Abstract
The quantitative trait locus controlling the number of primary rachis branches (PRBs) in rice was identified using backcrossed inbred lines of Sasanishiki/Habataki//Sasanishiki///Sasanishiki. The resultant gene was ABERRANT PANICLE ORGANIZATION 1 (APO1). Habataki-genotype segregated reciprocal recombinant lines for the APO1 locus increased both the number of PRB (12–13%) and the number of grains per panicle (9–12%), which increased the grain yield per plant (5–7%). Further recombination dividing this region revealed that different alleles regulated the number of PRB and the number of grains per panicle. The PRB1 allele, which includes the APO1 open reading frame (ORF) and the proximal promoter region, controlled only the number of PRB but not the number of grains per panicle. In contrast, the HI1 allele, which includes only the distal promoter region, increased the grain yield and harvest index in Habataki-genotype plants, nevertheless, the ORF expressed was Sasanishiki type. It also increased the number of large vascular bundles in the peduncle. APO1 expression occurred not only in developing panicles but also in the developing vascular bundle systems. In addition, Habataki plants displayed increased APO1 expression in comparison to Sasanishiki plants. It suggests that APO1 enhances the formation of vascular bundle systems which, consequently, promote carbohydrate translocation to panicles. The HI1 allele is suggested to regulate the amount of APO1 expression, and thereby control the development of vascular bundle systems. These findings may be useful to improve grain yield as well as quality through the improvement of translocation efficiency.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
Increased rice yield will play an important role in meeting the demands of an ever-increasing global population, because approximately half of the world’s population subsists primarily on rice. Researchers have recently intensified their genetic analyses of rice yield, through quantitative trait locus (QTL) analysis and map-based cloning of the genes related to rice yield. One of the most important factors is the number of grains per panicle. Many different gene/QTLs have been reported to determine the number of grains per panicle (Zhuang et al. 1997; Xiao et al. 1998; Yagi et al. 2001; Nagata et al. 2002; Mei et al. 2003; Kobayashi et al. 2004; Tian et al. 2006; Yoon et al. 2006). One of the most important QTLs is located on chromosome 1 (Gn1) and is identified as cytokinin oxidase (Ashikari et al. 2005). Another gene, DEP1, which increases the number of rachis branches, thereby increasing grain yield, has also been identified (Huang et al. 2009). A gene that controls grain kernel size has been identified as RING type E3 ubiquitin ligase (Song et al. 2007); a gene controlling grain kernel width has been identified in indica and japonica rice varieties (Shomura et al. 2008); a gene related to grain-filling has also been identified as cell wall invertase (Wang et al. 2008). Researchers have also begun to pyramid some of these QTLs into one strain (Ashikari and Matsuoka 2006; Ando et al. 2008).
Another approach to determine the genes that control the panicle architecture is to identify them through the analysis of mutants that have changed the panicle structure. The loss-of-function lax-1 mutation abrogating the axillary meristem initiation results in reduced branch formation in the panicle (Komatsu et al. 2003a). Conversely, the mutation of FZP promotes the formation of supernumerary axillary meristems (Komatsu et al. 2003b), while the overexpression of RCN1 also enhances the secondary branch numbers through the delayed phase change from the branch shoot to floral meristem (Nakagawa et al. 2002). Previous studies have reported that ABERRANT PANICLE ORGANIZATION 1 (APO1) was able to control spikelet numbers in rice (Ikeda et al. 2007; Ikeda-Kawakatsu et al. 2009). A loss-of-function apo1 mutation has been shown to lead to the precocious transition of rachis meristem and branch meristem into spikelet meristem resulting in a reduced number of spikelets (Ikeda et al. 2001, 2007), whereas gain-of-function Apo1-D mutants delayed these transitions for rice panicle architecture and increased the number of grains (Ikeda-Kawakatsu et al. 2009).
Important factors in improving rice yield include yield-determining components, such as the number of panicles per unit area, the number of spikelets per panicle, the percentage of fully ripened grains, and the weight of 1,000 mature kernels (Matsushima 1957). These factors are interrelated, so that increases in one component may lead to decreases in others. The plant, with finite metabolic resources, may compensate for increases in the number of grains per panicle by decreasing the ripening percentage. Therefore, scientists aim to identify genes/QTLs that increase one aspect of grain maturation without affecting others, in order to allow for easy manipulation of the potential yield. The goal is to prevent compensation effects, which may occur when the sink size is increased without a corresponding increase in the source supply. Simultaneous increases in sink size and source supply can be obtained by increasing the source supply and/or improving translocation efficiency. It is important to ensure that ripening traits and grain quality are not negatively impacted by increases in the number or size of grain kernels.
Using recombinant inbred lines, Nagata et al. (2002) identified two QTLs, on chromosomes 1 and 6 that determine the number of grains per panicle. The QTL on chromosome 6, PBN6 (Ando et al. 2008), increased the number of grains per panicle without reducing ripening percentage (Nagata et al. 2002). Ando et al. (2008) narrowed down the physical distance of this QTL into 390 kb between molecular markers S1306 and RM3430. This QTL increased the number of grains by increasing the number of primary rachis branches (PRBs) while keeping the number of grains per rachis unchanged (Nagata et al. 2002). Hence, increasing the number of PRBs may increase the translocation route from culm to grain. In addition, competition among grains within a rachis is reduced in comparison to the competition among secondary rachis branches. Therefore, increasing the number of PRBs is an important way to increase the number of grains per panicle without reducing ripening percentage, while minimizing the negative impact on grain quality. In contrast, Gn1 (Ashikari et al. 2005), located on chromosome 1, drastically increased the number of grain kernels (Yagi et al. 2001) but also negatively impacted ripening traits (Nagata et al. 2002). This QTL increased the number of grains mainly by increasing the number of secondary rachis branches, which may cause severe competition among the grains within a rachis and consequently cause a shortage in translocated carbohydrates (Nagata et al. 2002).
In this report, we identified a gene at the PBN6 QTL that determines the number of PRBs. We also analyzed the effects of this gene on yield and yield-determining components. Using rice lines that are recombinant at this locus, we further investigated how this locus controls each of the yield-determining components, such as harvest index, number of PRBs and number of grains per plant. This research should illuminate the factors that ultimately determine rice grain yield.
Materials and methods
Map-based cloning was performed to identify the gene responsible for the number of PRBs. We selected SHA48, which includes a heterogeneous region, in chromosome 6 as well as chromosome 1, between S10581EH and C112 markers (118.9–122.9 cM), from the BC2F5 mapping population of Sasanishiki/Habataki//Sasanishiki///Sasanishiki (Nagata et al. 2002). This line also includes Habataki-genotype region in chromosome 1 between S11122EH and R210 (28.5–33.1 cM) and chromosome 5 between R372EH and C1268E (26.1–33.1 cM). Reciprocally recombined progenies were identified continuously up to BC2F10 to narrow the heterogeneous region in chromosome 6, with the chromosome 1 heterogeneous region fixed to the Sasanishiki type. Around 10,000 plants were used to determine the genotype. The genotype was determined according to the method described in Nagata et al. (2002). The specific primers that were used for each marker are shown in Table 1. Rice plants were grown in the paddy field at Hokuriku Research Center, National Agricultural Research Center, Joetsu, Japan. We measured yield-determining components according to the method described in Nagata et al. (2002). After the plants were harvested and air-dried, we measured the number of panicles per plant, total plant length, stem length, total shoot weight, and head weight. Stems and leaves were dried in the drying oven at 80°C and dry weights were measured. For the measurement of panicle traits, five medium-sized panicles were obtained from each plant. We measured the panicle length, the number of PRBs, the number of secondary rachis branches, the number of grains per panicle, and the number of sterile grains. After the grains were detached, we measured the weight of the grain kernels with hulls attached by measuring the water content using a “Riceter J” grain moisture tester (Kett Electric Laboratory, Tokyo, Japan). To determine ripening percentage, the grains were water selected and the percentage of sunk grains with a relative density of >1.0 was measured.
