Genome-wide expression analysis of soybean NF-Y genes reveals potential function in development and drought response
Nuclear factor-Y (NF-Y), a heterotrimeric transcription factor, is composed of NF-YA, NF-YB and NF-YC proteins. In plants, there are usually more than 10 genes for each family and their members have been identified to be key regulators in many developmental and physiological processes controlling gametogenesis, embryogenesis, nodule development, seed development, abscisic acid (ABA) signaling, flowering time, primary root elongation, blue light responses, endoplasmic reticulum (ER) stress response and drought tolerance. Taking the advantages of the recent soybean genome draft and information on functional characterizations of nuclear factor Y (NF-Y) transcription factor family in plants, we identified 21 GmNF-YA, 32 GmNF-YB, and 15 GmNF-YC genes in the soybean (Glycine max) genome. Phylogenetic analyses show that soybean’s proteins share strong homology to Arabidopsis and many of them are closely related to functionally characterized NF-Y in plants. Expression analysis in various tissues of flower, leaf, root, seeds of different developmental stages, root hairs under rhizobium inoculation, and drought-treated roots and leaves revealed that certain groups of soybean NF-Y are likely involved in specific developmental and stress responses. This study provides extensive evaluation of the soybean NF-Y family and is particularly useful for further functional characterization of GmNF-Y proteins in seed development, nodulation and drought adaptation of soybean.
KeywordsNF-Y Nodulation Transcription factor Soybean Water stress
Days after flowering
Histone fold motif
Hidden Markov Model
Nuclear factor Y
Quantitative real-time PCR
Reads per kilobase per million
Nuclear factor Y (NF-Y) is a transcription factor (TF) complex that binds the CCAAT element to regulate gene expression. There are three distinct subunits: NF-YA, NF-YB and NF-YC, which contains evolutionary conserved domains for DNA binding and subunit interaction to form the heterotrimeric NF-Y complex. Subunit assembly is well studied in animals and is conserved in plants (Romier et al. 2003; Calvenzani et al. 2012; Hackenberg et al. 2012). Initially, NF-YB and NF-YC form a dimer complex in cytoplasm and then recruit the third subunit, NF-YA, to form a mature heterotrimemeric NF-Y complex TF (Frontini et al. 2004; Kahle et al. 2005). The binding of NF-Y can either activate or repress transcription of genes (Ceribelli et al. 2008). Sequence analysis of mammalian genome showed 25–30 % of the genes having NF-Y binding sites in their promoters (Mantovani 1998). The CCAAT box is typically found in both forward and reverse orientations between 60 and 100 bp upstream of transcription start site in mammalian genes and may be present in multiple copies with variable distances (Mantovani 1998). In a survey of a promoter sequence population, the CCAAT box element appeared highly associated with TATA-less promoters (Mantovani 1998).
In contrast to animals and yeast where there is a single gene encoding each subunit (Mantovani 1999), there are about 10 genes encoding each subunit in plants (Stephenson et al. 2007; Thirumurugan et al. 2008; Siefers et al. 2009; Liang et al. 2013) with temporal, spatial, universal and organ-specific expression patterns associating with various developmental and environmental response regulations. The roles of certain individuals and NF-Y complexes have been reported in regulation of embryogenesis, cellular signaling, flowering time control, nodulation and nitrogen nutrition and stress tolerance. Arabidopsis NF-YA genes NF-YA1, 3, 5, 6, 8, and 9 have been known to regulate gametogenesis, embryogenesis, seed morphology, and seed germination (Fornari et al. 2013; Mu et al. 2013). Specific members of NF-YB, particularly the LEC 1 group, have been reported to regulate embryogenesis and seed development (West et al. 1994; Parcy et al. 1997; Hawkins and Nakamura 1999; Kwong et al. 2003). LEC1 functions as an integrator of various regulatory events, involving light and hormone signaling, specifically during somatic and early zygotic embryogenesis (Petroni et al. 2012; Laloum et al. 2013) and fatty acid biosynthesis (Mu et al. 2008; Shen et al. 2010; Tan et al. 2011; Mendes et al. 2013). The third subunit, NF-YC, is also important in many developmental regulations. Arabidopsis NF-YC3, NF-YC4 and NF-YC9 physically interact in vivo with both NF-YB2 and NF-YB3 and are required to regulate flowering time.
