- Email: [email protected]
QTL mapping of grain weight in rice and the validation of the QTL qTGW3.2
Gene 527 (2013) 201–206
Contents lists available at SciVerse ScienceDirect
Gene journal homepage: www.elsevier.com/locate/gene
QTL mapping of grain weight in rice and the validation of the QTL qTGW3.2 Tang Shao-qing, Shao Gao-neng, Wei Xiang-jin, Chen Ming-liang, Sheng Zhong-hua, Luo Ju, Jiao Gui-ai, Xie Li-hong, Hu Pei-song ⁎ Chinese National Center for Rice Improvement/State Key Laboratory of Rice Biology, China National Rice Research Institute, Hangzhou 310006, PR China
a r t i c l e
i n f o
Article history: Accepted 27 May 2013 Available online 11 June 2013 Keywords: Rice Thousand grain weight QTL mapping
a b s t r a c t A recombinant inbred line (RIL) population bred from a cross between a javanica type (cv. D50) and an indica type (cv. HB277) rice was used to map seven quantitative trait loci (QTLs) for thousand grain weight (TGW). The loci were distributed on chromosomes 2, 3, 5, 6, 8 and 10. The chromosome 3 QTL qTGW3.2 was stably expressed over two years, and contributed 9–10% of the phenotypic variance. A residual heterozygous line (RHL) was selected from the RIL population and its selfed progeny was used to fine map qTGW3.2. In this “F2” population, the QTL explained about 23% of the variance, rising to nearly 33% in the subsequent “F2:3” generation. The physical location of qTGW3.2 was confined to a ~ 556 kb region flanked by the microsatellite loci RM16162 and RM16194. The region also contains other factors influencing certain yield-related traits, although it is also possible that qTGW3.2 affects these in a pleiotropic fashion. Crown Copyright © 2013 Published by Elsevier B.V. All rights reserved.
1. Introduction The determinants of grain yield in rice include a number of sub-traits, for example, the panicle number per plant (PN), the total number of spikelets formed per panicle (TNSP), the mean spikelet fertility (SF) and the thousand grain weight (TGW). The well developed genetic maps of rice have simplified the process of determining the inheritance of quantitative traits such as PN, TNSP, SF and TGW. Quantitative trait locus (QTL) mapping has to date placed loci responsible for variation in TGW on each of rice's 12 chromosomes (http:// www.gramene.org/), and a few of these have already been mapped at a fine scale (Li et al., 2004; Liu et al., 2010; Xie et al., 2006, 2008). The ability to perform fine-scale mapping has further facilitated the positional cloning of several such genes, for example GS3 on chromosome 3, a gene which conditions mainly grain length (Fan et al., 2006), GL3.1 and qGL3 (both on chromosome 3, and also conditions grain length) (Qi et al., 2012; Zhang et al., 2012), GS7 on chromosome 7, a gene which conditions mainly grain length and length to width ratio (Shao et al., 2012), and four grain weight QTLs mapping to Abbreviations: CIM, composite interval mapping; FLL, flag leaf length; FLL/W, flag leaf length to width ratio; FLW, flag leaf width; GL, grain length; GL/W, the ratio of grain length to width; GW, grain width; HD, heading date; HN, Hainan; HZ, Hangzhou; MAS, marker-assisted selection; Max, maximum; Min, minimum; PCR, polymerase chain reaction; PH, plant height; PL, panicle length; PN, panicle number per plant; PN, panicles number per plant; QTL, quantitative trait locus; R1–R9, recombinant individuals (1–9); R2, the phenotype variance explained by the QTL; RHL, residual heterozygous line; RHL-D, residual heterozygous line with D50 segment in the target region; RHL-H, residual heterozygous line with HB77 segment in the target region; RIL, recombinant inbred line; SF, spikelet fertility; SSR, simple sequence repeat; TGW, thousand grain weight; TNSP, total number of spikelets formed per panicle. ⁎ Corresponding author. Tel.: +86 571 63370221; fax: +86 571 63370080. E-mail addresses: [email protected], [email protected] (P. Hu).
chromosomes 2, 5 and 8 (Li et al., 2011; Song et al., 2007; Wang et al., 2012; Weng et al., 2008). Here, a mapping population derived from a javanica × indica cross was used to assess the inheritance of TGW. The most important QTL resided on chromosome 3, and fine-mapping was therefore carried out to define the genomic region harboring the gene. The region is distinct from the one in which either GS3 or GL3.1 resides. 2. Materials and methods 2.1. Plant materials An F7 recombinant inbred line (RIL) population of 178 lines has been developed from the cross cv. D50 (javanica) × cv. HB277 (indica) (Shao et al., 2010). The material was grown in experimental plots at Hangzhou City (Zhejiang Province). Each line was represented by four rows of eight plants each, grown as a randomized block design. Following the initial outcome of the QTL analysis, a line (“RHL”) heterozygous for the segment flanked by the SSR loci RM130 and RM85 (on chromosome 3), but homozygous elsewhere in the genome, was selected based on the genotyping with 102 informative SSR markers distributed across all 12 rice chromosomes (Fig. 1A). RHL carried the cv. D50 allele at 45 of the SSR loci and the cv. HB277 allele at 53, leaving four loci (RM143, RM130, RM148 and RM85) within the critical chromosome 3 region in the heterozygous state (Fig. 1B). A set of 116 selfed progeny of RHL (RHL “F2” population, RHL-F2) and the subsequent “F2:3” populations (RHL-F2:3) were field grown in Hangzhou (Zhejiang Province). Eleven informative SSR markers (Table 1) obtained from www. gramene.org were applied to the RHL “F2” to construct a linkage map of the critical region of chromosome 3 bounded by RM130 and
0378-1119/$ – see front matter. Crown Copyright © 2013 Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.gene.2013.05.063
202
S. Tang et al. / Gene 527 (2013) 201–206
A
B
qTGW3.2
Fig. 1. (A) The breeding strategy applied for QTL mapping, and (B) a graphical view of the genotype of the RHL used for mapping qTGW3.2.
RM85. The control lines R1 and R2, two RHL “F2” segregants bearing, respectively, cv. D50 and cv. HB277 alleles throughout the critical chromosome 3 segment, along with seven pairs of contrasting RHL “F2” segregants (R3 to R9) were used. A set of 60 “F2:3” individuals bred from each “F2” recombinant individuals (R3 to R9) were grown in the field and genotyped. Finally, the homozygous derivatives of each recombinant were used to assess the correlation between genotype and TGW. For these latter experiments, eight plants (“F4”) per recombinant line were grown in the field, in three replications. 2.2. Trait evaluation At crop maturity, the grain was allowed to dry conventionally before measuring TGW from a sample of 300 grains per RIL and 100 grains per RHL derivative in three replications. Other characteristics evaluated from eight field-grown plants of each RHL pair (including lines R1 and R2) were plant height (PH), heading date (HD), flag leaf length (FLL), flag leaf width (FLW), flag leaf length to width ratio (FLL/W), panicle length (PL), panicle number per plant (PN), the total number of spikelets per panicle (TNSP), and spikelet fertility (SF). PL was defined as the distance between the panicle's neck and tip (excluding the awn), while SF was calculated from the grain set in all panicles produced by the plants within each line. Grains were dehulled to determine their shape, and the rapid analysis system SC-E was applied to a sample of 20 grains per line to obtain mean values for grain length (GL), width (GW) and the length/width ratio (GL/W). 2.3. DNA extraction and molecular marker analyses DNA was extracted from seedlings, following Lu and Zheng (1992). Each 10 μL PCR contained 1 μL 10 × PCR buffer (25 mM MgCl2), 0.16 mM dNTP, 0.05 μM of each primer, 0.5 U Taq DNA polymerase and 1 μL template DNA (200 ng). The cycling regime consisted of an initial denaturation step of 94 °C/2 min, followed by 30 cycles of 94 °C/45 s, 55 °C/45 s, 72 °C/60 s, and a final extension step of 72 °C/8 min. Amplicons were electrophoretically separated through non-denaturating 6% polyacrylamide gels and visualized by silver staining (Shi et al., 2005). 2.4. Map construction and QTL analysis A localized linkage map spanning the region defined by RM130 and RM85 was constructed using MAPMAKER/Exp version 3.0 (Lincoln et al., 1992). Recombination fractions were converted into cM using the Kosambi function. The QTL analysis relied on composite
interval mapping (CIM) implemented within the software package QTL Cartographer V2.5 (statgen.ncsu.edu/qtlcart/WQTLCart.htm) (Wang et al., 2006). QTLs were called where their LOD value exceeded 3.0.