Further research on reciprocal recombination within the PBN6 locus was carried out using the 04SHA422-12-8.8-18.31 line. Yield-determining elements were identified in progeny grown in 2004 and 2005 (Table 2). The 2005 experimental design included a split plot with three replicates using five segregated lines that each included ten plants. The reciprocal recombinants identified were grown in the same field in 2006. Yield-determining elements were measured using segregated progeny. In the 2006 experiment on SHA-HI lines, the head alone was harvested for the measurement of yield components. Shoot biomass factors were measured during another experiment that included three replicates carried out at the same field.
Statistical analyses were carried out using the GLM procedure on SAS ver. 9.1 (SAS Institute, Cary, NC, USA).
Gene expression
We studied the gene expression of young panicles with RT-PCR. Young Sasanishiki and Habataki panicles were collected on July 4, 7, 10, and 13, with panicle sizes ranging from 0.2 to 2 mm. As the heading dates of the Sasanishiki plants and Habataki plants during the sampled year were July 29 and 30, respectively, these sampling dates correspond to 25, 22, 19 and 16 days before the heading of the Sasanishiki plants and 26, 23, 20 and 17 days before the heading of the Habataki plants.
RNA was prepared using the CTAB method (Chang et al. 1993). After treatment with RNase-free DNase (DNase I, amplification grade Invitrogen, Carlsbad, CA, USA) to remove DNA contamination, cDNA was synthesized using Superscript III reverse transcriptase (Invitrogen), following the protocol described in Hirose and Terao (2004). Specific 3628-67L1 and 3628-U2R1 primers were used for PCR to amplify the regions, including the intron (Table 1). Actin was used as the internal standard (Yamanouchi et al. 2002). Also the quantitative real-time PCR analysis was carried out using the Smart Cycler System (Cepheid Co., Sunnyvale, CA, USA), according to the method described in Hirose et al. (2006). Aliquots of the same cDNA prepared from young panicles of Habataki plants and Sasanishiki plants corresponding to 66 ng of total RNA, and the same primer set described above was used. Ubiquitin was used for the internal standard in this experiment. Since there was no difference in efficiency between APO1 and ubiquitin mRNA in the validation experiment, the DDCT method was used for relative quantification (Pfaffl 2001).
Counting the number of vascular bundles in the peduncle
Dried pieces of peduncles that had been cut about 1 cm below the panicle base node were used to count the number of large vascular bundles and measure their diameters. One piece was obtained from each of the five selected panicles for which yield components were measured. Peduncles were swelled with water, filled with epoxy resin, and sectioned along the transverse axis. Then the peduncle diameter was measured, and the number of large vascular bundles was counted.
In situ hybridization
In situ hybridization was carried out according to the method described in Hirose et al. (2002). Young panicles from Habataki and Sasanishiki plants were collected on July 5 and 9, about 24 and 20 days before heading, respectively. The hybridization probes were made using PCR primer sets 3628-67L1 and 3628-U2R2 (Table 1).
Transgenic plants
To suppress the expression of APO1, RNAi-APO1 was constructed using a p2K1+ binary vector, as described in Moritoh et al. (2005). A 373 bp Habataki APO1 fragment was PCR-amplified using the 3628-UFL1 and 3628-U2R1 primer sets (Table 1). As this fragment included a SacII restriction enzyme digestion site that would be digested in a later step, this site was corrupted by substituting CCGCGG into TCGCGG. This mutated fragment was sub-cloned into pGEM-Teasy (Promega Corporation, Madison, WI, USA) and its sequence was verified. A SacII–SpeI fragment and a SalI–ApaI fragment were inserted into the pGUS27 vector (Miki and Shimamoto 2004) in the inverted direction upstream and downstream of the gus linker fragment. Then a SacI–KpnI fragment was cloned into the p2K1+ binary vector. The construct was introduced into Sasanishiki using the R. radiobacter-mediated method described in Toki et al. (2006).
To investigate the differences of the functions between the Habataki-type APO1 and the Sasanishiki-type APO1, we introduced Habataki APO1 into a Sasanishiki plant with a native promoter of Habataki alleles. Using primer sets 3628-51L3 and 3628-53R3 (Table 1) and LA Taq polymerase (Takara Bio Inc., Shiga, Japan), we obtained a HindIII-digested PCR-amplified fragment from Habataki plants. This fragment included the APO1 open reading frame (ORF), with the promoter region up to −1,731 and a 789-bp 3′ region. This fragment was inserted into the HindIII restriction enzyme site in the pTN1 vector (Fukuoka et al. 2000). After checking the insertion direction, the construct was introduced into Sasanishiki using the R. radiobacter-mediated method as above.
Results
Map-based cloning of a gene that regulates the number of PRBs
The PBN6 QTL, which controls the number of PRBs, is located on chromosome 6, between R2549EH and G329. These loci are 69.3 and 107.3 cM, respectively, from the end of the short-arm side of chromosome 6. The region between these markers was mapped at a high resolution (Fig. 1a) and included an AP003628 PAC clone. Further fine mapping was carried out using CAPs, STS and SSR markers that were generated based on the genome sequence information from the japonica rice cultivar, Nipponbare (Fig. 1b) (http://rgp.dna.affrc.go.jp/). The line was heterogeneous in this region; eight reciprocal recombinant lines were found. We carried out progeny testing to determine the average number of PRBs. The number of PRBs segregated similarly to the genotype markers 3628-60, 3628-55 and 3628-41 (Fig. 1c). Accordingly, the QTL for the number of PRBs was determined to be located between markers 3628-41 and 3628-56, which contain about 11 kb. This result is much narrower than the region decided in the preceding report that included 390 kb (Ando et al. 2008). Only one predicted gene, Os06g0665400, existed at the 28357085–28358390 position on chromosome 6 (Fig. 1b). This gene was identical to APO1, a gene that has been cloned previously, during the analysis of a mutant plant called apo-1, and that is thought to regulate the transformation of the meristem into the panicle primordia (Ikeda et al. 2001, 2007). Notably, the apo-1 mutant displayed abnormally shaped panicles with reduced numbers of grains (Ikeda et al. 2001). It was found that these serious defects in the apo-1 mutant are due to the insertion of a stop codon in the APO1 gene (Ikeda et al. 2007). While this gene was found to be functional in Habataki and Sasanishiki plants, slight differences of the sequence may have affected the number of PRBs.
