Open Access

Dissecting combining ability effect in a rice NCII-III population provides insights into heterosis in indica-japonica cross

Rice201710:39

https://doi.org/10.1186/s12284-017-0179-9

Received: 18 June 2017

Accepted: 21 August 2017

Published: 29 August 2017

Abstract

Background

Combining ability is a measure for selecting elite parents that make the highest contributions to hybrid performance. However, the genetic bases of combining ability and how they contributed to heterosis is seldomly known.

Results

We constructed a both NCII and NCIII population derived from an indica-japonica cross to study the relationship among parental performance, combining ability and hybrid performance of 11 agronomic traits. Among them, specific combining ability is more important to grain yield than parental performance and general combining ability. We performed linkage analyses to phenotypic values and combining ability of all 11 traits in Doubled haploid lines and its two backcross populations and identified 108 QTLs in total. Among these QTLs, four known loci, Sd1, Ghd7, Ghd8 and DEP1 contribute a lot to GCA effects of agronomic traits except grain yield and seed setting rate. Three QTLs, Ghd8, S5 and qS12, contribute a lot to SCA effects of grain yield and present overdominace.

Conclusions

Our study provides insights into the genetic bases of combining ability and heterosis and will promote the improvements of indica-japonica hybrid breeding.

Keywords

Oryza sativa L.Combining abilityMating design indica-jaonica crossLinkage analysisHeterosis

Background

The phenomenon heterosis has been applied to rice breeding for improving grain yield since the early 1970s in China (Cheng et al. 2007; Darwin 1876; Li et al. 2007). To breed an ideal cross with the highest grain yield and great stress resistance is the ultimate goal of hybrid rice breeders. As parents with excellent agronomic traits do not always pass those traits on to their progenies, breeders often test the potential of a selfing line by cross it to several other lines. The potential for creating high heterosis progenies of an inbred line is called combining ability, a concept proposed by Sprague and Tatum and has been widely used in cross breeding for elite parent selection (Comstock and Robinson 1948; Griffing 1956; Sprague and Tatum 1942). Parent lines with high general combining ability (GCA) in grain yield and resistance to pests and diseases are more likely to form hybrids with satisfactory performance. On the other hand, some parent lines of general GCA also is able to form excellent hybrids and this is caused by another special effect called special combining ability (SCA). The both effects of GCA and SCA are able to create hybrids with high heterosis but the genetic bases of these effects are largely unknown.

The first attempt to unveil the genetic basis of combining ability was done by Griffing (1956). He proposed the methods of using diallel-cross to dissect the genetic variance into additive variance and non-additive variance, and estimated the GCA and SCA effects. His study provide theory basis for estimating genetic variance and combining ability effect in various mating design, including complete or incomplete diallel (Griffing 1956), North Carolina design (Comstock et al. 1949), and top crossing (Hill et al. 1998). Later study conducted transcriptome analysis and molecular markers to reveal the relationship between combining ability and genetic diversity (Ajmone et al. 1998; Bernardo 1996; Frisch et al. 2010; Stupar et al. 2008). Those studies revealed that the genetic distance between two parents are positively correlated with combining ability and hybrid performance.

Recently, with the development of molecular marker, linkage analysis has been used to dissect the GCA effects into quantitative trait locus (QTL) (Lv et al. 2012; Qi 2013; Qu 2012). Liu et al. (2015) utilized a NCII design, performed linkage analyses to GCA, and confirmed that Ghd7 and OsPRR37 (Koo et al. 2013; Xue et al. 2008) are major genes for GCA of heading date, plant height and spikelet per panicle in rice. Though a few QTLs contributed to combining ability has been indentified, the genetic bases of combining ability are still not clear and how they contributed to hybrid performance was totally unknown. In this study, we developed a both NCII and NCIII population (see Methods) derived from an indica-japonica cross, performed linkage analysis to both GCA and SCA effects, and explored how QTLs of combining ability contributed to hybrid performance of hybrid rice.

