Heterobeltiosis and interrelationship of some of important quantitative traits in oilseed rape genotypes

Abstract Half diallel crosses of eight spring genotypes of oilseed rape (Brassica napus L.) were considered to evaluate heterobeltiosis effects of plant height, yield component characters, seed yield and harvest index. Significant mean squares of general and specific combining abilities (GCA and SCA) were determined for all the traits except 1000-seed weight demonstrating prominence of additive and non additive genetic effects for the mentioned traits. Narrow-sense heritability estimates were high for siliquae on main raceme and 1000-seed weight representing the major importance of additive genetic effects for the characters. Most of the crosses with significant positive high parent heterosis for seed yield had also significant heterotic effects for siliquae per plant; therefore, this trait can be considered as indirect selection criterion for enhancing seed yield. Seed yield was significantly correlated with the traits including plant height, siliquae on main raceme and siliquae per plant based on mean performances of the traits and this result was confirmed with correlations based on heterobeltiosis. The crosses including L41×LF2 and L31×L401 with highly significant heterobeltiosis estimates of grain yield were superior combinations for breeding this trait. which proved good specific combiners for most of the traits.


Introduction
Oilseed rape (Brassica napus L.) is an important oilseed crop in the world, as a result of its high quality oil and meal and its potential as a renewable resource for bioproducts and biofuel (Downey and Rimer 1993; Oghan et al. 2016). The successes of hybrid breeding are the reason for its expansion in all major fields of agriculture (Mather and Jinks 1982;Rameeh 2014b). For developing a hybrid, as a first step information about genetic control of important characters is necessary. For this purpose, genetic information on heterosis is useful for developing breeding strategies to meet the demands of increased varieties (Mather and Jinks 1982). Heterosis and heterobeltiosis have comprehensively been explored and utilized for improving various quality traits in different crops. Therefore  Singh et al. 2010). Heterosis refers to the phenomenon that progeny of diverse genotypes of a species or crosses between species exhibit greater biomass, yield component characters and seed yield than both parents (Mather and Jinks 1982). Despite the successful establishment of heterosis in many agronomic crops, there is still challenge between the use of agricultural heterosis and our understanding of the genetic basis of heterosis, which inhibits the effective use of this biological occurrence (Brandle and McVetty 1990;Katiyar et al. 2000). Breeding for heterosis is one of the most successful technology options that involves improving the variety of prod-ucts. Heterosis can be visualized in terms of increase in strength, size, productivity, and resistance to diseases and pests of insects or climatic forces. One of the basic requirements for creating hybrid varieties in Brassica, has been proved to be heterosis. It is often observed that heterosis can be better expressed, although the crosses include the native or adapted and strange germplasm parents (Riaz et al. 2001;Rameeh 2014a). Improvement of hybrid varieties has been popular in many Brassica species. Mid-parents and heterobeltiosis have expansively been explored and utilized for enhancing various quantity and quality traits in oilseed rape (Teklwold and Becker 2005;Zhang and Zhu 2006). Heterosis is commercially exploited in oilseed rape and its potential use has been demonstrated in Indian mustard (B. juncea L.) and turnip rape (B. rapa L.) and for grain yield and most of the agronomic characteristics For seed yield in spring oilseed rape hybrids, an average heterobeltiosis of 30% with a range of 20 to 50% was detected, while for winter oilseed rape hybrids an average high parent heterosis of 50% was reported, ranging from 20 to 80% as reviewed by McVetty (1995).
Keeping in view the importance of edible oil and its inappropriate condition in the country, the present study was aimed to evaluate eight oilseed rape genotypes for heterotic effects and identify their potential hybrids for developing new genotypes.