2.5. Statistical analysis of data Mean phenotypic values were compared using the Student's t test. Correlations between genotype and phenotype were calculated using a generalized linear model implemented within the SAS statistical software package. A recurrent substitution mapping strategy as described by Paterson et al. (1990) was applied for the mapping of qTGW3.2.
3. Results 3.1. Segregation for TGW in the RIL population and QTL mapping The cvs. D50 and HB277 differed significantly from one another with respect to TGW (Table 2). The frequency distribution for TGW among the RILs was continuous (Fig. 2). When the linkage map derived from the D50/HB277 RILs was used to genetically dissect the variation for TGW, seven chromosomal regions were identified (Table 3; Fig. 3). Four of these QTLs, namely qTGW3.1, qTGW3.2 (chromosome 3 within, respectively, intervals RM338–RM168 and RM130– RM85), qTGW5 (chromosome 5, RM146–RM305) and qTGW6 (chromosome 6, RM587–RM3438) were all detectable in two seasons. qTGW3.2 and qTGW5 explained, respectively, 10.3% and 18.1% of the variance in TGW across the two seasons. Positive alleles were contributed by cv. HB277 at qTGW5, qTGW8 and qTGW10, while cv. D50 harbored the positive alleles for the other four QTLs.
Table 1 Primer sequences employed for the production of SSR amplicons. Markers
Forward primer
Reverse primer
RM130 RM16107 RM16121 RM16143 RM16162 RM16178 RM16194 RM148 RM16202 RM16217 RM85
TGTTGCTTGCCCTCACGCGAAG CTGTGCTACCGGGTCATAATTCC GACGTCAGCGATCTCCACTACG CCTTTGGTTGCGAGTGGTTGG CCAGCATCATCATTGTCATCATCG CCGTCTCCTTCTCGGTGTTCTGC TTAAGCCGATGCATGAAGGATGC ATACAACATTAGGGATGAGGCTGG GCCATGGTCGGTGAAGATCTGG GCTGCTGCCTAGCTCTATGTAATGG CCAAAGATGAAACCTGGATTG
GGTCGCGTGCTTGGTTTGGTTC AAGAGATTATGTGTGGCGTGATCC CAGCTGTTGCCTGTTGCATAGC CACGCTTCGAGAGAAGACGAAGG ATCCATCCAGCAGGAGAAACAGG AGAGGCTCCAGCGCCAAATCC ACTATCGACGAATGCACCAAACG TCCTTAAAGGTGGTGCAATGCGAG AGCTGCGGAAGCTGAACTGAACC GTGTTTGAATCGTCTCAACCATCG GCACAAGGTGAGCAGTCC
S. Tang et al. / Gene 527 (2013) 201–206
203
Table 2 Descriptive statistics for the segregation of TGW in the D50/HB277 RIL population. Place/year
Trait
Hangzhou/2006 Hangzhou/2007
TGW TGW
Parent
RIL population
D50
HB277
Mean
Max
Min
SD
Skewness
Kurtosis
23.38 25.27
25.17⁎⁎ 25.93⁎⁎
24.15 25.52
33.65 38.93
17.27 16.65
2.92 3.10
0.16 0.26
−0.08 1.26
⁎⁎ Asterisks indicate significant differences (P b 0.01; n = 8) between cv. D50 and cv. HB277.
Year 2007
Numbers
Year 2006
1,000 grain weight (g)
1,000 grain weight (g)
Fig. 2. The distribution of TGW among the D50/HB277 RIL population over two seasons.
3.2. Segregation for qTGW3.2 in the RHL “F2” and “F2:3” populations TGW segregated discontinuously in both the RHL “F2” and “F2:3” populations (Fig. 4), consistent with the presence of a single gene determining TGW. Lines homozygous for the cv. D50 allele of qTGW3.2 produced heavier grains than did the contrasting homozygotes, while heterozygous individuals had an intermediate TGW. The gene action of qTGW3.2 was thus mainly an additive effect. Win QTL Cartographer V2.5 based mapping confirmed the segregation of a single major QTL in both populations, lying within the interval RM16143– RM16217 (Figs. 5A,B). The relative LOD values were 9.7 and 15.8, and the proportion of the variance explained was 23.1% and 32.6% (Table 4). 3.3. Validation of qTGW3.2 using contrasting RHL pairs The R1 and R2 grains differed significantly from one another with respect to TGW (respectively, 22.20 ± 0.18 g and 21.49 ± 0.20 g, see Fig. 6), demonstrating that the presence of cv. D50 alleles in the RM130–RM85 interval increased TGW. Comparison of the contrasting homozygotes extracted from selections R3–R5 and R8 confirmed a significant association between genotype and TGW in the chromosome 3 region, but no such difference existed in the same contrast from selections R6, R7 and R9 (Figs. 6A,B). R4, R7 and R9 each carried a recombination between RM16162 and RM16194. Finally, qTGW3.2 co-segregated with RM16178, and its physical location was successfully narrowed to the interval defined by RM16162 and RM16194. This region represents a ~556 kb stretch of genomic DNA (35,226,547– 35,781,563bp) (http://www.gramene.org) and a genetic distance of 5.3 cM (Fig. 5C). It is defined by a five BAC clone contig (OSJNBa0042I09, Ba0075M12, Ba0032G11, Bb0043P23 and Bb0096M04) (Fig. 5D).
SF displayed a significant difference between R1 and R2 (Table 5). The cv. HB277 allele was associated with taller plants and a shorter panicle length, while the cv. D50 allele was associated with a higher level of spikelet fertility. The qTGW3.2 region may therefore either contain gene(s) exerting a positive effect on these traits, or qTGW3.2 itself, apart from its pleiotropic effect on grain size. 4. Discussion Food security for an ever-increasing global population depends heavily on increasing the productivity of the major crop plants (Gupta et al., 2006). A major focus in rice is the breeding of high yielding cultivars, and for this reason an understanding of the genetic basis of TGW is important. A large number of TGW QTLs have been detected, with every chromosome implicated. In the D50/HB277 population alone, it was possible to identify eight TGW QTLs distributed over six chromosomes; however, of these only three were detectable in both trial seasons. TGW QTLs have been identified in other populations in both the qTGW3.2 and qTGW5 regions. Thus, Septiningsih et al. (2003) distinguished two loci mapping within a region similar to that harboring qTGW3.1 and qTGW3.2 using a population bred from a cross between a cultivated type (cv. IR64) and an accession of the closely related wild species Oryza rufipogon, while both Moncada et al. (2001) and Xu et al. (2002) mapped TGW QTL in the same region
Table 3 Descriptive statistics for the various TGW QTL mapped in the D50/HB277 RIL population. Year
Trait
Locus
Chr.
