Botanical Studies (2011) 52: 1-14.
Mapping of quantitative trait loci for plant height and heading date in two inter-subspecific crosses of rice and comparison across Oryza genus
Yann-Rong LIN1'2'*, Shao-Chun WU2, Su-Er CHEN3, Tung-Hai TSENG4, Cheng-Sheng CHEN1, Su-Chen KUO1, Hong-Pang WU5, and Yue-Ie C. HSING5
lDepartment of Agronomy, National Taiiwan University, Taipei l06, Taiwan
2Department of Life Science, Fu-Jen Catholic University, Xinzhuang Dist., New Taipei City 24205, Taiwan
3Taoyuan District Agricultural Research and Extension Station, Council of Agriculture, Executive Yuan, Hsingwu, Taoyuan
  327, Taiwan
4Agriculture Research Institute, Council of Agriculture, Executive Yuan, Wufeng, Taichung 413, Taiwan
5Institute of Plant and Microbial Biology, Academia Sinica, Nankang, Taipei 115, Taiwan
(Received August 2, 2010; Accepted September 21, 2010)
ABSTRACT. Two inter-subspecific segregating populations were constructed to map quantitative trait loci (QTLs) for plant height (PH) and heading date (HD) in rice. A total of 99 and 95 polymorphic SSR and STS markers were used to establish linkage maps in japonica TK8 x indica IR1545-339 and japonica Nipponbare x IR1545-339, respectively. By employing composite interval mapping verified by single point analysis, ten PH and seven HD QTLs were located on seven and six chromosomes, respectively; only Qph1.2 and Qhd3 were common in the two populations. The phenotypic variance explained (PVE) by major PH and HD QTLs was 22.07% (Qph6.1) and 12.98% (Qph6.2), and 35.74% (Qhd3) and 15.54% (Qhd3), for TK8/ IR1545-339 and Nipponbare/IR1545-339, respectively. Comparative genetic analysis of QTL intervals on rice pseudomolecules showing the convergence of QTLs affecting PH and HD across the Oryza genus revealed all QTLs uncovered in the two populations possibly allelic to QTLs previously published. By comparative mapping and candidate gene approach, we specified three PH, Qph1.2, Qph4, and Qph6.1, and one HD QTL, Qhd10, without high-resolution mapping, and the reality of QTLs identified by interval mapping were dis­cussed herein.
Keywords: Comparative genetics; Candidate gene approach; Heading date; Plant height; Quantitative trait locus (QTL); Rice.
INTRODUCTION
mous historical milestone with a semi-dwarf variety, IR8, invoking the "Green Revolution" in the late 1960s. Variet­ies with reduced height can avoid wind and rain damage for resistance to lodging and for increase in yield with ad­equate fertilization by nitrogen. HD, or days to flowering, is one of the critical traits for rice adaptation in diverse environments and rice cultivation in various regions and cropping seasons.
Rice, domesticated 9000 years ago and grown from 55°N and 36°S, contains thousands of cultivated varieties belonging to two species, Oryza sativa grown mainly in Asia and Oryza glaberrima grown mainly in West Africa (Khush, 1997). Thus, rice, as the leading crop worldwide, provides 23% of the calories consumed by humans and feeds more than 50% of population. About 90% of rice is produced in Asia, and the demand for rice in Asia will increase by 40% by 2030 (Khush, 2005). Other than yield components, plant height (PH) and heading date (HD) are two important traits related to yield potential and are det­rimental in rice domestication and modern breeding pro­grams. PH plays an important role in yield improvement during breeding programs, as was shown in the most fa-
In general, plant height of rice is regulated by several genes and influenced by the environment. The dwarf genes, d-1 to d-60, and semi-dwarf genes, sd-1 to sd-7, were found to be induced artificially by radiation or chemi­cals or naturally identified; some were mapped by classical genetic analysis (Kinoshita, 1995). Dozens of PH genes were also detected in various interspecific, inter-subspecif­ic and intra-subspecific crosses (Li et al., 1995; Xiao et al., 1996; Zhuang et al., 1997; Yan et al., 1998; Ishimaru et al., 2001; Yu et al., 2002; Li et al., 2003; You et al., 2006). Quantitative trait loci (QTLs) for PH, isolated to elucidate