Map-based cloning of the QTL for the number of primary rachis branches. a Positioning of the QTL on PAC clone AP003628. b Predicted genes are marked with arrows between 40,000 and 95,000 bp on the AP003628 clone. Solid boxes represent the primer positions of CAPS, STS or SSR markers. Dotted boxes represent the predicted region of the QTL according to the results shown in C. The marker name is shown in C. c Graphical genotypes of the segregated progeny for which reciprocal recombination occurred within the above mentioned region; bar graphs of the number of primary rachis branches. Open boxes in the graphical genotype, Sasanishiki type; closed boxes Habataki type. Bar graphs represent the number of primary rachis branches in the segregated progeny shown in the graphical genotype as well as their parent cultivars Habataki (filled bars) and Sasanishiki (open bars). Determination of phenotype segregation is shown beside the bar graph. Possible location of the graphical genotype is shown in the dotted box; the genotype region of the markers and the phenotype of primary rachis branches matched. Error bars standard deviations, broken and dotted lines represent the mean and standard deviations of parent cultivars Habataki and Sasanishiki, respectively, for the comparison of values in segregated progeny
Further reciprocal recombination occurred in the heterogeneous region of the PBN6 locus in the 03SHA422-12-8-18.31 line (Fig. 2). This region was divided into two sub-regions between markers 3628-55 and 3628-41. The SHA-PRB line had a heterogeneous region that included markers 3628-55 and 3628-60, which corresponded to the ORF region of the APO1 gene and the proximal promoter region up to −2,287 bp, respectively. The SHA-HI line had a heterogeneous region including markers 3628-41, which corresponded to the distal promoter region past −2,832 bp (see below; Fig. 3).
The position at which the reciprocal recombination occurred within the heterogeneous region of the line 03SHA422-12-8-18.31. Thick arrow represents APO1 ORF and narrow arrows represent other predicted genes. The base numbers in AP003628 were shown to denote the position of recombination sites. Recombination site positions are a 61978–62124, b 68306–68904, c 69410–69439 and d 75540–75683
DNA differences at the PBN6 QTL: the PRB1 allele, which is responsible for the number of primary rachis branches; and the HI1 allele, which affects the number of grains per panicle and harvest index in Nipponbare, Habataki and Sasanishiki. The base count shown on the sequences represents Nipponbare base pair number, based on the ORF starting position. a–c Recombination sites shown in Fig. 2
DNA sequencing
The DNA sequence was established to identify differences in this region between Sasanishiki and Habataki plants, as well as to find the position of the reciprocal recombination. These sequences were compared to the published Nipponbare sequence (Sasaki et al. 2002, http://rgp.dna.affrc.go.jp/). Figure 3 shows the DNA sequences that differ among Habataki, Sasanishiki and Nipponbare plants in this region. The base number, counted from the start of the ORF, uses Nipponbare as a reference. Notably, 216 bp differed between Habataki and Sasanishiki plants out of a total of 11,086 bp between the 5′ side recombination site in SHA422-12-8-18.31 (a, −9,070 bp) and the 3628-56 marker (2,015 bp). The difference between Sasanishiki and Nipponbare plants was 14 bp, which is much smaller than the 202 bp difference between Habataki and Nipponbare plants. However, a higher frequency of base differences between Sasanishiki and Nipponbare plants occurred proximal promoter to the ORF region of APO1. Eight base pair differences occurred between Sasanishiki and Nipponbare, from −350 bp at the start of the ORF to the end of the ORF. This suggests that the origin of the Sasanishiki APO1 ORF region may differ from the origin of the Nipponbare region. Figure 3 also shows the site of recombination between SHA-HI and SHA-PRB (b and c, respectively). SHA-PRB recombination occurred between −2,687 and −2,287 bp, while SHA-HI recombination occurred between −2,827 and −2,798 bp.
The amino acid sequences deduced for the ORF in Nipponbare, Habataki and Sasanishiki APO1 are shown in Fig. 4. There were five sites of amino acid difference between Nipponbare and Habataki plants; there were four sites of difference between Nipponbare and Sasanishiki plants. As mentioned above, the sequence differences between Sasanishiki and Nipponbare plants seemed frequent in the ORF region, even though both varieties were japonica-type cultivars. One of the differences, a shift from isoleucine in Nipponbare and Sasanishiki plants to valine in Habataki plants, was located in the F-box protein motif. This motif may bind substrates for ubiquitin-mediated proteolysis.
Differences in the deduced amino acid sequence among Nipponbare, Habataki and Sasanishiki. The solid box represents the difference between Nipponbare and Habataki, while the broken box represents the difference between Nipponbare and Sasanishiki. F-box F-box motif
To identify promoter region differences between Habataki and Sasanishiki plants that may cause the difference in APO1 gene expression, we performed a PLACE search (Higo et al. 1999). This search revealed that one of the CarG8 motifs was missing in Habataki plants due to substitution of a G with a T at the −3,118 position in Nipponbare and Sasanishiki types (CarG in Fig. 3).
Yield and yield-determining components
The yield-determining elements were measured using the segregated progeny from the 03SHA422-12-8.8-18.31 line obtained in 2004 and 2005 (Table 2). These plants segregated the region including the APO1 gene, but both the 3′ and 5′ sides of this region were fixed in the Sasanishiki genotype (Fig. 1). Also, the other heterogeneous region in chromosome 1 was fixed into the Sasanishiki genotype. Thus, these segregated plants were mutually near-isogenic; they might be called isogenic, because the segregated region included only 1 gene, although they are not isogenic when compared to Sasanishiki plants. In both years, the traits showing significant increases in the Habataki genotype were shoot length (by 2% (2004) and 1% (2005)), head length (3% in both years), the number of PRBs per panicle (by 13%; 12%), the number of spikelets per panicle (by 9%; 13%), harvest indexes (by 4%; 7%), and the number of sunk grains (with hull) (by 7%; 11%). The number of secondary rachis branches per panicle increased in the Habataki plants, but the ratio of secondary to PRBs did not change in 2005 and even decreased in 2004. This suggests that the genotype difference in this region does not affect the number of secondary rachis branches. In contrast, the grain weight decreased in Habataki plants (by 2% (2004) and 3% (2005)). However, the magnitude of the reduction in grain weight (2–3%) was smaller than the magnitude of the increase in the number of grains per panicle (9–13%). Ripening percentage, as represented by the ratio of sunk grains to total grains, decreased significantly in the 2005 experiment. This parameter also decreased in the 2004 experiment, but the change was not significant. Notably, the magnitude of the decrease in the ripening percentage was small (2%) compared to the increase in the number of grains, even in the 2005 experiment. In Habataki-genotype plants, the weight of sunk grains per plant increased 5 and 7% in 2004 and 2005, respectively.
The yield-determining components of the segregated progeny of the SHA-PRB and SHA-HI lines, which have different heterogeneous regions within the PBN6 locus, were evaluated in 2006. The measurements of the SHA-PRB and SHA-HI lines are shown in Tables 3 and 4, respectively. The yield-determining components of the SHA-PRB progeny that segregated the ORF and proximal promoter region showed a difference in the number of PRBs. Habataki-type progeny (SHA-PRB-B) had more PRBs than Sasanishiki-type progeny (SHA-PRB-A). However, this increase in the number of PRBs did not directly increase the number of grains per panicle. Instead, the number of grains per primary rachis branch was reduced, and therefore the number of grains per panicle was not altered. This decrease in the number of grains per primary rachis branch occurred because the number of secondary rachis branches per primary rachis branch was reduced. Other yield-determining factors, such as, harvest index, grain weight, the number of panicles per plant and the percentage of fertile spikelets, did not change. Since these lines segregated at the ORF and proximal promoter region, this segregated region is considered to control the number of PRBs. The number of grains per panicle seems to be regulated by a different region. We designated this region the PRB1 allele because it is a part of PBN6 locus, which only acts to change the number of Primary Rachis Branches.