Results

The design of a both NCII and NCIII population

The F1 generation of an indica line ZS97B (ZS97) and a japonica line Wuyugeng2 (WYG) has heterosis in PH, HD, KGW and SP, but has hybrid weakness in SS (Table 1). This hybrid sterility phenomenon leads to the lower YD of F1 than both parents. To dissect the genetic basis of the heterosis and hybrid weakness of ZS97 × WYG, we constructed the both NCII and NCIII population using the doubled haploid lines (DHs) of ZS97 × WYG as male parents and taking the two parents and P64S as female parents (Fig. 1). By crossing the 190 DHs to the three tester lines, we finally get 149 F1 for ZS97 × DHs, 143 F1 for WYG × DHs and 145 F1 for P64S × DHs.
Table 1

The performances of an indica-japonica cross in 11 agronomic traits

 

GF

PH

HD

YD

TP

KGW

SP

GP

SS

PL

SDEN

ZS97

90.7

79.3

69.5

23.1

12.7

21.7

109.6

83.7

76.4

19.5

5.6

WYG

99.6

84.2

99.0

44.7

16.8

24.9

132.0

107.1

81.2

15.9

8.3

F1

99.7

121.1

109.0

22.6

14.2

27.0

147.4

58.8

39.9

19.9

7.4

GF grian-filling degree, PH plant height, HD heading date, YD yield, TP number of tillers per panicle, KGW grain weight per 1000 grains, SP spikelet per panicle, GP number of grains per panicle, SS seed-setting rate, PL panicle langth, SDEN grain density per panicle

Fig. 1

The experimental design and analysis procedure used in this study

Performance of the populations.

The phenotypic distributions of 11 agronomic traits measured in the DHs and their TC progenies are shown in Fig. 2. All these traits varied widely in the DHs and three TC populations and most of these traits showed normal distributions expect GF and SS.
Fig. 2

The phenotypic distributions of 11 agronomic traits in the DH population and their TC progenies. GF, grian-filling degree; PH, plant height; HD, heading date; YD, yield; TP, number of tillers per panicle; KGW, grain weight per 1000 grains; SP, spikelet per panicle; GP, number of grains per panicle; SS, seed-setting rate; PL, panicle langth; SDEN, grain density per panicle

The average levels of most traits, except HD, KGW and SS, in the TC populations were higher than that in the DHs. The average HD in DHs was higher than that of ZS97 × DHs, while higher than that of the other two TC populations. The average SS in ZS97 × DHs and WYG × DHs were both lower than that in DHs. These differences indicate that the three TC populations might have a high level of special combining ability and heterosis.

Relationship between traits values in DHs and TC populations

Table 1 shows the correlations among the phenotypic values and GCA effects of DHs, the phenotypic values and SCA effects of TC populations for the 11 agronomic traits. The phenotypes of DHs are significantly and positively correlated with GCA effects of DHs in most traits expect YD and SS. This indicated the genetic bases of GCA and trait value for YD and SS may be quite different. Both GCA and SCA effects are positively correlated with phenotypic values of all traits in all three TC populations, indicates that both GCA and SCA contribute a lot to hybrid performance.

Variance analysis of combining ability

We determined the effects of GCA (DHs or Testers) and SCA (Testers × DHs) by two-factor variance analysis and found these effects were significant in all traits. This also indicated that both kinds of gene effects were important for the inheritance of agronomic traits. We further calculated the additive variance (VA), non-additive variance (VNa) and narrow sense heritability (h 2 N ) for each trait. These traits with h 2 N larger than 0.5 indicates they are mainly controlled by additive effects, as PH, HD and KGW; conversely, those with h 2 N below 0.5 are mainly controlled by non-additive effects, as GF, YD and SS. This also indicates over-dominance contribute a lot to the heterosis of GF, YD and SS.

QTL mapping for combining ability

We applied linkage analysis to both phenotypic value and GCA effect of 190 DHs for 11 agronomic traits and identified many known and unknown QTLs for these traits (Fig. 3 and Additional file 1: Figure S1). Most loci detected in analysis of phenotype were also detected in GCA and we found these QTLs often explained larger proportion of phenotypic variations in DH population. Several QTLs for PH and HD also contribute to YD in DHs, such as Hd1 and Ghd8 (Yan et al. 2011; Yano et al. 2000). However, they have no effect on GCA and we detected no QTL for YD in analysis of GCA. This may be due to the low h 2 N of YD in our NCII population. Locations of QTLs for GCA on linkage map are show in Fig. 4 and detailed information is show in Additional file 2: Table S1.
Fig. 3

The linkage mapping of phenotypic values and GCA effect for 11 agronomic traits in DHs. For a description of agronomic traits see materials and methods

Fig. 4

The QTLs detected for GCA and SCA effect in NCII population. For a description of agronomic traits see materials and methods

We further performed linkage analysis to both phenotype and SCA effect for two BC populations (ZS97 × DHs and WYG × DHs) and indentified 66 QTLs for 11 agronomic traits. Most QTLs identified in BC populations were identified in DHs but with lower significance. Two new QTLs for SS and YD were detected on chromosome 6 and chromosome 12 in both phenotypic value and SCA effect of both BC populations. The QTL on chromosome 6 cover the known hybrid sterility gene S5 and has similar genetic effect (Yang et al. 2012). QTL spanning Ghd8 were identified for SP and YD in both phenotypic value and SCA effect of both BC populations. By sequencing these genes in ZS97 and WYG (Additional file 1: Figure S1), we found that ZS97 carried a S5-i allele of S5 and non-functional allele of Ghd8, and WYG carried a S5-j allele of S5 and functional allele of Ghd8 (Yan et al. 2011; Yang et al. 2012). So the two QTLs for SCA of grain yield are actually S5 and Ghd8. And the QTL on chromosome 12 is a new QTL for SS and YD. The location and detailed information for these QTLs are show in Fig. 4 and Additional files 3 and 4: Tables S2 and S3.