NF-Y proteins have been reported as a key factor regulating nodulation in the nitrogen-fixing plants and nitrogen assimilation in various plant species. The involvement in nodulation and nitrogen assimilation was identified by the early works on model legume Medicago truncatula for MtNF-YA1 (MtHAP2-1) which controls nodule meristem function (Combier et al. 2006). MtNF-YA1 controls infection thread progression from initial root infection through colonization of nodule tissues (Laporte et al. 2013). Later works on Lotus japonicus identified two NF-Y subunit genes, LjNF-YA1 and LjNF-YB1, which are targeted by nodule inception (NIN) and acted downstream of NIN in the nodule development regulation (Soyano et al. 2013). These two genes are expressed in the root nodule primordia and their protein product can form a NF-Y complex in plant cells. Knockdown of LjNF-YA1 inhibited root nodule organogenesis and ectopic expression of the LjNF-YA1 and LjNF-YB1 genes caused abnormal cell division during lateral root development, indicating that the Lotus NF-Y subunits can function to induce cortical cell division to initiate root nodule organogenesis. Recently, the C subunit from common bean (Phaseolus vulgaris) NF-YC1 was reported to play a key role in the improved nodulation seen by more efficient strains of rhizobia. RNAi-induced reduction of NF-YC1 in hairy roots resulted in the arrest of nodule development and defects in the infection process and the mechanisms appears to involve suppression of cortical cell divisions (Zanetti et al. 2010). Whether PvNF-YC1 functions independently or in combination with other subunits of NF-Y complex, however, has not been known.
In addition to the roles in plant growth and development, NF-Y proteins are important regulators of stress tolerance, especially in response to drought stress (Petroni et al. 2012; Laloum et al. 2013). Overexpression of AtNF-YA5, a strongly drought-induced gene with strong expression in the guard cells, improves drought tolerance and reduces water loss in the leaves (Li et al. 2008). In contrast to the overexpression of AtNF-YA5, the nfya5 knockout plants and plants overexpressing miR169a showed enhanced leaf water loss and were more sensitive to drought stress than wild-type plants. In soybean, GmNF-YA3 is induced by various stress treatments. The Arabidopsis transgenic plants overexpressing GmNF-YA3 show reduced leaf water loss and enhanced drought tolerance through regulating the common drought-responsive genes (Ni et al. 2013). In rice, the transgenic plants overexpressing HAP2E (NF-YA), a gene responsive to wounding, are tolerant to the pathogen Xanthomonas oryzae pv. oryzae, drought and salinity stresses, and improve photosynthesis and tiller numbers (Alam et al. 2014). Drought tolerance regulation is also reported in NF-YB subgroup. Overexpression of the drought-inducible NF-YB results in improved performance under drought conditions in Arabidopsis, maize and poplar (Nelson et al. 2007; Han et al. 2013). The transgenic plants have significant improvement in number of stress-related parameters, including chlorophyll content, stomatal conductance, leaf temperature, reduced wilting, maintenance of photosynthesis and the seed yield under water-limited conditions (Nelson et al. 2007). Further characterizations of drought-inducible NF-Y genes will facilitate genetic engineering for improvement of drought tolerance in crop plants.
Soybean is one of the most important crops for its nutrient sources for human and animal consumption, and potential biofuel and economic values. Extensive research has focused on oil seed development (Severin et al. 2010) and plant and symbiotic microbe interaction (Libault et al. 2010), making soybean an ideal plant species for both basic and applied research. Taking the advantage of broad understanding of the recent functional genomics of plant NF-Y genes and the current draft of soybean genome (Schmutz et al. 2010), we attempted to characterize the soybean NF-Y gene family. Together with analysis of NF-Y family in growth and development, we included an analysis of the family members in response to drought and Bradyrhizobium japonicum treatments which are particularly important to identify candidate genes involved in drought tolerance and nodulation and nitrogen fixation in soybean. Our study of transcript expression of soybean NF-Y gene families, therefore, would provide an initial step in understanding gene function that may facilitate functional characterization and genetic engineering to increase the efficiency of soybean in response to drought and B. japonicum. In this study, we identified and investigated 68 NF-Y (21 GmNF-YA, 32 GmNF-YB, 15 GmNF-YC) and 13 NC2 genes for their expression patterns in soybean tissues and in response to drought stress. Expression analysis showed that there are strong associations between gene expression patterns and homology-based predicted functions, allowing us to assign putative functions to the individual gene groups in regulation of plant growth, development and adaptation to drought stress.
Materials and methods
In silico identification of soybean NF-Y family members
All soybean protein sequences were downloaded from the Phytozome biomart, gene model V9.1 (http://www.phytozome.net/search.php) using PFAM ID PF02045 (CBFB/NFYA) and PF00808 (CBFD_NFYB_HMF) and KOG ID KOG0869, KOG0871, KOG1561, KOG1657 and KOG1659, for CCAAT-binding factor subunit A (HAP2), CCAAT-binding factor subunit B (HAP3), CCAAT-binding factor subunit C (HAP5), Class 2 transcription repressor NC2β subunit (Dr1) and Class 2 transcription repressor NC2α subunit (DRAP1), as queries. We used the updated classification (Petroni et al. 2012) of plant NF-Y, NC2 (DR1 and DRAP1) and Dpb3/Dpb4 from Arabidopsis, rice and wheat (both full-length and domain sequences) to classify the retrieved soybean proteins. This updated functional classification has corrected some of the previously identified genes in Arabidopsis, wheat and rice (Stephenson et al. 2007; Thirumurugan et al. 2008; Siefers et al. 2009) by re-grouping some of the previously annotated NF-YB and NF-YC genes to NC2β, NC2α and Dpb3/Dpb4 (Petroni et al. 2012). Local BLAST was performed with a threshold of E-10 to classify individuals into functional groups.