*Corresponding author: E-mail: ylin@ntu.edu.tw; Tel: +886­2-33664763.
2
Botanical Studies, Vol. 52, 2011
functions under molecular, biochemical and physiologi­cal levels, participate mostly in the metabolism and signal transduction of phytohormones. Deficiency in gibberellin acid (GA) and brassinosteroids, which can stimulate cell division and elongation, hinders plant growth and results in a dwarf stature. Examples include D18, D35, and sd-1 involving a GA synthetic pathway (Sasaki et al., 2002; Itoh et al., 2004); D1, SLR1, GID1 and GID2 involving a GA signal transduction pathway (Ashikari et al., 1999;
Sasaki et al., 2003; Ueguchi-Tanake et al., 2005); D2 and
D11 involving a brassinosteroid synthetic pathway (Hong et al., 2003; Tanabe et al., 2005); and D61 involving a brassinosteroid signal transduction pathway (Yamamuro et al., 2000). In addition, many genes involving cell division and elongation and development of apical meristem have great effect on PH. Recently, knowledge of the regula­tion mechanism of PH has been incorporated into breeding programs to generate short but high-yield rice (Ashikari et al., 2005).
In rice, HD, or flowering time, is controlled by both qualitative genes (i.e. photoperiodic and circadian genes) and quantitative genes of unknown functions and is influ­enced by conditions other than day length such as tem­perature and so on. A total of 15 QTLs affecting HD were detected by interval mapping of the F2 population and several advanced backcross populations of O. sativa ssp. japonica Nipponbare x O. sativa ssp. indica Kasalath, five of these-Hdl, Hd2, Hd3, Hd5, and Hd6- were regulated by day length (Yano et al., 2001; Yamamoto et al., 2000). Numerous QTLs were also identified from several inter­specific, inter-subspecific and intra-subspecific crosses (Kinoshita, 1995; Li et al., 1995; Zhou et al., 2001; Yu et al., 2002; Li et al., 2003; Thomson et al., 2003; Fujmo et al., 2005; You et al., 2006). To date, some rice HD genes, isolated via positional cloning and gene functions explored at the molecular level, were found to be involved in (1) light-controlled photoperiodic response, such as Hd6 en­coding the a subunit of casein kinase II; Hd3a, functioning as florigen similar to Arabidopsis FLOWERING LOCUS T (FT); and Ehdl, encoding a B-type response regulator acting as a floral inducer under short day; or (2) clock-controlled circadian response, such as Hd1 , which func­tions as a transcription similar to CONSTANS (CO), and Hd2, suspected as a pseudo response regulator (Yano et al., 2000; Takahashi et al., 2001; Kojima et al., 2002; Mu­rakami et al., 2005). The function and structure of genes involved in flowering time in rice, a short-day plant, and in of Arabidopsis, a long-day plant revealed a conserved flo­ral pathway but with minor difference, which the function of Ehd1 is unique to rice (Izawa, 2010).
Allelic variations in HD and PH exist in various natu­ral and bred varieties under natural or artificial selection pressure. QTLs are identified according to gene segrega­tion in a segregating population, which is based on the allelic polymorphism between parental lines. Therefore, continuing to discover QTLs by using different pedigrees will reveal not only novel genes but also genes expressed
consistently, which would be desirable for breeding pur­poses or for positional cloning for further investigation. Here we report on the identification of QTLs affecting PH and HD in two inter-subspecific crosses, O. sativa ssp. japonica cv TK8 x O. sativa ssp. indica IR1545-339 (TK8/IR1545-339), and O. sativa ssp. japonica cv Nip-ponbare x O. sativa ssp. indica IR1545-339 (Nippon-bare/IR1545-339) and their comparison with conservative QTLs across the Oryza genus in public databases. In ad­dition, we sequenced and aligned Nipponbare, TK8 and IR1545-339 alleles of three PH and two HD genes to con­firm the conservative QTLs by comparative mapping.
MATERIALS AND METHODS
Plant materials
Two O. sativa ssp. japonica varieties, TK8, an elite cultivar in Taiwan, and Nipponbare, a popular cultivar in Japan and used for rice genome sequencing, were crossed to an O. sativa ssp. indica variety, IR1545-339, a bacterial blight-resistant variety developed by the International Rice Research Institute (IRRI, Los Banos, Philippines). A total of 304 and 301 F2 individuals derived from TK8/IR1545-339 and Nipponbare/IR1545-339, respectively, were planted in the second crop season in 1999 for evaluating PH. Approximately 16 F2-derived F3 individuals for each F2 plant were grown in the first crop season in 2000 for evaluating HD. All experimental populations were planted at the Taoyuan District Agriculture Research and Exten-sion Station, Taoyuan, Taiwan.
Measurement of phenotypes
PH was measured from soil surface to the top of the panicle of the tallest tiller at maturity in centimeters. HD was from transplantation to the paddy field until complete exertion of one or more panicles with 50% of flowering spikelet, and the HD for each F2 was estimated from the mean HD of approximately 16 F3 progenies.
Molecular marker assay
Three grams of seeds of each F2 individual, about 100 seeds which could represent genotype of F2, were planted in pots for 3 to 4 weeks. The leaves of young seedlings were harvested, lyophilized, and then ground into fine powder for genomic DNA extraction. The procedures of DNA extraction followed Li et al. (1995) with minor modifications. About 0.5 gram of powdered leaf tissue was added to 9 ml of extraction buffer (100 mM Tris, pH 8.0; 50 mM EDTA, pH 8.0; 500 mM NaCl; 1.25% SDS; fresh-made 0.38% of NaHSO3), mixed well, and incubated in a 65°C water bath for 1 h with occasional inversion of the mix. The extraction mixture was added to 2.7 ml of 5 M potassium acetate, gently mixed by inversion, and then set on ice for 20 min. Supernatants were collected after centrifugation at 3,500 rpm (TJ-25 Centrifuge, Beckman Coulter, USA) for 20 min at 65°C and were added to 10 ml of ice-cold isopropanol to precipitate DNA. DNA pel-
LIN et al. ― QTL for rice plant height and heading date across Oryza genus
3
lets were hooked, soaked in 3 ml purifying buffer (70% of ethanol, 0.3 M sodium acetate) overnight, washed with cold 70% of ethanol, air dried, and then dissolved in 200 (il of TE (10 mM Tris, pH 8.0; 1 mM EDTA, pH 8.0).
Two types of PCR-based markers, SSRs and sequence tagged sites (STSs), previously mapped on the 12 rice chromosomes or discovered by bioinformatics (IRGSP, 2005), were used to establish linkage maps. Primers of SSR and STS markers were synthesized by GENSET (Singapore Biotech), and sequences are available from Gramene and the Rice Genome Research Program. The PCR reaction, containing 50 ng genomic DNA, 200 (M of each deoxynucleotide triphosphate (ABgene, UK), 0.2 (iM of forward and of reverse primers, and 0.5 unit of ProTaq and 1x Taq buffer (Protech, Taiwan) in a final volume of 25 (il, was amplified by use of a thermocycler (Model T1, Biometra, Germany) at 94°C for 1 min for 1 cycle; 94°C for 1 min, 55°C for 1 min, 72°C for 2 min for 35 cycles; and 72°C for 5 min for 1 cycle. Three (il of amplified DNA products was separated on 2.5% of SFR agarose (Amresco®, USA) by using RAGE electrophoresis system (Rapid Agarose Gel Electrophoresis, Cascade Biolog-ics, Oregon, USA) in 1 x TAE at 250 V for 9 to 23 min,depending on the size difference between amplified DNA fragments of SSR or STS alleles.
Comparative analyses of QTLs affecting PH and HD
The QTLs detected in this study underwent comparative genetic analysis across the Oryza genus. The complete sequencing of the rice genome allows for identifying ho-mology on the basis of overlapped intervals along physical maps (IRGSP, 2005). We performed a BLAST search of TIGR Rice Genome Pseudomolecules V.4 with the primer sequences of SSR and STS markers flanking QTLs and se­quences of cloned genes for PH and HD to delineate phys­ical chromosome segments of QTL intervals. In addition, physical positions of some published QTLs and markers are available from Gramene. Only physical intervals of QTLs overlapping with the identified QTLs in our studies were considered homology.
After comparative analysis by physical mapping, we identified numerous published PH and HD QTLs cor­responding to our discovery. Three PH related genes, Sd1 (accession no. AF465255), D35 (accession no. AK287607), and OsKSl (accession no. AB126933), and two HD genes, Ehd1 (accession no. AB092506) and Hd1 (accession no. AB041838), were subjected to analyze the alleles of IR1545-339 and TK8 to reveal the reality of QTL mapping and comparative QTL mapping. The recessive allele of sd1 is well-known as 383-bp deletion (Sasaki et al., 2002), the primer sequences flanking the deletion were designed, sd1-del-F: 5'-CACAgCgCTCACTTCTCATC-3' and sd1-del-R: 5'-CTTACATggCgTCgTCACAC-3'. For the other four genes, primer walking based on the pub-lished gene sequences of Nipponbare were employed to accomplish those gene sequences of IR1545-339 and TK8.
Analyses of linkage maps and QTLs affecting PH and HD
The linkage maps were constructed with 99 and 95 polymorphic markers for TK8/IR1545-339 and Nippon-bare/IR1545-339, respectively. Genetic distance between pairs of markers was estimated by use of MAPMAKER/ EXP, version 3.0 with the Kosambi function (Lander et al., 1987). The LOD threshold was fixed at 3.5 for unlinked markers, with genetic distance larger than 40 cM.
