A chitin-inducible gibberellin-responsive gene closely linked


Proper plant height is crucial for lodging resistance to improve the rice crop yield. The application of semi-dwarf 1 led to the Green Revolution in the 1960s, by predominantly increasing the rice yield. However, the frequent use of single sd1 gene sources may cause genetic vulnerability to pests and diseases. Therefore, it is necessary to develop an alternate or new source of dwarfs for broadening the genetic base of dwarfism. Identifying useful novel semi-dwarf genes is important for the genetic manipulation of plant architecture in practical rice breeding.

Methodology/Principal Findings

Introgression lines derived from two contrasting parents in plant height, Zhenshan 97 and Pokkali, were employed to locate a gene with an overwhelming effect on plant height by the bulk segregant analysis method. A major gene ph1 gene was mapped to a region closely linked to sd1 on chromosome 1; the additive effects of ph1 were more than 50 cm on the plant height and 2 days on the heading date in a BC4F2 population and its progeny. ph1 was then fine mapped to BAC AP003227.


Lady using a tablet
Lady using a tablet


Essay Writers

Lady Using Tablet

Get your grade
or your money back

using our Essay Writing Service!

Essay Writing Service

Gene annotation indicated that LOC_OS01g65990 encoding a chitin-inducible gibberellins-responsive protein (CIGR), which belongs to the GRAS family, might be the right candidate gene. Co-segregation analysis of the candidate gene derived marker finally confirmed its identity as the candidate gene. A higher expression level of ph1 was detected in all the tested tissues in tall plants compared to those of short plants. Ph1 with a tremendous genetic effect on plant height is distinct from sd1; this genetic effect can be a new resource for breeding semi-dwarf varieties. The evaluation of CIGR as the ph1 candidate will allow us to discover the mechanism underlying plant height determination.


Rice is the main food crop in many parts of the world, but predominantly in Asia and Africa. The rice yield is determined either directly or indirectly by many traits. Among them, plant height plays a pivotal role, and dwarfism is a valuable trait in crop breeding, because it increases lodging resistance and decreases crop damage due to wind and rain, thereby increasing the crop yield. The second half of the 20th century is very prominent in crop science, because of the famous 'Green Revolution', where the semi-dwarf variety IR8, also known as 'miracle rice', enabled dramatic yield increases and helped to avert predicted food shortages in Asia during that period [1]. During the same period in wheat, another dominant semi-dwarf cultivar, Rht, facilitated a considerable increase in productivity and led to the 'Green Revolution' in wheat [2]. The short stature of IR8 is due to a mutation in the plant's sd1 gene, which was later identified as encoding an oxidase enzyme (GA20ox-2) involved in the biosynthesis of gibberellin, a plant growth hormone [3]-[5]. Gibberellin is also implicated in green-revolution varieties of wheat, but the reduced height of those crops is conferred by defects in the hormone's signaling pathway [6].

Moreover with the availability of the complete rice sequence and advancement in molecular marker technology, conventional breeding was almost replaced by molecular breeding [7]. Different types of mapping populations such as double-haploids, F2, F3, recombinant inbred lines (RILs), backcross inbred lines (BILs) and introgression lines (ILs) were developed by molecular marker technology for quantitative trait locus (QTL) identification [8]-[11]. A limited number of near-isogenic lines (NILs), backcross inbred lines (BILs), advanced backcross (AB) lines or introgression lines (ILs) can be used to precisely identify QTLs instead of a large F2 or recombinant inbred line (RIL) population [12]-[14]. Since such lines are homozygous, numerous genetically identical plants can be evaluated, thus increasing the accuracy of phenotyping without increasing the efforts of genotyping. Moreover, ILs are effective as a tool for the genetic analysis of QTLs. Any differences between ILs and their parents must be due to a QTL located in the introgressed region. These IL series help the fine mapping of QTLs as well as the precise estimation of the effect of each QTL [15]. Several research groups have reported QTLs that control plant height in rice [16]-[19]. Among all the genes identified, sd1 is the dominant one being widely used in breeding to develop semi-dwarf varieties.