Interval
LOD score
Additive
R2 (%)
2006
TGW
2007
TGW
qTGW2 qTGW3.1 qTGW3.2 qTGW5 qTGW6 qTGW10 qTGW3.1 qTGW3.2 qTGW5 qTGW6 qTGW8
2 3 3 5 6 10 3 3 5 6 8
RM341–RM262 RM338–RM168 RM130–RM85 RM146–RM305 RM587–RM3438 RM258–RM590 RM338–RM168 RM130–RM85 RM146–RM178 RM587–RM3438 RM210–RM443
2.1824 3.8744 5.8133 8.8431 2.9858 5.2874 2.3321 5.5214 6.287 2.8815 2.4154
0.6232 0.7379 0.8993 −1.2767 0.6943 −0.9146 0.6746 1.0000 −1.0661 0.7371 −0.7593
4.28 6.12 9.37 18.13 5.56 9.62 4.52 10.31 11.43 5.46 4.47
3.4. Trait evaluation in the target region GL, GW and GL/W were evaluated to determine which aspect of grain shape was responsible for the increased TGW encoded by qTGW3.2. A comparison between the performance of lines R1 and R2 indicated that both GL and GL/W were primarily responsible. When other yield related traits (PH, HD, FLL, FLW, FLL/W, PL, PN, TNSP and SF) were also analyzed in the same way, only PH, PL and
204
S. Tang et al. / Gene 527 (2013) 201–206
Fig. 3. The chromosomal location of TGW QTL as detected in the D50/HB277 RIL population. HZ, Hangzhou; HN, Hainan.
as qTGW3.2. With respect to qTGW5, Hua et al. (2002) and Zhuang et al. (2002) were able to define a similar location in their populations. Here, we have defined the location of qTGW3.2 to a ~ 556 kb region of chromosome 3, at a site separated by 3.4 cM from that of qTGW3b (Liu et al., 2010) and by 100 cM from GL3.1 and qGL3 (Qi et al., 2012; Zhang et al., 2012). Many QTLs have proven to be pleiotropic, that is they exert an effect on more than one trait simultaneously. For example, Ghd7 controls the number of spikelets per panicle, plant height and heading data (Xue et al., 2008); qTGWT6-1, which associated with TGW and lies within a 125 kb region of chromosome 6, was found to co-segregate with the yield QTL qTNSP6-1 (Cheng et al., 2007); Xie et al. (2008) have shown that a 37.4 kb region of chromosome 9 harbors one (or more) QTL controlling both grain weight (gw9.1) and yield (sn9.1). A number of grain weight QTLs have also shown to exert pleiotropic effects, namely GL3.1, qGL3, GS3, GW2, GW5, GS5 and GS8 (Li et al., 2011; Qi et al., 2012; Song et al., 2007; Wang et al., 2012; Weng et al., 2008; Zhang et al., 2012). Unsurprisingly, therefore, there was evidence for pleiotropy associated with qTGW3.2,
40
RHL-F2 population
30
35
25
30
Numbers
Numbers
35
which formed a different grain length and length to width ratio. The target region may have pleiotropy on PH, PL and SF. The size of the qTGW3.2 segment is large enough, however, to allow the possibility that rather than this being a pleiotropy, there may instead be separate gene(s) for these other traits harbored within the same segment. Among the plethora of population types used for QTL mapping, RILs and doubled haploid lines have proven their superiority, as unlike F2 populations, they allow for replication across time and space, so are able to generate a reliable phenotype. In the context of fine-scale mapping, the requirement is to produce as isogenic a background as possible to avoid the statistical noise arising from variation in non-critical parts of the genome. While the using of near isogenic lines and chromosomal segment substitution lines have proved their worth for this purpose, their generation is a time- and resource-consuming activity. The suggestion of using an RHL as the parent of a fine-scale mapping population was initially made by Tuinstra et al. (1997), and has since been adopted by various researchers (Du et al., 2008; Shao et al., 2010; Yamanaka et al., 2005; Yu et al., 2008). Here, we were able to fine map qTGW3.2 by
20 15 10
25 20 15 10
5 0
RHL-F2:3 population
5 20.0-20.8 20.9-21.6 21.7-22.4 22.5-23.2 23.3-24.0 24.1-24.8 24.9-25.6
0
20.0-20.3 20.4-20.7 20.8-21.1 21.2-21.5 21.6-21.9 22.0-22.3 22.4-22.7
1,000 grain weight (g)
1,000 grain weight (g) Homozygous HB277 allele
Homozygous D50 allele
Heterzygous
Fig. 4. The distribution of TGW in the RHL “F2” and the derived RHL “F2:3” populations.
RM85
RM148
RM130
RM143
205
RM168
RM16
RM338
RM251
RM14586
A
RM175
RM14383
S. Tang et al. / Gene 527 (2013) 201–206
2.2
3.1
1
2
OSJNBa0042I09
OSJNBa0032G11
OSJNBa0075M12
RM85 RM16217
RM16217 RM16202
RM148 RM148
RM16194
RM16178
RM16162
C
D
RM16143
RM16121
RM130 RM16143
B
RM16107
Chr 3
OSJNBb0096M04
OSJNBb0043P23 556 Kb
Fig. 5. The chromosomal and physical locations of qTGW3.2. (A) Conventional QTL mapping in RIL population placed qTGW3.2 between the SSR loci RM130 and RM85. (B and C) This position was refined to between RM16162 and RM16194 using fine-scale mapping based on the RHL “F2” and the derived RHL “F2:3” populations. The genetic distances obtained from the RHL “F2” population between the closest markers and the QTL are shown above the relevant intervals, and the corresponding number of recombinants below the line. (D) The physical location of qTGW3.2 is within a ~556 kb contig defined by five BAC clones.
Table 4 Descriptive statistics for qTGW3.2 mapped in the RHL “F2” and the derived RHL “F2:3” populations. Trait
Population
Interval
LOD Score
Additive
Dominance
R2 (%)
TGW
F2 F2:3
RM16143–RM16217 RM16143–RM16217
9.67 15.80
0.0568 0.0872
0.0156 0.0406
23.13 32.63
A
exploiting the residual heterozygosity remaining in an F7 RIL, and the resolution achieved may be adequate to progress to the isolation of the relevant coding sequence. A number of rice genes/QTLs involved in the control of grain weight have already been identified and their effect characterized. This includes GL3.1, qGL3, GS3, and GS7, which are all largely responsible for the determination of grain length (Li et al., 2010; Qi et al., 2012; Shao et al., 2012; Zhang et al., 2012). While qTGW3.2 is mainly increased by grain length, so the isolation
B
qTGW3.2
Homozygous HB277 allele
Homozygous D50 allele
Heterozygous region
Recombinant region
Fig. 6. The genotype of the RHL pairs used to validate the effect on TGW of qTGW3.2. (A) Graphical genotypes of the selected nine RHL pairs. (B) Comparison of the contrasting homozygotes extracted from the selections R1–R9. R1 and R2 are two RHL “F2” segregants bearing, respectively, cv. D50 and cv. HB277 alleles throughout the critical chromosome 3 segment. The presence of a statistical difference in TGW within each RHL pair is indicated by ** (P b 0.01; n = 8). a, nine selected RHL F2 individuals; b, the SSR marker RM16178 was used for genotype analysis in their sixty RHL F2:3 populations. Finally, the homozygous derivatives (“F4”) of each recombinant were used to assess the correlation between genotype and TGW.