QTL analysis was employed by use of QTL Cartogra­pher, version 2.0 (Wang et al., 2001). To investigate the association between phenotypic variation and marker gen­otype, we first used single point analysis (SPA) to detect the linkage of quantitative traits to markers individually by F-test at four significance levels: a= 0.05, 0.01, 0.001, and 0.0001. Then, composite interval mapping (CIM) was used to detect 90% confidence intervals of chromo­somal regions of putative QTLs. Model 6, which uses a specified number of markers that fall outside a window around the flanking markers, was applied. A window size of 2 cM was used to scan the whole genome with forward stepwise regression. A LOD threshold of 2.4 was needed p = 0.05 for declaring the significance of a putative QTL. QTLs detected through CIM analysis were assigned with an abbreviation of the trait name (HD for heading date and PH for plant height) and the chromosome number, and numbers were affixed at the end if more than one locus was identified on the same chromosome. The phenotypic variance explained by the putative QTLs was estimated by the coefficient of determination (R2), with maximum like­lihood used for CIM (Basten et al., 2001).
RESULTS
Genetic variation for plant height and heading date
The PH of the tallest tiller of parents, TK8, Nipponbare,
and IR1545-339 was 108, 90, and 100 cm, respectively.
The mean PH was 108.3 (±12.7) cm (range 73 to 141 cm) and 101.0 (±12.2) cm (range 70 to 132 cm) for the F2 pop­ulations of TK8/IR1545-339 and Nipponbare/IR1545-339,
respectively (Figure 1). The male parent IR1545-339 was shorter than the female parent TK8 by 8 cm but was taller than the female parent Nipponbare by about 10 cm, and transgressant progenies were found in both F2 populations of the two pedigrees (Figure 1). The phenotypic segrega­tions in these two F2 populations exhibited normal distri­bution, a typical phenomena of quantitative a trait, which indicates that PH was regulated by several genes and in­fluenced by the environment.
Days to heading, the date of exertion of at least one panicle with 50% of the flowering spikelet, of the parents
TK8, Nipponbare, and IR1545-339 were 93, 69, and 103
days, respectively (Figure 1). Mean HD of approximately 16 F3 progenies for each F2 individual was 95.6 (±5.3) days (range 80 to 122 days) for TK8/IR1545-339 and 84 (±4.5) days (range 72 to 99 days) for Nipponbare/IR1545-339. The male parent IR1545-339 showed delayed heading, by approximately 10 and 34 days, as compared with TK8
4
Botanical Studies, Vol. 52, 2011
and Nipponbare, respectively, and transgressant progenies were observed only in TK8/IR1545-339 and not in Nip-ponbare/IR1545-339. The frequency distributions for HD in these two crosses were normal (Figure 1), which indi­cates that HD was a polygenic inheritance.
serve as anchor markers to align two linkage maps for further comparative analysis of QTL map position. How­ever, 22 and 18 markers were specifically polymorphic to TK8/IR1545-339 and Nipponbare/IR1545-339 and are indicated with superscripts T and N, respectively (Figure 2). The differential marker polymorphism revealed al-lelic variation among TK8, Nipponbare, and IR1545-339, which implied various alleles of QTLs affecting PH and HD and different QTLs with some common QTLs would be detected in these two crosses.
Linkage maps
A total of 99 polymorphic markers, 88 SSR and 11 STS, comprising 90% of the rice genome as referred to the established map published by IRGSP (http://rgp.dna.affrc. go.jp/E/IRGSP/index.html), were employed in the 304 F2 progenies of TK8/IR1545-339. Twelve linkage groups spanning 1342.9 cM, with an average distance of 16.65 cM, were established (Figure 2). The linkage map of Nip-ponbare/IR1545-339 was composed of 95 polymorphic markers, 75 SSR and 20 STS, applied to 301 F2 progenies. The map included 12 linkage groups spanning 1027.4 cM, with an average distance of 10.81 cM, and covered 85% of the rice genome. All marker orders of these two linkage maps agreed with the physical maps obtained by BLAST search against rice genome sequences (IRGSP, 2005), ex­cept three intervals of RM1387/RM104, RM1367/RM263, and R20049/RM16 located on chromosomes 1, 2 and 3, respectively.
QTL analyses
We employed SPA and CIM to detect segregating QTLs affecting PH and HD. Large QTLs with more phenotypic variance explained (PVE) effects deduced from CIM analysis usually show significance. However, two of ten PH QTLs, Qph4 and Qphll, and one of seven HD QTLs, Qhd9, did not show significance by SPA; these three QTLs were detected by relatively small LOD and contributed less PVE (Table 1). This discrepancy of QTL results is commonly found in several QTL studies involving SPA, interval mapping (IM), CIM, and multiple interval map­ping (MIM) because of different algorithms and statistical methods used. Since interval mapping involves the use of genotype information of two markers simultaneously rather than one marker a time, we used PH and HD QTLs with intervals of significant LOD after CIM analysis.
A total of 117 polymorphic markers were applied in these two F2 populations for QTL analyses. Seventy-seven (66%) markers were commonly polymorphic between TK8/IR1545-339 and Nipponbare/IR1545-339 and could
Mapping QTLs for plant height
Four QTLs affecting PH were identified in TK8/
IR1545-339 (Figure 2), with PVE ranging from 4.17% to
22.07% (Table 1). Two of the four, Qph1.2 and Qph6.1, contributed large PVE of 19.16% and 22.07%, respective­ly, and the other two, Qph2.1 and Qph9, contributed small PVE of 4.17% and 5.28%, respectively. All four showed partial dominance. The three alleles from TK8, Qph1.2TK,
Qph2.1TK, and Qph9TK, increased PH by 9.63, 4.08, and
4.05 cm, respectively. The allele Qph6.1IR, in contrast, in­creased PH by 8.38 cm (Table 1).
Seven PH QTLs were detected in Nipponbare/IR1545-
339, with PVE ranging from 3.28% to 12.98% (Table 1).
Two of seven QTLs contributed PVE greater than 12% and the other five QTLs contributed PVE less than 5%. The Nipponbare alleles of four PH QTLs, QphL1NBQphl.f, Qph4m, and Qphl严,provided increased height, and the
IR1545-339 alleles for the other three QTLs, Qph2.2IR,
Qph5IR, and Qph6.2IR, provided increased height. These three QTLs displayed an overdominant manner, with the other four an additive manner.
We uncovered 10 QTLs affecting PH, including one mapped on chromosome 1, with flanking markers E60551 and RRM1387 identified in both populations. In addition, the interval for Qph6.1 identified in TK8/IR1545-39 was mapped between R3879 and RM30 on chromosome 6, which is the next adjacent interval to Qph6.2, RM30 and RM340. Qph6.1 and Qph6.2 could belong to the same PH gene because of unresolved mapping with the genotyping
Figure 1. Frequency distributions for plant height and heading date in segregating populations derived from the intra-subspecies crosses between japonica TK8 x indica IR1545-339 and japon­ica Nipponbare x indica IR1545-339. The arrows indicate the intervals of trait values of parents TK8 Nipponbare (™*), and IR1545-339 (―), and of F2 The numbers in parenthe­ses represent the standard deviation for plant height and heading date.
LIN et al. ― QTL for rice plant height and heading date across Oryza genus                                                                                           5
Figure 2. QTLs for plant height and heading date along the genetic linkage map of two F2 populations derived from TK8/IR1545-339 and Nipponbare/IR1545-339. The 90% of confidence intervals of QTLs is plotted with the LOD score peak indicated by a solid tri­angle.
and phenotyping data analyzed by CIM. However, these two QTLs mapped on chromosome 6 identified in these two populations independently contributed the largest PVE with incongruity (Table 1).
three acted as partial dominance in TK8/IR1545-339. We identified four QTLs affecting HD in Nipponbare/
IR1545-339, with PVE ranging from 5.06% to 15.54%.
The largest, Qhd3, mapped between RM3203 and RM545 on chromosome 3, promoted heading and explained
15.54% of the PVE with an additive effect of 4.58 days
and dominant deviation of 1.5 days. However, two al-leles, Qhd8.2NB and Qhd9NB, postponed heading for about two days each (Table 1). Three QTLs, Qhd1, Qhd8.2, and Qhd9, performed as partial dominance and one, Qhd8.1, performed as overdominance. The QTL intervals of Qhd8.1 and Qhd8.2 did not overlap, and gene actions and allele contributions differed, which supports that the two HD QTLs were detected on chromosome 8.
We uncovered seven QTLs affecting HD. Qhd3 contrib­uted the largest PVE of HD, and Qhd3m delayed flowering for approximately five days in both crosses with incongru­ity.
Mapping QTLs for heading date
Four QTLs affecting HD were identified in TK8/
IR1545-339 (Figure 2), with PVE ranging from 3.46%
to 35.74% (Table 1). The largest, Qhd3, was mapped between RM3203 and RM545 on chromosome 3 and con­tributed up to 35.74% of the PVE. The QTLs, Qhd7 and Qhd10, also had large effects on HD, with 12.66% and 12.91% of PVE, respectively (Table 1). Two TK8 alleles,
Qhd3TK and Qhd7TK, and two IR1545-339 alleles, Qhd2IR
and Qhdl0IR, promoted heading in the first crop season in 2000; the two TK8 alleles promoted heading, for a total of approximately 9 days, whereas the IR1545-339 alleles pro­moted heading for three days only. One of the four QTLs, Qhd2, exhibited significant underdominance, but the other
6
Botanical Studies, Vol. 52, 2011
Table 1. Parameters of QTLs for plant height and heading date detected in the crosses of TK8 x IR1545-339 and Nipponbare x IR1545-339.