The yield-determining components of the segregated progeny of SHA-HI showed that the panicle weight, the number of grains per plant, and the number of grains per panicle were increased in Habataki-genotype progeny (SHA-HI-B). The number of secondary rachis branches increased ~12% among the Habataki genotype; this difference was highly significant. The number of PRBs increased only 2% among the Habataki genotype. This 2% increase was significant at P < 0.05, but the extent of the increase was reduced compared to that of secondary rachis branches. Accordingly, the increase in the number of spikelets per panicle in SHA-HI-B resulted from an increased number of secondary rachis branches. The difference in the grain weights between SHA-HI-A and SHA-HI-B was not significant. The increased number of grains per panicle did not reduce the grain weights but increased the sunk grain weight per plant and the harvest index. Hence, we designated this region as the HI1 allele because it is a part of the PBN6 locus that can control the Harvest Index. This allele included the distal promoter region of APO1 beyond −2,832 bp; thus, the grain yield per panicle seems to be regulated in this region.
These results indicate that two important factors in panicle architecture, the number of PRBs per panicle and the number of spikelets per panicle, are controlled independently. This is demonstrated by the fact that the increased number of PRBs per panicle did not directly increase the grain yield per panicle. These components are controlled by different alleles on the same gene. Moreover, the effects of these two alleles were additive. The line with a Habataki genotype PBN6 that includes Habataki genotypes at both PRB1 and HI1 alleles exhibited a higher number of PRBs, demonstrating an effect at the PRB1 allele. This line also exhibited increases in the number of spikelets per panicle, harvest index, the number of vascular bundles in each peduncle and peduncle diameter, demonstrating an effect at the HI1 allele.
Gene expression
The expression of APO1 during young panicle development (Fig. 5a) was studied using semi-quantitative RT-PCR (Fig. 5). Because the amplified fragment included an intron, we amplified a shorter band when the template was cDNA (Fig. 5, lane cDNA) as compared to genomic DNA (Fig. 5, lane genome DNA). Both PCR product sizes corresponded to the sizes expected without and with the intron, respectively. APO1 expression in the Habataki plants was consistently higher than that in the Sasanishiki plants. After 30 cycles of PCR amplification, APO1 expression was evident only in Habataki cDNA sampled on July 7, corresponding to ~23 days before heading. When the number of cycles was increased to 32, the Habataki samples showed obvious APO1 expression on July 4, and showed the highest expression on July 7, while reduced but still obvious expression was seen on July 10 and 13. In the Sasanishiki plants, APO1 expression was not detectable in samples obtained on July 4, 10 or 13 and was only detectable in the sample obtained on July 7, although the expression level was still lower than in Habataki plants. After 35 cycles, the Sasanishiki plants expressed APO1 at all stages, with the lowest levels observed in samples obtained on July 4 and 10. To determine by how much the expression levels of APO1 differed between Habataki and Sasanishiki plants, quantitative real-time PCR analyses were performed using the same samples of Habataki and Sasanishiki plants (Fig. 5b). The expression levels in Habataki plants were 3.5, 13, 7 and 4 times (21.8, 23.8, 22.9 and 22.0 times) higher than that of Sasanishiki plants on July 4, 7, 10 and 13, respectively. Thus, the peak strength of APO1 expression in Habataki plants was estimated to be around ten times higher than that of Sasanishiki plants.
APO1 expression at the young panicle of Habataki (H) and Sasanishiki (S) plants, sampled on July 4, July, 7 July 10 and July 13, measured by semi-quantitative RT-PCR. The size of the amplified cDNA was 82 bp smaller than that of genomic DNA because an intron was included between primer sets 67L1 and U2R1. Data for 30, 32 and 35 PCR cycles are shown. Actin was used for the internal standard. a Sampled part of panicle. b Quantification of APO1 expression level in the same sample measured by real-time PCR
As presented in Fig. 6, we also measured expression of APO1 at the region beneath the developing panicle, the next block below the samples used for the analysis presented in Fig. 5 (Fig. 6a). In this case, samples obtained on July 7, July 10 and July 13 were used. The APO1 expression levels were almost the same as the young panicle. The expression in Habataki plants was higher than that in Sasanishiki plants, and the expression was highest in the July 7 samples. Quantitative real-time PCR analyses revealed that APO1 expression in Habataki plants was 6, 9 and 9 times (22.6, 23.2 and 23.2) higher than that of Sasanishiki plants on July 7, 10 and 13, respectively (Fig. 6b). Previous studies reported that APO1 is expressed exclusively in the panicle primodia and young panicles, particularly in the developing rachis branches (Ikeda et al. 2007). However, our results suggest that this gene is expressed in other parts of developing panicles, and that it affects processes other than the developmental change from meristem to panicle primodia. Figure 6b also shows that APO1 expression in the region beneath the young panicle was much slowly decreased compared to that in the panicle. Also the expression levels in this region in Habataki plants on July 10 and 13 was about nine times higher than that in Sasanishiki plants, whereas the difference in developing panicle reduced to seven and four times. This suggests that APO1 expression in Habataki plants is not only higher than that in Sasanishiki plants but also keeps higher expression to the later stage particularly in the region other than the growing panicle.
APO1 expression at the region beneath the developing panicle in Habataki (H) and Sasanishiki (S) plants, sampled on July 7, July 10 and July 13. The other conditions were the same as those described in Fig. 5
Transgenic plants
We cultivated T0 and T1 generations of RNAi-APO1 plants, and counted the numbers of PRBs as well as grains per panicle. The averages of the number of PRBs in the two differently transformed T0 generation plants were 6.22 (n = 21) and 5.71 (n = 17), which are slightly lower than the 7.57 (n = 5) observed in the non-transgenic Sasanishiki plants grown in the same conditions. In the T1 generation segregated plants, each ten segregated T1 plants from six independent T0 plants were grown and the number of PRBs per panicle as well as the number of grains per panicle were measured. The average number of PRBs in the plants, either heterogeneous or homogeneous RNAi-APO1, was 7.44, which was significantly lower than that of RNAi-null plants with an average of 8.52 (Table 6). Also, the average number of grains per panicle in the RNAi-APO1 plants was 58.01, significantly lower than the 71.07 of the RNAi-null plants (Table 6). We could not measure the expression level of APO1 in the RNAi-APO1 plants, because the destructive sampling of young panicles required for the measurement of mRNA would prevent the measurements of the number of PRBs and of the grains per panicle thereafter. However, significant differences were detected both in the number of PRBs and of grains per panicle between the segregated plants, both with RNAi-APO1 and without. This strongly suggests that the suppression of APO1 expression in RNAi plants caused the reduction of both the numbers of PRBs and of grains per panicle.
In the experiment that introduced Habataki APO1 into Sasanishiki plants, we cultivated a T3 generation of transgenic plants by introducing Habataki APO1 with a 1,731-bp promoter region and a 789-bp 3′ region into Sasanishiki plants. We compared T3-segregated progeny of null plants and two-copy introduced plants, which might be considered homo-introduced plants derived from single-copy introduced plants. Each of the 12, 10 and 8 T3 plants from three independent lines was analyzed. Null plants had an average of 8.4 PRBs, and two-copy plants (homo) had an average of 8.8; these values are not significantly different. The number of grains per panicle in the homo plants was 98.1, compared to 91.1 in the null plants; this difference is also not significant. These data suggest that the exogenous Habataki APO1 gene was not functional in the transgenic plants. Interestingly, Ikeda et al. (2007) reported that transformation with the APO1 gene from wild-type plants could not rescue mutant plants when the introduced gene contained a 2.8-kb promoter region, but the same gene could rescue mutant plants when introduced with a 5-kb promoter region. Durfee et al. (2003) showed that an Arabidopsis T-DNA insertion mutant occurred at the −4,000-bp position in the promoter region of the UNUSUAL FLORAL ORGAN (UFO) gene. We therefore concluded that the length of the promoter region is not sufficient in the introduced Habataki gene, which prevents its effective function in transgenic plants.