Retrieved sequences were aligned using CrustalW with a gap open cost of 10.0 and gap extension cost of 1.0. Phylogenetic tree was constructed by MEGA5 software (Tamura et al. 2011) using neighbor-joining method (Poisson model with uniform rates among sites) and the relationship was estimated using a bootstrap analysis with 1,000 replicates. MultiExperiment Viewer TM4 software (Aryee et al. 2009) was used to generate heat map and hierarchical clustering analysis results.
Plant materials and drought treatment
Soybean cultivar Williams 82 was grown in greenhouse (28/20 °C day/night temperature, 14-h day-length photoperiod, 800 µmol m−2 s−1 light intensity and 60 % humidity) with a density of four plants per 1-gallon pot, containing a mixture of 1 sand:1 turface. The plants were watered every 2 days. Drought treatments started when the plants reached the first leaf stage (V1—first fully open trifoliolate leaf stage, day 14th) by withholding watering and the plants were harvested when the desired stem water potential was reached. There were three drought treatments with stem water potentials of −0.5 (day 18th), −1.0 (day 22th) and −1.5 MPa (day 34th) for drought and the corresponding control treatments had a stem water potential of ~−0. 2 MPa. To measure stem water potential, the stems were cut at the hypocotyl segment and immediately measured for water potential using a pressure chamber (PMS Instrument Co. Albany, CA, USA) at around 4am. Predawn tissue collection was selected because it was considered equivalent to whole soil water potentials. Three biological replicate experiments were conducted following a complete randomized design. Each replicate consisted of samples pooled from three plants.
RNA isolation, cDNA synthesis and mRNA expression analyses for drought responses
Each frozen tissue collected from the three-biological repeat drought experiment was separated into two technical repeats and grinded in liquid nitrogen. RNA was extracted using Trizol (Invitrogen Inc.). RNA integrity was verified by electrophoresis using 1 % agarose in buffer TAE. The RNA was purified from DNA contamination using the Turbo DNA-free DNaseI kit (Ambion, Austin, TX, USA). One microgram of total RNA and the primer mix of oligodT and random hexamer were used to synthesize cDNA using iScript cDNA Synthesis Kit (Bio-Rad, Hercules, CA, USA) in a reaction volume of 20 µL. Quantitative real-time PCR (qRT-PCR) was performed to quantify mRNA expression under drought conditions. Initially, PRIMEGENS (Srivastava and Xu 2007) was used to design primers for soybean NF-Y gene family and all the obtained sequences were manually verified for specific alignment by Blast against the soybean genome database housed in Phytozome to have the final primer list (Supplementary Table S2). Primer pairs with acceptable PCR efficiency and the unique melting disassociation curve were used for qRT-PCR. All reference genes (Supplementary Table S3) were selected based on recommendations from current literatures (Jian et al. 2008; Libault et al. 2008; Hu et al. 2009; Le et al. 2012). qRT-PCR results were analyzed using a common PCR efficiency for each primer obtained from linregPCR analysis software (Ramakers et al. 2003). The obtained efficiencies were consistently higher than 1.8 with average of about 1.95 for whole primer sets. GeNorm (Vandesompele et al. 2002) was used to test the stability (M index) to determine a set of reference genes to be used to normalize individual expression data. In our experiment, we calculated relative expression by normalizing data to the four best reference genes: IDE, UNK1, UNK2 and ACT (Supplementary Figure S1). Fold change of gene expression was calculated only when there was a significant difference in means of relative gene expression between the drought and control samples from a Student T test.
Tissue-specific expression and responses to B. japonicum treatments
Data were retrieved from previously reported datasets (Libault et al. 2010; Severin et al. 2010) and detailed experimental procedures were described therein and briefly summarized below. Tissue-specific expression in 14 tissues was performed on soybean genotype P-C609-45-2-2 (Severin et al. 2010). The soybean plants were growing in pots containing B. japonicum-inoculated soil and full-nutrient fertilizer (Osmocote 14-14-14) in growth chambers under photoperiod of 14/10 h, thermocycle of 22/10 °C, relative humidity of 50–60 %, and light intensity of 550–740 μE m−2 s−1. Samples were pooled from a minimum of three plants. Young leaf tissue samples were collected at flowering stage and pods and seeds were harvested by seed weight and pod lengths. The 1-cm pod was approximately harvested at 7 days after flowering (DAF), and the 4- and 5-cm pods were harvested at 10–13 DAF and 14–17 DAF, respectively, and were divided into seed and pod-shell components. Seed 21-, 25-, 28- and 35-DAF had seed weights between 10 and 25 mg, 25 and 50 mg, 50 and 100 mg, 100 and 200 mg, and greater than 200 mg, respectively. Root and nodule tissues were from plants grown in growth chambers under 16-h photoperiods and light intensities between 310 and 380 μE m−2 s−1. Root tissue was harvested after 12 days and nodules were harvested at 20–25 days after bacterial inoculation. For analysis of gene expression in root hairs in response to B. japonicum treatments, normalized data were retrieved from a previous report (Libault et al. 2010) and the experimental procedures are described briefly below. At least three independent biological replicates were produced to ensure the reproducibility of the plant tissues analyzed. Soybean (cv. Williams 82) seeds were germinated on nitrogen-free B&D agar medium in growth chambers (dark conditions, 80 % humidity, 27 °C). Three days after sowing, a B. japonicum strain USDA 110 cell suspension or water (mock inoculation) was sprayed on the seedlings and the treated seedlings were returned to the original growth chamber for incubation. Root hair cells were isolated after 12, 24 and 48 h of bacterial inoculation. For each condition, similar quantities of total RNA isolated from three independent biological replicates were pooled together to synthesize cDNA which were then ligated to Solexa adaptors. Complementary DNAs were sequenced using Solexa platform and data were analyzed using Illumina Genome Analyzer II, and the data were analyzed and normalized to reads per kilobase per million (RPKM).