Trait Chr

QTLs

Flanking markersa


TK8 x IR1545-339


Nipponbare x IR1545-339

LOD

ab

d

d/[a]b

PVE%c

LOD

ab

d

d/[a]b

PVE%c

Plant Height 1

Qphl.1

RM580 - RM246**






2.64

4.73

0.57

0.12

3.28
1
Qph1.2
E60551 - RM1387***
7.84
9.63
3.51
0.36
19.16
7.96
6.46
-2.22
-0.34
12.86
2
Qph2.1
RM211 - RM5356*
3.03
4.08
2.05
0.50
4.17
2
Qph2.2
RM263 - RM250***
3.38
-0.74
4.83
6.52
4.78
4
Qph4
RM252 - RM567NS
2.96
6.36
3.02
0.47
4.64
5
Qph5
RM430 - RM480***
2.80
-1.97
2.89
1.47
3.87
6
Qph6.1
RM541 - RM30***
16.49
-8.38
1.12
0.13
22.07
6
Qph6.2
RM3879 - RM340***
9.34
-2.32
6.77
2.92
12.98
9
Qph9
RM3912 - RM278*
3.11
4.05
2.91
0.72
5.28
11
Qphll
E21117 - RM21NS
2.42
5.42
2.14
0.39
3.94
Heading Date 2
Qhd2
RM240 -RM207*
3.29
0.19
-2.00
-10.26
3.46
3
Qhd3
RM3203 - RM545***
23.47
-5.21
-3.01
-0.58
35.74
10.62
-4.58
-1.50
-0.33
15.54
7
Qhd7
RM214 - RM11*
5.68
-3.61
-2.65
-0.73
12.66
8
Qhd8.1
RM1019 - RM5432***
3.50
-0.40
2.27
5.67
7.89
8
Qhd8.2
RM331 - RM447**
2.89
2.35
1.30
0.55
5.06
9
Qhd9
RM3912 - RM278NS
2.70
2.44
1.84
0.76
5.39
10
Qhd10
C51124 - RM258**
9.46
2.16
-2.00
-0.92
12.91

a The significance levels of nearest markers detected by SPA; NS, *, **, and *** indicate nonsignificant, p<0.05, 0.001, and 0.0001, respectively.
b The direction of additive effects was from either TK8 or Nipponbare in the two population. d/[a] indicates degree of dominance.
cPhenotypic variance explained by individual QTL.
Comparative QTL analyses of plant height and heading date
Positional cloning and sequencing of Sd-1 alleles re­vealed that a recessive allele of miracle rice IR8, derived from a native Taiwanese variety, Dee-geo-woo-gen, had a 383-bp deletion, which led a mutation of GA 20-oxidase (Sasaki et al., 2002). By using the primers flanking the deletion, the amplicons of IR1545-339 was truncated but TK8 and Nipponbare were in full-length (Figure 3). As the evidence, Qphl.2, identified in both populations, was because of segregation of Sd-1. After annotation of D35 sequences, one amino acid substitution was found in exon III, which isoleucine in IR1545-339 but valine in TK8 and Nipponbare. However, Qph6.1 was only found in TK8/IR1545-339 but not in Nipponbare/IR1545-339 (Table 2, Figure 3). It was possible that the adjacent PH QTL with largest effect, Qph6.2, impeded the identification of D35 segregation in Nipponbare/IR1545-339. The sequence alignment of OsKS1 showed no polymorphism at the cod­ing region among three parents. Nevertheless, Nipponbare possessed different allele to IR1545-339 and TK8 at up­stream of 5'-UTR region with 23 SNPs (Figure 3), which was presumed to influence gene expression of OsKSl lead-ing to plant height variation because of these two alleles segregating in the F2 population of Nipponbare/IR1545-339 (Table 2, Figure 2).
By QTL convergence comparison, we found 10 PH QTLs potentially homologous to other QTLs by signifi­cant overlap of intervals, and none were novel. Half of the 10 PH QTLs we identified corresponded to three dwarf genes, d30, d57, d27, and two semi-dwarf genes, sd-1 and d35. The chromosome intervals of PH QTLs correspond­ing to Sd-1 were identified from at least eight pedigrees besides our studies (Table 2). In our studies, one PH QTL, Qph1.2, corresponding to Sd-1, was identified in both populations, with a main contribution to PVE (Table 1). Qph6.1, mapped in TK8/IR1545-339 only, was cor­responded to another isolated PH gene is D35, encoding ent-Kaurene oxidase, and the recessive allele d35 is defec­tive in GA synthesis (Itoh et al., 2004). The intervals of Qph11, corresponding to d27 and Qph4 encompass OsKS8 and OsKS1 , respectively, which both encode ent-Kaurene synthase, catalyzing the early step in GA biosynthetic pathway. Additionally, the intervals of Qph1.1 and Qph5 encompass OsGA2ox2 and OsGA20ox4, respectively, which both encode GA 2- and GA 20-oxidase, catalyzing the later steps in the GA biosynthetic pathway (Sakamoto et al., 2004).
LIN et al. ― QTL for rice plant height and heading date across Oryza genus                                                                                           7
Our comparative analysis with HD QTLs detected by independent research groups allowed us to speculate on the function or effects of QTLs we identified herein. On searching chromosome segments of rice pseudomolecules of these seven HD QTLs for possible orthologous QTLs from other pedigrees, only Qhd3, segregated in TK8/
IR1545-339 and Nipponbare/IR1545-339, corresponded to
at least two QTLs, including Hd9. Qhd10 might be allelic to Hd14, dth10.1, and Ehd1, on the basis of the physical interval of C51124 and RM258. We also found possible allelic QTLs identified from previous studies to five HD QTLs, Qhd2, Qhd7, Qhd8.1, Qhd8.2, and Qhd9 (Table 3).
Table 2. The convergence of QTLs affecting plant height identified in this study to those identified in previous studies.

Chr

QTLs

Flanking markers

Physical positiona

Cross designb

Corresponding QTLs or nearest marker

Reference

1

Qphl.l

RM580

-RM246

9,603,452 - 27,663,181

N

Phl

Xiao et al., 1996
Qphl
Li et al., 2003
Phl.l
Septiningsih et al., 2003
Phl.l
Thomson et al., 2003
Qphlb
Mei et al., 2005
1
Qphl.2
E6055l
-RMl387
35,012,879 ~ 40,534,591
T, N
Sd-l
Xiao et al., 1992
Phl
Yan et al., 1998
qPHl
He et al., 2001
R24l4
Ishimaru et al., 2001
Phl.l, phl.2
Moncada et al., 2001
qPHT-l
Hittalmani et al., 2002
Phl.l,
Septiningsih et al., 2003
Phl.2
Thomson et al., 2003
Qphla
Mei et al., 2005
2
Qph2.l
RM2ll -
- RM5356
2,020,685 ~ 9,432,636
T
d30
Xiao et al., 1992
Ph2
Xiao et al., 1996
2
Qph2.2
RM263
- RM250
25,865,402 ~ 32,774,538
N
Ph2
You et al., 2006
4
Qph4
RM252
- RM567
28182082 ~ 34,500,023
N
Ph4
Zhuang et al., 1997
Ph4.l
Moncada et al., 2001
QPh4c
Li et al., 2003
5
Qph5
RM430
- RM480
18,670,780 ~ 27,292,849
N
Ph5
Xiao et al., 1996
Ph5-2
Yan et al., 1998
QPh5
Li et al., 2003
Ph5.l
Thomson et al., 2003
Ph5
You et al., 2006
6
Qph6.l
RM54l
- RM30
19,513,541 - 27,252,381
T
D35
Itoh et al., 2004
Ph6
Xiao et al., 1996
Ph6.l
Thomson et al., 2003
6
Qph6.2
R3879 -
RM340
25,926,402 - 28,599,297
N
Ph6
Xiao et al., 1996
Qph6
Mei et al., 2003
9
Qph9
R39l2 -
RM278
10,826,231 ~ 27,062,430
T
d57
Kinoshita, 1995
QPh9a
Li et al., 2003
Ph9b
You et al., 2006
11
Qphll
E2lll7
- RM2l
4,553,531 - 23,239,202
N
d27
Abenes et al., 1994
Phlla
You et al., 2006