Number of large vascular bundles in the peduncle
We studied the effects of the PRB1 and HI1 alleles on the formation of vascular bundles in the peduncle by counting the number of large vascular bundles in Habataki-type and Sasanishiki-type segregated progeny of the SHA-PRB and SHA-HI lines (Table 5). Habataki-type segregated progeny with the HI1 allele (SHA-HI-B) had 10.5 large vascular bundles on average, compared to 9.1 among Sasanishiki-type progeny (SHA-HI-A). This 15% increase is statistically significant. The effect of the PRB1 allele on the number of large vascular bundles was much smaller but still significant. Habataki-type progeny (SHA-PRB-B) averaged 9.8 vascular bundles, compared to 9.3 among Sasanishiki type (SHA-PRB-A), an increase of ~5%. Habataki-type progeny that had been segregated with the HI1 allele (SHA-HI-B) displayed increased peduncle diameter. SHA-HI-B peduncles were about 9% thicker than those of the SHA-HI-A plants (1.34 vs. 1.23 mm, respectively). No difference in the diameter of the peduncle was observed between SHA-PRB-A- and SHA-PRB-B-type progenies. Accordingly, we concluded that the HI1 allele affected vascular bundle formation in the peduncle. In the Habataki genotype, the HI1 allele increased the number of large vascular bundles, and consequently increased peduncle diameter. Since the HI1 allele corresponds to a promoter region further than 2.8 kb past the start of APO1, we hypothesize that differential expression of the APO1 gene affected vascular bundle formation as well as peduncle formation.
The relationship between the number of vascular bundles and the number of PRBs in segregated progeny carrying the HI1 or PRB1 allele is shown in Figs. 7 and 8, respectively. For both SHA-HI-A and SHA-PRB-A, i.e., the Sasanishiki-type progeny of HI1 and PRB1 alleles, the ratio of the number of vascular bundles to the number of PRBs was about 1 (1.0008 in SHA-HI-A and 1.0129 in SHA-PRB-A). This indicates that the number of vascular bundles corresponded to the number of PRBs. SHA-HI-B, the Habataki-type progeny carrying the HI1 allele, displayed significantly increased numbers of vascular bundles. Hence, this ratio was reduced to 0.9, although we also observed a slight but significant increase in the number of PRBs (Table 4A). This means that the increased number of large vascular bundles in SHA-HI-B did not directly cause the proportional increase in the number of PRBs. Rather, the ratio in SHA-PRB-B, Habataki-type progeny carrying the PRB1 allele, increased to 1.1. This was due to the dramatic increase in the number of PRBs, which outweighed the slight but significant increase in the number of vascular bundles (Table 5). This also indicates that the increase of PRBs in SHA-PRB-B was not caused by the increased number of large vascular bundles.
Correlation of the number of vascular bundles and the number of primary rachis branches in the segregated progeny of SHA-HI. Closed square and solid line represent segregated progeny carrying the HI1-Sasanishiki allele; associated regression line passes through the origin (SHA-HI-A). Open triangles and dotted line represent segregated progeny carrying the HI1-Habataki allele (SHA-HI-B). Equations for regression lines are also shown. Thin broken line shows a 1:1 correlation
Correlation of the number of vascular bundles and the number of primary rachis branches in the segregated progeny of SHA-PRB. Closed squares and solid line represent segregated progeny carrying the PRB1-Sasanishiki allele; associated regression line passes through the origin (SHA-PRB-A). Open triangles and dotted line represent segregated progeny carrying the PRB1-Habataki allele (SHA-PRB-B). Other markings are the same as in Fig. 7
In situ hybridization
We performed in situ hybridization, to localize APO1 expression in developing Habataki panicles. Figure 9 shows a transverse section of a young panicle from a Habataki plant sampled on July 9. The APO1 gene was expressed in the developing PRB, as reported by Ikeda et al. (2007). The APO1 gene also appeared to be expressed in the vascular bundles of developing leaves. We sought to determine whether this vascular bundle staining represented true APO1 gene expression in vascular bundle cells, or artifact staining due to the higher density of small developing cells. Therefore, we used a sense probe as a control (Fig. 10). While the antisense probe clearly deposited color in the developing vascular bundles, the sense probe did not stain vascular bundles. We therefore concluded that the APO1 gene is expressed not only in the developing panicle but also in developing vascular bundles.
In situ hybridization of a transverse section of young panicle obtained from a Habataki plant using the APO1 anti-sense probe
In situ hybridization of the transverse section of the basal part of a young panicle from a Habataki plant with developing vascular bundles. a Anti-sense probe, b sense probe
Discussion
Identification of the PBN6 gene and its effects on yield and yield components
We identified the gene containing the PBN6 QTL, which determines the number of PRBs, to be APO1. We also showed that the Habataki allele, in comparison to the Sasanishiki allele, of this QTL/APO1 increases the number of PRBs, the number of grains per panicle, and the harvest index, thereby increasing grain yield. Mutating this gene results in aberrant panicle organization (Ikeda et al. 2007; Ikeda-Kawakatsu et al. 2009), similar to UFO in Arabidopsis thaliana, PROLIFERATING FLORAL ORGANS in Lotus japonicus, STAMINA PISTILLOIDA in Pisum sativum, and FIMBRIATA in Antirrhinum (Levin and Meyerowits 1995; Wilkinson and Haughn 1995; Simon et al. 1994; Pouteau et al. 1998; Taylor et al. 2001; Zhang et al. 2003). Previous mutational analysis and the complementation experiment showed that this gene controls the production of spikelets and inflorescence structures. The present study also confirmed that RNAi-APO1 plants reduced the number of PRB and the number of grains per panicle (Table 6). Accordingly, it is important to improve the Sasanishiki grain yield as well as that of other japonica rice cultivars. Notably, our findings suggest that this gene has the potential to contribute higher grain yield in widely grown rice cultivars, both in indica rice cultivars as well as in the newly developed indica–japonica crossed cultivars. Recently, a gain-of-function mutant of APO1 gene apo1-D, which increased the expression level of APO1 and hence increased the number of PRB as well as the number of grains per panicle, was reported (Ikeda-Kawakatsu et al. 2009). Ikeda-Kawakatsu et al. (2009) also reported that an Undulate rachis 1 (Ur1) mutant (Nagao et al. 1958) occurred in the same mechanism: the insertion of nDart1 transposon (Tsugane et al. 2006). Those gain-of-function mutants increased the number of PRBs and secondary rachis branches and hence the number of grains per panicle. Although it has not been proven that apo1-D mutant itself is directly usable for breeding high yielding cultivar, the mutation on this gene is usable for breeding of high yielding cultivars (Murai et al. 2005).
Two alleles of PBN6 differentially control yield-determining component
The ability of APO1 to alter the number of PRBs follows from the reported role of APO1 in regulating the initiation of panicle and rachis primodia (Ikeda et al. 2007). As expected from this APO1 ability, the Habataki allele of the PBN6 locus clearly increased the number of PRBs as well as the number of grains per panicle, hence increasing grain yield per plant (Table 2). Because of this, we first considered that the increase of the number of PRBs should be the key factor of increasing the grain yield per plant, since increasing the number of PRBs may increase the potential number of grains per panicle. However, during later experiments, we divided the locus into two sub-regions; PRB1 and HI1 alleles, and obtained results that did not support the original hypothesis. The correlation that we observed between an increased number of grains per panicle and an increased number of PRBs was an additive effect of activity at those two alleles. The PRB1-Habataki allele increased the number of PRBs but decreased the number of grains per rachis and hence caused no net increase of the number of grains per panicle. The HI1-Habataki allele is less effective in increasing the number of PRBs but did increase the number of grains per primary rachis branch, and hence increased the number of grains per panicle. At both alleles, the positive and negative effects canceled out, with regard to the number of grains per rachis. What is left is the increase in the number of PRBs, caused by the PRB1-Habataki allele, and the increase in the number of grains per panicle, caused by the effect of HI1-Habataki allele, which seem correlated but that are regulated separately. The fact that the PRB1-Habataki allele had no effects on increasing the number of grains per panicle suggests that an increase in the potential sink size may not increase grain yield unless there is a sufficient source supply. Contrarily, the HI1-Habataki allele had the effect of increase the number of grains per panicle. This might be caused through the improvement of translocation efficiency (see below).