Plant material, study site and experimental design
Eight genotypes of oilseed rape including L41, Zafar, L56, L31, L22, LF2, L420 and L401, which were prepared by Mazandaran Agricultural and Natural Resources Research and Education Center, were crossed in half diallel mating design at Bayekola Agriculture Research Station, located in Neka, Iran (13º53 ′ E longitude and 43º36 ′ N latitude, 15 m above sea level) during winter during 2011-12.
Twenty-eight F 1 offspring along with eight parents were planted in a randomized complete block design with three replications during 2012-13. The plots consisted of four rows of 5 meters and a distance of 40 cm and a spacing of plants in each row of 5 cm. The soil was classified as a deep loam soil (Typic Xerofluents, USDA classification) enclosed an average of 280 g clay kg -1 , 560 g silt kg -1 , 160 g sand kg -1 , and 22.4 g organic matter kg -1 with a pH of 7.3. Soil samples were found to have 45 kg ha -1 of mineral nitrogen (N) in the profile above 30 cm. The farm experiment received 50 kg ha -1 P, 75 kg ha -1 K and 100 kg ha -1 N. All the plant protection measures were adopted to make the crop free from insects.

Variables studied
Grain yield (kg ha -1 ) was recorded based on two middle rows of each plot. Plant height and yield components including siliquae in the main raceme, siliquae per plant, seed per siliqua and 1000-grain weight were measured based on 10 randomly selected plants per plot Harvest index was determined by using the following formula: Harvest index = [(economical or seed yield/ biological yield)] × 100

Statistical analysis
The combining ability analysis was conducted as outlined by Griffing (1956) method-II with mixed-B model. The least significant difference (LSD) was used to test the significance of the heterosis from high parent values (heterobeltiosis). All the analyses were performed by using MS-Excel and SAS software version 9.2.
Ethical approval: The conducted research is not related to either human or animal use.

Diallel analysis of variance
Significant mean squares of GCA for the parents were found for the characters including plant height, siliquae per main raceme, siliquae per plant, seeds per siliquae, 1000-seed weight, seed yield and harvest index indicating the importance of additive genetic effects for these traits (Table 1). Significant mean squares of SCA of the crosses were found for all the traits except 1000-seed weight, demonstrating non-additive genetic effects also had important role for controlling the traits except 1000seed weight.
High value of narrow-sense heritability estimates for 1000-seed weight, signifying that additive effect was controlling the inheritance of the trait. Similarly, significant GCA and SCA effects were reported for some important agronomic traits in Brassica napus (

Means of the parents and their cross combinations
Means of parents and their F 1 combinations are shown in Table 2 and Table 3, respectively. The means of parents for plant height ranged from 139 to 180 cm in L401 and L420, respectively. The genotypes with low mean value of plant height will be more resistance to lodging, therefore, the parent including L401, L31 and LF2 with 139, 147 and 156 cm and also the crosses such as LF2×L401, L41×L31 and L22×L401 with 134, 141 and 144 cm of plant height, respectively, will be preferred. Siliquae on main raceme was significantly correlated with seed yield (Table 4), therefore high mean value of siliquae on main raceme will be leading up high mean value of seed yield. The genotypes such as L56, L22 and Zafar with 64, 63 and 57 siliquae on main raceme are suitable for improving siliquae on main raceme. The cross combinations maintain Zafar×L40, L41×L22, L41×L56, L41×LF2, Zafar×L56, Zafar×L22 and L56×L22 with 65, 63, 62,62, 62, 62 and 61 siliquae on main raceme considered as merit combinations for this trait ( Table  3). Most of crosses with high mean performance of siliquae on main raceme had at least one parent with high mean value of the trait. Siliquae per plant was significantly correlated with seed yield, therefore this trait can be good criterion for improving seed yield ( Table  4). The parents comprising L22, L41 and Zafar with 158, 140 and 137 siliquae per plant were good candidate for improving seed yield. The crosses Zafar×L401, Zafar×L420, L41×LF2 and L31×LF2 with 177, 173, 172 and 172 siliquae per plant were superior combinations for improving this trait. Seeds number per siliqua is one of three components (including siliqua number and  grain weight) of yield and an important target characteristic of breeding in oilseed rape. As with other crops, seeds number per siliqua is usually negatively correlated with the other two yield components. These trade-offs are usually clarified as competition among sinks. Genetically, the correlation between traits is generally considered to be either genetic linkage or pleiotropy (Mather and Jinks 1982). LF2, L41 and Zafar with 29, 28 and 27 seeds per siliquae were superior parents and also the crosses including L41×L22, Zafar×L22, L41×L401, L41×Zafar, L56×LF2 and L22×L420 with 29, 29, 28, 28, 28 and 28 seeds per siliquae were superior genotypes.1000-seed weight ranged from 3.73 to 4.31 in L22 and L56, respectively. L56, LF2 and Zafar with 4.31, 4.29 and 4.08g of 1000-seed weight were detected as good parents for breeding this trait. Most of the crosses with high mean value of 1000-seed weight had at least one parent with high mean value of this trait. Seed yield of parents varied from 2083 to 3121 kg ha -1 in L401 and L22, respectively, and it ranged from 2197