aThe physical map of QTLs are based on TIGR rice genome pseudomolecules V.4
bT and N of the cross design are indicated TK8/IR1545-339 and Nipponbare/IR1545-339, respectively.
8
Botanical Studies, Vol. 52, 2011
The function study of Ehdl , a photoperiodic gene regu­lating the transition of vegetative stage to reproductive stage, revealed that the mutation of glycine substituted by arginine in GARP domain was photoperiod insensi­tive under short day condition (Doi et al., 2004). After annotation analysis of Ehd1 sequences, both IR1545-339 and Nipponbare possessed the same wild type allele but TK8 possessed the mutated allele. Thus, Qhdl0 was only identified in TK8/IR1545-339 (Table 1, Figure 2). Hdl, mapped on chromosome 6, is another rice photoperiodic gene corresponding to CONSTANS in Arabidopsis (Yano et al., 2000). Three different alleles were identified that Nipponbare had the full length of wild type allele, but IR1545-339 and TK8 had four and two-bp deletion at the exon 2, respectively (Figure 3). No QTL was uncovered in Nipponbare/IR1545-339, which was discordant to the expectation of the segregation of function alleles of Nip-ponbare and null-function of IR1545-339. Nevertheless, it
was consistent to the no QTL detected in TK8/IR1545-339
because of null function by deletion.
DISCUSSION
Assessments of PH and HD QTLs in TK8/ IR1545-339 and Nipponbare/IR1545-339
Quantitative traits such as plant height and flower­ing time are regulated by multiple genes, environments, and interactions between genes (epistasis) and between genes and environments (GxE). Nevertheless, in specific crosses, QTLs can be detected as segregating loci only, not nonsegregating loci, even loci involved in regulating morphology. More QTLs can be identified by using vari­ous cross combinations of elite cultivars, wild accessions, and relative species. In this study, we crossed two elite japonica cultivars, TK8 and Nipponbare, to a wild indica accession IR1545-339, and used SPA and CIM analysis
Table 3. The convergence of QTLs affecting heading date identified in this study to those identified in previous studies.

Chr

QTLs

Flanking markers

Physical positiona

Cross designb

Corresponding QTLs or nearest marker

Reference

2

Qhd2

RM240 -RM207

31,497,147 ~ 35,369,659

T, N

C560

Lin et al., 1998
Hd7
Yamamoto et al., 2000
C560
Ishimaru et al., 2001
dth2.l
Moncada et al., 2001
dth2.2
Septiningsih et al., 2003
QHd2b
Li et al., 2003
3
Qhd3
RM3203 - RM22
776,123 ~ 1,499,104
T, N
QHd3
Mei et al., 2003
Hd9
Yano et al., 2001
QHd3a
Mei et al., 2005
7
Qhd7
RM2l4 - RMll
12,782,791 ~ 19,256,338
T
Hd4
Yano et al., 1997
dth7.l
Moncada et al., 2001
dth7.l
Thomson et al., 2003
qDTH-7-l
Fujino et al., 2005
hd7
You et al., 2006
8
Qhd8.l
RMl0l9 - RM5432
195,984 ~ 4,371,994
N
Cl66
Wang et al., 2002
qDTH8
Miyata et al., 2007
8
Qhd8.2
RM33l - RM447
12,288,944 ~ 26,416,973
N
QHd8a
Li et al., 2003
QHd8
Mei et al., 2005
9
Qhd9
RM39l2 - RM278
10,826,231 ~ 27,062,430
N
FTLQ3
Sarma et al., 1998
QTL9
Zhou et al., 2001
QHd9
Li et al., 2003
10
Qhdl0
C5ll24 - RM258
7,180,112 - 17,756,238
T
Hdl4
Yano et al., 2001
dthl0.l
Thomson et al., 2003
Ehdl
Doi et al., 2004

