Genome-wide identification and analysis of the growth-regulating factor family in Chinese cabbage (Brassica rapa L. ssp. pekinensis)

Background Growth regulating factors (GRFs) have been shown to play important roles in plant growth and development. GRF genes represent a large multigene family in plants. Recently, genome-wide structural and evolutionary analyses of the GRF gene families in Arabidopsis, rice, and maize have been reported. Chinese cabbage (Brassica rapa L. ssp. pekinensis) is one of the most important vegetables for agricultural production, and a full genome assembly for this plant has recently been released. However, to our knowledge, the GRF gene family from Chinese cabbage has not been characterized in detail. Results In this study, genome-wide analysis was carried out to identify all the GRF genes in Chinese cabbage. Based on the complete Chinese cabbage genome sequence, 17 nonredundant GRF genes, named BrGRFs, were identified and classified into six groups. Phylogenetic analysis of the translated GRF protein sequences from Chinese cabbage, Arabidopsis, and rice indicated that the Chinese cabbage GRF proteins were more closely related to the GRF proteins of Arabidopsis than to those of rice. Expression profile analysis showed that the BrGRF genes had uneven transcript levels in different organs or tissues, and the transcription of most BrGRF genes was induced by gibberellic acid (GA3) treatment. Additionally, over-expression of BrGRF8 in transgenic Arabidopsis plants increased the sizes of the leaves and other organs by regulation of cell proliferation. Conclusions The data obtained from this investigation will contribute to a better understanding of the characteristics of the GRF gene family in Chinese cabbage, and provide a basis for further studies to investigate GRF protein function during development as well as for Chinese cabbage-breeding programs to improve yield and/or head size. Electronic supplementary material The online version of this article (doi:10.1186/1471-2164-15-807) contains supplementary material, which is available to authorized users.


Background
Growth regulating factors (GRFs) are plant-specific proteins that play important roles in regulating plant growth and development. A GRF gene was first identified in rice where it encodes a protein that functions in regulating gibberellic acid (GA)-induced stem elongation [1].
The deduced protein products of GRF genes contain two conserved domains in the N-terminal regions, the QLQ and WRC domains. The QLQ domain interacts with GRF interacting factors (GIF) and the resulting complex acts as a transcriptional co-activator [2], while the WRC domain comprises a functional nuclear localization signal (NLS) and a zinc-finger motif that functions in DNA binding [3].
Currently, the GRF gene family consists of nine Arabidopsis thaliana genes [3], 12 Oryza sativa genes [4], 14 Zea mays genes [5], and 10 Brachypodium distachyon genes [6]. In these plants, the GRF genes are strongly expressed in actively growing and developing tissues, such as shoot tips, flower buds, and immature leaves, but weakly expressed in mature tissues or organs. GRF genes have been reported to act as positive regulators of leaf size through the promotion and/or maintenance of cell proliferation activity in leaf primordia [2,3,7,8]. In Brassica napus, GRF2 was found to enhance seed oil production by regulating cell number and plant photosynthesis [9]. GRF genes may act by regulating cell proliferation through the suppression of KNOX gene expression [10], which inhibits GA biosynthesis in the S-adenosyl methionine (SAM) cycle by down-regulating the key biosynthetic gene GA20 oxidase [11][12][13][14][15], or by controlling the level of GA2 oxidase 1, which degrades GA [16]. Recently, many studies have reported the involvement of GRF genes in the regulation of flower development [17][18][19].
Chinese cabbage (Brassica rapa L. ssp. pekinensis) is one of the most important vegetables for agricultural production in Asia and, although it originated in China, it is increasingly popular in other countries around the world. Yield enhancement is now one of the critical targets of plant breeding programs focused on genetic improvement. Understanding the functions of GRF genes in regulating organ size in Chinese cabbage will achieve this objective. However, knowledge about the molecular features of GRF genes in Chinese cabbage is limited. Therefore, identifying and characterizing GRF genes in Chinese cabbage is of great interest.
In this study, 17 putative GRF genes of Chinese cabbage were identified from the Brassica database (http://brassicadb.org/brad/) [20]. The expression patterns of these BrGRFs and the molecular features of the translated BrGRF proteins were analyzed, and the function of BrGRF8 was studied further. The results indicated that GRF genes had higher expression levels in immature organs or tissues than in mature ones, and their transcription was induced by gibberellic acid (GA3). Further analysis showed that cell proliferation was enhanced in transgenic plants.