Heterobeltiosis
The result of heterobeltiosis effects of crosses for the studied traits are presented in Table 6. Out of 28 crosses, 8 crosses had significant negative heterobeltiosis effects for plant height. Due to low mean value of plant height has direct effect on resistance to lodging, the crosses including L41×L31, L56×L420, L56×L401, L31×L420 and LF2×L420 with highly significant negative heterobeltiosis effects considered as superior combinations (Table 5). Significant positive correlation was detected between heterobeltiosis effects of siliquae on main raceme and seed yield (Table  6), therefore, heterobeltiosis effects of siliquae on main raceme can be suitable selection criterion for improving seed yield. Out of 28 crosses, 6 crosses had significant positive heterobeltiosis effects of siliquae per plant. The crosses viz. L41×LF2, L41×L420, Zafar×L420, Zafar×L401, L31×LF2 and L31×L401 with highly significant positive heterobeltiosis effects for siliquae per plant were considered as preferred crosses for improving this trait. Non of the cross combinations had significant positive heterobeltiosis effects for seeds per siliqua. Due to high narrow-sense heritability estimates for 1000-seed weight, none of the cross combinations had also significant positive heterobeltiosis effects for this trait. L41×Zafar and Zafar×L56 had positive heterobeltiosis effects for seeds per siliqua and 1000-seed weight, simultaneously. Out of 28 crosses, 7 crosses had significant heterobeltiosis effects for seed yield. The crosses including L41×LF2 and L31×L401 with high significant positive heterobeltiosis effects for seed yield were superior combinations for improving this trait. Mid parents heterosis and heterobeltiosis have been widely investigated and used to increase the qualitative and qualitative characteristics of oilseed rape (Nassimi et al. 2006a). Harvest index is a complex polygenic phenomenon that is strongly affected by both environmental and genotype factors. The implications of these results are that HI can be increased by reducing plant height or decreasing inefficient transport from siliquae to grains in oilseed rape. Harvest index is a complex polygenic phenomenon that is strongly influenced by both environmental and genotype factors. The implications of these results are that HI can be increased by decreasing plant height or reducing inefficient transport from siliquae to seeds in oilseed rape. Positive and significant correlation was determined between high parent heterosis effects of harvest index and grain yield (Table 6), therefore heterobeltiosis effects of harvest index can be suitable selection criterion for improving seed yield. The crosses including L41×L56, L41×LF2, L41×L420, Zafar×L56, Zafar×L420, L56×L420 and LF2×L420 with significant positive heterobeltiosis effects for harvest index considered as merit combinations for improving this trait.

Conclusion
Significant mean squares of SCA of the crosses were determined for all the traits except for 1000-seed weight demonstrating that except for 1000-seed weight, non-additive genetic effects had also an important role for controlling the traits. High narrow-sense heritability estimates for 1000-seed weight indicating prime importance of additive genetic effects, therefore, none of crosses had significant heterobeltiosis effect for this trait. The same trend was determined for correlation between both traits based on mean performance and heterobeltiosis. Siliquae on main raceme and siliquae per plant were highly correlated with seed yield from aspect of correlation of mean performance and heterobeltiosis.
Acknowledgements. The author wishes to thank Mazandaran Agricultural and Natural Resources Research and Education Center and Seed and Plant Improvement Based on least significant test (LSD); * and **: significant at 5% and 1% levels of probability, respectively.
Institute (SPII) for providing genetic materials and facilities for conducting the research. * and **: significant at 5% and 1% levels of probability, respectively.