Identification of NF-Y genes from soybean
NF-Y and NC2/Dpb genes in soybean
For verification purposes, soybean NF-Y protein sequences were downloaded from three current databases of soybean transcription factors. The first database, SoyDB (http://soykb.org/), provides the predicted transcription factors from InterProScan searches against 11 databases integrated in InterPro to find transcription factor domains (Wang et al. 2010). The second soybean transcription database, SoybeanTFDB (http://soybeantfdb.psc.riken.jp/), used Hidden Markov Model (HMM) of PFAM to search against the modeled proteome data of annotated genes in Glyma1 from Phytozome database with a threshold of E < 1e-5 and a verification from domain searches using InterProScan and Blastp for putative homologous genes (Mochida et al. 2010). The third transcription factor database, PlantTFDB (http://planttfdb.cbi.pku.edu.cn/), used 48 HMM-Pfam profiles and 16 additional HMM constructed from GenPept and Swiss-Prot protein databases to construct 64 TF families in Arabidopsis and other plant species (Guo et al. 2008). Alignment analysis of these obtained sequences resulted in a number of genes incorrectly classified to NF-YB and NF-YC subfamilies (data not shown) mainly due to the fact that the classification did not have a separate description and classification for NF-YB, NF-YC, NC2 or Dpb group, but instead, have a common description for these two groups with a name NF-YB/HMF domain. Given the critical differences among NC2, Dpb and NF-Y groups, it is necessary to annotate them accurately for the updated research.
Tissue differential expression and drought and B. japonicum inducibility of NF-Y genes
Domain structure and putative function of soybean NF-Y subunits
The residues required for the subunit interaction and DNA binding (Xing et al. 1993; Sinha et al. 1996) are highly conserved among GmNF-YB proteins except for relatively few alterations by similar functional residues (Fig. 5). Arginine (R55) and aspartate (D65) are suggested to have a crucial role in protein interaction due to possession of an important intramolecular hydrogen bond within the HFM (Romier et al. 2003; Hackenberg et al. 2012). These arginine–aspartate pairs are highly conserved through evolution in almost plants and animal species and are important in subunit interactions between NF-YB and NF-YC (Romier et al. 2003; Hackenberg et al. 2012). They are present in all soybean NF-YB proteins except GmNF-YB16 and GmNF-YB22 which have deletions in this domain and might not be able to form the NF-Y complex. Given the number of the GmNF-YB is almost three times while the numbers of GmNF-YA and GmNF-YC are about two times of corresponding genes in Arabidopsis, it is likely that the GmNF-YB proteins have gone through largest functional alteration.
Analysis of domain sequence of Arabidopsis NF-YB proteins reveals that the translocation of plant NF-Y complex into nucleus follows the piggyback transport model (Hackenberg et al. 2012) in which NF-YB and NF-YC heterodimer is formed first by head-to-tail annealing of their HFMs (Thon et al. 2010). A disulfide bond between two cysteines of NF-YB (C33 and C37) controls the heterodimerization of two subunits and further controls the spatial distribution of subunits and subcellular localization of the NF-Y complex upon a redox state. In soybean, all NF-YBs apparently lack the critical cysteine at position 33 and cannot form the disulfide bond (Fig. 5) which resembles the model of nuclear translocation of NF-Y complex in Arabidopsis (Hackenberg et al. 2012). Therefore, the redox-regulated assembly of NF-YB and NF-YC of human and Aspergillus nidulans is not present in the soybean NF-Y.