The PRB1 allele corresponds to the ORF and proximal promoter region of the APO1 gene. On the other hand, the HI1 allele corresponds to the distal promoter region of the same gene. The HI1 allele controls the number of grains, number of secondary rachis branches per panicle, the harvest index, the number of large vascular bundles, and the peduncle diameters, and limits the increase in the number of PRBs to 2%. These findings demonstrate that the distal promoter region, but not the ORF and proximal promoter region, regulates the yield. Interestingly, differences in this region affect yield but not the translated product. Distal promoter region activity was also able to rescue apo1 mutation in rice (Ikeda et al. 2007) and Arabidopsis (Durfee et al. 2003). Thus, promoter region variation may control each part of the panicle structure in different ways, even when the same gene is expressed. This might be caused through the difference of special distribution of expression such as rachis and vascular bundle primordia as well as the difference of expression strength.
Effect of increasing harvest index on the grain yield
In addition to increase the number of PRBs and grain yield, APO1 has two more important functions that we identified. These are the ability to change the harvest index and the ability to alter the number of vascular bundles. Increasing the harvest index is one of the most effective ways to increase grain yield (Snyder and Carlson 1984; Zhang and Kokubun 2004). This study has shown that the PBN6 locus and HI1 allele represent the QTL for harvest index and that these QTLs are regulated by APO1. Few reports in the literature have examined QTLs in rice for their harvest index. Mao et al. (2003) reported four main-effect QTLs for the harvest index (on chromosomes 1, 4, 8 and 11) and one epistatic QTL on chromosome 5. Lanceras et al. (2004) described harvest index QTLs on chromosomes 1 and 3, but did not report the harvest index QTL on chromosome 6. Notably, increasing the harvest index may be the most effective way to increase the yield per unit area because high-yielding bigger plants may require larger areas or lower optimum planting density. Therefore, increasing the harvest index by improving yield without altering plant size allows farmer to increase grain yield per unit area. A higher harvest index also increases yield per unit input of solar radiation or soil nutrients. Our research indicates that breeding cultivars of high harvest index requires Habataki alleles at PBN6 locus and HI1-Habataki allele. The improvement in the harvest index and possibly translocation efficiency may be important for researchers working to pyramid genes that improve grain yield with genes that increase sink size. It may be necessary to collect genes with compensatory roles, such as those for sink size and source potential, in addition to collecting genes with similar roles, such as those that control the number of grains and grain size, which was now being examined (Ashikari and Matsuoka 2006; Ando et al. 2008).
Effect of HI1 allele on vascular bundle system formation
How does the promoter region of APO1 affect the harvest index as well as grain yield? The underlying mechanism may be increased translocation efficiency. Sasahara et al. (1999) showed that one of the QTLs for the number of vascular bundles in the peduncle was located near the C962 marker on chromosome 6. This location is very close to the PBN6 locus, as well as to APO1. We therefore considered the possibility that the number of vascular bundles in the peduncle was also controlled by APO1. Our results clearly showed that the HI1 allele controls the number of vascular bundles in the peduncle as well as peduncle diameter. In contrast, the PRB1 allele has no effect on peduncle diameter and a limited effect on the number of large vascular bundles. Thus, attaching the Habataki-type promoter to APO1 increased the number of vascular bundles, which might be favorable to enhance the translocation of carbohydrate from leaves to panicles. Ikeda-Kawakatsu et al. (2009) also showed that the number of vascular bundles was increased in the gain-of-function mutant apo1-D1, in which the APO1 transcription level increased because of the insertion of nDart1-0 transposon into the negative regulation region of APO1 promoter. The apo1-D1 mutant increased the number of PRBs and the number of vascular bundles, and constitutive overexpressed APO1 plants showed a much higher number of PRBs and vascular bundles. In that same study, Ikeda-Kwakatsu et al. suggested that the increase of meristem size and subsequent increase of panicle branching occurred according to the increase of the vascular bundles. However, our results clearly revealed that the number of vascular bundles and the number of rachis branches are regulated independently. The apparently synchronized increase in the number of PRBs and the number of vascular bundles shown in Ikeda-Kawakatsu et al. (2009) was considered to have occurred because of the somehow constitutive expression of APO1, not only in the constitutive APO1 overexpressing transgenic plants but also in the apo1-D1 mutant. Instead, it is suggested that APO1 expression in the panicle primordia and vascular bundle primordia are regulated independently, and this consequently caused the different development of PRBs and vascular bundles.
Published research has established that, in japonica-type cultivars, each of the large vascular bundles is connected to a primary rachis branch. However, in indica-type cultivars (e.g., Habataki types), the number of large vascular bundles in the peduncle is much higher than the number of PRBs; therefore, some of the large vascular bundles are directly connected to the secondary rachis branches (Fukushima and Akita 1997). Increased numbers of large vascular bundles without accompanying increases in the number of PRBs result in more large vascular bundles that are directly connected to secondary rachis branches. Alternatively, having too few vascular bundles for a given number of PRBs may cause the branching of vascular bundles, as they struggle to connect to two PRBs. This may reduce translocation efficiency.
If the expression site, duration, and strength of the APO1 gene differed while the expressed gene remained the same, and expression still resulted in the formation of vascular bundle systems, one would expect that the expression strength and site should differ between Habataki and Sasanishiki plants. We therefore investigated the expression of the APO1 gene at the panicle initiation stage. We observed a higher expression of APO1 in Habataki plants compared to Sasanishiki plants throughout panicle development (Fig. 5). Furthermore, APO1 was expressed not only in the panicle, but also in the region beneath the developing panicle (Fig. 6). In this region, only the vegetative stem and leaf sheath was included, but this region is also the place where vascular bundles develop. The expression of APO1 in the developing vascular bundles was also confirmed using in situ hybridization (Figs. 9, 10). Moreover, APO1 expression in this region beneath the panicle was kept higher in Habataki plants than in Sasanishiki plants in the later developing stage. This might be consistent with the higher number of vascular bundles developed in this region in Habataki plants. Therefore, it appears that higher APO1 expression is necessary to increase the number of vascular bundles that develop in the peduncle, which may improve translocation efficiency as well as harvest index.
Effects of gene structure on panicle formation
The number of PRBs was regulated by the PRB1 allele, which includes the ORF and promoter region up to −2,686 bp. Several differences were observed between Habataki and Sasanishiki alleles in this region. One difference that may be of functional significance is a substitution of valine (in Habataki type) for isoleucine (in Nipponbare and Sasanishiki types) in the F-box protein motif, which may bind substrates for ubiquitin-mediated proteolysis. This variation may produce different numbers of PRBs. However, this switch may not affect the number of PRBs because both valine and isoleucine were observed in several dicot UFO homologs. Ikeda et al. (2007) suggested that the F-box motif in UFO homologs carrying APO1 is non-functional because of the lack of a protein–protein interaction motif at the C-terminal.
