Huang Ze-su,Laosuwan Paisan,Machikowa Thitiporn,Chen Ze-hui,Dai Wen-dong,Tang Rong,and Li Dezhen

1 Guizhou Institute of Oil Crops,Guiyang 550006,China

2 Faculty of Science and Technology,Hatyai University,Hatyai,Songkla 90110,Thailand

3 School of Crop Production,Institute of Agriculture Technology,Suranaree University of Technology,111 University Avenue,Nakhon Ratchasima 30000,Thailand

4 Guizhou Academy of Agriculture Science,Guiyang 550006,China

Introduction

China is the largest producer in both the planting area and total production of rapeseed.Among three main species of rapeseed,Brassica napus L.accounts for about 95% of total planting area in China (Wang et al.,2007).Prior to 1985,pure lines were grown with a commercial scale.Until recently,more than 70% of planted areas are grown to hybrids (Zhou and Fu,2007;Li,1999).Many research institutes in China have been working on the development of hybrid varieties to replace pure line cultivars.

For breeding of hybrid varieties in rapeseed,one of the most important materials is male sterile line.This is because rapeseed is largely a self-pollination and androgynous crop (Rakow and Woods,1987),and it is difficult to obtain hybrid by emasculation.Therefore,using of male sterile system is the most practical method in the production rapeseed hybrid seeds.Male sterile line could be found in nature or might be induced by mutation.It could be transferred from one species or one cultivar to another by backcrossing.For example,Ogu cytoplasmatic male sterile (Ogu CMS) in rapeseed was transferred from a male sterile radish (Barnnerot et al.,1974),Pol CMS was found in a rapeseed variety named Polima (Fu et al.,1995).After the CMS system had been found,several other systems of male sterility such as genetic male sterile (GMS),ecotype sensitive male sterile or environment sensitive male sterile (EMS),gametocide (GC) and self-incompatible (SI) systems were developed or reported (Yu and Hu,2007).

Now male sterile system is widely used in the production of hybrid in rapeseeed.The CMS system is the most popular system.However,in Guizhou Province,China,GMS system is more popular than CMS system.The GMS lines used are of two types,recessive genetic male sterile (RGMS) and dominant genetic male sterile (DGMS).The RGMS is more widely used than DGMS,because RGMS has extensively restorers.Therefore,in the breeding for hybrid varieties,it is necessary that the combining ability of RGMS lines have to be evaluated.These are general combining ability (GCA) and specific combining ability (SCA).

The objective of this study was to evaluate the relative importance of additive and non-additive genetic effects and the combining ability of the RGMS lines of rapeseed (Brassica napus L.) for oil content,seed yield and other characters related to them.

Materials and Methods

Plant materials

Plant materials used in this study consisted of 10 RGMSs with low erucic acid and glucosinolate contents and varied oil contents.These lines were IIAB,Qianyou 3AB,Qianyou 5AB,Qianyou 7AB,Qianyou 8AB,You 2894AB,Qianyou 6AB,QH303-4AB,You 157AB and You 2341AB.These parents were planted in Sept.2006,and crossed in a half diallel (Griffing,Method 2) in spring 2007.A total of 45 F1crosses and 10 parents were obtained.

Field experiment

Experiment was carried out in a randomized complete block design with three replications in Guiyang,Guizhou,China,during Sept.,2007 to May,2008.Plots consisted of two rows of 5-m in length with 45-cm inter-row and 33.3-cm intra-row spacings.Plots were prepared carefully and 600 kg • hm-2N,P and K fertilizers and 15 kg • hm-2borax were applied in hills before sowing.All the 45 crosses and 10 parents were planted in hills on 27,Sept.,2007,and thinned to two plants per hill within 45 days after planting.Each plot contained 60 plants.The total amount of 375 kg • hm-2urea was used by applying in hills for two times,pesticide application was done three times,and weeding was made twice during growing period of rapeseed.Supplement irrigations were made as needed.The crosses and parents were harvested from May 7 to May 19,2008.

Data collections

Ten plants,five male steriles and five male fertiles,were selected randomly and tagged for each plot at flowering.These 10 tagged plants were measured for plant height,branches per plant,pods per plant and seeds per pod at maturity.Means of these characters were used for data analysis.1000-seed weight was measured by using a bulk of seed from each plot.Days to flowering,days to maturity,and yield were based on plot observation.Oil content was analyzed by using open pollinated bulked seeds.Data for characters were recorded as the followings:

Days to flowering (DF): days from sowing until 50% of the plants flowered.

Days to maturity (DM): days from sowing until 90% of the pods matured.

Branches per plant (B/P): productive branches originating from the main stem.

Pods per plant (P/P): productive pods borne on all branches of a plant.

Plant height (PH): main stem length measured from the cotyledonary node to the top of the plant.