206
S. Tang et al. / Gene 527 (2013) 201–206
Table 5 Relative performance of a set of agronomic traits between the control lines R1 and R2. GL, grain length (mm); GW, grain width (mm); L/W, the ratio GL/GW; TGW, thousand grain weight (g); PH, plant height (cm); HD, heading date; FLL, flag leaf length (cm); FLW, flag leaf width (cm); FLL/W, the ratio FLL/FLW; PL, panicle length (cm); PN, panicle number per plant; TNSP, total number of spikelets per panicle; SF, spikelet fertility (%). Trait
RHL-D
GL (mm) GW (mm) L/W TGW (g) PH (cm) HD FLL (cm) FLW (cm) FLL/W PL (cm) PN TNSP SF (%)
7.58 1.99 3.82 22.20 104.0 110.0 31.20 1.38 22.60 29.3 9.3 167.0 71.28
± ± ± ± ± ± ± ± ± ± ± ± ±
RHL-H 0.07 0.01 0.05 0.18 3.0 2.0 3.16 0.09 1.54 1.36 2.0 12.0 2.50
7.31 1.98 3.69 21.49 92.0 111.0 31.00 1.39 22.29 27.20 8.8 160.0 78.30
± ± ± ± ± ± ± ± ± ± ± ± ±
0.12⁎⁎ 0.03 0.06⁎⁎ 0.20⁎⁎ 2.0⁎⁎ 3.0 2.98 0.06 1.78 1.42⁎⁎ 2.4 18.0 2.80⁎⁎
⁎⁎ Asterisks indicate significant differences (P b 0.01; n = 8) between R1 and R2.
of qTGW3.2 will serve to improve our understanding of the determination of grain weight in rice. In addition, our continuing work will also certainly be used to help improve grain yield in crops through approaches such as marker-assisted selection (MAS). Conflict of interest None. Acknowledgments This research was supported by the 863 Program of China (Grant no. 2011AA10A101), the China Natural Science Foundation (Grant Nos. 31161140348 and 31000702), and the China Department of Agriculture (Grant Nos. 2011ZX08001-001, 2011ZX08001-002 and 2011ZX08001-006). References Cheng, S.H., Zhuang, J.Y., Fan, Y.Y., Du, J.H., Cao, L.Y., 2007. Progress in research and development on hybrid rice: a super-domesticate in China. Ann. Bot. 100, 959–966. Du, J.H., Fan, Y.Y., Wu, J.R., Zhuang, J.Y., 2008. Dissection of QTLs for yield traits on the short arm of rice chromosome 6. Sci. Agric. Sin. 41, 513–520. Fan, C.C., et al., 2006. GS3, a major QTL for grain length and weight and minor QTL for grain width and thickness in rice, encodes a putative transmembrane protein. Theor. Appl. Genet. 112, 1164–1171. Gupta, P.K., Rustgi, S., Kumar, N., 2006. Genetic and molecular basis of grain size and grain number and its relevance to grain productivity in higher plants. Genome. 49, 565–571.
Hua, J.P., Xing, Y.Z., Xu, C.G., Sun, X.L., Yu, S.B., Zhang, Q.F., 2002. Genetic dissection of an elite rice hybrid revealed that heterozygotes are not always advantageous for performance. Genetics. 162, 1885–1895. Li, J.M., Thomason, M., McCouch, S.R., 2004. Fine mapping of a grain-weight quantitative trait locus in the pericentromeric region of rice chromosome 3. Genetics. 168, 2187–2195. Li, Y.B., et al., 2011. Natural variation in GS5 plays an important role in regulating grain size and yield in rice. Nat. Genet. 43, 1266–1269. Lincoln, E., Daly, M., Lander, E.S., 1992. Constructing Genetic Maps with MAPMAKER/ EXP 3.0//Whitehead Institute Technical Report. Whitehead Institute, Cambridge, MA, USA. Liu, T.M., Shao, D., Kovi, M.R., Xing, Y.Z., 2010. Mapping and validation of quantitative trait loci for spikelets per panicle and 1000-grain weight in rice (Oryza sativa L.). Theor. Appl. Genet. 120, 933–942. Lu, Y.J., Zheng, K.L., 1992. A simple method for isolation of rice DNA. Chin. J. Rice Sci. 6, 47–48 (in Chinese with an English abstract). Moncada, P., et al., 2001. Quantitative trait loci for yield and yield components in an Oryza sativa × Oryza rufipogon BC2F2 population evaluated in an upland environment. Theor. Appl. Genet. 102, 41–52. Paterson, A.H., Deverna, J.W., Lanini, B., Tanksley, S.D., 1990. Fine mapping of quantitative trait loci using selected overlapping recombinant chromosomes, in an interspecies cross of tomato. Genetics. 124, 735–742. Qi, P., et al., 2012. The novel quantitative trait locus GL3.1 controls rice grain size and yield by regulating Cyclin-T1;3. Cell Res. 22 (12), 1666–1680. Septiningsih, E.M., et al., 2003. Identification of quantitative trait loci for yield and yield components in an advanced backcross population derived from the Oryza sativa variety IR64 and the wild relative O. rufipogon. Theor. Appl. Genet. 107, 1419–1432. Shao, G.N., et al., 2010. Mapping of qGL7-2, a grain length QTL on chromosome 7 of rice. J. Genet. Genomics. 37, 523–531. Shao, G.N., et al., 2012. Allelic variation for a candidate gene for GS7, responsible for grain shape in rice. Theor. Appl. Genet. 125, 1303–1312. Shi, Y.F., Ying, J.Z., Wang, L., Zhu, Z., Zhuang, J.Y., 2005. Screening SSR markers for rice variety identification. Chin. J. Rice Sci. 19, 195–201 (in Chinese with English abstract). Song, X.J., Huang, W., Shi, M., Zhu, M.Z., Lin, H.X., 2007. A QTL for rice grain width and weight encodes a previously unknown RING-type E3 ubiquitin ligase. Nat. Genet. 39, 623–630. Tuinstra, M.R., Ejeta, G., Goldsbrough, P.B., 1997. Heterogeneous inbred family (HIF) analysis: a method for developing near-isogenic lines that differ at quantitative trait loci. Theor. Appl. Genet. 95, 1005–1011. Wang, S., Basten, C.J., Zeng, Z.B., 2006. Win QTL Cartographer 2.5. Department of Statistics, North Carolina State University, Raleigh, USA. Wang, S.K., et al., 2012. Control of grain size, shape and quality by OsSPL16 in rice. Nat. Genet. 44, 950–954. Weng, J.F., et al., 2008. Isolation and initial characterization of GW5, a major QTL associated with rice grain width and weight. Cell Res. 18, 1199–1209. Xie, X.B., et al., 2008. Fine mapping of a yield enhancing QTL cluster associated with transgressive variation in an Oryza sativa × O. rufipogon cross. Theor. Appl. Genet. 116, 613–622. Xie, X.B., et al., 2006. Fine mapping of a grain weight quantitative trait locus on rice chromosome 8 using near-isogenic lines derived from a cross between Oryza sativa and Oryza rufipogon. Theor. Appl. Genet. 113, 885–894. Xu, J.L., Xue, Q.Z., Luo, L.J., Li, Z.K., 2002. Genetic dissection of grain weight and its related traits in rice (Oryza sativa L.). Chin. J. Rice Sci. 16 (1), 6–10. Xue, W.Y., et al., 2008. Natural variation in Ghd7 is an important regulator of heading date and yield potential in rice. Nat. Genet. 40, 761–767. Yamanaka, N., Watanabe, S., Toda, K., et al., 2005. Fine mapping of the FT1 locus for soybean flowering time using a residual heterozygous line derived from a recombinant inbred line. Theor. Appl. Genet. 110, 634–639. Yu, S.W., Yang, C.D., Fan, Y.Y., Zhuang, J.Y., Li, X.M., 2008. Genetic dissection of a thousand-grain weight quantitative trait locus on rice chromosome 1. Chin. Sci. Bull. 53, 2326–2332. Zhang, X.J., et al., 2012. Rare allele of OsPPKL1 associated with grain length causes extra-large grain and a significant yield increase in rice. PNAS. 109, 21534–21539. Zhuang, J.Y., Fan, Y.Y., Rao, Z.M., Wu, J.L., Xia, Y.W., Zheng, K.L., 2002. Analysis on additive effects and additive-by-additive epistatic effects of QTLs for yield traits in a recombinant inbred line population of rice. Theor. Appl. Genet. 105, 1137–1145.