Furthermore, the region containing the ORF varied more between Sasanishiki and Nipponbare plants, which suggests the possibility of functional differentiation among japonica cultivars. These differences in PRB1 are considered to be non-critical for the fundamental function of this gene because both of the expressed genes result in the development of normal panicles. However, the number of PRBs was ~10% higher in the Habataki genotype.
There must be some variations in the distal promoter region beyond −2,827 bp that differentially affect the number of spikelets per panicle as well as the harvest index. One of the candidate sources of this variation is the CArG box. Durfee et al. (2003) suggested that the CArG box might be the recognition site for the MADS box DNA-binding protein. There are five CArGCW8AT motifs in the Nipponbare and Sasanishiki rice cultivar, at −720, −1,122, −2,707, −3,118, and −4,273 in the promoter region of APO1. The Habataki plants lack a CArGCW8AT motif at −3,118 bp, which was instead included in the HI1 allele. Ikeda et al. (2007) showed that a version of the APO1 gene with a 5-kb promoter region rescued the apo-1 mutant, while the version with a 2,924-bp promoter region did not. This CArG motif at −3,126 bp was placed within the functional promoter region. Therefore, this CArG motif may regulate APO1 gene expression and increase the number of spikelets per panicle, and the harvest index. This CArG motif may act as a repressor or negatively regulate APO1 gene expression because the lack of this motif in Habataki plants increased the expression of APO1. When Ikeda et al. (2007) introduced APO1 with a 2,924-bp promoter and longer 3′ flanking region to mutant plants, these plants overexpressed the APO1 gene and displayed abnormal shaped panicles with a higher numbers of spikelets per panicle. These authors hypothesized that there is a repressor region in the 3′ flanking region. However, it is also possible that the repressor is located in the promoter region, between −2,924 and −5,000 bp. Ikeda-Kawakatsu et al. (2009) showed that insertion of nDart1-0 3.5 kb upstream region of the APO1 enhance APO1 expression, which supports the regulatory role of distal promoter region of this gene.
Higher expression of APO1 in Habataki increased grain number
APO1 mRNA expression in the young developing panicle was about ten times higher in Habataki plants as in Sasanishiki plants, at the peak expressing stage. Also, the RNAi-APO1 reduced both the number of PRBs and the number of grains per panicle. Thus, higher APO1 expression in Habataki may increase the number of PRBs and the number of grains per panicle. However, Ikeda et al. (2007) showed that the overexpression of APO1, produced by the introduction of the 3-kb promoter APO1 gene fragment into apo1 mutants, resulted in the development of abnormal panicles that had too many PRBs and sterile spikelets. They consider that a moderate increase of APO1 expression is necessary to increase the panicle size in the mutants as well as the transformed plants although expression levels were not quantified (Ikeda-Kawakatsu et al. 2009). It is not clear whether this ten times higher APO1 expression in Habataki plants to be called moderate; this level of increase in APO1 expression observed in Habataki plants in concert with non-constitutive expression may increase the number of grains per panicle while keeping normal panicle architecture.
Conclusions
The APO1 gene is responsible for the increase of the number of PRBs, the harvest index, and the number of grains per panicle, thereby increasing grain yield. It also controls not only the development of panicle primordia but also the development of vascular bundle systems. Vascular bundle system development was regulated by the HI1 allele, which may correspond to the distal promoter region of the APO1 gene. This effect on developing vascular bundle systems increased grain yield through the increase of harvest index, with only a limited reduction in grain quality or ripening percentage.
References
Ando T, Yamamoto T, Shimizu T, Ma XF, Shomura A, Takeuchi Y, Lin SY, Yano M (2008) Genetic dissection and pyramiding of quantitative traits for panicle architecture by using chromosomal segment substitution lines in rice. Theor Appl Genet 116:881–890
Ashikari M, Matsuoka M (2006) Identification, isolation and pyramiding of quantitative trait loci for rice breeding. Trends Plant Sci 11:344–350
Ashikari M, Sakakibara H, Lin S, Yamamoto T, Takashi T, Nishimura A, Angeles ER, Qian Q, Kitano H, Matsuoka M (2005) Cytokinin oxidase regulates rice grain production. Science 309:741–745
Chang S, Puryear J, Cairney J (1993) A simple and efficient method for isolating RNA from pine trees. Plant Mol Biol Rep 11:113–116
Durfee T, Roe JL, Sessions RA, Inouye C, Serikawa K, Feldmann KA, Weigel D, Zambryski PC (2003) The F-box-containing protein UFO and AGAMOUS participate in antagonistic pathways governing early petal development in Arabidopsis. Proc Natl Acad Sci 100:8571–8576
Fukuoka H, Ogawa T, Mitsuhara I, Iwai T, Isuzugawa K, Nishizawa Y, Gotoh Y, Nishizawa Y, Tagiri A, Ugaki M, Ohshima M, Yano H, Murai N, Niwa Y, Hibi T, Ohashi Y (2000) Agrobacterium-mediated transformation of monocot and dicot plants using the NCR promoter derived from soybean chlorotic mottle virus. Plant Cell Rep 19:815–820
Fukushima A, Akita S (1997) Varietal differences of the course and differentiation time of large vascular bundles in the rachis of rice. Jpn J Crop Sci 66:24–28
Higo K, Ugawa Y, Iwamoto M, Korenaga T (1999) Plant cis-acting regulatory DNA elements (PLACE) database. Nucleic Acids Res 27:297–300
Hirose T, Terao T (2004) A comprehensive expression analysis of the starch synthase gene family in rice (Oryza sativa L.). Planta 220:9–16
Hirose T, Takano M, Terao T (2002) Cell wall invertase in developing rice caryopsis: molecular cloning of OsCIN1 and analysis of its expression in relation to its role in grain filling. Plant Cell Physiol 43:452–459
Hirose T, Ohdan T, Nakamura Y, Terao T (2006) Expression profiling of genes related to starch synthesis in rice leaf sheaths during the heading period. Physiol Plant 128:425–435
Huang X, Qian Q, Liu Z, Sun H, He S, Luo D, Xia G, Chu C, Li J, Fu X (2009) Natural variation at the DEP1 locus enhances grain yield in rice. Nat Genet 41:494–497
Ikeda K, Nagasawa N, Natago Y (2001) ABERRANT PANICLE ORGANIZATION 1 temporally regulates meristem identity in rice. Dev Biol 282:349–360
Ikeda K, Ito M, Nagasawa N, Kyozuka J, Nagato Y (2007) Rice ABERRANT PANICLE ORGANIZATION 1, encoding an F-box protein, regulates meristem fate. Plant J 51:1030–1040
Ikeda-Kawakatsu K, Yasuno N, Oikawa T, Iida S, Nagato Y, Maekawa M, Kyozuka J (2009) Expression level of ABERRANT PANICLE ORGANIZATION 1 determines rice inflorescence form through control of cell proliferation in the meristem. Plant Physiol 150:736–747
Kobayashi S, Fukuta Y, Yagi T, Sato T, Osaki M, Khush GS (2004) Identification and characterization of quantitative trait loci affecting spikelet number per panicle in rice (Oryza sativa L.). Field Crops Res 89:253–262