However, the frequent use of single sd1 gene sources may cause genetic vulnerability to pests and diseases [20]-[22]. Therefore, it is necessary to develop alternate or new sources of dwarfs for broadening the genetic base of dwarfism. Undoubtedly, the identification of more plant height related genes/QTLs will provide us with more opportunities to breed diverse semi-dwarf varieties that can resist lodging. In addition, the functional dissection of more gene-regulated plant height traits will be helpful to further understand the molecular mechanism involved in semi-dwarfism. Identifying useful novel semi-dwarf genes is important for the genetic manipulation of plant architecture in practical rice breeding.

Lady using a tablet
Lady using a tablet


Writing Services

Lady Using Tablet

Always on Time

Marked to Standard

Order Now

In this study, we identified a major QTL on chromosome 1, ph1, which controls plant height and heading date from introgression lines derived from two contrasting parents in plant height, Zhenshan 97 and Pokkali. Further, we report here the fine mapping of ph1 and validation of its candidate gene by comparative sequencing and expression analysis between its parents.


Phenotypic variations of plant height in BC4F2 population and its progeny

The two parents, Zhenshan 97 and Pokkali, showed highly significant differences in plant height. Zhenshan 97 was of short stature with an average height of 88 cm and ranging from 85.0 to 89.5 cm, while Pokkali was tall with an average height of 196 cm and ranging from 195.0 to 218 cm (Table 1). In the BC4F2 population, the difference in average plant height between short and tall plants (carrying homozygous Zhenshan 97 alleles and Pokkali alleles at the targeted gene, respectively) was large, reaching up to 100 cm (Figure 1A). The major differences were caused by both the length of internodes and the number of internodes (Figure 1B). The short plants had an average of four elongated internodes, while tall plants had six. All the other internodes, except the first, of the tall plants were significantly longer than the counterparts of the short plants. Meanwhile, the tall plants had a longer panicle compared to the short plants (Figure 1C). In tall plants, the first and second internodes were of similar length and together contributed approximately 60% to the total culm, while in short plants, the first internode alone contributed more than 60% to the total culm (Figure 1D).

In BC4F2, plant height varied widely and ranged from 70 to 233 cm. Among the 172 BC4F2 plants, there were 40 short and 132 tall plants, which was in agreement with the expected segregation ratio (1:3) of a single Mendelian gene (χ2=0.22, P=0.639). Further progeny tests (BC4F3) confirmed that 40 and 44 plants were homozygous allele for Zhenshan 97 and Pokkali at the locus of target plant height gene; whose progeny expressed identical short and tall plants, respectively. Whereas the progenies of 88 plants showed varied plant height, indicating that they are heterozygous. The mean height for short, heterozygote and tall plants was 90.2 cm, 177.9 cm and 204.4 cm, respectively (Figure 2). The frequencies of the three genotypes matched the expected Mendelian ratio (1:2:1) for single locus segregation (χ2=0.28, P=0.869). This analysis suggested that one gene (termed ph1) controlled the variation of plant height in the BC4F2 population.

Primary mapping of ph1

Bulk segregant analysis was used to detect the major plant height gene. A total of 150 polymorphic simple sequence repeat (SSR) markers distributed evenly on all the 12 chromosomes between the parents were selected and screened for two bulks, the tall bulk and the short bulk. Among the 150 SSR markers, there was no polymorphism between the two bulks at the loci of 143 markers; the bulks are identical to Zhenshan 97 genotype at 135 marker loci, while the bulks are heterozygotes at eight markers. Polymorphisms were identified between the tall and short bulks at the loci of seven markers (RM3304, RM3825, RM6439, RM472, RM5382, RM1339 and RM1387) located on chromosome 1 (Figure S1). Consequently, these seven markers were regarded as being linked to the plant height gene (termed ph1) and were used to genotype the 172 BC4F2 plants. A local linkage map covering 35.8-cM was constructed (Figure 3A). The individual BC4F2 genotypes at ph1 were determined by progeny (BC4F3) tests. Then, ph1 treated as a marker was directly localized into the linkage group, where it is located in a 0.6-cM region flanked by markers RM5382 and RM1339 and co-segregated with RM1339 (Figure 3A). It is closely linked to sd1.