Contents lists available at SciVerse ScienceDirect
Gene journal homepage: www.elsevier.com/locate/gene
QTL mapping of grain weight in rice and the validation of the QTL qTGW3.2 Tang Shao-qing, Shao Gao-neng, Wei Xiang-jin, Chen Ming-liang, Sheng Zhong-hua, Luo Ju, Jiao Gui-ai, Xie Li-hong, Hu Pei-song ⁎ Chinese National Center for Rice Improvement/State Key Laboratory of Rice Biology, China National Rice Research Institute, Hangzhou 310006, PR China
a r t i c l e
i n f o
Article history: Accepted 27 May 2013 Available online 11 June 2013 Keywords: Rice Thousand grain weight QTL mapping
a b s t r a c t A recombinant inbred line (RIL) population bred from a cross between a javanica type (cv. D50) and an indica type (cv. HB277) rice was used to map seven quantitative trait loci (QTLs) for thousand grain weight (TGW). The loci were distributed on chromosomes 2, 3, 5, 6, 8 and 10. The chromosome 3 QTL qTGW3.2 was stably expressed over two years, and contributed 9–10% of the phenotypic variance. A residual heterozygous line (RHL) was selected from the RIL population and its selfed progeny was used to fine map qTGW3.2. In this “F2” population, the QTL explained about 23% of the variance, rising to nearly 33% in the subsequent “F2:3” generation. The physical location of qTGW3.2 was confined to a ~ 556 kb region flanked by the microsatellite loci RM16162 and RM16194. The region also contains other factors influencing certain yield-related traits, although it is also possible that qTGW3.2 affects these in a pleiotropic fashion. Crown Copyright © 2013 Published by Elsevier B.V. All rights reserved.
1. Introduction The determinants of grain yield in rice include a number of sub-traits, for example, the panicle number per plant (PN), the total number of spikelets formed per panicle (TNSP), the mean spikelet fertility (SF) and the thousand grain weight (TGW). The well developed genetic maps of rice have simplified the process of determining the inheritance of quantitative traits such as PN, TNSP, SF and TGW. Quantitative trait locus (QTL) mapping has to date placed loci responsible for variation in TGW on each of rice's 12 chromosomes (http:// www.gramene.org/), and a few of these have already been mapped at a fine scale (Li et al., 2004; Liu et al., 2010; Xie et al., 2006, 2008). The ability to perform fine-scale mapping has further facilitated the positional cloning of several such genes, for example GS3 on chromosome 3, a gene which conditions mainly grain length (Fan et al., 2006), GL3.1 and qGL3 (both on chromosome 3, and also conditions grain length) (Qi et al., 2012; Zhang et al., 2012), GS7 on chromosome 7, a gene which conditions mainly grain length and length to width ratio (Shao et al., 2012), and four grain weight QTLs mapping to Abbreviations: CIM, composite interval mapping; FLL, flag leaf length; FLL/W, flag leaf length to width ratio; FLW, flag leaf width; GL, grain length; GL/W, the ratio of grain length to width; GW, grain width; HD, heading date; HN, Hainan; HZ, Hangzhou; MAS, marker-assisted selection; Max, maximum; Min, minimum; PCR, polymerase chain reaction; PH, plant height; PL, panicle length; PN, panicle number per plant; PN, panicles number per plant; QTL, quantitative trait locus; R1–R9, recombinant individuals (1–9); R2, the phenotype variance explained by the QTL; RHL, residual heterozygous line; RHL-D, residual heterozygous line with D50 segment in the target region; RHL-H, residual heterozygous line with HB77 segment in the target region; RIL, recombinant inbred line; SF, spikelet fertility; SSR, simple sequence repeat; TGW, thousand grain weight; TNSP, total number of spikelets formed per panicle. ⁎ Corresponding author. Tel.: +86 571 63370221; fax: +86 571 63370080. E-mail addresses: [email protected], [email protected] (P. Hu).
chromosomes 2, 5 and 8 (Li et al., 2011; Song et al., 2007; Wang et al., 2012; Weng et al., 2008). Here, a mapping population derived from a javanica × indica cross was used to assess the inheritance of TGW. The most important QTL resided on chromosome 3, and fine-mapping was therefore carried out to define the genomic region harboring the gene. The region is distinct from the one in which either GS3 or GL3.1 resides. 2. Materials and methods 2.1. Plant materials An F7 recombinant inbred line (RIL) population of 178 lines has been developed from the cross cv. D50 (javanica) × cv. HB277 (indica) (Shao et al., 2010). The material was grown in experimental plots at Hangzhou City (Zhejiang Province). Each line was represented by four rows of eight plants each, grown as a randomized block design. Following the initial outcome of the QTL analysis, a line (“RHL”) heterozygous for the segment flanked by the SSR loci RM130 and RM85 (on chromosome 3), but homozygous elsewhere in the genome, was selected based on the genotyping with 102 informative SSR markers distributed across all 12 rice chromosomes (Fig. 1A). RHL carried the cv. D50 allele at 45 of the SSR loci and the cv. HB277 allele at 53, leaving four loci (RM143, RM130, RM148 and RM85) within the critical chromosome 3 region in the heterozygous state (Fig. 1B). A set of 116 selfed progeny of RHL (RHL “F2” population, RHL-F2) and the subsequent “F2:3” populations (RHL-F2:3) were field grown in Hangzhou (Zhejiang Province). Eleven informative SSR markers (Table 1) obtained from www. gramene.org were applied to the RHL “F2” to construct a linkage map of the critical region of chromosome 3 bounded by RM130 and
0378-1119/$ – see front matter. Crown Copyright © 2013 Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.gene.2013.05.063
202
S. Tang et al. / Gene 527 (2013) 201–206
A
B
qTGW3.2
Fig. 1. (A) The breeding strategy applied for QTL mapping, and (B) a graphical view of the genotype of the RHL used for mapping qTGW3.2.
RM85. The control lines R1 and R2, two RHL “F2” segregants bearing, respectively, cv. D50 and cv. HB277 alleles throughout the critical chromosome 3 segment, along with seven pairs of contrasting RHL “F2” segregants (R3 to R9) were used. A set of 60 “F2:3” individuals bred from each “F2” recombinant individuals (R3 to R9) were grown in the field and genotyped. Finally, the homozygous derivatives of each recombinant were used to assess the correlation between genotype and TGW. For these latter experiments, eight plants (“F4”) per recombinant line were grown in the field, in three replications. 2.2. Trait evaluation At crop maturity, the grain was allowed to dry conventionally before measuring TGW from a sample of 300 grains per RIL and 100 grains per RHL derivative in three replications. Other characteristics evaluated from eight field-grown plants of each RHL pair (including lines R1 and R2) were plant height (PH), heading date (HD), flag leaf length (FLL), flag leaf width (FLW), flag leaf length to width ratio (FLL/W), panicle length (PL), panicle number per plant (PN), the total number of spikelets per panicle (TNSP), and spikelet fertility (SF). PL was defined as the distance between the panicle's neck and tip (excluding the awn), while SF was calculated from the grain set in all panicles produced by the plants within each line. Grains were dehulled to determine their shape, and the rapid analysis system SC-E was applied to a sample of 20 grains per line to obtain mean values for grain length (GL), width (GW) and the length/width ratio (GL/W). 2.3. DNA extraction and molecular marker analyses DNA was extracted from seedlings, following Lu and Zheng (1992). Each 10 μL PCR contained 1 μL 10 × PCR buffer (25 mM MgCl2), 0.16 mM dNTP, 0.05 μM of each primer, 0.5 U Taq DNA polymerase and 1 μL template DNA (200 ng). The cycling regime consisted of an initial denaturation step of 94 °C/2 min, followed by 30 cycles of 94 °C/45 s, 55 °C/45 s, 72 °C/60 s, and a final extension step of 72 °C/8 min. Amplicons were electrophoretically separated through non-denaturating 6% polyacrylamide gels and visualized by silver staining (Shi et al., 2005). 2.4. Map construction and QTL analysis A localized linkage map spanning the region defined by RM130 and RM85 was constructed using MAPMAKER/Exp version 3.0 (Lincoln et al., 1992). Recombination fractions were converted into cM using the Kosambi function. The QTL analysis relied on composite
interval mapping (CIM) implemented within the software package QTL Cartographer V2.5 (statgen.ncsu.edu/qtlcart/WQTLCart.htm) (Wang et al., 2006). QTLs were called where their LOD value exceeded 3.0.