Seeds per pod (S/P): Every seed of 12 individual pods per plant.

1000-seed weight (TSW): weight of 1 000 seeds taken randomly,average of three samples.

Where,Y=yield in kg • hm-2;X=moisture content (measured);Ys=standard moisture content (9%);F.W.=harvested yield in kg • plot-1and A=area harvested (m2).

Percentage of oil content (OC): proportion of oil content from dried seeds determined by NIRS.

Statistical analysis

Diallel analysis method of Griffing's Approach Method II Model I (Griffing,1956) was used to analyze for the combining abilities and other parameters of the 10 male sterile lines and their crosses.

The mathematical model for combining ability analysis was as the followings:

Where,u was the population mean;gi(gj) was the GCA effect;sijwas the SCA effect such that sij=sji;andwas the error effect to the ijklth observation;i,j were numbers of parents;k was number of replication and l was number of observation.

GCA and SCA effects and their respective standard errors were estimated as the followings:

To evaluate the relative importance of additive and non-additive genetic effects for each character,the ratio of MSgca/MSsca was used.

The analyses of data for GCA and SCA effects were made by using DPS 9.50 data processing system that copyright belonged to Tang Qiyi,China.

Results

Growing condition

Growing conditions during 2007-2008 were quite unfavorable for rapeseed.Early winter in Guiyang was quite dry and late winter was quite cold due to ice rain during Jan 12,2008-Feb 4,2008.The ice rain damaged early buds.Spring in 2008 was late,but the temperature was good for rapeseed pollination.Therefore,seed yield was not much affected.However,due to the cold weather,the flowering and maturity periods of rapeseed were longer than usual.

Combining ability analysis of variance

The results for analyses of variance for GCA and SCA are presented in Table 1.The analyses indicated that both GCA and SCA variances were highly significant (P＜0.01) for all characters except SCA variance of seeds per pod,which suggested that both additive and non-additive gene effects were important for the expression of these characters.The relative importance of GCA and SCA was judged from the ratio of mean squares GCA to SCA which helped to indicate the predominant presence of either additive and nonadditive effects.The ratio showed that additive gene action was predominant for all characters.These ratios for seed yield (1.48),pods per plant (1.92),seeds per pod(2.25) and plant height (2.44) were quite low,but very high for certain characters such as oil content (20.70).

GCA effects

Estimates of general combining ability for seed yield and other characters of each parent are presented in Table 2.

Highly significant GCA effects for seed yield were found for three RGMS lines.They were lines Qianyou 8AB and You 2894AB with positive GCA effects of 230.94 and 127.65 kg • hm-2,and Qianyou 3AB with negative GCA effects (Table 2),respectively.Although other lines were not significant,positive GCA effects were found in lines Qianyou 6AB and You 157AB.The GCA effects of the parents were found to associate with mean of crosses.These results indicated that Qianyou 8AB and You 2894AB and their crosses should give good yield performance,in contrast,Qianyou 3AB and its crosses should give poor yield performance.

Table1 Mean squares for GCA and SCA effects of nine characters of diallel crosses of rapeseed involving 10 parents

Table2 Estimates of GCA effects for nine characters of rapeseed lines

The highly significant GCA effects for oil content (OC) were found for all RGMS lines except Qianyou 6AB.Among them,lines You 2894AB,QH303-4AB,You 157AB and You 2341AB gave significantly positive GCA effects of 0.99,1.62,1.20 and 1.53% while lines IIAB,Qianyou 3AB,Qianyou 5AB,Qianyou 7AB and Qianyou 8AB gave significantly negative GCA effects (Table 2).The results showed that lines You 2894AB,QH303-4AB,You 157AB,You 2341AB and their crosses should give good oil contents.

Among 10 RGMS lines,the significantly negative GCA effects for pods per plant (P/P) were found for Qianyou 5AB and positive GCA effects for lines You 2894AB,You 157AB and You 2341AB,while no significant GCA effects were found for all other lines (Table 2).The results indicated that You 2894AB,You 157AB,You 2341AB and their crosses gave more pods per plant.

Lines Qianyou 3AB and Qianyou 8AB gave significantly positive GCA effects for seeds per pod (S/P) at P＜0.01,while lines Qianyou 7AB and You 2341AB gave significantly negative GCA effects for the same character at P＜00.05 and P＜00.01,respectively.Other lines were not significant for GCA effects for seeds per pod (Table 2).

For 1000-seed weight,lines IIAB and Qianyou 5AB showed significantly positive GCA effects at P＜0.01,while lines Qianyou 3AB and Qianyou 8AB gave significantly negative GCA effects at P＜0.01 and P＜0.05,respectively.No significant GCA effects were found for other lines (Table 2).The results indicated that lines IIAB,Qianyou 5AB and their crosses gave good performance for 1000-seed weight.