The interval mapping method was conducted to estimate its genetic effects by using the phenotypic data from BC4F2 and its progeny. A major plant height QTL (ph1) was detected in the 0.6-cM region between the markers RM5382 and RM1339 (Fig 4A). ph1 explained 90.3% of the phenotypic variance with additive and dominant effects of 57.1 cm and 30.6 cm, respectively. In the progeny test (BC4F3), ph1 explained 92.7% of plant height variation, but the additive effect was less than that of the F2 population. ph1 acted as partially dominant in both generations; the Pokkali allele increased plant height (Table 2). Variations in heading date were also observed in the BC4F3 population (Figure S2). Tall plants always flowered 3-4 days earlier than short ones, indicating the pleiotropic effect of ph1 on heading date. One QTL for heading date was also detected in the same interval where ph1 is located. This QTL could explain 31.2% of the total phenotypic variance with LOD scores of 13.7. The additive effect was 1.5 days. Pokkali allele promoted heading date (Table 2).

Fine mapping of ph1

Lady using a tablet
Lady using a tablet

This Essay is

a Student's Work

Lady Using Tablet

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Examples of our work

The revolutionary gene sd1 is located on the right side of the marker RM1339. RM1339 is more than 180 kb from sd1. Therefore, to provide evidence that the large effect of ph1 on plant height is not the allelism of sd1, we performed sequencing analysis of Zhenshan 97 and Pokkali with sd1. The sequencing analysis of sd1 showed no sequence difference in the three exons and two introns (Figure S3). However, we identified one SNP in the promoter region of sd1. To further investigate the role of sd1, we performed expression analysis. The quantitative real-time polymerase chain reaction (qRT-PCR) results showed that there was no difference in the expression levels of sd1 in root, leaf sheath and leaf blade tissues in both Zhenshan 97 and Pokkali (Figure S4). Thus, clear molecular level evidence determined that ph1 is distinct from sd1.

It is worth isolating ph1 because it is a novel plant height gene. In total, 1250 extremely short plants with a plant height of less than 100 cm, which were assumed to be Zhenshan 97 homozygotes at Ph1, were selected from 6400 plants of the BC4F2 population to screen recombinants between Ph1 and markers RM5382 and RM1339. Nine and three recombinants were identified between RM5382 and ph1, and RM1339 and ph1, respectively. In case the trait measurement yielded any false positives in the Zhenshan 97 homozygous plants, progeny tests of the twelve recombinants between RM5382 and RM1339 were conducted. Each recombinant progeny showed an identical short plant height, which was highly significantly shorter than the control Pokkali homozygotes, but showed no difference with the control Zhenshan 97 homozygotes (Table 3). This result confirmed the homozygous identity of the twelve Zhenshan 97 recombinants at ph1. To enhance the resolution of the ph1 local linkage map, we used one SSR (RM11960) and four InDel markers (Table S1) to screen the twelve recombinants; seven recombinants were identified between RM11960 and Ph1, six between PHA and Ph1, four between PHB and Ph1 and one between PHC and Ph1 (Figure 3B). Thus, Ph1 was narrowed down to the region of approximately 92 kb bounded by markers PHB and PHC (Figure 3B). One Nipponbare BAC (AP003227) spanned the region exactly.

The candidate gene in the 92-kb target region

There are 17 predicted genes in the 92-kb region according to the rice genome automated annotation database (http://rice.plantbiology.msu.edu/) (Table S2). Of these, 12 genes have homology with rice full-length cDNAs. Among these 17 genes, five are of unknown function; the functional annotations of 12 remaining genes are given in Table S2. Gibberellin-responsive genes were reported to be associated with plant height [4], [23]-[26]. Among all the 12 putative genes, LOC_Os01g65900, which encodes a chitin induced gibberellin responsive protein (CIGR), was gibberellin responsive; thus, CIGR was regarded as the ph1candidate.