2.5. Statistical analysis of data Mean phenotypic values were compared using the Student's t test. Correlations between genotype and phenotype were calculated using a generalized linear model implemented within the SAS statistical software package. A recurrent substitution mapping strategy as described by Paterson et al. (1990) was applied for the mapping of qTGW3.2.
3. Results 3.1. Segregation for TGW in the RIL population and QTL mapping The cvs. D50 and HB277 differed significantly from one another with respect to TGW (Table 2). The frequency distribution for TGW among the RILs was continuous (Fig. 2). When the linkage map derived from the D50/HB277 RILs was used to genetically dissect the variation for TGW, seven chromosomal regions were identified (Table 3; Fig. 3). Four of these QTLs, namely qTGW3.1, qTGW3.2 (chromosome 3 within, respectively, intervals RM338–RM168 and RM130– RM85), qTGW5 (chromosome 5, RM146–RM305) and qTGW6 (chromosome 6, RM587–RM3438) were all detectable in two seasons. qTGW3.2 and qTGW5 explained, respectively, 10.3% and 18.1% of the variance in TGW across the two seasons. Positive alleles were contributed by cv. HB277 at qTGW5, qTGW8 and qTGW10, while cv. D50 harbored the positive alleles for the other four QTLs.
Table 1 Primer sequences employed for the production of SSR amplicons. Markers
Forward primer
Reverse primer
RM130 RM16107 RM16121 RM16143 RM16162 RM16178 RM16194 RM148 RM16202 RM16217 RM85
TGTTGCTTGCCCTCACGCGAAG CTGTGCTACCGGGTCATAATTCC GACGTCAGCGATCTCCACTACG CCTTTGGTTGCGAGTGGTTGG CCAGCATCATCATTGTCATCATCG CCGTCTCCTTCTCGGTGTTCTGC TTAAGCCGATGCATGAAGGATGC ATACAACATTAGGGATGAGGCTGG GCCATGGTCGGTGAAGATCTGG GCTGCTGCCTAGCTCTATGTAATGG CCAAAGATGAAACCTGGATTG
GGTCGCGTGCTTGGTTTGGTTC AAGAGATTATGTGTGGCGTGATCC CAGCTGTTGCCTGTTGCATAGC CACGCTTCGAGAGAAGACGAAGG ATCCATCCAGCAGGAGAAACAGG AGAGGCTCCAGCGCCAAATCC ACTATCGACGAATGCACCAAACG TCCTTAAAGGTGGTGCAATGCGAG AGCTGCGGAAGCTGAACTGAACC GTGTTTGAATCGTCTCAACCATCG GCACAAGGTGAGCAGTCC
S. Tang et al. / Gene 527 (2013) 201–206
203
Table 2 Descriptive statistics for the segregation of TGW in the D50/HB277 RIL population. Place/year
Trait
Hangzhou/2006 Hangzhou/2007
TGW TGW
Parent
RIL population
D50
HB277
Mean
Max
Min
SD
Skewness
Kurtosis
23.38 25.27
25.17⁎⁎ 25.93⁎⁎
24.15 25.52
33.65 38.93
17.27 16.65
2.92 3.10
0.16 0.26
−0.08 1.26
⁎⁎ Asterisks indicate significant differences (P b 0.01; n = 8) between cv. D50 and cv. HB277.
Year 2007
Numbers
Year 2006
1,000 grain weight (g)
1,000 grain weight (g)
Fig. 2. The distribution of TGW among the D50/HB277 RIL population over two seasons.
3.2. Segregation for qTGW3.2 in the RHL “F2” and “F2:3” populations TGW segregated discontinuously in both the RHL “F2” and “F2:3” populations (Fig. 4), consistent with the presence of a single gene determining TGW. Lines homozygous for the cv. D50 allele of qTGW3.2 produced heavier grains than did the contrasting homozygotes, while heterozygous individuals had an intermediate TGW. The gene action of qTGW3.2 was thus mainly an additive effect. Win QTL Cartographer V2.5 based mapping confirmed the segregation of a single major QTL in both populations, lying within the interval RM16143– RM16217 (Figs. 5A,B). The relative LOD values were 9.7 and 15.8, and the proportion of the variance explained was 23.1% and 32.6% (Table 4). 3.3. Validation of qTGW3.2 using contrasting RHL pairs The R1 and R2 grains differed significantly from one another with respect to TGW (respectively, 22.20 ± 0.18 g and 21.49 ± 0.20 g, see Fig. 6), demonstrating that the presence of cv. D50 alleles in the RM130–RM85 interval increased TGW. Comparison of the contrasting homozygotes extracted from selections R3–R5 and R8 confirmed a significant association between genotype and TGW in the chromosome 3 region, but no such difference existed in the same contrast from selections R6, R7 and R9 (Figs. 6A,B). R4, R7 and R9 each carried a recombination between RM16162 and RM16194. Finally, qTGW3.2 co-segregated with RM16178, and its physical location was successfully narrowed to the interval defined by RM16162 and RM16194. This region represents a ~556 kb stretch of genomic DNA (35,226,547– 35,781,563bp) (http://www.gramene.org) and a genetic distance of 5.3 cM (Fig. 5C). It is defined by a five BAC clone contig (OSJNBa0042I09, Ba0075M12, Ba0032G11, Bb0043P23 and Bb0096M04) (Fig. 5D).
SF displayed a significant difference between R1 and R2 (Table 5). The cv. HB277 allele was associated with taller plants and a shorter panicle length, while the cv. D50 allele was associated with a higher level of spikelet fertility. The qTGW3.2 region may therefore either contain gene(s) exerting a positive effect on these traits, or qTGW3.2 itself, apart from its pleiotropic effect on grain size. 4. Discussion Food security for an ever-increasing global population depends heavily on increasing the productivity of the major crop plants (Gupta et al., 2006). A major focus in rice is the breeding of high yielding cultivars, and for this reason an understanding of the genetic basis of TGW is important. A large number of TGW QTLs have been detected, with every chromosome implicated. In the D50/HB277 population alone, it was possible to identify eight TGW QTLs distributed over six chromosomes; however, of these only three were detectable in both trial seasons. TGW QTLs have been identified in other populations in both the qTGW3.2 and qTGW5 regions. Thus, Septiningsih et al. (2003) distinguished two loci mapping within a region similar to that harboring qTGW3.1 and qTGW3.2 using a population bred from a cross between a cultivated type (cv. IR64) and an accession of the closely related wild species Oryza rufipogon, while both Moncada et al. (2001) and Xu et al. (2002) mapped TGW QTL in the same region
Table 3 Descriptive statistics for the various TGW QTL mapped in the D50/HB277 RIL population. Year
Trait
Locus
Chr.