For branches per plant (B/P),significant GCA effects were found for all lines except lines Qianyou 7AB,You 2894AB and Qianyou 6AB.Among them,lines Qianyou 3AB,QH303-4AB,You 157AB and You 2341AB gave positive GCA effects,while lines IIAB,Qianyou 5AB and Qianyou 8AB gave negative GCA effects (Table 2).This indicated that lines Qianyou 3AB,QH303-4AB,You 157AB,You 2341AB and their crosses had more primary branches than others.

Significantly positive GCA effects for plant height were found for lines Qianyou 8AB,You 2894AB and You 2341AB,while significantly negative GCA effects was found for Qianyou 3AB.The results showed that Qianyou 3AB should be good for breeding of short varieties,and lines Qianyou 8AB,You 2894AB and You 2341AB were good for tall varieties.

Lines IIAB,Qianyou 8AB and You 2341AB expressed significantly positive GCA effects for days to flowering (DF),while lines Qianyou 3AB,QH303-4AB and You 157AB gave significantly negative GCA effects for this character.Similar results were found for days to maturity (DM).Lines IIAB,Qianyou 8AB,QH303-4AB and You 2341AB gave significantly positive GCA effects,while lines Qianyou 3AB,Qianyou 5AB and You 157AB gave significantly negative GCA effects (Table 2).Significantly positive GCA effects for these two characters indicated that these lines were good for late flowering and maturity.On the other hand,significantly negative GCA effects were suitable for breeding for early flowering and maturity.

SCA effects

Estimates of SCA effects for seed yield were positively significant in seven crosses as shown in Table 4.The big magnitude of SCA values indicated that lines used in these crosses were of more diverse in seed yield than other crosses,and may be developed to be candidate hybrid varieties.Crosses between lines Qianyou 8A×You 2894B,You 2894A×Qianyou 6B,Qianyou 3A×Qianyou 8B and Qianyou 8A×QH303-4B which gave significant SCA effects were also high yielders among other crosses.

Significantly positive SCA effects for oil content were found in six crosses (Table 4).The cross of Qianyou 7A×Qianyou 6B showed the highest positive SCA effect for this character;however,the mean oil content given in Table 3 was not the highest.However,this combination could be used for line improvement or in case that the level of seed yield was acceptable.

Significantly positive SCA effects were found for pods per plant in four crosses (Table 4).Crosses among lines IIA×You 2894B,Qianyou 3A×Qianyou 8B,Qianyou 8A×QH303-4B and You 2894A×Qianyou 6B gave highly significant SCA effects.These crosses also gave high means of pods per plant(Table 3).This character was associated with seed yield;therefore,it was usually found in high yield lines and crosses.

Table3 Means of nine characters of single cross hybrids of rapeseed

Table4 Estimates of SCA effects for nine characters of rapeseed crosses

Continued

Significantly positive SCA effects of seeds per pod were found in five crosses.The highest SCA effect was found in the cross between lines IIA×QH303-4B (Table 4) which also showed the highest seeds per pod (Table 3).

Estimate effects of SCA were found significant for seed size (TSW).Among 11 crosses,seven showed positive values.The highest SCA effect was found in the cross between lines Qianyou 6A×You 157B (Table 4) which seed size was also significantly larger than others (Table 3).

Significantly positive SCA effects for branches per plant were found in two crosses.The highest SCA effect was found in the cross between lines You 2894A×Qianyou 6B (Table 4) which also showed high branches per plants (Table 3).

For plant height,10 crosses gave significantly positive SCA effects and three crosses gave significantly negative SCA effects (Table 4).The highest positive value was found in the cross of lines Qianyou 5A×You 2341B which also gave high plant height (Table 3).The maximum value of negative SCA effect was found in the cross between lines You 2894A×QH303-4B which also gave relatively lower plant height.The results indicated that crosses with significantly positive and negative SCA effects were good for developing respective tall and short hybrids. For days to flowering,11 crosses gave significantly positive SCA effects,four crosses gave significantly negative SCA effects (Table 4).The highest positive value was found in the cross of lines QH303-4A×You 157B.The maximum value of negative was found in the cross between lines IIA×Qianyou 8B.For days to maturity,significantly positive SCA effects were detected in nine crosses,while significantly negative SCA effects were found in eight crosses (Table 4).The highest positive value was found in the cross of lines Qianyou 3A×QH303-4B.The maximum value of negative SCA effect was found in the cross between lines Qianyou 7A×You 2894B.The results indicated that crosses with significantly positive SCA effects were good for developing late hybrids,while crosses with significantly negative SCA effects were good for developing early hybrids.