Three QTLs for SCA effect of grain yield

We analyzed the effects of S5, qS12 and Ghd8 on grain yield in both BC populations and identified the overdominant effect of these QTLs. In ZS97 × DHs (Fig. 5a), heterozygous genotypes (H) of S5 and qS12 both show lower grain yield than correspondent homozygous ZS97 genotype (A); similarly in WYG × DHs (Fig. 5b), heterozygous genotypes (H) of S5 and qS12 both show lower grain yield than correspondent homozygous WYG genotype (B). This indicates S5 and qS12 show hybrid weakness in grain yield. On the contrary, the heterozygous genotype of Ghd8 show higher grain yield than both homozygous genotypes and that indicates Ghd8 show hybrid vigour in grain yield (Fig. 5a, b).
Fig. 5

The effects of S5, qS12 and Ghd8 on agronomic traits in two BC populations. a The average grain yield of ‘A’ and ‘H’ genotype of the three QTLs in ZS97 × DHs; b The average grain yield of ‘B’ and ‘H’ genotype of the three QTLs in WYG × DHs; c The average seed setting rate of ‘A’ and ‘H’ genotype and combination genotype of S5 and qS12 in ZS97 × DHs; d The average spikelet number per panicle of ‘A’ and ‘H’ genotype of Ghd8 in ZS97 × DHs; e The average seed setting rate of ‘B’ and ‘H’ genotype and combination genotype of S5 and qS12 in WYG × DHs; f The average spikelet number per panicle of ‘B’ and ‘H’ genotype of Ghd8 in WYG × DHs; ‘A’ and ‘B’ represent the different homozygous allelic type from ZS97 and WYG, respectively. ‘H’ represents the hybrid allelic type of the two parents

Dosage effects of QTLs on yield traits.

S5 and qS12 both effected on SS and YD in our NCII population and they also show dosage effect on SS and YD (Fig. 5c, e). The both homozygous genotype (AA or BB) show much higher SS and YD than the both heterozygous genotype (HH). The heterozygous genotype of Ghd8 has more spikelet number than homozygous genotype which lead to the heterozygous advantage in grain yield (Fig. 5d, f). Though the genetic effect of Ghd8 is different to S5 and S12, the three genes show dosage effect on YD. Comparing the top 10 and bottom 10 lines of grain yield in BC populations, we found that advantage genotypes were accumulated in the top 10 lines and disadvantage genotypes were accumulated in the bottom 10 lines (Fig. 6). The F1 of ZS97 × WYG has heterozygous genotypes in all S5, S12 and Ghd8, and the weakness in seed setting rate weight more than the heterosis in total spikelet number and grain weight (Table 1).
Fig. 6

The advantage genotypes of S5, qS12 and Ghd8 were accumulated in the top 10 lines and the disadvantage genotypes were accumulated in the bottom 10 lines. For S5 and qS12, homozygous genotypes are advantageous. For Ghd8, heterozygous genotype is advantageous

Discussion

In this study, GCA and SCA effects in an indica-japonica cross were estimated using a NCII population. It is interesting to note that, the performance of GCA was not significantly or strongly (−0.35 < r < 0.35) correlated to the performance of DHs per se in several traits (Table 2). And these traits all have low heritability in our NCII population (Table 3). So traits with low heritability, as GF, YD and SS, are hardly to be predicted by parental performance. Weak correlations between the performance of inbred lines and their GCA effects were also detected in a previous study (Lv et al. 2012), so the performance of testcrosses was important for the evaluation and selection of elite inbred lines. On the other hand, only QTLs with large effects on agronomic traits can pass their effects from parents to hybrids. The major QTLs, sd1, Ghd8 and DEP1 (Sasaki et al. 2002; Yan et al. 2011; Yano et al. 2000), detected in DHs were also major QTLs for GCA effects (Fig. 3). However, several QTLs (Hd1, Ghd7 and Hd2) have large effects in DHs but have little effects in GCA effects (Koo et al. 2013; Xue et al. 2008; Yano et al. 2000). And other QTLs have little effects in DHs can hardly be detected in GCA effects. These differences among QTLs also reflect the different genetic basis between parental performance and GCA.
Table 2