aThe physical map of QTLs are based on TIGR rice genome pseudomolecules V.4
bT and N of the cross design are indicated TK8/IR1545-339 and Nipponbare/IR1545-339, respectively.
LIN et al. ― QTL for rice plant height and heading date across Oryza genus
9
Figure 3. The alleles of three plant height genes (A) and two heading date genes (B) among Nipponbare, IR1545-339, and TK8. From the comparative analysis, the plant height genes, Sd-l, D35, and OsKSl, were corresponding to Qphl.2, Qph6.l, and Qph4, while the heading date gene, Ehdl, was corresponding to Qhdl0. The * above nucleotide or amino acid indicates variation.
to identify PH and HD QTLs. In both TK8/IR1545-339 and Nipponbare/IR1545-339, only 66% of 117 mark­ers were commonly polymorphic, and nearly all markers displayed three alleles among three parents. In addition, 22 and 18 specific polymorphic markers were employed independently to each cross, respectively. Various allelic polymorphisms in the two crosses implied various QTLs segregating in the F2 populations, which can be detected by phenotyping, genotyping, and linkage analysis. Only one PH QTL, Qphl.2, and one HD QTL, Qhd3, were identified in both populations, and nine PH and six HD QTLs were unique to either population (Table 1). Thus, using more pedigrees of various parent combinations can help uncover
more QTLs.
The phenotypic frequency distribution of quantitative traits is continuous with transgressive progenies whose phenotypes are beyond the phenotypes of parents. Most identified QTLs in this study performed as partial domi­nance, except for four PH QTLs in Nipponbare/IR1545-339 and one HD QTL in both populations (Table 1). The major reason for transgressive segregation could be com­plementary gene action for QTLs with opposite effects of parents (Thomson et al., 2003; Yano et al., 1997). In our study, the IR1545-339 alleles of PH and HD QTLs could either increase or decrease PH and HD in both F2 popula­tions, which is consistent with the above hypothesis.
10
Botanical Studies, Vol. 52, 2011
The PVE of PH QTLs, four in TK8/IR1545-339 and
seven in Nipponbare/IR1545-339, ranged from 4.17% to
22.07% and 3.28% to 12.98%, respectively. The pheno-
typic variation explained by the four HD QTLs in TK8/
IR1545-339 and Nipponbare/IR1545-339 ranged from 3.46% to 35.74% and from 5.06% to 15.54%, respec­tively. More QTLs might have segregated in these two F2 populations but were unable to be detected because of environmental or GxE influences, masking by QTLs with large effects, or resolution of statistical analysis. Use of advanced backcross designs by fixing QTLs with large ef­fects could detect more QTLs with minor effects. From our study, no HD QTL corresponding to Hdl was detected Nipponbare/IR1545-339 even if the functional alleles of Nipponbare and null function of IR1545-339 should segregate in the F2 population (Figure 3). Both Hdl and Ehdl, regulating the mobile florigen Hd3a expression, are involved in the photoperiod control of flowering in rice and promote heading under short day condition (Izawa, 2010). From the study of heading genes from three seg­regating population of BC3F2 derived from crosses of donor parent Tainung 67 and recurrent parent Koshihikari, Ehdl could explain 21.6~36.8% of PVE but Hdl could explain 2.5~6.5% only (Y.-R. Lin et al. unpublished data). The plausible possibility that the small effect of Hdl was masked by Ehdl for detection since the natural day length less than 13.5 hours in Taiwan is short day for heading. Because of adjacent to Qph6.2, Qph6.l, corresponding to
D35 was only found in TK8/IR1545-339 but not in Nip-
ponbare/IR1545-339 even the same allele of TK8 and Nip-ponbare (Table 2, Figure 3). By candidate gene approach, we identified three PH and one HD gene without fine map­ping and validate the reality of QTLs by interval mapping in agreement with the proposals of confirmation (Price, 2006; Yamamoto et al., 2009). Nevertheless, we cannot rule out the possibility that the phenotypic variation was caused by other genes resided in the same QTL intervals of 5~10 Mb. For example, RFTl and Hd3a are essential for rice flowering and are 11.5 kb apart chromosome 6 (Komiya et al., 2008). Even if there are some variations of DNA and amino acid after comparison, gene expression of the candidate genes for the three PH and one HD gene could provide further confirmation in three parental variet­ies.
In the grass family Poaceae, flowering always termi­nates vegetative growth, which results in a highly signifi­cant relation between plant height and flowering time. However, we found nearly no phenotypic correlation
between PH and HD in TK8/IR1545-339 and Nipponbare/
IR1545-339 (Spearman's rho = 0.01 and = 0.11, respec­tively) and no overlapped or closely linked QTLs affecting PH and HD from CIM analysis. Even though all QTLs, ten PH and seven HD, were uncovered independently in these two crosses, only two QTL intervals overlapped. Phenotypic correlation and gene identification analyses demonstrated no correlation between PH and HD in the two pedigrees.
Convergence of PH and HD QTLs across the Oryza genus
QTLs affecting PH and HD in rice have been exten­sively mined by linkage analyses of classical morpholo­gies and molecular markers. A total of 1011 PH and 618 HD QTLs, annotated by Gramene in 2008, were published between 1994 and 2006 (Yamamoto et al., 2009). Four and seven QTLs affecting PH were identified in TK8/
IR1545-339 and Nipponbare/IR1545-339, respectively
(Table 1, Figure 2). By QTL convergence comparison, all 10 PH QTLs identified in this study showed potentially homologous to other published QTLs. The convergence of QTLs identified across varieties and species provide addi­tional confirmation of QTL existence and also indicate the main contributions against evolution or artificial selection during breeding, and might be worth cloning for further investigation of biological function and application in rice breeding (Price, 2006; Yamamoto et al., 2009).
PH is significantly correlated with grain yield, so re­duction in height could improve lodging resistance and harvest index. Qphl.2, identified in both populations with a main contribution to PVE, was corresponding to Sd-l which also uncovered from at least eight pedigrees (Tables 1, 2). The amplicons of Sd-l revealed Sd-l al-leles, Qphl.2TK and Qphl.2NB, were dominant and in­creased plant height by 9.63 and 6.46 cm, respectively, and Qphl.2IR was recessive resulting in reduced PH (Table 1, Figure 3). Three other dwarfing alleles encoding pro­teins with one amino acid substitution also arose from natural mutation or from gamma-ray (Sasaki et al., 2002; Spielmeyer et al., 2002). Other research groups have mapped PH QTLs corresponding to Sd-l (Table 2); the sd-l alleles from these studies might be the same four al-leles mentioned above and be introgressed to these variet­ies of QTL mapping from the same lineage, or be different alleles. Even the recessive allele with 383-bp deletion of the Sd-l locus originating from a native Taiwanese variety, 'Dee-geo-woo-gen', none popular semi-dwarf varieties were derived from this sd-l allele in Taiwan. However, se­lecting different sd-l alleles for a desired effect on PH had been incorporated into rice breeding programs for produc­ing ideotype varieties (Asano et al., 2007).
