various factors such as seed variety, degree of milling, extraction method, and agro-climatic conditions (Ghosh, 2007).

Rice bran oil is a healthy edible oil that has a balanced source of 39% polyunsaturated (linoleic acid), 19% saturated (palmitic acid), and 42% monounsaturated (oleic acid) fatty acid, so RBO is one of the most nutritious edible oils (Mezouari et al., 2006). Bran has a sufficient amount of antioxidant (γ-oryzanol) and nutritionally essential components such as minerals, vitamin E (tocopherols and tocotrienols), phytosterols, and phenolic components ( Sohail et al., 2017). It also helps to reduce LDL cholesterol, protects against cardiovascular diseases, and has anticancer effects (Awad-Allah et al., 2022, Punia et al., 2021). RBO has a significantly longer shelf life than other cooking oils because of its antioxidants. Due to its low viscosity, it absorbs less oil during cooking, resulting in fewer calories overall (Chakrabarty, 1989). It is commonly blended with other vegetable oils to enhance the cooking quality of rice bran oil due to its relatively low smoking point.

The prime objective of plant breeding programmes is genetic variability, which results from genetic differences among individuals within a population. Proper management of diversity can result in a permanent improvement in the plant's performance and can serve as a buffer against seasonal instabilities. Genetic variability among traits is essential for breeding and in selecting desirable types (Sadhana et al., 2022). Estimates of heritability help plant breeders in selecting traits for which selection would be effective. Heritability, along with genetic advance, helps to identify the gene action. The association studies between quantitative and quality traits are helpful for the breeders in their selection strategies for crop improvement. This study evaluates the genetic variability and trait associations among rice genotypes to identify those with superior rice bran oil (RBO) content and quality traits.

Rice bran was extracted from all the genotypes with a laboratory miller and kept immediately at 4ºC to control the growth of rice bran free fatty acid (FFA). The rice bran oil content was determined using a Soxhlet apparatus, 5 grams of bran were placed in a packed bed within a thimble, which was then inserted into a distillation flask in a Soxhlet apparatus containing 10 ml of petroleum ether. The extraction process continued until completion. The petroleum ether was evaporated after extraction to separate it from the oil. The weight of the extracted rice bran oil was then measured and recorded, and standardized in the near-infrared spectroscopy (NIR) instrument (M/s ZEUTEC, Germany; Model: SPA 1.0), which was calibrated with the oil content of rice bran ranging from 12 – 23%. R² value was 92.8. The analysis of unknown oil content samples of each genotype was recorded by scanning samples in a wave number range of 1400 – 2400 wave length/nm.

Statistical analysis

ANOVA for the randomized block design (RBD) was computed at the 5% significance level. Biometric analyses like PCV, GCV, heritability (h²), and GAM were calculated. Path coefficient analysis was carried out as per the standard method suggested by Dewey and Lu (1959). The statistical analysis, viz., variability and path analysis, was carried out using the TNAUSTAT statistical package (Manivannan, 2014). The correlation coefficient was computed by Pearson’s method using R software version 4.2.3.

The bran oil percentage ranged from 12.3% to 20.2%, indicating significant differences in the percentage of oil content concerning selected varieties. The highest mean bran oil percentage of 20.2% and 20.1% was observed in the varieties ADT54 and Kichadi samba, respectively. The results indicate that the two varieties had a potential for improving bran oil content in rice. In the experiment, the varieties viz., ASD16 (19.5%), CR 1009 (19.6%), and C053 (19.1%) exhibited relatively higher mean bran oil percentages and were further considered for bran oil characteristics improvement. On the lower end of the spectrum, the varieties Boothakali karupan, Moota kuruva, and Kallan samba possessed a lower mean bran oil percentage of 12.5% (Figure 1).

Figure 1.  Mean Performance of 32 Rice Genotypes for Bran Oil Percentage

The success of genetic improvement depends upon the existing genetic variability and the efficiency in selecting traits. The quantification of variation in the genotypes is indispensable for improving the traits. Therefore, quantifying genotypic variance helps enhance the trait of interest. The assessment of existing variation can be made by discriminating the genetic variance from the phenotypic variance by avoiding the environmental variance. The variance components estimated were PCV, GCV, heritability (h²), and genetic advance as a cent of the mean and tabulated in Table 2.

              The difference between PCV and GCV was found to be minimum, and it indicated the lesser influence of the environment on the expression of traits. The PCV was considerably higher than the GCV for all the traits. The traits, viz., number of productive traits, flag leaf length, total spikelets per panicle, thousand grain weight, and grain yield per plant, revealed high PCV and GCV. It showed the presence of a significant amount of genetic variability for these traits, and similar results were reported by Sadhana et al. (2022)  and Zahid et al. (2006).  These traits are suitable for selection in a crop improvement programme. Whereas, moderate GCV was observed for plant height traits, bran, and bran oil %. This result was in agreement with earlier reports GUPTA et al. (2016) and (Babu et al., 2017). The traits panicle length, brown rice, milling % and head rice recovery displayed a low PCV and GCV. The results represent the need for special breeding procedures for improving the traits exhibiting low PCV and GCV. Similar results were obtained by Adjah et al. (2020).