Phenotypic correlation (r) coefficients for agronomic traits in DHs and TC populations

Correlations

GF

PH

HD

YD

TP

KGW

SP

GP

SS

PL

SDEN

DHs and GCADHs

0.32**

0.86**

0.75**

0.06

0.36**

0.55**

0.66**

0.38**

0.12

0.67**

0.73**

ZS97 × DHs and DHs

0.34**

0.83**

0.83**

0.42**

0.40**

0.58**

0.69**

0.57**

0.36**

0.69**

0.79**

ZS97 × DHs and GCADHs

0.78**

0.89**

0.83**

0.50**

0.82**

0.77**

0.68**

0.47**

0.48**

0.87**

0.74**

ZS97 × DHs and SCAZS97×DHs

0.78**

0.52**

0.44**

0.80**

0.73**

0.40**

0.57**

0.70**

0.79**

0.38**

0.42**

WYG × DHs and DHs

0.23*

0.60**

0.44**

−0.31**

0.22*

0.44**

0.31**

−0.17*

−0.16*

0.56**

0.34**

WYG × DHs and GCADHs

0.52**

0.76**

0.48**

0.67**

0.64**

0.65**

0.68**

0.58**

0.67**

0.71**

0.69**

WYG × DHs and SCAWYG×DHs

0.74**

0.17*

0.17*

0.81**

0.62**

0.33**

0.50**

0.74**

0.85**

0.25**

0.48**

P64S × DHs and DHs

0.25**

0.82**

0.57**

−0.05

0.22*

0.58**

0.43**

0.33**

0.11

0.62**

0.59**

P64S × DHs and GCADHs

0.73**

0.91**

0.88**

0.62**

0.70**

0.73**

0.79**

0.64**

0.59**

0.87**

0.83**

P64S × DHs and SCAP64S×DHs

0.63**

0.33**

0.61**

0.58**

0.66**

0.22*

0.65**

0.46**

0.24*

0.31**

0.63**

GCADHs and SCAZS97×DHs

0.22*

0.21*

0.17*

−0.08

0.24*

0.02

−0.08

−0.10

−0.11

0.03

−0.13

GCADHs and SCAWYG×DHs

−0.20*

−0.35**

−0.56**

0.19*

−0.18*

−0.15*

−0.15*

0.08

0.29**

−0.26**

−0.15*

GCADHs and SCAP64S×DHs

−0.06

0.10

0.36**

−0.21*

−0.07

−0.11

0.24*

−0.07

−0.45**

0.00

0.26**

GCADHs denote the general combining ability of DH lines; SCAZS97×DHs, SCAWYG×DHs and SCAP64S×DHs indicate the special combining ability for hybrids of ZS97 × DHs, WYG × DHs and P64S × DHs

GF grian-filling degree, PH plant height, HD heading date, YD yield, TP number of tillers per panicle, KGW grain weight per 1000 grains, SP spikelet per panicle, GP number of grains per panicle, SS seed-setting rate, PL panicle langth, SDEN grain density per panicle

* P < 0.05, ** P < 0.01

Table 3

Variance analysis of combining ability and genetic parameters estimation of agronomic traits

Trait

DHs

Testers

Testers × DHs

VA

VNa

h 2 N

GF

37**

193**

27**

1.4

9.15

0.13

PH

1116**

18515**

230**

138.01

76.67

0.64

HD

924**

47864**

366***

164.84

121.91

0.57

YD

326**

8498**

392**

10.19

130.64

0.07

TP

35**

157**

19**

2.03

6.33

0.24

KGW

25**

3626**

9**

9.57

3.13

0.75

SP

3866**

140348**

2155**

489.26

718.35

0.41

GP

2308**

183467**

2793**

337.13

931.13

0.27

SS

618**

21437**

846**

19.26

281.87

0.06

PL

23**

823**

6**

3.68

1.89

0.66

SDEN

11**

419**

5**

1.57

1.59

0.5

V A , additive variance, V Na non-additive variance, h2 N narrow-sense heritability, GF grian-filling degree, PH plant height, HD heading date, YD yield, TP number of tillers per panicle, KGW grain weight per 1000 grains, SP spikelet per panicle, GP number of grains per panicle, SS seed-setting rate, PL panicle langth, SDEN grain density per panicle