Interval
LOD score
Additive
R2 (%)
2006
TGW
2007
TGW
qTGW2 qTGW3.1 qTGW3.2 qTGW5 qTGW6 qTGW10 qTGW3.1 qTGW3.2 qTGW5 qTGW6 qTGW8
2 3 3 5 6 10 3 3 5 6 8
RM341–RM262 RM338–RM168 RM130–RM85 RM146–RM305 RM587–RM3438 RM258–RM590 RM338–RM168 RM130–RM85 RM146–RM178 RM587–RM3438 RM210–RM443
2.1824 3.8744 5.8133 8.8431 2.9858 5.2874 2.3321 5.5214 6.287 2.8815 2.4154
0.6232 0.7379 0.8993 −1.2767 0.6943 −0.9146 0.6746 1.0000 −1.0661 0.7371 −0.7593
4.28 6.12 9.37 18.13 5.56 9.62 4.52 10.31 11.43 5.46 4.47
3.4. Trait evaluation in the target region GL, GW and GL/W were evaluated to determine which aspect of grain shape was responsible for the increased TGW encoded by qTGW3.2. A comparison between the performance of lines R1 and R2 indicated that both GL and GL/W were primarily responsible. When other yield related traits (PH, HD, FLL, FLW, FLL/W, PL, PN, TNSP and SF) were also analyzed in the same way, only PH, PL and
204
S. Tang et al. / Gene 527 (2013) 201–206
Fig. 3. The chromosomal location of TGW QTL as detected in the D50/HB277 RIL population. HZ, Hangzhou; HN, Hainan.
as qTGW3.2. With respect to qTGW5, Hua et al. (2002) and Zhuang et al. (2002) were able to define a similar location in their populations. Here, we have defined the location of qTGW3.2 to a ~ 556 kb region of chromosome 3, at a site separated by 3.4 cM from that of qTGW3b (Liu et al., 2010) and by 100 cM from GL3.1 and qGL3 (Qi et al., 2012; Zhang et al., 2012). Many QTLs have proven to be pleiotropic, that is they exert an effect on more than one trait simultaneously. For example, Ghd7 controls the number of spikelets per panicle, plant height and heading data (Xue et al., 2008); qTGWT6-1, which associated with TGW and lies within a 125 kb region of chromosome 6, was found to co-segregate with the yield QTL qTNSP6-1 (Cheng et al., 2007); Xie et al. (2008) have shown that a 37.4 kb region of chromosome 9 harbors one (or more) QTL controlling both grain weight (gw9.1) and yield (sn9.1). A number of grain weight QTLs have also shown to exert pleiotropic effects, namely GL3.1, qGL3, GS3, GW2, GW5, GS5 and GS8 (Li et al., 2011; Qi et al., 2012; Song et al., 2007; Wang et al., 2012; Weng et al., 2008; Zhang et al., 2012). Unsurprisingly, therefore, there was evidence for pleiotropy associated with qTGW3.2,
40
RHL-F2 population
30
35
25
30
Numbers
Numbers
35
which formed a different grain length and length to width ratio. The target region may have pleiotropy on PH, PL and SF. The size of the qTGW3.2 segment is large enough, however, to allow the possibility that rather than this being a pleiotropy, there may instead be separate gene(s) for these other traits harbored within the same segment. Among the plethora of population types used for QTL mapping, RILs and doubled haploid lines have proven their superiority, as unlike F2 populations, they allow for replication across time and space, so are able to generate a reliable phenotype. In the context of fine-scale mapping, the requirement is to produce as isogenic a background as possible to avoid the statistical noise arising from variation in non-critical parts of the genome. While the using of near isogenic lines and chromosomal segment substitution lines have proved their worth for this purpose, their generation is a time- and resource-consuming activity. The suggestion of using an RHL as the parent of a fine-scale mapping population was initially made by Tuinstra et al. (1997), and has since been adopted by various researchers (Du et al., 2008; Shao et al., 2010; Yamanaka et al., 2005; Yu et al., 2008). Here, we were able to fine map qTGW3.2 by
20 15 10
25 20 15 10
5 0
RHL-F2:3 population
5 20.0-20.8 20.9-21.6 21.7-22.4 22.5-23.2 23.3-24.0 24.1-24.8 24.9-25.6
0
20.0-20.3 20.4-20.7 20.8-21.1 21.2-21.5 21.6-21.9 22.0-22.3 22.4-22.7
1,000 grain weight (g)
1,000 grain weight (g) Homozygous HB277 allele
Homozygous D50 allele
Heterzygous
Fig. 4. The distribution of TGW in the RHL “F2” and the derived RHL “F2:3” populations.
RM85
RM148
RM130
RM143
205
RM168
RM16
RM338
RM251
RM14586
A
RM175
RM14383
S. Tang et al. / Gene 527 (2013) 201–206
2.2
3.1
1
2
OSJNBa0042I09
OSJNBa0032G11
OSJNBa0075M12
RM85 RM16217
RM16217 RM16202
RM148 RM148
RM16194
RM16178
RM16162
C
D
RM16143
RM16121
RM130 RM16143
B
RM16107
Chr 3
OSJNBb0096M04
OSJNBb0043P23 556 Kb
Fig. 5. The chromosomal and physical locations of qTGW3.2. (A) Conventional QTL mapping in RIL population placed qTGW3.2 between the SSR loci RM130 and RM85. (B and C) This position was refined to between RM16162 and RM16194 using fine-scale mapping based on the RHL “F2” and the derived RHL “F2:3” populations. The genetic distances obtained from the RHL “F2” population between the closest markers and the QTL are shown above the relevant intervals, and the corresponding number of recombinants below the line. (D) The physical location of qTGW3.2 is within a ~556 kb contig defined by five BAC clones.
Table 4 Descriptive statistics for qTGW3.2 mapped in the RHL “F2” and the derived RHL “F2:3” populations. Trait
Population
Interval
LOD Score
Additive
Dominance
R2 (%)
TGW
F2 F2:3
RM16143–RM16217 RM16143–RM16217
9.67 15.80
0.0568 0.0872
0.0156 0.0406
23.13 32.63
A
exploiting the residual heterozygosity remaining in an F7 RIL, and the resolution achieved may be adequate to progress to the isolation of the relevant coding sequence. A number of rice genes/QTLs involved in the control of grain weight have already been identified and their effect characterized. This includes GL3.1, qGL3, GS3, and GS7, which are all largely responsible for the determination of grain length (Li et al., 2010; Qi et al., 2012; Shao et al., 2012; Zhang et al., 2012). While qTGW3.2 is mainly increased by grain length, so the isolation
B
qTGW3.2
Homozygous HB277 allele
Homozygous D50 allele
Heterozygous region
Recombinant region
Fig. 6. The genotype of the RHL pairs used to validate the effect on TGW of qTGW3.2. (A) Graphical genotypes of the selected nine RHL pairs. (B) Comparison of the contrasting homozygotes extracted from the selections R1–R9. R1 and R2 are two RHL “F2” segregants bearing, respectively, cv. D50 and cv. HB277 alleles throughout the critical chromosome 3 segment. The presence of a statistical difference in TGW within each RHL pair is indicated by ** (P b 0.01; n = 8). a, nine selected RHL F2 individuals; b, the SSR marker RM16178 was used for genotype analysis in their sixty RHL F2:3 populations. Finally, the homozygous derivatives (“F4”) of each recombinant were used to assess the correlation between genotype and TGW.