Table 2: Estimates of genetic variability for thirteen characters in rice genotypes

(PH-plant height, NPT-number of productive tillers, FLL- flag leaf length, FLW- flag leaf width, PL-panicle length, NGP-number of grains per panicle, TGW-thousand grain weight, SPY-single plant yield, BR- brown rice recovery, MILL- milling %, HRR-head rice recovery) 

The heritability ranged from 73.68 - 99.49 %. High heritability was noticed for all the evaluated traits viz., plant height (92.73 %), number of productive tillers (81.96 %), flag leaf length (86.59 %), flag leaf width (73.68 %), panicle length (89.18%), TSP (98.06%), thousand grain weight (97.22%), grain yield per plant (92.67%), brown rice (95.27%), milling % (96.51%), head rice recovery (95.40%), bran (99.12%) and bran oil % (99.49%). Sadhana et al. (2022) obtained similar results and reported high heritability for the characters viz., plant height, number of productive tillers per plant, panicle length, number of grains per panicle, thousand grain weight, grain yield per plant, milling percentage, and head rice recovery. Arya et al. (2024) reported the same results for the trait bran oil content, which had high heritability.

The genetic advance as percent of mean (GAM) varied between 10.80 and 91.81%. The high GAM was obtained for the traits viz., plant height (30.26 %), number of productive tillers (48. 24 %), flag leaf length (39.04), TSP (91.81 %), thousand grain weight (45.11 %), grain yield per plant (53.02), bran (28.72 %) and bran oil % (25.89 %). A similar result was observed by Sadhana et al. (2022) only for the traits thousand grain weight and grain yield per plant. Other traits, viz., flag leaf width (19.69%), panicle length (18.28%), brown rice (11.46%), milling% % (10.48%), and head rice recovery (11.97%), recorded moderate GAM in the genotypes.  Similarly, Adjah et al. (2020) reported the same results for brown rice and milling percentage traits. The high heritability along with high GAM was accompanied by the traits plant height, number of productive tillers, flag leaf length, TSP, thousand grain weight, grain yield per plant, bran, and bran oil %.  The obtained outcomes indicated that these characters were controlled by additive gene action, the environmental effect was minimal, and the selection could be effective for the genetic enhancement of the traits.

Assessing the relationship between yield and its component traits aids the plant breeders to improve the genetic potential of the genotypes in a desirable direction. The correlation coefficient among different traits is presented in Figure 2. Bran oil% % showed a positive and significant correlation with brow rice recovery (r = 0.610), productive tillers (r = 0.433), milling percentage (r = 0.497), and NGP (r = 0.301). Similar results were reported by Adjah et al. (2020). These outcomes figured out that the traits, viz., productive tillers, number of grains per panicle, and milling% were the important selection indices to improve the bran oil% trait. The bran oil percentage trait was non-significant and positively related to flag leaf length (r = 0.045), flag leaf width (r = 0.212), grain yield per plant (r = 0.196), and head rice recovery (r = 0.210). The characters viz., plant height (r = -0.065), panicle length (r = -0.003), thousand grain weight (r = -0.233), and bran (r = -0.135) were non-significant and negatively associated with bran oil% %. Similar results were provided by the authors Manivelan et al. (2022)  for the traits panicle length and head rice recovery.

Figure 2.  The Pearson’s Correlation co efficient among quantitative and quality traits *, ** Significant at 5% and 1% level, respectively

              Regarding inter-correlation, the number of productive tillers displayed a significant and positive association with the number of grains per panicle, brown rice recovery, milling %, and a significant negative correlation with thousand grain weight. Similarly Sadhana et al. (2022) reported for the traits, brown rice, and milling percentage. The trait brown rice recovery expressed significant and positive inter-correlation with milling%, head rice recovery, and significant negative correlation with Bran. Milling% exposed positive and significantly associated with head rice recovery and negatively associated with bran. Number of grains per panicle showed a positive and significant correlation with milling %, brown rice, and a negative and significant correlation with thousand-grain weight. Lakshmi and Chamundeswari (2021) reported positive and significant correlation for brown rice traits, head rice recovery and milling percentage.

Table 3: Path co-efficient analysis of direct (diagonal) and indirect effects of thirteen traits on bran oil % in rice genotypes

(PH-plant height, NPT-number of productive tillers, FLL- flag leaf length, FLW- flag leaf width, PL-panicle length, NGP-number of grains per panicle, TGW-thousand grain weight, SPY-single plant yield, BR- brown rice recovery, MILL- milling %, HRR-head rice recovery) 

Residuals 0.3871

The minimal residual effect implies that the chosen traits were highly suitable for conducting path analysis to determine their impact on bran oil %. Path analysis is the partitioning correlation coefficient into direct and indirect effects (Table 3). This analysis was made to study the cause and effect of the dependent and independent traits. The traits, flag leaf width (0.413), grain yield per plant (0.469), head rice recovery (0.300), and BR (0.979) exhibited a high positive direct effect on bran oil%. Sadhana et al. (2022) also reported the same results for grain yield traits and head rice recovery. Bran showed a moderately positive direct effect on bran oil %. The traits viz., head rice recovery (0.300) and bran (0.203) exhibited a moderate positive effect on bran oil %. The characters viz., flag leaf length (-0.184), panicle length (-0.145), NGP (-0.604), thousand grain weight (-0.573), and milling% (-0.443) revealed an adverse direct effect on the bran oil% % trait. The residual effect was observed as 0.387. These fallouts indicated that the studied traits reliable in contributing to the trait bran oil %. However, 38.71% of the variability remains unexplored in this study, which may contribute to other traits for the enhancement of the trait bran oil %. 

In this study, Bran and bran oil yield percentage showed high heritability coupled with genetic advance as a percentage of the mean. The expression indicates additive action of genes; further selection and exploitation of traits are beneficial and improve bran oil properties in rice. The per se performance for the trait rice bran oil was observed to be high for genotypes ADT 54, Kichadi samba, and CR 1009. These genotypes can be utilized in breeding programmes to improve the rice bran oil percentage. The obtained results infer that the selection of high bran oil-yielding varieties helps in reducing the demand for rice bran oil. These results provide valuable insights for selecting varieties with higher oil content for specific applications in the bran oil industry.