Genome-Wide Analysis and Expression Profiling of the Phospholipase C Gene Family in Soybean (Glycine max)

Phosphatidylinositol-specific phospholipase C (PI-PLC) hydrolyses phosphatidylinositol-4,5-bisphosphate to produce diacylglycerol and inositol 1,4,5-trisphosphate. It plays an important role in plant development and abiotic stress responses. However, systematic analysis and expression profiling of the phospholipase C (PLC) gene family in soybean have not been reported. In this study, 12 putative PLC genes were identified in the soybean genome. Soybean PLCs were found on chromosomes 2, 11, 14 and 18 and encoded 58.8–70.06 kD proteins. Expression pattern analysis by RT-PCR demonstrated that expression of the GmPLCs was induced by PEG, NaCl and saline-alkali treatments in roots and leaves. GmPLC transcripts accumulated specifically in roots after ABA treatment. Furthermore, GmPLC transcripts were analyzed in various tissues. The results showed that GmPLC7 was highly expressed in most tissues, whereas GmPLC12 was expressed in early pods specifically. In addition, subcellular localization analysis was carried out and confirmed that GmPLC10 was localized in the plasma membrane in Nicotiana benthamiana. Our genomic analysis of the soybean PLC family provides an insight into the regulation of abiotic stress responses and development. It also provides a solid foundation for the functional characterization of the soybean PLC gene family.


Introduction
The PLC (phospholipase C) gene family is a subfamily of the phospholipase superfamily, which includes phospholipase D, phospholipase C, phospholipase A 1 and phospholipase A 2 . PLCs are divided into phosphatidylinositol-specific phospholipase C (PI-PLC) and non-specific phospholipase (NPC) by different hydrolysis sites [1]. NPC hydrolyzes the membrane lipids phosphatidylcholine and phosphatidylethanolamine in a calcium-independent manner [1]. It plays an important role in the supply of inorganic phosphate in the root plasma membrane during phosphate deprivation [2]. PI-PLC, also named PLC, exists in both animals and plants.
PLC-catalyzed production of the second messengers inositol 1,4,5-trisphosphate (IP 3 ) and diacylglycerol (DAG) is a key event in the transduction of agonist-dependent signals over the plasma membrane [3,4]. IP 3 was shown to release Ca 2+ from internal stores to the cytoplasm in animals. However, an IP 3 receptor has not been found in plants. Furthermore, protein kinase C (PKC), which is the target of DAG in animals, is lacking from plant genomes [5]. In plants, DAGs can be phosphorylated by DAG kinase (DGK) and converted into phosphatidic acids (PAs), which have emerged as a new class of lipid mediators involved in diverse cellular functions in plants [6]. IP 3 is rapidly converted into IP 6 and causes the release of Ca 2+ in plants [7]. The function of PLC is carried out mainly by the PA and IP 6 pathways in plants.
Two different types of PI-PLC enzymes, soluble and membrane-bound, were identified in the cells of several plants in early studies [8][9][10][11][12][13]. Only membrane-bound PI-PLCs were characterized to hydrolyze phosphatidylinositol-4,5-bisphosphate (PI(4,5)P 2 ) and phosphatidylinositol 4-phosphate (PI4P) [14], although the activities of these PI-PLCs were measured from crude protein extracts. Nonetheless, in vitro PLC activity has been confirmed for recombinant Arabidopsis PLCs. All five AtPLC fusion proteins tested required micromolar levels of Ca 2+ for PI(4,5)P 2 hydrolyzing activity with the exception of AtPLC4 [15]. However, our research showed that the activity of OsPLC1 was inhibited under high Ca 2+ concentrations (more than 1 mM) (data not shown). This should be caused mainly by feedback in the PLC signal pathway. Two messengers produced by PLC are thought to play signaling roles in guard cells [7], pathogen responses, Nod factor signaling and carbon fixation in C4 plants [16][17][18]. In Arabidopsis, there are nine known PI-PLC sequences in the genome [19]. Pokotylo et al. [20] characterized six NPCs in Arabidopsis and NPC involvement has been identified in the regulation of diverse cellular processes including root development and hormone signaling [21], Al stress signaling [22], abscisic acid (ABA) sensing, and tolerance to hyperosmotic and salt stresses [1,23]. Both PI-PLCs and NPCs have been found and characterized in the rice and tomato genomes [24,25].
Soybean PLC was first cloned and characterized in 1995 [26]. Soluble enzyme fractions that hydrolyzed phosphoinositides were separated from soybean sprouts using Matrex green gel column chromatography [27]. Functional analysis of soybean PLC was carried out using the PI-PLC specific inhibitor U-73122 and suggested that PI-PLC may negatively regulate the expression of defense genes [28]. In this study, we identified and characterized 12 PLC genes in the soybean genome. The expression profiles of soybean PLC genes were examined in different tissues and under various abiotic stresses. Moreover, subcellular localization analysis was carried out and showed that GmPLC10 was located in the plasma membrane.

Genomic search and sequence analysis
We searched for soybean PLC genes in the Phytozome v9.1 database (http://www.phytozome. org/). The key words "phospholipase C" were also searched using the browse tool on the NCBI website (www.ncbi.nlm.nih.gov/). The conserved domains of GmPLC proteins were identified by CDSEARCH/cddv3.13 (http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi). Arabidopsis PLC sequences were obtained from TAIR 10.0 (The Arabidopsis Information Resource). Rice PLC sequences were obtained from RGAP (Rice Genome Annotation Project). The loci of PLCs from Arabidopsis and rice are shown in S1 Table. A phylogenetic tree was constructed by MEGA5.1 with the default parameters [29].