Among three NF-Y subfamilies, only members of plant NF-YB have been identified to involve photosynthesis so far. Rice OsHAP3 genes regulate the expression of nuclear-encoded chloroplast-targeted genes and normal development of chloroplasts (Miyoshi et al. 2003). Down-regulation of OsHAP3A, OsHAP3B and OsHAP3C resulted in reduced chlorophyll and degenerated chloroplasts which failed to develop lamella and accumulate starch. These plants showed reduced expression of nuclear-encoded photosynthesis genes. In accordance with this, a wheat TaNF-YB3 was identified to co-express with photosynthetic genes and the transgenic plants showed increased leaf chlorophyll content and photosynthesis (Stephenson et al. 2011). Although close orthologs can be found in soybean, these genes do not belong to a certain phylogenetic tree branch (Fig. 6) making it unreliable to predict soybean NF-YB genes functioning in photosynthesis.
Our analysis also shows that the NC2α and Dpb groups share some homology to NF-YC proteins and NC2β shares homology to NF-YB (Petroni et al. 2012). Although there is sequence similarity, NF-YC cannot pair with NC2β and NF-YB cannot pair with NC2α (Zemzoumi et al. 1999). This binding preference might be explained by the difference in sequence structure, specifically the W44 of NF-YC at the end of helix alpha2 (Fig. 7), which is not present in NC2α. Additional to this difference, the hydrophobic cores organizing L1 and L2 regions are also different from that of NC2 (Romier et al. 2003). Although it is sharing some homology, all NC2 and Dpb3/Dpb4 genes are classified in a separate clade from phylogenetic tree analysis (Figs. 6, 8). In addition, NC2 and Dpb3/Dpb4 appear to have different functions such as NC2 associates with TATA box-binding protein to bind TATA box in the core promoters and the Dpb4/Dpb4 complex with DNA polymerase ε and the chromatin-remodeling complex CHRAC (Kamada et al. 2001; Hartlepp et al. 2005).
In the present report, we attempted to incorporate all the current annotations to explore the soybean NF-Y genes. Based on phylogenetics and up to date annotations, we identify 21 NF-YA, 32 NF-YB and 15 NF-YC and assign putative roles in plant growth and development, rhizobium-soybean symbiosis and drought adaptation for each subgroups. The updated clarification between NC2, Dpb3/4 and NF-Y (Petroni et al. 2012) allows accurate classification of NF-Y genes and updates databases and literature of previously characterized ones in plants including A. thaliana, O. sativa, T. aestivum, Brachypodium distachyon, Selaginella moellendorffii and Brassica napus (Stephenson et al. 2007; Thirumurugan et al. 2008; Siefers et al. 2009; Cao et al. 2011; Liang et al. 2013; Saha et al. 2013). Although there is high primary sequence similarity among these gene families, NC2 and Dpb3/4 groups are the outliners in the reported analyses from wheat, Arabidopsis and B. distachyon (Stephenson et al. 2007; Siefers et al. 2009; Cao et al. 2011). There is also a distinct functionality of NC2 groups: NC2 complex works as a global transcriptional repressor that targets TBP, thereby blocking formation of the transcription preinitiation complex (Kim et al. 1997). To this end, it is also necessary to update latest annotations for proper classification for the soybean transcription databases that are currently active (Guo et al. 2008; Mochida et al. 2010; Wang et al. 2010). With the recent reports describing the involvement of NF-Y in nodule and seed development, and nitrogen-fixing bacterium interactions (Combier et al. 2006; Libault et al. 2010; Severin et al. 2010; Soyano et al. 2013; Battaglia et al. 2014; Laloum et al. 2014; Ripodas et al. 2014; Singh et al. 2014) and our current transcriptional analysis under drought treatments, our analyses suggest possible roles of soybean NF-Y genes in seed and nodule development and drought responses.