Identification of the BrGRFs
Based on the recently sequenced B. rapa line Chiifu genome and annotated genes [20], 17 BrGRFs were identified from the Brassica database (http://brassicadb.org/ brad/) and designated BrGRF1-BrGRF17 according to their distribution in the genome ( Table 1). The coding sequence (CDS) lengths of the BrGRFs varied widely; BrGRF15 was the longest (1617 bp) and BrGRF11 was the shortest (1089 bp). The intron/exon structures of the BrGRFs were determined by aligning the CDSs to the genomic sequence. The results indicated that all the BrGRF gene sequences contained introns in their CDSs. The number of introns varied from one to nine ( Figure 1B); however, most of the genes (10 out of 17) had three introns, followed by four introns (3 out of 17), two introns (2 out of 17), and one and nine introns (1 each out of 17). This result was similar to the findings from previous studies on Arabidopsis (a dicot) where most AtGRFs contained three introns (7 out of 9) [4], and on rice (a monocot) where most OsGRFs had two introns (6 out of 12), followed by three introns (4 out of 12), and four introns (2 out of 12). Additionally, the number of introns in the CDSs of GRF genes in the same subfamily was different. For example, BrGRF4, BrGRF15, and BrGRF17, which belong to subfamily A ( Figure 1A), had three, nine, and three introns respectively.
Simple sequence repeat (SSR) markers are useful for a variety of applications in plant genetic mapping and molecular breeding because of their genetic codominance, abundance, dispersal throughout the genome, multi-allelic variation, high reproducibility, and high level of polymorphisms [21]. In this study, 16 SSR markers, including nine di-, six tri-, and one hexa-nucleotide motifs were detected in the 17 identified BrGRFs using the online SSR identification tool SSRIT (Table 2). Further, BrGRF10, BrGRF13, BrGRF15, and BrGRF16 had two SSR markers each, BrGRF2-BrGRF6, BrGRF11, BrGRF12, and BrGRF17 genes had only one SSR marker each, and BrGRF1, BrGRF7-BrGRF9, and BrGRF14 genes had no SSR markers. Of the detected SSRs in these genes, seven were detected in exons and nine were detected in introns.
The BrGRFs were unevenly distributed across the chromosomes (Figure 2), which is similar to previous results in Arabidopsis and rice [4]. The chromosome 03, 01, 02 and 04 has five, three, two and two genes, respectively, whereas the chromosome 06, 07 and 09 each includes only one.

Conserved domains and motifs in the predicted BrGRF proteins
The modular structures of GRF proteins have been studied thoroughly in Arabidopsis [3], rice [4], maize [5], and Brachypodium distachyon [6]. Based on this information, we used the MEME web server (http://meme.nbcr.net/ meme/cgi-bin/meme.cgi) to analyze the domain distribution in BrGRFs. Motif 2, specified as the QLQ domain, was predicted to be present in 16 of the 17 BrGRFs, BrGRF3 was the exception (Figure 3). We found that although BrGRF3 could encode the QLQ domain, the deletion of two adenine residues at positions 62 and 63 bp in the BrGRF3 sequence compared with the BrGRF8 sequence caused a frameshift that introduced a TAG stop codon that truncated the BrGRF3 protein sequence (Figure 4). Motif 1, specified as the WRC domain, was predicted in all 17 BrGRF proteins. The results also indicated that AtGRF9 and BrGRF12 contained a second motif 1 downstream of the first. Additionally, motif 3, specified as the GGPL domain, was observed in 12 of the 17 BrGRF proteins, including BrGRF1, BrGRF2, BrGRF4-BrGRF7, BrGRF9-BrGRF11, BrGRF13, BrGRF15, and BrGRF17 ( Figure 3).
Analyzing the duplication events that may have occurred in the Chinese cabbage genome during evolution will help in understanding the evolutionary mechanisms of the BrGRFs. Paralogous relationships for these BrGRFs have been displayed in http://brassicadb.org/brad/. The results indicated that there were three set of triplicated genes in the Chinese cabbage genome: BrGRF3, 8, and 14 in the F block; BrGRF6, 11, and 13 in the J block; and BrGRF15, 4, and 17 in the N block ( Table 3). The other BrGRF genes were either duplicated or singletons. No tandem duplication event was detected in the BrGRF family.