Sequencing analysis of the 3640-bp CIGR gene showed single-base substitutions in the promoter region and exon 1 of both Zhenshan 97 and Pokkali, and also detected a 4-bp deletion in the intron region of Zhenshan 97 (Figure 4). Between the parents, four SNPs were found at 706 bp, 1048 bp, 1151 bp and 1377 bp upstream of ATG, and two SNPs were located in exon 1. CIGR contains three exons; it encodes a protein with 553 amino acids, but modifications in two amino acids were observed between the parents (Figure S6).

Co-segregation of the candidate

In order to confirm whether the candidate gene (CIGR) co-segregated with plant height, we developed a gene specific marker (C1) covering the SNP region in the first exon of ph1 and screened all five recombinants (32, 58, 67, 83 and 121). The genotype data showed that all the five recombinants are homozygous to Zhenashan 97 at C1 (Figure S5) and confirmed that CIGR co-segregated with plant height. Thus, CIGR might be the real candidate for ph1. In addition, our qRT-PCR results showed that Pokkali had significantly higher expression levels of CIGR in root, leaf sheath and leaf blade tissues than Zhenshan 97 (Figure 5).


Comparison of the genetic effects of ph1 to sd1

In this study, it was clearly confirmed that ph1 is distinct from sd1 by fine mapping with a large BC4F2 population. Comparative sequencing between the parents further confirmed that sd1 did not contribute to plant height variation. We detected no difference in the expression level of sd1 between ZS97 and Pokkali, and this was in agreement with the result that ph1 is a novel gene distinct from sd1. Although ph1 was not the allele of sd1, it showed similar genetic characters on plant height but with larger effects. ph1 has an overwhelming genetic effects; two homozygotes of ph1 showed a huge difference of 110 cm in plant height.

The variation in the final plant height in the BC4F2 population between tall and short plants is mainly contributed by the difference in the length of the panicle, the second, third and fourth uppermost stem internodes; in particular, the second and third showed differences of about 25 cm; But the uppermost stem length showed no significant difference (Figure 1B). While variation in the final plant height by sd1 results from differences in stem length of the first (subtending panicle) and second (subtending flag leaf) stem internodes [5], [27]. Meanwhile, the average number of elongated internodes of tall and short lines is different (Figure 1C). That is to say, ph1 mainly controls the number of internodes and the length of uppermost internodes except the first one. In addition, ph1 has a pleiotropic effect on the heading date. Pokkali alleles increase the plant height and decrease the days to heading, which is in agreement with their negative correlation (r=-0.55, P< 0.01). Taken together, these results confirm that ph1 is a novel gene based on phenotypic variation.

Chitin-inducible gibberellin-responsive gene might be the probable candidate gene

First, we fine mapped ph1 to the BAC clone AP003227 (Figure 3B) containing twelve putative known functional genes. Then, with the help of bioinformatics analyses, we inferred the CIGR gene (LOC_OS01g65900) encoding a chitin-induced gibberellins-responsive protein as the candidate of ph1. Finally, the identity of the candidate gene was validated by co-segregation analysis.

It is a well known fact that gibberellin plays a key role in the plant height. Many important genes like sd1 and d18 are influenced by gibberellins [5], [28]. CIGR genes are gibberellin response genes and were reported to play key roles in defense and plant development [29]-[33]. Moreover, the CIGR gene belongs to the GRAS family. Comparative analysis showed important genes like SCARECROW LIKE 1 (SCL1) in Arabidopsis, Dwarf8 in maize and MOC1 (MONOCULM 1) in rice belong to this family and are related to plant growth and development. CIGR is 85% similar to SCL in Aeluropus littoralis and 77% to the SCL in Arabidopsis. Recently, Wang et al. [34] reported that SCL6 is involved in shoot branching and plant height by testing double and triple mutants of SCL6-III, SCL6-IV and SCL6-V in Arabidopsis.

In addition, it was reported that the induction of CIGR1 and CIGR2 was observed only upon application of phytoactive GA, and the accumulation of mRNAs for CIGR1 and CIGR2 is correlated with the bioactivities of GA in suspension-cultured cells of rice [31]. CIGR1 and CIGR2 are excellent markers of GA signal transduction [31]. In fact, high levels of CIGR1 and CIGR2 mRNAs accompany with a high concentration of phytoactive GA, which promotes plant height. In this study, the expression level of ph1 was significantly higher in root, leaf sheath and leaf blade tissues in tall plants compared with short plants; this finding indicates that tall plants contain a higher concentration of phytoative GA that resulted in increased plant height. Transformation of the candidate gene will ultimately reveal its biological functions on plant height.