Plant fertility, embryogenesis and seed development
The involvement of NF-Y genes in developmental processes, particularly seed development and embryogenesis, has been identified in all three subfamilies (Petroni et al. 2012; Laloum et al. 2013). As expected, our transcript analysis identified several groups of soybean NF-YA, NF-YB and NF-YC which have highly specific or differential expression in soybean seeds and close relationship to the known genes. GmNF-FA subfamily appears to have more genes (8 out of total 16 GmNF-YA genes with available expression data) highly and differentially expressed in the seeds (Fig. 1). Their Arabidopsis orthologs, AtNF-YA1, AtNF-YA3, AtNF-YA5, AtNF-YA6, AtNF-YA8 and AtNF-YA9, have conserved and redundant function in male gametogenesis, embryogenesis and seed development (Fornari et al. 2013; Mu et al. 2013). Because the majority of the NF-YA genes specifically or differentially expressed in seeds are also induced by drought treatment (Fig. 1), they might be involved in the dehydration-response maturation process during seed development. Similar to the soybean seed development-related GmNF-YA, the soybean NF-YB (GmNF-YB03, GmNF-YB10, GmNF-YB26 and GmNF-YB30, Figs. 5, 6) are highly similar to the Arabidopsis AtLEC1 and L1L (Meinke 1992; West et al. 1994; Parcy et al. 1997; Lotan et al. 1998; Kwong et al. 2003; Junker et al. 2012). LEC1 and L1L are particularly different from other NF-YB proteins and appear to possess unique residues that are important to regulate hormonal and light signaling during embryogenesis (Lee et al. 2003; Junker et al. 2012). Indeed, their orthologs in other plant species showed the highly conserved function in regulation of the transition from embryo to adult status (Zhang et al. 2002; Yazawa et al. 2004; Alemanno et al. 2008; Mu et al. 2008; Schellenbaum et al. 2008; Shen et al. 2010; Cao et al. 2011; Tan et al. 2011). In addition, overexpression of LEC1, L1L and their orthologs resulted in seed oils via activation of various genes regulating fatty acid biosynthesis (Mu et al. 2008; Shen et al. 2010; Tan et al. 2011). Seed-specific expression is also seen in soybean GmNF-YC group for NF-YC05, NF-YC09, NF-YC10, NF-YC11. These genes share sequence similarity with the Arabidopsis AtNF-YC12, a flower differentially expressed gene (Siefers et al. 2009). During seed developmental stages, AtNF-YC12 is highly expressed in chalazal endosperm and chalazal seed coat of green mature seed (http://www.seedgenenetwork.net/). AtNF-YC12 interacts bilaterally with LEC1 and L1L in yeast—two hybrid assays (Hackenberg et al. 2012) indicating that they co-regulate seed development and embryogenesis. Therefore, it would be interesting to investigate the functions of soybean gene combinations in the control of embryogenesis and seed oil production in soybean.
Increasing evidences have showed that NF-Y genes are key regulators of flowering time control in plants. Arabidopsis AtNF-YB2 and AtNF-YB3 promote flowering in response to inductive long-day condition (Kumimoto et al. 2008) through regulation of expression of FLOWERING LOCUS T (FT), a key gene that controls vegetative to floral meristem conversion (Mathieu et al. 2007). AtNF-YB2 and AtNF-YB3 interact with three AtNF-YC (NF-YC3, NF-YC4 and NF-YC9), which physically interact with CONSTANS (CO), and are required for CO-mediated floral promotion to induce FT expression in the phloem tissue of long-day young leaves (Kumimoto et al. 2010). CO protein shares some homology to the Arabidopsis NF-YA subunits. Since overexpression of NF-YA1 and NF-YA4 caused delayed flowering, which may due to the impaired formation of NF-YB/NF-YC/CO (Wenkel et al. 2006). It is noted that the genes required for flowering time control often expressed in leaf vascular where CO genes are expressed (Kumimoto et al. 2010; Cao et al. 2011). Among the soybean NF-Y genes that are differentially expressed in young leaf, GmNF-YB02, GmNF-YB13, GmNF-YC04, GmNF-YC06 and GmNF-YC14 are particularly of interest for their role in regulation of flowering time in soybean due to close sequence relationship with Arabidopsis and rice flowering time control genes AtNF-YB2, AtNF-YB3, OsHAP3H, AtNF-YC3 and AtNF-YC9 (Miyoshi et al. 2003; Kumimoto et al. 2008, 2010).
Nitrogen nutrition and nodulation
Nodulation is the result of a mutualistic interaction between the plant and symbiotic soil bacteria rhizobia, initiated by the infection of plant root hair cells by the symbiont. Initially, flavonoids secreted by the plant lead to the synthesis and secretion of a bacterial Nod factor (NF), which composes of chitooligosaccharide. Upon perception of NF, the symbiotic system activates various factors, of which, the nodule inception (NIN) and its downstream gene early nodulin (ERN) transcription factors are required for initiating nodulation-specific symbiotic processes (Madsen et al. 2010). In L. japonicus, NIN activates of the NF-Y subunit genes, LjNF-YA1 and LjNF-YB1 to induce cortical cell division, which is an initial step in root nodule organogenesis (Soyano et al. 2013). Similarly, in M. truncatula, the MtNF-YA1 (MtHAP2-1) gene which shares homology to LjNF-YA1 is important in nodule development and plant–bacterial symbiotic interaction (Combier et al. 2006; Laporte et al. 2013; Laloum et al. 2014). It is highly expressed in mature nodules, first induced at the onset of symbiotic development during preparation for, and initiation and progression of, symbiotic infection. MtNF-YA1 controls rhizobial infection progression by regulating the formation and the wall of infection thread. In M. truncatula, MtNF-YA1 and MtNF-YA2 have a redundant function in nodulation development via ERN1 activation (Laloum et al. 2014). Our analysis in soybean shows that the M. truncatula MtNF-YA1 and L. japonicus LjNF-YA1 share strong homology to several soybean NF-YA genes (group A: GmNF-YA01, GmNF-YA03, GmNF-YA10 and GmNF-YA21). These genes are expressed specifically in nodules and induced by B. japonicum (Fig. 4), indicating a function in the symbiotic interaction and nodule development. We also found that genes in group F of GmNF-YB subfamily are specifically expressed in nodule and share a high sequence homology to L. japonicus LjNF-YB1 (Fig. 6), which regulates nodule development in interaction with LjNF-YA1, downstream of NIN (Soyano et al. 2013). Within this group, GmNY-YB15 is of particular interest for its nodule differential and rhizobium induced, suggesting a role in initial root nodule organogenesis. Members of subfamily NF-YC, on the other hand, are least known for the involvement in the nodulation development process. Among 15 soybean NF-YC genes, only GmNF-YC08 is induced by B. japonicum treatments (Supplementary Figure S2), suggesting an involvement in initial process of nodule development. Up to date, the Phaseolus vulgaris PvNF-YC1 is the first and the only NF-YC known to be required, and plays a key role in the nodule organogenesis and rhizobium infection (Zanetti et al. 2010; Mazziotta et al. 2013). Overexpression of PvNF-YC1 improves nodulation efficiency on the less-efficient rhizobium strains, possibly through activation of the genes in G2/M transition. NF-YC1 could interact physically with SIN1, a GRAS-type transcription factor family which is required for efficient nodule development.