Discussion

The variance analyses for GCA and SCA indicated that both GCA and SCA were important for all characters studied in this experiment.However,the ratio of MSgca/MSsca indicated that additive gene effects were more important than non-additive gene effects.The small ratio of MSgca to MSsca for seed yield and pods per plant might indicate the high heterosis of these characters because heterosis related proportionally to non-additive gene actions.For other characters,especially oil content,the ratio of MSgca/MSsca was very high (20.70).The high magnitude of mean square GCA relative to SCA indicated that the diversity within the materials studied was not high and line improvement for these characters was effective.

The analyses of GCA did not show that any single line was a high general combiner for all the characters simultaneously.Outstanding lines such as You 2894AB gave significant GCA estimates for seed yield,oil content,pods per plant and plant height should be favourable for any breeding programs.Moreover,most of the crosses involving You 2894AB were good yielders (Table 3).Therefore,it was the best choice for breeding of rapeseed hybrids.However,Qianyou 8AB was a good combiner for seed yield.It gave significant GCA effects and some of its crosses expressed high yield.Though Qianyou 3AB was not a good combiner for seed yield;it was good for short plant height,early flowering and early maturity;and its cross sometimes also could result in high yield as the cross lines Qianyou 3A×Qianyou 8B (Table 3).It can be used in breeding program for short plant height and early maturity.

Analysis of specific combining ability in this study revealed that a number of crosses showed significant SCA effects for each character,but none showed the best SCA effects simultaneously.Cross lines Qianyou 8A×You 2894B gave the highest seed yield and also gave significant SCA effect for this character which indicated that it should be the best hybrid for seed yield.Cross lines Qianyou 3A×Qianyou 8B gave the highest SCA effect,the yield was also high,and should be the second choice for seed yield.The highest and positive SCA effect for oil content was found in the cross of lines Qianyou 7A×Qianyou 6B,which gave oil content of 38.9%.This indicated that the magnitude of SCA effect might not correspond to the expression of the character as the characters was the sum of mean,GCA and SCA effects and environmental effect.

Most crosses with significant SCA effects for seed yield also showed significant SCA effects for yield related traits.For example,cross lines Qianyou 3A×Qianyou 8B,You 2894A×Qianyou 6B and Qianyou 8A×QH303-4B also showed significant SCA effects for pods per plant.Cross lines IIA×QH303-4B,Qianyou 6A×You 2341B and You 2894A×Qianyou 6B also showed significant SCA effects for seeds per pod.Cross lines IIAB×Qianyou 5AB also showed significant SCA effects for 1000-seed weight.These indicated that there were some relationships between the magnitude of SCA effects for seed yield and that of yield related traits.

Parents of some crosses with high SCA effects were both negative,or one negative and one positive,or both positive in GCA effects.These indicated that high SCA effects could be resulted from any parents with high or low GCA effects.This result was not agreed with the report of Sheikh et al.(1998) who found that high SCA effects of crosses in Brassica juncea was resulted from high GCA effect parents crossed with low GCA effects parents.

The crosses of lines Qianyou 3A×You 157B and Qianyou 3A×Qianyou 7B showed significantly negative SCA effects,and both crosses gave the shortest days to flowering and days to maturity,which indicated that these two crosses could be used for improving early flowering and early maturity lines.

Barnnerot H,Loulidard L,Cauderon Y,et al.1974.Transfer of cytoplasmic male sterility from Raphanus stativus to Brassica oleracea.Proc Eucarpia Meeting-Crucifera Crop Sect,25: 52-54.

Fu T D,Yang G S,Yang X N,et al.1995.Discovery,study and utilization of cytoplasm male sterility Polima in Brassica napus L.Improvement in Nature Science-Communication of National Key Laboratory,5(3): 287-293.

Griffing B.1956.Concept of general and specific combining ability in relation to diallel crossing systems.Aust J Biol Sci,9: 463-493.

Li D R.1999.Hybrid seed production in rapeseed.In: Basra A S.Heterosis and hybrid seed production in agronomic crops.Food Products Press,New York-London-Oxford,pp.217-260.

Rakow G,Woods D L.1987.Outcrossing in rape and mustard under saskatchewan prairie conditions.Can J Plant Sci,67: 147-151.

Sheikh I A,Singh J N.1998.Combining ability of seed yield and oil content in Brassica juncea.(L.).Indian J Genet,58(4): 507-511.

Wang H Z,Guan C Y,Zhang C L.2007.Studies on rapeseed production and cultivation science ＆technology in China.Proc.12thInt.Rapeseed Congress,Wuhan,China.pp.2-7.

Yu C Y,Hu S W.2007.Utility of heterosis of two-line hybrid rapeseed in China: approaches and achievements.Proc.12th Int.Rapeseed Congress,Wuhan,China.pp.128-129.

Zhou Y M,Fu T D.2007.Genetic improvement of rapeseed in China.Proc.12thInt.Rapeseed Congress,Wuhan,China.pp.210-212.