Plant height gene cluster on chromosome 1

In rice, disease resistant (R) gene clusters contributing significantly to plant defense were observed and studied extensively to elucidate their mechanisms [35], [36]. The characteristic clustering of R genes has been proposed to facilitate the evolution of novel resistance specificities via recombination or gene conversion [37]. Therefore, the organization of functionally related genes in clusters is expected to have an evolutionary advantage to the organism. Besides, many genes controlling the same agronomic traits were located in a very small region, making it difficult to determine whether they are one or two genes in a primary mapping population. For example, Gn1, which is regarded as one gene controlling rice grains per panicle, is located to the nearest marker BB-85; but, it was dissected into the two closely linked genes Gn1a and Gn1b, which both control the trait in advanced populations [38]. Later, SPP1, which controls the number of spikelets per panicle, was also detected in the nearby region [39]. A similar situation is observed for the genes controlling plant height. qCL1 is located 1.4 cM and 2.6 cM away from two plant height genes, d18 and d2, which are located with 1.2 cM distance on chromosome 1 [40]. QTLph1 and sd1 are linked on chromosome 1 with a 1.7-Mb physical distance [19], [41]. The important plant height genes sd1, QTLph1, d2, d18, qCL1 and ph1 from this study are all located on chromosome 1; many of these genes are located at the distal end, indicating a hot spot region for plant height genes. However, the mechanism or functions of plant growth or development gene cluster genes are not clear. Activating these clusters will uncap a vast resource of novel enzymes, compounds, pathways and diverse chemistries, which can be exploited for a wide range of applications in plant height adaptation.

Materials and Methods

Plant materials and development of mapping population

A set of introgression lines (BC4F2) was developed from a cross between two contrasting Oryza sativa L. indica varieties in plant height, namely Zhenshan 97, a Chinese native, and Pokkali, an Indian native (Fig 1A); Zhenshan 97 was used as the recurrent parent. One introgression line (BC4F2 family) showed a clear segregation in plant height in 2003. The family, which consisted of 172 plants, and its progeny (BC4F3) were grown in summer 2004 and 2005 in a bird-net-equipped experimental farm of Huazhong Agricultural University, Wuhan, China (29°58'N, 113°41'E). Field experiments were carried out in a randomized complete block design.

For the progeny test, 24 plants of each BC4F3 family were grown in a two-row plot in Wuhan in 2006. The 20 plants growing in the middle were investigated for plant height and heading date. Field management, including irrigation, fertilizer application and pest control, followed essentially the normal agricultural practices.

Fine mapping population

A large BC4F2 population of 6400 plants was grown for plant height in 2005 in Wuhan. The progenies of twelve recombinants between the gene Ph1 and two closely linked flanking markers were sowed in summer 2007 in the same field where the progeny of BC4F2 population grown.

DNA extraction and marker selection

DNA was extracted from fresh leaf samples collected at the seedling stage using a modified CTAB protocol [42]. PCR was performed using a hot start Taq polymerase (TaKaRa) under the following conditions: 95°C for 4 min, 35 cycles of 95°C for 40 s, 55-58°C for 40 s and 72°C for 1 min, followed by 72°C for 6 min. SSR markers were used for developing a genetic linkage map. In addition to SSR markers, three newly developed InDel markers for fine mapping are listed in Supplemental table 1, according to the publicly available rice genome sequences (http://www.rgp.dna.affrc.go.jp). Markers of the Monsanto Rice Genome (MRG) series were designed according to the rice genome sequences of the Monsanto Company [43] and those of the Rice Microsatellite (RM) series according to Temnykh et al. [44], [45]. The SSR assay was conducted by polyacrylamide gel electrophoresis, as described by Wu and Tanksley [46], with minor modifications.