Putative functional analysis of the BrGRF proteins
The Gene Ontology (GO) database (http://www.geneontology.org/) is an international standardized gene functional classification system that offers a dynamically  updated controlled vocabulary that comprehensively describes the properties of genes and their products in any organism under three main categories: biological process, molecular function, and cellular component. In this study, the 17 predicted BrGRFs were assigned GO terms. The results indicated that most of BrGRFs (except BrGRF3, which had no GO hits) were annotated with terms in the same GO groups, including ATP binding; hydrolase activity, acting on acid anhydrides, in phosphorus-containing anhydrides under the molecular function category, regulation of transcription under the biological process category, and nucleus under the cellular component category (Additional file 1). These data implied that these BrGRF proteins could have similar biological functions. BrGRF15 was also annotated with terms related to hydrolase activity, acting on ester bonds; nuclease activity; DNA binding; recombinase activity under the molecular function category, and response to DNA damage stimulus; DNA recombination; DNA repair; nucleobase, nucleoside, nucleotide and nucleic acid metabolic process under the biological process category. This result suggested that BrGRF15 may be involved in nucleic acid metabolic processes.
Expression patterns of the BrGRFs and BrGIFs (GRF interacting factor genes) GRF proteins play important roles in the growth and development in plants [2,3,[7][8][9][17][18][19]. To understand which GRF genes may be involved in regulating specific tissue or organ growth in Chinese cabbage, the expression patterns of the BrGRFs in various tissues were investigated by realtime quantitative PCR (RT-qPCR). The results indicated that all 17 BrGRFs had higher expression levels in young leaves compared with in old leaves, except for the BrGRF3 and BrGRF14, which were undetectable in all the tissues examined ( Figure 5A). In addition, BrGRF1, 4, 5, 6, 8, and 13 had higher expression levels in buds than in blooming flowers, whereas BrGRF2, 7, 9, 11, 12, 15, and 16 had higher expression levels in blooming flowers than in buds. The expression levels of BrGRF16 were highest in root tissues. The BrGRFs that were expressed mainly in certain organs or tissues might play important roles in the growth and development of these organs or tissues. The response to GA3 treatment was also investigated. Compared with distilled water (DW) treatment, the transcript levels of BrGRF5, 8,9,11,12,13,15,16, and 17 were increased more than 5-fold and BrGRF2, 4, and 7 were increased 2-to 5-fold in response to 100 μM GA3 application ( Figure 6A). The expression levels of the other BrGRFs were only slightly affected (less than 2-fold) or not affected by the GA3 treatment.
GRF and GIF proteins may act as transcription activators or coactivators, respectively, as part of a complex involved in the growth and development in plants [4,22]. Therefore, the expression patterns of BrGIF genes, including Bra010002, Bra020616, Bra036131, Bra032623 and Bra033281 were also investigated by RT-qPCR.
The results indicated that the Bra010002, Bra020616, and Bra036131 genes had the highest expression levels in buds, while Bra032623 and Bra033281 had the highest expression levels in young leaves ( Figure 5B). Additionally, the transcript expression levels of BrGIFs could be increased by GA3 treatment. At 3 h after GA3 treatment, the transcript levels were 11.9-, 199.6-, 4.3-, 2.3-, and 2.8-fold higher for Bra010002, Bra020616, Bra032623, Bra033281 and Bra036131, respectively, compared with their levels after DW treatment ( Figure 6B).  Table 4). The increase in leaf area in the BrGRF8 overexpressers was mediated directly by an increase in cell number because the adaxial epidermal surface cell sizes were normal (Figure 8).