Stress response and adaptation
Increasing evidences have shown that individual NF-Y subunits, particularly NF-YA and NF-YB, are important to regulate drought and ER stress tolerance in both dicots and monocots with various mechanisms including tolerance to reduced water loss and low tissue water potential (Nelson et al. 2007; Li et al. 2008; Liu and Howell 2010; Han et al. 2013; Ni et al. 2013). The present investigation identified a large number of soybean NF-YA and NF-YB genes responsive to drought treatments and possibly involved in drought adaptation (Figs. 1, 2). Soybean NF-YA family appears to have more genes induced by drought (7 of 21 genes). Most these genes belong to the groups that have differential expression in the seeds, suggesting that they are involved in maturation and dehydration signaling. Three soybean genes: GmNF-YA09, GmNF-YA17 and GmNF-YA19 have sequence homology to the drought tolerance GmNF-YA12 [or GmNF-YA3 (Ni et al. 2013)], AtNF-YA5 and AtNF-YA6 proteins (Li et al. 2008; Fornari et al. 2013; Mu et al. 2013). GmNF-YA12 was induced by abscisic acid (ABA) and abiotic stresses, such as NaCl and cold, and has a role in drought tolerance by reduced leaf water loss via regulation of the ABA signaling pathway. AtNF-YA5 is induced by ABA, NaCl, cold and PEG treatment, and control water loss in Arabidopsis (Li et al. 2008). AtNF-YA5 also has a redundant role with AtNF-YA1, AtNF-YA6 and AtNF-YA9 which are downstream of LEC1 and regulate of embryogenesis and seed development in Arabidopsis (Mu et al. 2013). Most of drought-inducible soybean genes (5/7 genes) in this group are predicted to be targets of microRNA169 (data not shown), which is known to cleave GmNF-YA12 and AtNF-YA5 in response to drought stress (Li et al. 2008; Ni et al. 2013), indicating a conserved function in seed development and drought-response regulation.
Our phylogenetic analysis shows that two clades of soybean homology to the drought-tolerant Arabidopsis AtNF-YB1 and maize ZmNF-YB2 (Nelson et al. 2007). AtNF-YB1 and ZmNF-YB2 function in drought tolerance by maintaining leaf water potential, stomatal conductance and photosynthesis (Nelson et al. 2007). Surprisingly, most of the genes sharing homology to AtNF-YB1 (group F) are repressed by drought treatment. The majority of these genes are in fact specific/differential to nodule. Therefore, it is unlikely that they are involved in drought tolerance in soybean. GmNF-YB01, the only homolog of drought-tolerant genes AtNF-YB1 and ZmNF-YB2, is expressed ubiquitously in various tissues and induced by drought (Fig. 6) suggesting that it is potentially involved in drought tolerance.
We found that the majority of genes that are highly specific to nodules are downregulated by drought stress. Soybean nodulation and nodule activity are strongly reduced under the stress (Sinclair et al. 2007) suggesting that NF-Y transcription factors are important in the regulation of nodule activity. Further characterization of these transcription factors will be useful in understanding of molecular regulation of nodulation and improvement of nitrogen assimilation during drought stress, as maintenance of nitrogen fixation is highly associated with maintenance of soybean productivity under the stress (Sinclair et al. 2007).
Funding supports from Missouri Soybean Merchandising Council (MSMC) to the soybean genomics drought research program at the University of Missouri are gratefully appreciated.
