Introduction
Dwarf genes play an important role in the improvement of crop yield. In the 1960 s, plant breeders developed cereal varieties with shorter stems, which improved lodging resistance and, in turn, increased yield (Khush Reference Khush2001). The new varieties prevent many people across the world from starving, which was well known as the ‘Green Revolution’. Currently, there are a large number of modern wheat varieties that contain semi-dwarfing alleles (Borner et al. Reference Borner, Plaschke, Korzun and Worland1996).
Molecular identification has revealed that the genes responsible for the Green Revolution in wheat (Peng et al. Reference Peng, Richards, Hartley, Murphy, Devos, Flintham, Beales, Fish, Worland, Pelica, Sudhakar, Christou, Snape, Gale and Harberd1999) and rice are involved in the gibberellin (GA) biosynthetic/signalling pathway (Hedden Reference Hedden2003, Monna et al. Reference Monna, Kitazawa, Yoshino, Suzuki, Masuda, Maehara, Tanji, Sato, Nasu and Minobe2002). It has been found that GA plays important roles in the control of dwarf and plant development, including seed germination, leaf expansion, stem elongation (Plackett et al. Reference Plackett, Thomas, Wilson and Hedden2011, Sun and Gubler Reference Sun and Gubler2004) and stress (Achard et al. Reference Achard, Cheng, De Grauwe, Decat, Schoutteten, Moritz, Van der Straeten, Peng and Harberd2006).
It has been predicated that the current crop production must be doubled by 2050 to meet the food demand of the increasing world population. Soybean [Glycine max (L.) Merr.], one of the most important crops, is a main source of protein and oil for both humans and animals. Dwarf mutant analysis is essential for soybean yield improvement. In this study, a severe soybean dwarf mutant screened from a γ-ray-treated population was characterized. RNA-seq analysis revealed that many genes related to hormone metabolism exhibited significant differences in the wild type and dwarf mutant. Treatment with GA3 (Gibberellin 3) could partially restore the mutant phenotype. Our results indicate that the responding gene may play essential roles in many hormone pathways.
Materials and methods
Screening of Gmdwarf1
Gmdwarf1 was screened from a γ-ray-treated (with 250 Gy) population of the soybean variety Huaxia 3 in 2006. After treatment under short-day conditions and with 10 mg/l GA3, M2 (indicate second-generation) mature seeds were obtained at the Campus Farm, Chinese Academy of Agricultural Sciences in Beijing.
Plant growth conditions and material collection
The soybean plants were grown in the normal season of 2012 and 2013 at the Experimental Station of the Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, in Beijing. Four seeds were planted in each pot. Nine pots were used for each experiment.
For RNA-seq analysis, 10-day-old roots and cotyledons were collected from the wild-type and mutant plants after germination and immediately frozen in liquid nitrogen. Each sample was collected from at least five independent plants and pooled together.
RNA-seq library construction and sequencing
Total RNA was isolated using the TRIzol reagent (Invitrogen, http://www.lifetechnologies.com/). RNA-seq libraries were constructed following the method described previously (Severin et al. Reference Severin, Woody, Bolon, Joseph, Diers, Farmer, Muehlbauer, Nelson, Grant, Specht, Graham, Cannon, May, Vance and Shoemaker2010). RNA sequencing was carried out on a Hi-Sequation 2000 analyser (illumina, http://www.illumina.com/systems.ilmn).
Computational analysis of sequencing data and determination of differentially expressed genes (DEGs)
After quality control, the raw sequencing data were mapped to the soybean reference genome (http://www.phytozome.net/soybean) using TopHat2 (Trapnell et al. Reference Trapnell, Pachter and Salzberg2009) with a default parameter. DEGs were determined using edgeR (Robinson et al. Reference Robinson, McCarthy and Smyth2010). Gene Ontology (GO) analysis was carried out using agriGO (Du et al. Reference Du, Zhou, Ling, Zhang and Su2010) and WEGO (Ye et al. Reference Ye, Fang, Zheng, Zhang, Chen, Zhang, Wang, Li, Li and Bolund2006).