Plant height and heading date measurement

Plant height was recorded from the field surface to the top of the highest panicle of each plant for BC4F2 and its progeny BC4F3. At the seed maturation stage, the average panicle length, number of internodes and their lengths were recorded for both Zhenshan 97 and Pokkali. The heading date was recorded as the days from sowing to the first panicle appearance for each plant in BC4F3.

Bulk segregant analysis for primary mapping

Five extremely tall individuals with a plant height greater than 200 cm and five extreme short individuals with plant height less than 90 cm were selected from the BC4F2 population for developing two DNA bulks. Equal quantities of leaf tissue from each individual were bulked, and DNA was extracted from the bulk. To identify the polymorphic markers, the two parents Zhenshan 97 and Pokkali were genotyped with 440 SSR markers covering 12 rice chromosomes. Markers showing polymorphism in the form of a clearly visible difference in band intensity were selected for genotyping these two tail bulks.

Genetic linkage map and genetic effect analysis

The molecular linkage map was constructed using Mapmaker/EXP 3.0 [47]. The Kosambi function was used to calculate the genetic distance. Interval mapping was conducted using Mapmaker/QTL [48]. The LOD threshold was of 3.0, determined by 1,000 random permutations at a false positive rate of 0.05 for the trait.

Sequencing analysis of sd1 gene and probable candidate gene

Based on the BAC clones AP003561 and AP003227 sequences (http://www.ncbi.nlm.nih.gov), eight and five specific primer pairs were designed to sequence the promoter and coding regions of the sd1 gene and the probable candidate gene, respectively (Table S1). The two genes were amplified using LA Taq (Takara) from genomic DNA of Zhenshan 97 and Pokkali, and the PCR products were purified. These purified PCR fragments were sequenced using the Big Dye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA). Data were collected using the ABI Prism 3730 DNA Analyzer (Applied Biosystems, Foster City, CA, USA) and interpreted using SEQUENCHER 4.6 (Gene Codes Corporation, Ann Arbor, MI, USA). The sequence alignment was performed with the BLAST network service (http://www.ncbi.nlm.nih.gov), National Center for Biotechnology Information, (NCBI) and ClustalW (http://www.ebi.ac.uk/Tools/clustalw2/index.html), European Bioinformatics Institute, EBI).

RNA extraction and quantitative real time PCR analysis (qRT-PCR)

Total RNA was extracted from duplicate biological samples leaf sheath, root and leaf blade tissues of Zhenshan 97 and Pokkali using the TRIzol reagent (Invitrogen, Carlsbad, CA, USA), according to the manufacturer's instructions. The quality of the RNA samples was evaluated by absorbance measurements using a Nanodrop Spectrophotometer ND-1000 (Nanodrop Technologies, USA). All the RNA samples used in the qRT-PCR reactions showed a 260/280 nm absorbance ratio of 1.9-2.2. Prior to qRT-PCR, the total RNA samples were pretreated with RNase-free DNase I to eliminate any contaminating genomic DNA.

First-strand cDNAs were synthesized from the DNaseI-treated total RNA samples using Superscript reverse transcriptase (Invitrogen, Carlsbad, CA, USA) in a reaction volume of 40 μl, according to the manufacturer's instructions. qRT-PCR was performed in an optical 96-well plate with an ABI PRISM 7500 real-time PCR system (Applied Biosystems, Foster City, CA, USA). Gene-specific primers were designed for the sd1 gene and the candidate gene (CIGR) (Table S1). Each reaction contained 12.5 µL of SYBR Premix Ex Taq (TAKARA), 0.5 µL of ROX Reference Dye (TAKARA), 5.0 µL of cDNA samples and 10 µM gene specific primers in a final volume of 25 µL. The rice actin1 gene was used as the endogenous control; it was amplified with the primers 5´-TGGCATCTCTCAGCACATTCC-3´ and 5´-TGCACAATGGATGGGTCAGA-3´. The thermal cycle used was as follows: 95°C for 10 s, 45 cycles of 95°C for 5 s and 60°C for 35 s. All qRT-PCR reactions were carried out in triplicate. The relative expression levels were analyzed as described by Livak and Schmittgen [49].


The authors kindly thank farm technician Mr. J.B. Wang for his excellent field work.