Ectopic expression of BrGRF8 in Arabidopsis increases organ size
The data were downloaded from the Brassica Database (http://brassicadb.org/brad/).

Discussion
Previous studies have shown that GRF genes have positive functions in regulating organ size by promotion and/or maintenance of cell proliferation activity [2,7,8]. However, it is not known whether or how GRF proteins in Chinese cabbage regulate organ size. Because GRF proteins in different species may have different regulatory roles during the process of plant development, it is necessary to study the GRF genes in each species to understand the mechanisms of plant organ size control. In this study, a comprehensive set of 17 non-redundant GRF proteins were identified and characterized. Although the B. rapa genome is approximately 4-fold larger than the Arabidopsis genome (485 Mb and125 Mb, respectively), the gene number in B. rapa is only twice that of Arabidopsis (17:9), suggesting that there was extensive gene loss during genome duplication [20,23]. Pseudogenes are non-functional copies of gene fragments incorporated into the genome either by retrotransposition of mRNA or by duplication of genomic DNA. Pseudogenes are widely distributed in eukaryotic genomes. Sequence alignment showed that the BrGRF3 and BrGRF8 gene sequences were highly similar, and that there was a deletion of two A residues in the BrGRF3 CDS compared with the BrGRF8 CDS that lead to the introduction of a premature stop codon. Because, BrGRF3 gene expression was not detected in the various tissues, even after GA3 treatment, we suggest that BrGRF3 may be a pseudogene in the Chinese cabbage genome.
Phylogenetic analysis of the predicted GRF protein sequences from Chinese cabbage, Arabidopsis, and rice indicated that the BrGRFs shared higher similarity with AtGRFs than with OsGRFs, which is consistent with their evolutionary relationships: Chinese cabbage and Arabidopsis are both dicots of the Brassicaceae family, while rice is a monocot of the Poaceae family. Additionally, the phylogenetic and gene duplication analyses revealed that the BrGRFs contained three triplets, two duplicates, and four singletons. However, none of the BrGRFs exhibited tandem duplication. These results suggest that the expansion of the BrGRFs could be explained by the ancestor genome of B. rapa experiencing a whole-genome triplication and evolving finally into Chinese cabbage [20,23].
The BrGRFs were expressed mainly in specific organs and tissues, suggesting that they may play important roles in the growth and development of these organs or tissues. We found that the transcript levels of the BrGRF1, 4, 6, 8, 10, 15, and 17 genes were higher in young leaves than in the other tested tissues, suggesting that these genes are involved mainly in the growth and development of young leaves in Chinese cabbage. Thus, improving the expression levels of these genes might help improve Chinese cabbage leaf head yield.
GA3 is involved in various physiological processes in plants and KNOX proteins contribute to the regulation of meristem maintenance by negatively regulating the production of gibberellins [11][12][13][14][15][16]. Therefore, downregulation of KNOX gene expression at the flanks of the SAM cycle was reported to increase the level of GA, resulting in organized cell proliferation and determination of cell fate [13]. A previous study showed that GRF proteins act as repressors and down-regulators of KNOX gene expression [10]. As reported previously in rice [4], the transcription of most BrGRFs was induced by GA3 treatment. These results suggested that the GRF genes may function in maintaining or promoting cell proliferation in plants by a feedback regulation mechanism, in which the GRF genes positively regulate the production of gibberellins, and GAs in turn upregulate GRF gene expression.
In previous studies, the over-expression of GRF genes in Arabidopsis was found to result in bigger organ size than the vector control [3,7], while reduction of AtGRF gene expression by the overexpression of the microRNA miR396 in transgenic Arabidopsis caused narrow-leaf phenotypes due to a reduction in cell number [24][25][26]. Here, we found that the ectopic expression of BrGRF8 also increased leaf size in Arabidopsis. In addition, our histological results revealed a significant increase in cell number but not in cell size in the 35S:BrGRF8 transgenic Arabidopsis plants compared with the vector control plants, suggesting that BrGRF8 may control the growth of plant organs by regulating cell proliferation rather than by enlarging cell volume.