Komatsu K, Maekawa M, Ujiie S, Satake Y, Furutani I, Okamoto H, Shimamoto K, Kyozuka J (2003a) LAX and SPA: major regulators of shoot branching in rice. Proc Natl Acad Sci 100:11765–11770
Komatsu M, Chujo A, Nagato Y, Shimamoto K, Kyozuka J (2003b) FRIZZY PANICLE is required to prevent the formation of axillary meristems and to establish floral meristem identity in rice spikelets. Development 130:3841–3850
Lanceras JC, Pantuwan G, Jongdee B, Toojinda T (2004) Quantitative trait loci associated with drought tolerance at reproductive stage in rice. Plant Physiol 135:384–399
Levin JZ, Meyerowitz EM (1995) UFO: an Arabidopsis gene involved in both floral meristem and floral organ development. Plant Cell 7:529–548
Mao B-B, Cai W-J, Zhang Z-H, Hu Z-L, Li P, Zhu L-H, Zhu Y-G (2003) Characterization of QTLs for harvest index and source-sink characters in a DH population of rice (Oryza sativa L.). Acta Genet Sin 30:1118–1126
Matsushima S (1957) Analysis of developmental factors determining yield and yield prediction in lowland rice. Bull Natl Inst Agric Sci A 5:1–271
Mei HW, Luo LJ, Ying CS, Wang YP, Yu XQ, Guo LB, Paterson AH, Li ZK (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
Miki D, Shimamoto K (2004) Simple RNAi vectors for stable and transient suppression of gene function in rice. Plant Cell Physiol 45:490–495
Moritoh S, Miki D, Akiyama M, Kawahara M, Izawa T, Maki H, Shimamoto K (2005) RNAi-mediated silencing of OsGEN-L(OsGEN-like), a new member of the RAD2/XPG nuclease family, causes male sterility by defect of microspore development in rice. Plant Cell Physiol 46:699–715
Murai M, Nakamura K, Saito M, Nagayama A, Ise K (2005) Yield-increasing effect of a major gene, Ur1(Undulate rachis-1) on different genetic backgrounds in rice. Breed Sci 55:279–285
Nagao S, Takahashi M, Kinoshita T (1958) Inheritance on a certain ear type in rice. Genetical studies on plant, XXIII. Mem Fac Agric Hokkaido Univ 3:38–47
Nagata K, Fukuta Y, Shimizu H, Yagi T, Terao T (2002) Quantitative trait loci for sink size and ripening traits in rice (Oryza sativa L.). Breed Sci 52:259–273
Nakagawa M, Shimamoto K, Kyozuka J (2002) Overexpression of RCN1 and RCN2, rice TERMINAL FLOWER 1/CENTRORADIALIS homologs, confers delay of phase transition and altered panicle morphology in rice. Plant J 29:743–750
Pfaffl MW (2001) A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 29:e45
Pouteau S, Nicholls D, Tooke F, Coen E, Battey N (1998) Transcription pattern of a FIM homologue in Impatiens during floral development and reversion. Plant J 14:235–246
Sasahara H, Fukuta Y, Fukuyama T (1999) Mapping of QTLs for vascular bundle system and spike morphology in rice, Oryza sativa L. Breed Sci 49:75–81
Sasaki T, Matsumoto T, Yamamoto K et al (2002) The genome sequence and structure of rice chromosome 1. Nature 420:312–316
Shomura A, Izawa T, Ebana K, Ebitani T, Kanegae H, Konishi S, Yano M (2008) Deletion in a gene associated with grain size increased yields during rice domestication. Nat Genet 40:1023–1028
Simon R, Carpenter R, Doyle S, Coen E (1994) Fimbriata controls flower development by mediating between meristem and organ identity genes. Cell 78:99–107
Snyder FW, Carlson GE (1984) Selecting for partitioning of photosynthetic products in crops. Adv Agron 37:47–72
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
Taylor S, Hofer J, Murfet I (2001) Stamina pistilloida, the pea ortholog of Fim and UFO, is required for normal development of flowers, inflorescences, and leaves. Plant Cell 13:31–46
Tian F, Zhu Z, Zhang B, Tan L, Fu Y, Wang X, Sun CQ (2006) Fine mapping of a quantitative trait locus for grain number per panicle from wild rice (Oryza rufipogon Griff.). Theor Appl Genet 113:619–629
Toki S, Hara N, Ono K, Onodera H, Tagiri A, Oka S, Tanaka H (2006) Early infection of scutellum tissue with Agrobacterium allows high-speed transformation of rice. Plant J 47:969–976
Tsugane K, Maekawa M, Takagi K, Takahara H, Qian Q, Eun CH, Iida S (2006) An active DNA transposon nDart causing leaf variegation and mutable dwarfism and its related elements in rice. Plant J 45:46–57
Wang E, Wand J, Zhu X, Hao W, Wang L, Li Q, Zhang L, He W, Lu B, Lin H, Ma H, Zhang G, He Z (2008) Control of rice grain-filling and yield by a gene with a potential signature of domestication. Nat Genet 40:1370–1374
Wilkinson MD, Haughn GW (1995) UNUSUAL FLORAL ORGANS Controls meristem identity and organ primordia fate in Arabidopsis. Plant Cell 7:1485–1499
Xiao J, Li J, Grandillo S, Ahn SN, Yuan L, Tanksley SD, McCouch SR (1998) Identification of trait-improving quantitative trait loci alleles from a wild rice relative, Oryza rufipogon. Genetics 150:899–909
Yagi T, Nagata K, Fukuta Y, Tamura K, Ashikawa I, Terao T (2001) QTL mapping of spikelet number in rice (Oryza sativa L.). Breed Sci 51:53–56
Yamanouchi U, Yano M, Lin H, Ashikari M, Yamada K (2002) A rice spotted leaf gene, Spl7, encodes a heat stress transcription factor protein. Proc Natl Acad Sci 99:7530–7535
Yoon D-B, Kang K-H, Kim H-J, Ju H-G, Kwon S-J, Suh J-P, Jeong O-Y, Ahn S-N (2006) Mapping quantitative trait loci for yield components and morphological traits in an advanced backcross population between Oryza grandiglumis and the O. sativa japonica cultivar Hwaseongbyeo. Theor Appl Genet 112:1052–1062
Zhang W-H, Kokubun M (2004) Historical changes in grain yield and photosynthetic rate of rice cultivars released in the 20th century in Tohoku region. Plant Prod Sci 7:36–44
Zhang S, Sandal N, Plolwick PL, Stiller J, Stougaard J, Fobert PR (2003) Proliferating Floral Organs (Pfo), a Lotus japonicus gene required for specifying floral meristem determinacy and organ identity, encodes an F-box protein. Plant J 33:607–619
Zhuang J-Y, Lin H-X, Lu J, Qian H-R, Hittalmani S, Huang N, Zheng K-L (1997) Analysis of QTL × environment interaction for yield components and plant height in rice. Theor Appl Genet 95:799–808
Acknowledgments
This work was supported by a Grant-in-aid from the Ministry of Agriculture, Forestry and Fishery (MAFF) of Japan (Bio-design project 1998–2007). SAS submission was performed with the assistance of the Computer Center for Agriculture, Forestry and Fisheries Research, MAFF, Japan. We thank Prof. K Shimamoto of Nara Inst. Sci Tec. for providing p2K1+ vector. We are also grateful to Mr. Tokuya Genba, Mr. Tuyoshi Kotake, Mr. Tetsuya Aota, Mr. Masanori Ichihashi, Mr. Shinobu Yuminamochi, Ms. Keiko Nozaki, Ms. Kiiko Takatsuto and Ms. Setsuko Hayashi for their excellent technical assistance.
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Y. Xu.
Rights and permissions
About this article
Cite this article
Terao, T., Nagata, K., Morino, K. et al. A gene controlling the number of primary rachis branches also controls the vascular bundle formation and hence is responsible to increase the harvest index and grain yield in rice. Theor Appl Genet 120, 875–893 (2010). https://doi.org/10.1007/s00122-009-1218-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00122-009-1218-8