HD is one of critical characters for rice adaptation in various environments, and one of pivotal characters for rice breeding because of its major association with grain filling, yield, and quality. Two major factors affecting HD are basic vegetative growth (BVG) and photoperiod sen­sitivity, of which day length is a key environmental factor for triggering photoperiod-sensitive varieties to flower in rice. Photoperiod-insensitive cultivars such as TK8, Tainung67 (TNG67), and Taichung65 (TC65) were de­veloped for allowing two crops a year in Taiwan, but Nip-ponbare is a photoperiod-sensitive cultivar in Japan. From sequenced and function-identified photoperiodic genes, alleles of TC65, TK8, and TNG67 for Hdl and Ehdl pos­sessing loss of function led these Taiwanese cultivars into photoperiod insensitivity (Doi et al., 2004; Chen at al.,
LIN et al. ― QTL for rice plant height and heading date across Oryza genus
11
2010). Ehdl has major effect on BVG that wild type allele promote heading under short day condition. The recessive ehdl allele of one amino substitution (G^R) of GARP domain was unique to TC65 and other Taiwan modern cultivars (Figure 3; Saito et al., 2009). This explained why Qhdl0 was only identified in TK8/IR1545-339 but not in Nipponbare/IR1545-339 (Table 1, Figure 2). Nevertheless, environmental factors other than day length, such as tem­perature and gene interaction (Luan et al., 2009), might account in part for failure of detection Hdl gene in Nip-ponbare/IR1545-339.
The QTLs identified by interval mapping and then by comparative mapping for QTL convergence in this study provide additional information about the natural varia­tion of PH and HD QTLs during evolution and breeding. Through the candidate gene approach, we isolated four conserved QTLs without high-resolution mapping. QTLs conserved across many pedigrees might be worth isolat­ing by candidate gene approach for further investigation and applied to breeding programs. The novel HD QTL we found might be enriched our understanding about flower­ing time in rice.
Doi, K., T. Izawa, T. Fuse, U. Yamanouchi, T. Kubo, Z. Shi-matani, M. Yano, and A. Yoshimura. 2004. Ehdl, a B-type response regulator in rice, confers short-day promotion of flowering and controls FT-like gene expression indepen­dently of Hdl. Genes Dev. 18: 926-936.
Fujino, K. and H. Sekiguchi. 2005. Mapping of QTLs conferring extremely early heading in rice (Oryza sativa L.). Theor.
Appl. Genet. 111: 393-398.
He, P., J.Z. Li, X.W. Zheng, L.S. Shen, C.F. Lu, Y. Chen, and L.H. Zhu. 2001. Comparison of molecular linkage maps and agronomic trait loci between DH and RIL populations derived from the same rice cross. Crop Sci. 41: 1240-1246.
Hittalmani, S., H.E. Shashidhar, P.G. Bagali, N. Huang, J.S. Sid-
hu, VP. Singh, and G.S. Khush. 2002. Molecular mapping of quantitative trait loci for plant growth, yield and yield related traits across three diverse locations in a doubled haploid rice population. Euphytica 125: 207-214.
Hong, Z., M. Ueguchi-Tanaka, K. Umemura, S. Uozu, S. Fujio-ka, S. Takatsuto, S. Yoshida, M. Ashikari, H. Kitano, and M. Matsuoka. 2003. A rice brassinosteroid- deficient mutant, ebisu dwarf (d2), is caused by a loss of function of a new member of cytochrome P450. Plant Cell 15: 2900-2910.
IRGSP, International Rice Genome Sequencing Project. 2005. The map-based sequence of the rice genome. Nature 436:793-800.
Ishimaru, K., M. Yano, N. Aoke, K. Ono, T. Hirose, S.Y. Lin, L. Monna, T. Sasaki, and R. Ohsygi. 2001. Toward the map­ping of physiological and agronomic characters on a rice function map: QTL analysis and comparison between QTL and expressed sequence tags. Theor. Appl. Genet. 102: 793­800.
Itoh, H., T. Tatsumi, T. Sakamoto, K. Otomo, T. Toyomasu, H. Kitano, M. Ashikari, S. Ichihara, and M. Matsuoka. 2004. A rice semi-dwarf gene, Tan-Ginbozu (D35), encodes the gibberellin biosynthesis enzyme ent-kaurene oxidase. Plant Mol. Biol. 54: 533-547.
Izawa, T. 2010. Photoperiodic control of flowering in the short-day plant Oryza sativa. In R. J. Nelson et al. (eds.), Photo-periodism the Biological Calendar, Oxford, pp. 38-58.
Khush, G.S. 1997. Origin, dispersal, cultivation and variation of
rice. Plant Mol. Biol. 35: 25-34.
Khush, G.S. 2005. What it will take to feed 5.0 billion rice con­sumers in 2030. Plant Mol. Biol. 59: 1-6.
Kinoshita, T. 1995. Report of committee on gene symbolization. Nomenclature and linkage groups. Rice Genet. Newsl. 12:
9-153.
Kojima, S., Y. Takahashi, Y. Kobayashi, L. Monna, T. Sasaki, T. Araki, and M. Yano. 2002. Hd3a, a rice ortholog of the Arabidopsis FT gene, promotes transition to flowering downstream of Hdl under short-day conditions. Plant Cell
Physiol. 43: 1096-1105.
Komiya, R., A. Ikegami, S. Tamaki, S. Yokoi, and K. Shimamo-to. 2008. Hd3a and RFTl are essential for flowering in rice. Development 135: 767-774.
Acknowledgements. This work was supported by the Na­tional Science Council (89 to 91-B-030-010-) and the The­matic Project, Academia Sinica, Taipei, Taiwan. We also appreciate Chih-Chi Lin and Ming-san Wu for technical assistance and three anonymous reviewers for their valu­able comments.
LITERATURE CITED
Abenes, M.L.P., R.E. Tabien, S.R. McCouch, R. Ikeda, P. Ronald, G.S. Khush, and N. Huang. 1994. Orientation and integration of the classical and molecular genetic maps of chromosome 11 in rice. Euphytica 76: 81-87.
Asano, K., T. Takashi, K. Miura, Q. Qian, H. Kitano, M. Mat-
suoka, and M. Ashikari. 2007. Genetic and molecular analy­sis of utility of sdl alleles in rice breeding. Breed. Sci. 57:53-58.
Ashikari, M., H. Sakakibara, S. Lin, T. Yamamoto, T. Takashi, A. Nishimura, E.R. Angelels, Q. Qian, H. Kitano, and M. Matsuoka. 2005. Cytokinin oxidase regulates rice grain pro­duction. Science 309: 741-745.
Ashikari, M., J. Wu, M. Yano, T. Sasaki, and A. Yoshimura. 1999. Rice gibberellin- insensitive dwarf mutant gene Dwarfl encodes the alpha-subunit of GTP-binding protein. Proc. Natl. Acad. Sci. USA 96: 10284-10289.
Basten, C.J., B.S. Weir, and Z.B. Zeng. 2001. QTL Cartogra­pher, a reference manual and tutorial for QTL mapping. North Carolina State University, USA.
Chen, C.-S., R.-K. Chen, H.-H. Chin, and Y.-R. Lin. 2010. Intro-
gression of three heading date genes, hdl, Hd6, and ehdl, to Oryza sativa L. japonica cv. Koshihikari by marker-assisted selection. Crop, Environment & Bioinformatics 7: 1-20 (Chinese with English abstract).
12
Botanical Studies, Vol. 52, 2011
Lander, E.S., P. Green, J. Abrahamson, A. Barlow, M.J. Daly, S.E. Lincoln, and L. Newburg. 1987. Mapmaker: an inter­active computer package for constructing primary genetic linkage maps of experimental and natural populations. Ge-
nomics 1: 174-181.
Li, Z.K., S.B. Yu, H.R. Lafitte, N. Huang, B. Courtois, S. Hit-
talmani, C.H.M. Vijayakumar, G.F. Liu, G.C. Wang, H.E. Shashidhar, J.Y. Zhuang, K.L. Zheng, V.P. Singh, J.S. Sidhu, S. Srivantaneeyakul, and G.S. Khush. 2003. QTL x
environment interactions in rice. I. Heading date and plant
height. Theor. Appl. Genet. 108: 141-153.
Li, Z.K., S.R.M. Pinson, M.A. Marchetti, J.W. Stansel, and W.D. Park. 1995. Identification of QTL for heading date and plant in rice using RFLP markers. Theor. Appl. Genet. 91:
382-388.
Lin, S.Y., T. Sasaki, and M. Yano. 1998. Mapping quantitative trait loci controlling seed dormancy and heading date in rice, Oryza sativa L., using backcross inbred lines. Theor.
Appl. Genet. 96: 997-1003.
Luan, W., H. Chen, Y. Fu, H. Si, W. Peng, S. Song, W. Liu, G. Hu, Z. Sun, and D. Xie. 2009. The effect of the crosstalk
between photoperiod and temperature on the heading-date
in rice. PloS One 4: e5891. Mei, H.W., L.J. Luo, C.S. Ying, Y.P. Wang, X.Q. Yu, L.B. Guo,
A.H. Paterson, and Z.K. Li. 2003. Gene actions of QTLs affecting several agronomic traits resolved in a recombinant inbred rice population and two testcross populations. Theor.
Appl. Genet. 107: 89-101. Mei, H.W., Z.K. Li, Q.Y. Shu, L.B. Guo, Y.P. Wang, X.Q. Yu,