Conclusions
We identified 17 members of the Chinese cabbage GRF gene family that encoded putative GRF proteins that fell into six subfamilies. The phylogenetic relationships among Chinese cabbage, rice, and Arabidopsis GRF genes, suggested that the BrGRFs were more closely allied with AtGRFs than with OsGRFs. Further, phylogenetic and duplication event analysis suggested that whole genome duplication may have been the main contributor to the expansion of the BrGRFs. Additionally, the ectopic expression of BrGRF8 in Arabidopsis positively controlled organ size by regulating cell proliferation. The expression profiles obtained by RT-qPCR showed that the BrGRFs may be involved in immature organ or tissue growth and development via the GA pathway. Together, these data will not only contribute to a further understanding of the characteristics and functions of the GRF family in different species, but will also provide a promising strategy for Chinese cabbage breeding programs to improve yield/head size.

Identification and analysis of GRF genes in Chinese cabbage
The nucleotide and protein sequences of BrGRFs were identified based on the B. rapa line Chiifu genome sequence (http://brassicadb.org) [20]. The GRF nucleotide sequences were aligned using DNAMAN 6.0.40 (Lynnon Biosoft, USA). Intron/exon structure analysis was performed using the Gene Structure display Server (GSDS) (http:// gsds.cbi.pku.edu.cn/). Phylogenetic trees were constructed with the MEGA 4.0 software using the neighbor-joining method and a bootstrap test that was replicated 1000 times [27]. The GC content was calculated by DNASTAR (Madison, WI, USA). The number of amino acids, molecular weight (MW), and theoretical isoelectric point (pI) were computed using the ProtParam tool (http://web. expasy.org/protparam/). The SSR markers were detected using the SSRIT software (http://archive.gramene.org/ db/markers/ssrtool) with the parameters adjusted for identification of perfect di-, tri-, tetra-, penta-, and hexanucleotide motifs with a minimum of 6, 5, 5, 4 and 4 repeats, respectively. The GO analysis of BrGRF proteins was carried out using the GO term analysis tool in Gramene (http://www.geneontology.org/) developed by the GO consortium Conserved motifs in the full-length amino acid sequences of GRF proteins from Chinese cabbage, Arabidopsis, and rice were identified using MEME [28]. The Arabidopsis and rice GRF protein sequences were  For the GA3 treatment, uniformly sized seedlings were selected when they had developed four fully opened leaves, and then treated with 100 μM GA3 or distilled water (DW). The leaves of the seedlings were harvested after 0, 1, and 3 h of GA3 treatment. For analysis of GRF genes expression in different tissues, roots (R), stems (S), expanded rossete leaves (ERL), young folding leaves (YFL) beginning to fold at the early folding stage (about 24-25 leaves), buds (B), blooming flowers (BF), and immature siliques (IS) 15 d after fertilization were collected. All materials were frozen immediately in liquid nitrogen, and stored at −80°C until RNA isolation.

Real-time quantitative PCR
Total RNA was extracted from each sample using Trizol reagent (Invitrogen, Carlsbad, CA, USA) and treated with RNase-free DNase I (Takara, Dalian, China) for 45 min according to the manufacturer's protocol. Firststrand cDNA was synthesized from 1 μg of total RNA using a PrimeScript 1 st Strand cDNA Synthesis Kit (Takara). RT-qPCR was carried out using a SYBR Green Master mix (Takara) on an IQ5 Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA). The gene-specific primers designed for the BrGRF and BrGIF genes are listed in Additional file 2. The actin gene was used as a constitutive expression control in the RT-qPCR experiments. The PCR cycling conditions comprised an initial polymerase activation step of 95°C for 1 min, followed by 40 cycles of 95°C for 10 s and 60°C for 30 s. After each PCR run, a dissociation curve was designed to confirm the specificity of the product and to avoid the production of primer dimers. The relative amounts of the amplification products were calculated by the comparative 2 −ΔΔCт method [29].