** P < 0.01

Indica-japonica cross were reported to has higher heterosis than indica-indica cross and japonica-japonica cross (Yuan 1994). Although indica generally has better yield performance, japonica carries many beneficial alleles that are uncommon in indica gene pools (for example, DEP1, IPA1 and NAL1) (Fujita et al. 2013; Huang et al. 2009; Jiao et al. 2010). Positive partial dominance and overdominance effects have served as the major causes of heterosis in indica-japonica F1 hybrids (Huang et al. 2016). Matting design in classic genetics helps to dissect the contributions of different genetic effects and QTL mapping in molecular genetics helps to identify important loci for hybrid performance (George 2012; Zeng 1994). We combined both methods of classic and molecular genetics to detect alleles of additive and non-additive effects contribute to inter-subspecies hybrid performance. This new method is better than simply perform QTL mapping to hybrid performance, as we identified more loci with higher significances and more clear effects (Additional files 2, 3, and 4: Tables S1-S3). For example, sd1 is the major locus with additive effects (a = −7.54) on plant height of hybrids and Ghd8 is the major locus with overdominace effect on grain yield of hybrids (Sasaki et al. 2002; Yan et al. 2011). The same locus may have different effect on different traits as Ghd8 is also a major locus with additive effect (a = −9.42) on heading date of hybrids. The S5 (Yang et al. 2012) and qS12 are two loci conferring the wide-compatibility between indica and japonica. The two QTLs will accelerate the development of intersubspecific hybrids with high heterosis.

Prediction of hybrid performance is important in hybrid rice breeding. Heterosis and combining ability are two main indexes for hybrid performance. In this study, we dissect the hybrid performance of ZS97 × WYG and the NCII population in to GCA and SCA effects (Table 3). Among 11 agronomic traits, PH, HD, KGW and PL were mainly controlled by GCA effects, and major QTLs were detected for these traits (sd1, Ghd8, qKGW2 and DEP1 (Che et al. 2015; Huang et al. 2009; Sasaki et al. 2002; Yan et al. 2011)) (Figs. 3 and 4). On the contrary, GF, YD and SS were mainly controlled by SCA effects, and three QTLs (S5, qS12 and Ghd8) contributed to these traits (Fig. 4). With the knowledge of these information, we will be able to predict the hybrid performance of an indica-japonica cross.

Conclusion

We dissected the effects of GCA and SCA in an indica-japonica cross and identified lots of known and unknown QTLs for them. Among these QTLs, Ghd8, S5 and qS12 largely contributed to the hybrid grain yield of indica-japonica cross. These results provide insights into the genetic bases of combining ability and heterosis and will provide valuable information for the improvements of indica-japonica hybrid breeding.

Methods

Materials and field planting

The NCII and NCIII populations in our study were constructed according to the North carolina design II and III (George 2012). In a NCII design, each member of a group of parents used as males is mated to each member of another group of parents used as females. NCII design is used to evaluate GCA for inbred lines and SCA for every cross. In a NCIII design, a random sample of F2 plants (as well as RILs and DH lines) is backcrossed to the two inbred lines from which the F2 was descended.

Our study included 3 rice mapping populations (Fig. 1). The first one is a set of 190 Doubled haploid lines (DHs) derived from the anther culture of the F1 from a cross between an indica variety ZS97 and a japonica variety WYG (Jiang et al. 2004). Subsequently, taking the two parents as two testers, two BC populations were developed by crossing all 190 DHs to each parent. The gather of the two BC populations is also called a NCIII population. We took an indcia variety P64S as the third tester, and cross all 190 DHs to P64S. Finally, the NCII population was composed of the three testcross (TC) populations. The parents of the DHs (ZS97 and WYG), the F1 hybrid (ZS97 × WYG), were used as control.

The phenotypic performance was evaluated at the experimental field of Huazhong Agricultural University, Wuhan, China. All materials for these populations and parents were sown in the seedling nursery, and 27-day-old seedlings were transplanted into tow-row plots, 10 plants with 16.5 cm of space between plants with a row and 26.4 cm of space between the rows. The plots were arranged in a randomized complete block design with two replications.

Trait measurements and statistical analyses

Agronomic traits measured in this study included grain yield per plant (YD), number of tillers per plant (TP), number of grains per panicle (GP), 1000-grain weight (KGW), spikelet per panicle (SP), panicle length (PL), grain density per panicle (SDEN), grain-filling degree (GF), and seed-setting rate (SS). Trial means were determined from eight randomly selected plants in the middle of the rows of all lines. The former four traits, YD, TP, GP, and KGW, were essentially as described previously by Yu et al. (1997). The SP was scored as the total number of spikelets divided by the number of reproductive tillers of a plant. The PL was measured as the average length from the bottom neck of three main panicles to their tips for each plant. The SDEN was scored as the number of grains divided by the PL, with average grain number per centimeter for three main panicles representative of each plant. The GF was scored as the percentage of the average weight of a single fertilized grain compared to the weight of single grain with mass density > 1 in each plant, essentially as previously described by Niu et al. (2004) and Zhu et al. (1995). SS was scored as the number of grains divided by the total number of spikelets from the reproductive tillers of a plant. Phenotypic distribution of 11 agronomic traits were drew by boxplot() function of R software (Ihaka and Gentleman 1996). P-values for phenotypic coefficients were calculated with a two-sided t-test using cor.test() function in R.