Results and discussion
Phenotypic characterization of the Gmdwarf1 mutant in soybean
A dwarf mutant, Gmdwarf1, was screened from a γ-ray-treated population of the soybean variety Huaxia 3. The Gmdwarf1 mutant germinated later than the wild type (Fig. 1(a)). On average, germination was delayed by 5 day in the mutant (Fig. 1(b)). After germination, the Gmdwarf1 mutant exhibited an extremely dwarf phenotype and no obvious internode. The leaves of the mutant were much smaller and darker green than those of the wild type (Fig. 1(c)). In addition, elongated roots were observed in the mutant (Fig. 1(d)).
Fig. 1 Phenotypes of soybean Gmdwarf1 mutant and wild-type (WT) plants. (a) Phenotypes of 3-week-old Gmdwarf1 mutant and WT plants (scale bar 5 cm). (b) Difference in the germination date between Gmdwarf1 mutant and WT plants. (c) Four-week-old Gmdwarf1 mutant and WT plants (scale bar 5 cm). (d) Difference in the root phenotype of 10-day-old Gmdwarf1 mutant and WT plants (scale bar 5 cm).
The phenotype of Gmdwarf1 indicated that it might be caused by a mutation related to the GA biosynthetic or response pathway (Peng and Harberd Reference Peng and Harberd2002, Thomas and Sun Reference Thomas and Sun2004). To determine the genes that would be affected by the mutation, the RNA of 10-day-old roots and cotyledons of the wild type and dwarf mutant were extracted and used for RNA sequencing.
Gmdwarf1 led to different expression of hormone pathway-related genes
In total, more than 263 M reads (~26.3 Gb) were obtained from RNA-seq. The average reads for each sample were 65.8 M (~6.58 Gb; Fig. 2(a)). After trimming of adaptor sequences and filtering of low-quality reads, 52–73% of high-quality unique reads (excluding multiple mapped reads) were aligned to the soybean genome. Among the four samples, the mapped read ratios in the cotyledons were higher than those in the roots (Fig. 2(a)).
Fig. 2 Differentially expressed genes (DEGs) in Gmdwarf1 mutant and wild-type (WT) plants detected through RNA-seq analysis. (a) Statistics of the mapped read ratios from RNA-sequation (b) Venn diagram of DEGs in the samples. (c) Heat map of DEG expression in Gmdwarf1 mutant and WT plants. (d) DEGs related to endogenous stimulus response. (e) Phenotype of Gmdwarf1 mutant and WT plants after treatment with 140 μM GA3 (scale bar 5 cm). (f) Phenotype of Gmdwarf1 mutant and WT plants after treatment with water (scale bar 5 cm).
DEG detection revealed that there were 793 DEGs in the roots of the wild type and mutant. In the mutant, 460 DEGs were up-regulated and 333 were down-regulated (Fig. S1(A), available online). In total, 3186 DEGs were found in the cotyledons of the mutant, including 1490 that were up-regulated and 1696 that were down-regulated (Fig. S1(B), available online). Of the total DEGs, 52 that were up-regulated and 77 that were down-regulated were shared by both the cotyledons and roots (Fig. 2(b)). Their expression patterns are shown in Fig. 2(c). GO annotation demonstrated that the DEGs were involved in different biological processes (Fig. S1(C), available online). Further analysis revealed that these DEGs were enriched in the process of endogenous stimulus response (GO:0009717; Fig. S1(D), available online), in which GA-related genes only make up a small portion. However, many genes involved in other hormone pathways were detected (Fig. 2(d)). This may indicate that Gmdwarf1 may not play essential roles in only a single hormone biosynthetic pathway.
Treatment with 140 μM GA3 could partially restore the mutant phenotype (Fig. 2(e)), whereas that with water could not (Fig. 2(f)). Genetic analysis through crossing with the wild type revealed that the segregating ratio of the wild type to the mutant in the F2 population was almost equal to 3:1 (65:31;
$$\chi _{3:1}^{2} = 2.72 \lt \chi _{0.05}^{2} = 3.84 $$
), indicating the Gmdwarf1 phenotype was controlled by a single recessive gene. Future cloning and functional analysis of Gmdwarf1 would help us to understand the underlying mechanisms in the plant hormone pathways.
Supplementary material
To view supplementary material for this article, please visit http://dx.doi.org/10.1017/S1479262114000306
Acknowledgements
This study was supported by the open project of the State Key Laboratory of Plant Cell and Chromosome Engineering, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, National Natural Science Foundation of China and China Agriculture Research System (grant no. 31222042, grant no. 91131005 and CARS-04).