C.S. Ying, and L.J. Luo. 2005. Gene actions of QTLs af­fecting several agronomic traits resolved in a recombinant inbred rice population and two backcross populations.
Theor. Appl. Genet. 110: 649-659. Miyata, M., T. Yamamoto, T. Komori, and N. Nitta. 2007.
Marker-assisted selection and evaluation of the QTL for stigma exsertion under japonica rice genetic background.
Theor. Appl. Genet. 114: 539-548.
Moncada, P., C.P. Martinez, J. Borreor, M. Chatel, H. Gauch, E. Guimaraes, J. Tohme, and S.R. McCouch. 2001. Quantita­tive trait loci for yield and yield components in an Oryza sativa x Oryza rufipogon BC2F2 population evaluated in an upland environment. Theor. Appl. Genet. 102: 41-52.
Murakami, M., A. Matsushika, M. Ashikari, T. Yamashino, and T. Mizuno. 2005. Circadian-associated rice pseudo response regulators (OsPRRs): insight into the control of flowering time. Biosci. Biotechnol. Biochem. 69: 410-414.
Price, A.H. 2006. Believe it or not, QTLs are accurate! Trends
Plant Sci. 11: 213-216.
Saito, H., Q. Yuan, Y. Okumoto, K. Doi, A. Yoshimura, H. Inoue, M. Teraishi, T. Tsukiyama, and T. Tanisaka. 2009. Multiple alleles at Early flowering 1 locus making variation in the basic vegetative growth period in rice (Oryza sativa L.).
Theor. Appl. Genet. 119: 315-323.
Sakamoto, T., K. Miura, H. Itoh, T. Tatsumi, M. Ueguchi-Tana-ka, K. Ishiyama, M. Kobayashi, G.K. Agrawal, S. Takeda,
K. Abe, A. Miyao, H. Hirochika, H. Kitano, M. Ashikari, and M. Matsuoka. 2004. An overview of gibberellin metab­olism enzyme genes and their related mutants in rice. Plant
Physiol. 134: 1642-1653. Sarma, R.N., B.S. Gill, T. Sasaki, G. Galiba, J. Sutka, D.A.
Laurie, and J.W. Snape. 1998. Comparative mapping of the wheat chromosome 5A Vrn-Al region with rice and its rela­tionship to QTL for flowering time. Theor. Appl. Genet. 97:
103-109.
Sasaki, A., H. Itoh, K. Gomi, M. Ueguchi-Tanaka, K. Ishiyama, M. Kobayashi, D.H. Jeong, G. An, H. Kitano, M. Ashikari, and M. Matsuoka. 2003. Accumulation of phosphorylated repressor for gibberellin signaling in an F-box mutant. Sci­ence 299: 1896-1898.
Sasaki, A., M. Ashikari, K. Ueguchi-Tanaka, H. Itoh, A. Nishimura, D. Swapan, K. Ishiyama, T. Saito, M. Kobayas-
hi, G.S. Khush, H. Kitano, and M. Matsuoka. 2002. Green
revolution: a mutant gibberellin-synthesis gene in rice. Na­ture 416: 701-706.
Septiningsih, E.M., J. Prasetiyono, E. Lubis, T.H. Tai, T. Tjubaryat, S. Moeljopawiro, and S.R. McCouch. 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.
Spielmeyer, W., M.H. Ellis, and P.M. Chandler. 2002. Semidwarf (sd-l), "green revolution" rice, contains a defective gibber-
ellin 20-oxidase gene. Proc. Natl. Acad. Sci. USA 99: 9043­9048.
Takahashi, Y., A. Shomura, T. Sasaki, and M. Yano. 2001. Hd6,
a rice quantitative trait locus involved in photoperiod sen­sitivity, encodes the alpha subunit of protein kinase CK2.
Proc. Natl. Acad. Sci. USA 98: 7922-7927.
Tanabe, S., M. Ashikari, S. Fujuoka, S. Takatsuto, S. Yoshida, M. Yano, A. Yoshimura, H. Kitano, M. Matsuoka, Y. Fuju-sawa, H. Kato, and Y. Iwasaki. 2005. A novel cytochrome P450 is implicated in brassinosteroid biosynthesis via the characterization of a rice dwarf mutant, dwarfll, with re­duced seed length. Plant Cell 17: 776-790.
Thomson, M.J., T.H. Tai, A.M. McClung, X.-H. Lai, M.E. Hin-ga, K.B. Lobos, Y. Xu, C.P. Martinez, and S.R. McCouch.
2003. Mapping quantitative trait loci for yield, yield com­ponents and morphological traits in an advanced backcross population between Oryza rufipogon and the Oryza sativa
cultivar Jefferson. Theor. Appl. Genet. 107: 279-493.
Ueguchi-Tanake, M., M. Ashikari, M. Nakajima, H. Itoh, E. Katoh, M. Kobayashi, T.-Y. Chow, Y.-I. Hsing, H. Kitano,
I. Yamaguchi, and M. Masouka. 2005. GIBBERELLIN INSENTITVE DWARFl encodes a soluble receptor for gib-
berellins. Nature 437: 693-698.
Wang, C.-M., H. Yasui, A. Yoshimaru, J.-M. Wan, and H.-Q. Zhai. 2002. Identification of quantitative trait loci control­ling F2 sterility and heading date in rice. Yi Chuan Xue Bao
29: 339-342.
Wang S., C.J. Basten, and Z.B. Zeng. 2001. Windows QTL Car-
LIN et al. ― QTL for rice plant height and heading date across Oryza genus
13
tographer Version 2.0. Statistical Genetics, North Carolina State University, USA.
Xiao, J., J. Li, L. Yuan, and S.D. Tanksley. 1996. Identifica­tion of QTL affecting traits of agronomic importance in a recombinant inbred population derived from a subspecific rice cross. Theor. Appl. Genet. 92: 230-244.
Xiao, J., T. Fulton, S. McCouch, S.D. Tanksley, N. Kishimoto,
R. Ohsawa, Y. Ukai, and A. Aaito. 1992. Progress in inte­gration of the molecular maps of rice. Rice Genet. Newsl. 9:124-128.
Yamamoto, T., H. Lin, T. Sasaki, and M. Yano. 2000. Identifica­tion of heading date quantitative trait locus Hd6 and charac­terization of its epistatic interactions with Hd2 in rice using advanced backcross progeny. Genetics 154: 885-891.
Yamamoto, T., J. Yonemaru, and M. Yano. 2009. Towards the understanding of complex traits in rice: substantially or su­perficially? DNA Res. 16: 141-154.
Yamamuro, C., Y. Ihara, and X. Wu. 2000. Loss of function of a rice brassinosteroid insensitive homolog prevents internode enlongation and bending of the lamina joint. Plant Cell 191:
1591-1605.
Yan, J., J. Zhu, C. He, M. Benmoussa, and P. Wu. 1998. Mo­lecular dissection of developmental behavior of plant height in rice (Oryza sativa L.). Genetics 150: 1257-1265.
Yano, M., S. Kojima, Y. Takahashi, H. Lin, and T. Sasaki. 2001. Genetic control of flowering time in rice, a short-day plant.
Plant Physiol. 127: 1425-1429.
Yano, M., Y. Harushima, Y. Nagamura, N. Kurata, Y. Minobe, and T. Sasaki. 1997. Identification of quantitative trait loci controlling heading date in rice using a high-density linkage
map. Theor. Appl. Genet. 95: 1025-1032.
Yano, M., Y. Katayose, M. Ashikari, Y. Yamanouchi, L. Monna, T. Fuse, T. Baba, K. Yamamoto, Y. Umehara, Y. Nagamura, and T. Sasaki. 2000. Hdl, a major photoperiod sensitivity quantitative trait locus in rice, is closely related to the Arabidopsis flowering time gene CONSTANS. Plant
Cell 12: 2473-2483. You, A., X. Lu, H. Jin, X. Ren, K. Liu, G. Yang, H. Yang, L.
Zhu, and G. He. 2006. Identification of quantitative trait loci across recombinant inbred lines and testcross popula­tions for traits of agronomic importance in rice. Genetics
172: 1287-1300. Yu, S.B., J.X. Li, C.G. Xu, Y.F. Tan, X.H. Li, and Q. Zhang.
2002. Identification of quantitative trait loci and epistatic interactions for plant height and heading date in rice. Theor.
Appl. Genet. 104: 619-625. Zhou, Y., W. Li, W. Wu, Q. Chen, D. Mao, and A.J. Worland.
2001. Genetic dissection of heading time and its compo­nents in rice. Theor. Appl. Genet. 102: 1236-1242.
Zhuang, J.Y., H.X. Lin, J. Lu, H.R. Qian, S. Hittalmani, N.
Huang, and K.L. Zheng. 1997. Analysis of QTL x environ­ment interaction for yield components and plant height in
rice. Theor. Appl. Genet. 95: 799-808.
14
Botanical Studies, Vol. 52, 2011
水稻株高和抽穗期數量性狀基因座於兩個亞種雜交族群之圖譜
分析及稻屬間之比較
林彥蓉1,2 吳少鈞2 陳素蛾3 曾東海4 陳正昇1郭素真1 鄔宏潘5 邢禹依5
1國立台灣大學農藝學系
2天主教輔仁大學生命科學系
3農業委員會桃園區農業改良場作物改良課
4農業委員會農業試驗所作物組
5中央研究院植物暨微生物學研究所
本研究建構兩個水稻亞種間的雜交族群,用於株高與抽穗期之數量性狀基因座圖譜分析,分別以
99個和95個多型性的SSRSTS分子標幟於梗稻台梗TK8 x秈稻IR1545-339和梗稻日本晴x IR1545-
339
建立遺傳連鎖圖譜,經由組合區間定位法分析和單點分析。於七條和六條染色體上偵測出十個株
高和七個抽穗期數量性狀基因座,其中只有Qphl.2是共通的,主要株高基因座於TK8/IR1545-339
Nipponbare/IR1545-339分別可解釋22.07% (Qph6.l)12.98% (Qph6.2)之株高變異,而主要抽穗期基因
Qhd3則可分別解釋35.74%15.54%之抽穗日數變異。經由於染色體偽分子之物理圖譜進行稻屬間
之株高與抽穗期數量性狀基因座之比較分析本研究發現數量性狀基因座皆與目前已發表之數量性狀基
因座具有保守性,隨後利用比較數量性狀基因座的結果與候選基因策略,在未進行高解析之圖譜分析而
成功的達到三個株高基因座(Qphl.2, Qph4, and Qph6.l)和一個抽穗期基因座(Qhdl0之確認,本文並討
論區間定位法之正確性。
關鍵詞:比較遺傳學;候選基因策略;抽穗期;株高;數量性狀基因座;水稻。