Genotyping and linkage map construction

A linkage map consisting of 179 SSR markers (Fig. 4) covering the 12 chromosomes with a total length of 1849.4 cM (an average length of 9.4 cM) between two markers was constructed as described by Jiang et al. (2004). All the 179 markers are public makers and their sequence can be queried on GRAMENE website (http://www.gramene.org/).

Variance analysis and QTL mapping

Two-factor variance analysis were performed to the NCII population to calculated the effects of DHs, testers and DHs × testers. The GCA variance effects of the parents and the SCA variance effects of the hybrids were estimated by the fixed model described by Mo (1982).

The mathematical representation of the relationship between phenotype and combining ability for each cross is: Y ij =‾Y + G i  + Gj + S ij ; where Y ij is the phenotype value of the hybrid derived from the i th male parent and j th female parent,‾Y is the mean phenotype value of all hybrids, G i is the general combining ability (GCA) of the i th male parent, G j is the GCA of the j th female parent, S ij is the specific combining ability (SCA) of the hybrid derived from the i th male parent and the j th female parent. As Y i is the mean phenotype of the hybrid derived from the i th male parent and Y j is the mean phenotype of the hybrid derived from the j th female parent, the combining ability was calculated by the following equations:
$$ {\mathrm{G}}_i={\overline{\mathrm{Y}}}_i-\overline{\mathrm{Y}};{\mathrm{G}}_j={\overline{\mathrm{Y}}}_j-\overline{\mathrm{Y}};{\mathrm{S}}_{ij}={\overline{\mathrm{Y}}}_{ij}-{\overline{\mathrm{Y}}}_i-{\overline{\mathrm{Y}}}_j-\overline{\mathrm{Y}} $$

QTL analysis was performed separately for the DH and BC populations. For the DH population, the mean trait values and GCA effect were used as input data. For each of the BC populations, the mean trait values and SCA effect of the BC hybrids were used as input data. The SCA data corresponding to the ZSF1 population were referred to as SCAZS97, and the data from the WYF1 population were referred to as SCAWYG. Analyses of main-effect QTL (M-QTL) was conducted in each mapping population by composite interval mapping using WinQTLCart version 2.5 software (Zeng 1994). In the analyses, the likelihood ratio (LR) and t test were combined to test the significance of the single-locus QTL additive effects. The LR and t values corresponding to P = 0.001, equivalent to LOD (log likelihood value) ≥3.0 threshold according to calculated results, were used as the threshold for claiming the putative M-QTLs. The peak points of the LR in the linkage map were considered the putative positions of the QTLs. The relative contribution of a genetic component (R 2 ; %) was calculated as the proportion of phenotypic variance explained by that component in the selected model.

Candidate gene confirmation

Many QTLs were colocated with known genes and have similar effects on agronomic traits. To confirm whether these QTLs were identical to these cloned genes, we sequenced these genes in ZS97B and Wuyugeng2 and compared their allelotype (Additional file 1: Figure S1) according to previous studies (Huang et al. 2009; Kojima et al. 2002; Koo et al. 2013; Sasaki et al. 2002; Xue et al. 2008; Yan et al. 2011; Yang et al. 2012; Yano et al. 2000).

Abbreviations

DHs: 

double haploid lines

GCA: 

general combining ability

QTLs: 

quantitative trait locus

SCA: 

specific combining ability

Declarations

Acknowledgements

This work was supported by grants from the National Natural Science Foundation (91435202, 91635302), the National Program on R&D of Transgenic Plants (2016ZX08001002-002), the National 863 Project (2012AA10A303) and the earmarked fund for the China Agriculture Research System (CARS-01-03) of China and the Bill & Melinda Gates Foundation.

Author contributions

HZ conducted analyzed the data and wrote the paper. DX, JZ and GHJ participated in the field management and logistic work. YH designed and supervised the study.