- Alam MM, Tanaka T, Nakamura H, Ichikawa H, Kobayashi K, Yaeno T, Yamaoka N, Shimomoto K, Takayama K, Nishina H, Nishiguchi M (2014) Overexpression of a rice heme activator protein gene (OsHAP2E) confers resistance to pathogens, salinity and drought, and increases photosynthesis and tiller number. Plant Biotechnol J. doi:10.1111/pbi.12239 PubMedGoogle Scholar
- Aryee MJ, Gutierrez-Pabello JA, Kramnik I, Maiti T, Quackenbush J (2009) An improved empirical bayes approach to estimating differential gene expression in microarray time-course data: BETR (Bayesian Estimation of Temporal Regulation). BMC Bioinformatics 10:409CrossRefPubMedCentralPubMedGoogle Scholar
- Battaglia M, Ripodas C, Clua J, Baudin M, Aguilar OM, Niebel A, Zanetti ME, Blanco FA (2014) A nuclear factor Y interacting protein of the GRAS family is required for nodule organogenesis, infection thread progression, and lateral root growth. Plant Physiol 164:1430–1442CrossRefPubMedCentralPubMedGoogle Scholar
- Combier JP, Frugier F, de Billy F, Boualem A, El-Yahyaoui F, Moreau S, Vernie T, Ott T, Gamas P, Crespi M, Niebel A (2006) MtHAP2-1 is a key transcriptional regulator of symbiotic nodule development regulated by microRNA169 in Medicago truncatula. Genes Dev 20:3084–3088CrossRefPubMedCentralPubMedGoogle Scholar
- Junker A, Monke G, Rutten T, Keilwagen J, Seifert M, Thi TM, Renou JP, Balzergue S, Viehover P, Hahnel U, Ludwig-Muller J, Altschmied L, Conrad U, Weisshaar B, Baumlein H (2012) Elongation-related functions of LEAFY COTYLEDON1 during the development of Arabidopsis thaliana. Plant J 71:427–442PubMedGoogle Scholar
- Kim I-S, Sinha S, De Crombrugghe B, Maity SN (1996) Determination of functional domains in the C subunit of the CCAAT-binding factor (CBF) necessary for formation of a CBF-DNA complex: CBF-B interacts simultaneously with both the CBF-A and CBF-C subunits to form a heterotrimeric CBF molecule. Mol Cell Biol 16:4003–4013PubMedCentralPubMedGoogle Scholar
- Laloum T, Baudin M, Frances L, Lepage A, Billault-Penneteau B, Cerri MR, Ariel F, Jardinaud MF, Gamas P, de Carvalho-Niebel F, Niebel A (2014) Two CCAAT box-binding transcription factors redundantly regulate early steps of the legume-rhizobia endosymbiosis. Plant J 79:757–768CrossRefPubMedGoogle Scholar
- Le DT, Aldrich DL, Valliyodan B, Watanabe Y, Ha CV, Nishiyama R, Guttikonda SK, Quach TN, Gutierrez-Gonzalez JJ, Tran LS, Nguyen HT (2012) Evaluation of candidate reference genes for normalization of quantitative RT-PCR in soybean tissues under various abiotic stress conditions. PLoS One 7:e46487CrossRefPubMedCentralPubMedGoogle Scholar
- Libault M, Farmer A, Brechenmacher L, Drnevich J, Langley RJ, Bilgin DD, Radwan O, Neece DJ, Clough SJ, May GD, Stacey G (2010) Complete transcriptome of the soybean root hair cell, a single-cell model, and its alteration in response to Bradyrhizobium japonicum infection. Plant Physiol 152:541–552CrossRefPubMedCentralPubMedGoogle Scholar
- Nelson DE, Repetti PP, Adams TR, Creelman RA, Wu J, Warner DC, Anstrom DC, Bensen RJ, Castiglioni PP, Donnarummo MG, Hinchey BS, Kumimoto RW, Maszle DR, Canales RD, Krolikowski KA, Dotson SB, Gutterson N, Ratcliffe OJ, Heard JE (2007) Plant nuclear factor Y (NF-Y) B subunits confer drought tolerance and lead to improved corn yields on water-limited acres. Proc Natl Acad Sci USA 104:16450–16455CrossRefPubMedCentralPubMedGoogle Scholar
- Schmutz J, Cannon SB, Schlueter J, Ma J, Mitros T, Nelson W, Hyten DL, Song Q, Thelen JJ, Cheng J, Xu D, Hellsten U, May GD, Yu Y, Sakurai T, Umezawa T, Bhattacharyya MK, Sandhu D, Valliyodan B, Lindquist E, Peto M, Grant D, Shu S, Goodstein D, Barry K, Futrell-Griggs M, Abernathy B, Du J, Tian Z, Zhu L, Gill N, Joshi T, Libault M, Sethuraman A, Zhang XC, Shinozaki K, Nguyen HT, Wing RA, Cregan P, Specht J, Grimwood J, Rokhsar D, Stacey G, Shoemaker RC, Jackson SA (2010) Genome sequence of the palaeopolyploid soybean. Nature 463:178–183CrossRefPubMedGoogle Scholar
- Zanetti ME, Blanco FA, Beker MP, Battaglia M, Aguilar OM (2010) A C subunit of the plant nuclear factor NF-Y required for rhizobial infection and nodule development affects partner selection in the common bean-Rhizobium etli symbiosis. Plant Cell 22:4142–4157CrossRefPubMedCentralPubMedGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.