206
S. Tang et al. / Gene 527 (2013) 201–206
Table 5 Relative performance of a set of agronomic traits between the control lines R1 and R2. GL, grain length (mm); GW, grain width (mm); L/W, the ratio GL/GW; TGW, thousand grain weight (g); PH, plant height (cm); HD, heading date; FLL, flag leaf length (cm); FLW, flag leaf width (cm); FLL/W, the ratio FLL/FLW; PL, panicle length (cm); PN, panicle number per plant; TNSP, total number of spikelets per panicle; SF, spikelet fertility (%). Trait
RHL-D
GL (mm) GW (mm) L/W TGW (g) PH (cm) HD FLL (cm) FLW (cm) FLL/W PL (cm) PN TNSP SF (%)
7.58 1.99 3.82 22.20 104.0 110.0 31.20 1.38 22.60 29.3 9.3 167.0 71.28
± ± ± ± ± ± ± ± ± ± ± ± ±
RHL-H 0.07 0.01 0.05 0.18 3.0 2.0 3.16 0.09 1.54 1.36 2.0 12.0 2.50
7.31 1.98 3.69 21.49 92.0 111.0 31.00 1.39 22.29 27.20 8.8 160.0 78.30
± ± ± ± ± ± ± ± ± ± ± ± ±
0.12⁎⁎ 0.03 0.06⁎⁎ 0.20⁎⁎ 2.0⁎⁎ 3.0 2.98 0.06 1.78 1.42⁎⁎ 2.4 18.0 2.80⁎⁎
⁎⁎ Asterisks indicate significant differences (P b 0.01; n = 8) between R1 and R2.
of qTGW3.2 will serve to improve our understanding of the determination of grain weight in rice. In addition, our continuing work will also certainly be used to help improve grain yield in crops through approaches such as marker-assisted selection (MAS). Conflict of interest None. Acknowledgments This research was supported by the 863 Program of China (Grant no. 2011AA10A101), the China Natural Science Foundation (Grant Nos. 31161140348 and 31000702), and the China Department of Agriculture (Grant Nos. 2011ZX08001-001, 2011ZX08001-002 and 2011ZX08001-006). References Cheng, S.H., Zhuang, J.Y., Fan, Y.Y., Du, J.H., Cao, L.Y., 2007. Progress in research and development on hybrid rice: a super-domesticate in China. Ann. Bot. 100, 959–966. Du, J.H., Fan, Y.Y., Wu, J.R., Zhuang, J.Y., 2008. Dissection of QTLs for yield traits on the short arm of rice chromosome 6. Sci. Agric. Sin. 41, 513–520. Fan, C.C., et al., 2006. GS3, a major QTL for grain length and weight and minor QTL for grain width and thickness in rice, encodes a putative transmembrane protein. Theor. Appl. Genet. 112, 1164–1171. Gupta, P.K., Rustgi, S., Kumar, N., 2006. Genetic and molecular basis of grain size and grain number and its relevance to grain productivity in higher plants. Genome. 49, 565–571.
Hua, J.P., Xing, Y.Z., Xu, C.G., Sun, X.L., Yu, S.B., Zhang, Q.F., 2002. Genetic dissection of an elite rice hybrid revealed that heterozygotes are not always advantageous for performance. Genetics. 162, 1885–1895. Li, J.M., Thomason, M., McCouch, S.R., 2004. Fine mapping of a grain-weight quantitative trait locus in the pericentromeric region of rice chromosome 3. Genetics. 168, 2187–2195. Li, Y.B., et al., 2011. Natural variation in GS5 plays an important role in regulating grain size and yield in rice. Nat. Genet. 43, 1266–1269. Lincoln, E., Daly, M., Lander, E.S., 1992. Constructing Genetic Maps with MAPMAKER/ EXP 3.0//Whitehead Institute Technical Report. Whitehead Institute, Cambridge, MA, USA. Liu, T.M., Shao, D., Kovi, M.R., Xing, Y.Z., 2010. Mapping and validation of quantitative trait loci for spikelets per panicle and 1000-grain weight in rice (Oryza sativa L.). Theor. Appl. Genet. 120, 933–942. Lu, Y.J., Zheng, K.L., 1992. A simple method for isolation of rice DNA. Chin. J. Rice Sci. 6, 47–48 (in Chinese with an English abstract). Moncada, P., et al., 2001. Quantitative trait loci for yield and yield components in an Oryza sativa × Oryza rufipogon BC2F2 population evaluated in an upland environment. Theor. Appl. Genet. 102, 41–52. Paterson, A.H., Deverna, J.W., Lanini, B., Tanksley, S.D., 1990. Fine mapping of quantitative trait loci using selected overlapping recombinant chromosomes, in an interspecies cross of tomato. Genetics. 124, 735–742. Qi, P., et al., 2012. The novel quantitative trait locus GL3.1 controls rice grain size and yield by regulating Cyclin-T1;3. Cell Res. 22 (12), 1666–1680. Septiningsih, E.M., et al., 2003. Identification of quantitative trait loci for yield and yield components in an advanced backcross population derived from the Oryza sativa variety IR64 and the wild relative O. rufipogon. Theor. Appl. Genet. 107, 1419–1432. Shao, G.N., et al., 2010. Mapping of qGL7-2, a grain length QTL on chromosome 7 of rice. J. Genet. Genomics. 37, 523–531. Shao, G.N., et al., 2012. Allelic variation for a candidate gene for GS7, responsible for grain shape in rice. Theor. Appl. Genet. 125, 1303–1312. Shi, Y.F., Ying, J.Z., Wang, L., Zhu, Z., Zhuang, J.Y., 2005. Screening SSR markers for rice variety identification. Chin. J. Rice Sci. 19, 195–201 (in Chinese with English abstract). Song, X.J., Huang, W., Shi, M., Zhu, M.Z., Lin, H.X., 2007. A QTL for rice grain width and weight encodes a previously unknown RING-type E3 ubiquitin ligase. Nat. Genet. 39, 623–630. Tuinstra, M.R., Ejeta, G., Goldsbrough, P.B., 1997. Heterogeneous inbred family (HIF) analysis: a method for developing near-isogenic lines that differ at quantitative trait loci. Theor. Appl. Genet. 95, 1005–1011. Wang, S., Basten, C.J., Zeng, Z.B., 2006. Win QTL Cartographer 2.5. Department of Statistics, North Carolina State University, Raleigh, USA. Wang, S.K., et al., 2012. Control of grain size, shape and quality by OsSPL16 in rice. Nat. Genet. 44, 950–954. Weng, J.F., et al., 2008. Isolation and initial characterization of GW5, a major QTL associated with rice grain width and weight. Cell Res. 18, 1199–1209. Xie, X.B., et al., 2008. Fine mapping of a yield enhancing QTL cluster associated with transgressive variation in an Oryza sativa × O. rufipogon cross. Theor. Appl. Genet. 116, 613–622. Xie, X.B., et al., 2006. Fine mapping of a grain weight quantitative trait locus on rice chromosome 8 using near-isogenic lines derived from a cross between Oryza sativa and Oryza rufipogon. Theor. Appl. Genet. 113, 885–894. Xu, J.L., Xue, Q.Z., Luo, L.J., Li, Z.K., 2002. Genetic dissection of grain weight and its related traits in rice (Oryza sativa L.). Chin. J. Rice Sci. 16 (1), 6–10. Xue, W.Y., et al., 2008. Natural variation in Ghd7 is an important regulator of heading date and yield potential in rice. Nat. Genet. 40, 761–767. Yamanaka, N., Watanabe, S., Toda, K., et al., 2005. Fine mapping of the FT1 locus for soybean flowering time using a residual heterozygous line derived from a recombinant inbred line. Theor. Appl. Genet. 110, 634–639. Yu, S.W., Yang, C.D., Fan, Y.Y., Zhuang, J.Y., Li, X.M., 2008. Genetic dissection of a thousand-grain weight quantitative trait locus on rice chromosome 1. Chin. Sci. Bull. 53, 2326–2332. Zhang, X.J., et al., 2012. Rare allele of OsPPKL1 associated with grain length causes extra-large grain and a significant yield increase in rice. PNAS. 109, 21534–21539. Zhuang, J.Y., Fan, Y.Y., Rao, Z.M., Wu, J.L., Xia, Y.W., Zheng, K.L., 2002. Analysis on additive effects and additive-by-additive epistatic effects of QTLs for yield traits in a recombinant inbred line population of rice. Theor. Appl. Genet. 105, 1137–1145.