Competing interests

The authors declare that they have no competing financial interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Authors’ Affiliations

(1)
National Key Laboratory of Crop Genetic Improvement and National Centre of Plant Gene Research, Huazhong Agricultural University

References

  1. Ajmone MP, Castiglioni P, Fusari F, Kuiper M, Motto M (1998) Genetic diversity and its relationship to hybrid performance in maize as revealed by RFLP and AFLP markers. Theor Appl Genet 96:219–227View ArticleGoogle Scholar
  2. Bernardo R (1996) Testcross additive and dominance effects in best linear unbiased prediction of maize single-cross performance. Theor Appl Genet 93:1098–1102View ArticlePubMedGoogle Scholar
  3. Che R, Tong H, Shi B, Liu Y, Fang S, Liu D, Xiao Y, Hu B, Liu L, Wang H (2015) Control of grain size and rice yield by GL2-mediated brassinosteroid responses. Nature plants 2:15195View ArticlePubMedGoogle Scholar
  4. Cheng SH, Zhuang JY, Fan YY, Du JH, Cao LY (2007) Progress in research and development on hybrid rice: a super-domesticate in China. Ann Botany 100:959–966View ArticleGoogle Scholar
  5. Comstock RE, Robinson H, Harvey P (1949) Breeding procedure designed to make maximum use of both general and specific combining ability. Agron J 41:360–367View ArticleGoogle Scholar
  6. Comstock RE, Robinson HF (1948) The components of genetic variance in populations of biparental progenies and their use in estimating the average degree of dominance. Biometrics 4:254–266View ArticlePubMedGoogle Scholar
  7. Darwin C (1876) The effects of cross and self fertilisation in the vegetable kingdom. John Murray, LondonView ArticleGoogle Scholar
  8. Frisch M, Thiemann A, Fu J, Schrag TA, Scholten S, Melchinger AE (2010) Transcriptome-based distance measures for grouping of germplasm and prediction of hybrid performance in maize. Theor Appl Genet 120:441–450View ArticlePubMedGoogle Scholar
  9. Fujita D, Trijatmiko KR, Tagle AG, Sapasap MV, Koide Y, Sasaki K, Tsakirpaloglou N, Gannaban RB, Nishimura T, Yanagihara S (2013) NAL1 allele from a rice landrace greatly increases yield in modern indica cultivars. Proc Natl Acad Sci U S A 110:20431–20436View ArticlePubMedPubMed CentralGoogle Scholar
  10. George A (2012) Principles of plant genetics and breeding. John Wiley & Sons, New YorkGoogle Scholar
  11. Griffing B (1956) Concept of general and specific combining ability in relation to diallel cross systems. AustJBiolSci 9:463–493Google Scholar
  12. Hill J, Becker HC, Tigerstedt PM (1998) Quantitative and ecological aspects of plant breeding. Plant BreedingView ArticleGoogle Scholar
  13. Huang X, Qian Q, Liu Z, Sun H, He S, Luo D, Xia G, Chu C, Li J, Fu X (2009) Natural variation at the DEP1 locus enhances grain yield in rice. Nat Genet 41:494–497View ArticlePubMedGoogle Scholar
  14. Huang X, Yang S, Gong J, Zhao Q, Feng Q, Zhan Q, Zhao Y, Li W, Cheng B, Xia J (2016) Genomic architecture of heterosis for yield traits in rice. Nature 537:629–633View ArticlePubMedGoogle Scholar
  15. Ihaka R, Gentleman R (1996) R: a language for data analysis and graphics. J Comp &Graph Stat 5:299–314Google Scholar
  16. Jiang GH, He YQ, Xu CG, Li XH, Zhang Q (2004) The genetic basis of stay-green in rice analyzed in a population of doubled haploid lines derived from an indica by japonica cross. Theor Appl Genet 108:688–698View ArticlePubMedGoogle Scholar
  17. Jiao Y, Wang Y, Xue D, Wang J, Yan M, Liu G, Dong G, Zeng D, Lu Z, Zhu X (2010) Regulation of OsSPL14 by OsmiR156 defines ideal plant architecture in rice. Nat Genet 42:541–544View ArticlePubMedGoogle Scholar
  18. Kojima S, Takahashi Y, Kobayashi Y, Monna L, Sasaki T, Araki T, Yano M (2002) Hd3a, a rice ortholog of the Arabidopsis FT gene, promotes transition to flowering downstream of Hd1 under short-day conditions. Plant & Cell Physiology 43:1096View ArticleGoogle Scholar
  19. Koo BH, Yoo SC, Park JW, Kwon CT, Lee BD, An G, Zhang Z, Li J, Li Z, Paek NC (2013) Natural variation in OsPRR37 regulates heading date and contributes to rice cultivation at a wide range of latitudes. Mol Plant 6:1877–1888View ArticlePubMedGoogle Scholar
  20. Li SQ, Yang DC, Zhu YG (2007) Characterization and use of male sterility in hybrid rice breeding. J Integr Plant Biol 49:791–804View ArticleGoogle Scholar
  21. Liu C, Song G, Zhou Y, Qu X, Guo Z, Liu Z, Jiang D, Yang D (2015) OsPRR37 and Ghd7 are the major genes for general combining ability of DTH, PH and SPP in rice. Sci Rep 5:12803View ArticlePubMedPubMed CentralGoogle Scholar
  22. Lv AZ, Zhang H, Zhang ZX, Tao YS, Yue B, Zheng YL (2012) Conversion of the statistical combining ability into a genetic concept. J Integr Agr 11:43–52View ArticleGoogle Scholar
  23. Mo H (1982) The analysis of combining ability in p×q mating pattern. Journal of Yangzhou UniversityGoogle Scholar
  24. Niu AL, Lu XL, Song C (2004) Mapping of QTLs controlling grain plumpness of rice using doubled haploid lines. Journal of Wuhan Botanical Research 22:477–481Google Scholar
  25. Qi H (2013) Identification of combining ability loci for five yield-related traits in maize using a set of testcrosses with introgression lines. Theor Appl Genet 126:369–377View ArticlePubMedGoogle Scholar
  26. Qu Z (2012) QTL mapping of combining ability and heterosis of agronomic traits in rice backcross recombinant inbred lines and hybrid crosses. P Lo S One 7:e28463View ArticleGoogle Scholar
  27. Sasaki A, Ashikari M, Ueguchi-Tanaka M, Itoh H, Nishimura A, Swapan D, Ishiyama K, Saito T, Kobayashi M, Khush GS (2002) Green revolution: a mutant gibberellin-synthesis gene in rice. Nature 416:701–702View ArticlePubMedGoogle Scholar
  28. Sprague GF, Tatum LA (1942) General vs specific combining ability in single cross of corn. J Amer Soc Agron 34:923–932View ArticleGoogle Scholar
  29. Stupar RM, Gardiner JM, Oldre AG, Haun WJ, Chandler VL, Springer NM (2008) Gene expression analyses in maize inbreds and hybrids with varying levels of heterosis. BMC Plant Biol 8:33View ArticlePubMedPubMed CentralGoogle Scholar
  30. Xue W, Xing Y, Weng X, Zhao Y, Tang W, Wang L, Zhou H, Yu S, Xu C, Li X, Zhang Q (2008) Natural variation in Ghd7 is an important regulator of heading date and yield potential in rice. Nat Genet 40:761–767View ArticlePubMedGoogle Scholar
  31. Yan WH, Wang P, Chen HX, Zhou HJ, Li QP, Wang CR, Ding ZH, Zhang YS, Yu SB, Xing YZ, Zhang QF (2011) A major QTL, Ghd8, plays pleiotropic roles in regulating grain productivity, plant height, and heading date in rice. Mol Plant 4:319–330View ArticlePubMedGoogle Scholar
  32. Yang J, Zhao X, Cheng K, Du H, Ouyang Y, Chen J, Qiu S, Huang J, Jiang Y, Jiang L (2012) A killer-protector system regulates both hybrid sterility and segregation distortion in rice. Science 337:1336–1340View ArticlePubMedGoogle Scholar
  33. Yano M, Katayose Y, Ashikari M, Yamanouchi U, Monna L, Fuse T, Baba T, Yamamoto K, Umehara Y, Nagamura Y (2000) Hd1, a major photoperiod sensitivity quantitative trait locus in rice, is closely related to the Arabidopsis flowering time gene CONSTANS. Plant Cell 12:2473–2483View ArticlePubMedPubMed CentralGoogle Scholar
  34. Yu SB, Li JX, Xu CG, Tan YF, Gao YJ, Li XH, Zhang Q, Maroof MAS (1997) Importance of epistasis as the genetic basis of heterosis in an elite rice hybrid. Proc Natl Acad Sci U S A 94:9226–9231View ArticlePubMedPubMed CentralGoogle Scholar
  35. Yuan L (1994) Increasing yield potential in rice by exploitation of heterosis. Hybrid rice technology: New developments and future prospects:1–6Google Scholar
  36. Zeng ZB (1994) Precision mapping of quantitative trait loci. Genetics 126:999–1009Google Scholar
  37. Zhu QS, Wang ZQ, Zhang ZJ, Hui DF (1995) Study on indicators of grain-filling of rice. Journal of Jiangsu agricultural college 16:1-4 (Chinese with English abstract)Google Scholar

Copyright

© The Author(s). 2017