Extraction of total RNA and cDNA synthesis
Total RNA was isolated from the collected tissue samples separately using Trizol reagent (Invitrogen, Carlsbad, CA, USA) following the manufacturer's protocols. The quality and quantity of the total RNA were determined using a NanoDrop 2000 (ThermoFisher Scientific, Beijing, China). cDNA was synthesized from 2 μg of total RNA using the PrimeScript RT reagent kit (Takara, Japan).

Quantitative real-time PCR
Gene-specific primers for the GmPLCs and reference genes were designed using Primer Express and are shown in S2 Table [30]. For expression analysis under abiotic stresses, Actin11 was selected as the reference gene. EF1A was chosen for the expression analysis in different tissues. RT-PCR was performed on a Stratagene Mx3000P thermocycler (Agilent) with the following program: 95°C for 15 s, followed by 40 cycles of 95°C for 15s and annealing at 60°C for 30 s. The expression level was calculated using the 2 −ΔΔCt method for abiotic stress treatments and the 2 ΔCt method for different tissues. Triplicates of each reaction were performed.

Promoter analysis
The promoters of the GmPLCs, which were 1 kb upstream from the translation start site, were mapped in the soybean genome. The cis-acting regulatory elements were identified in each promoter. Among these, we chose abiotic stress response and development related elements for expression analysis.

Nicotiana benthamiana transient transformation and subcellular localization
For subcellular localization analysis of soybean PLC, GmPLC10 was introduced into the binary expression vector pCAMBIA1302. The expression vector was transformed into Agrobacterium tumefaciens cells of the EHA105 strain. The transient transformation of N. benthamiana leaves was carried out following Marion et al. [31], with minor revision. For confocal microscopy, tobacco leaves were imaged using an inverted TCS-SPE spectral confocal laser scanning microscope (Leica, Germany). GFP fluorescence was excited with 488 nm argon laser lines with an emission band of 495-540 nm. Triplicate tobacco infiltrations were performed for the subcellular localization experiment.

Identification of PLC genes in soybean
The genome of soybean was sequenced recently [32]. We searched for soybean phospholipase C in both the genome and GenBank. Twelve PLCs were identified after removing redundant sequences. We named these 12 PLC genes GmPLC1 to GmPLC12 according to their physical locations on chromosomes 1-20 (Table 1). There are four, six and nine PLCs in the rice, tomato and Arabidopsis genomes, respectively [24,25,33]. The greater number of PLCs in soybean implied the importance and complexity of GmPLCs. The soybean PLCs were spread over chromosomes 2, 11, 14 and 18 equally (S1 Fig). Interestingly, all three PLC genes in each chromosome were located adjacent to one another, with the exception of GmPLC9 in chromosome 14. A similar phenomenon was found in Arabidopsis; AtPLC1, AtPLC4 and AtPLC5 are located in a 12 kb fragment on chromosome 5 [33]. The open reading frames (ORFs) of the soybean PLC genes ranged from 1623 to 1833 bp and encoded proteins of 540 to 610 amino acids. The theoretical isoelectric points of the soybean PLC proteins ranged from 5.57 to 9.11 with molecular masses ranging from 58.80 to 70.06 kD. The average length of PLC proteins is about 600 amino acids in plants. The smallest PLC is OsPLC2 from rice, which is composed of 491 residues [24]. In soybean, GmPLC5 and GmPLC11 were smaller than the others (Table 1). Furthermore, the coding sequences were analyzed and we found that all of the soybean PLC genes contained nine exons, except GmPLC4 and GmPLC12. The number of exons in the soybean PLCs was greater than that in rice and Arabidopsis (S2 Table).

GmPLC domain analysis and phylogenetic analysis
To characterize the soybean PLC genes, their domain structure was analyzed and a phylogenetic tree was constructed. PLC proteins contain a Pleckstrin Homology (PH) domain, EF hand, catalytic domain and C2 domain in animals. Some also contain a PDZ binding motif, Ras-binding domain, guanine-nucleotide-exchange factor for the Ras domain and Src homology domain [5]. However, in plants, such as soybean, PLCs only contain an EF hand, catalytic domain and C2 domain (Fig 1). The EF hand, which also found in DGK, has a key role in Ca 2+ binding [34]. The catalytic domain, which contains X and Y domains, is conserved among eukaryotic PLCs [35], especially soybean PLC (S2 Fig). The C2 domain can bind Ca 2+ and other effectors, including phospholipids, inositol phosphates, and proteins [36][37][38]. The C2 domain appears at the beginning of phospholipase D [39], but at the end of soybean PLC (Fig 1). There are two types of phospholipase C (PI-PLC and NPC) in plants. However, no NPCs were found in the soybean genome. Phylogenetic analysis was carried out on the plant PLC and NPC gene families in Arabidopsis, rice and soybean. Soybean PLCs showed high homology to PI-PLCs from Arabidopsis and rice. On the other hand, OsNPCs and AtNPCs shared similarity with each other (Fig 2). The differences between the PLC types derived from diversity in structure and amino acid composition. These results suggest that soybean PLC genes belong to the PLC gene family and should be classified as PI-PLCs.

Expression analysis of GmPLCs under abiotic stresses
To investigate the expression patterns of the soybean PLC genes, qRT-PCR was used to detect their transcript levels in leaves and roots under PEG, ABA, salt, alkali and saline-alkaline treatments. Our result indicated that the expression of GmPLC1, GmPLC2, GmPLC4 in leaves and GmPLC2, GmPLC4, GmPLC8 in roots cannot be detected. In leaves, GmPLC6 transcripts accumulated dramatically under saline-alkali and ABA treatments (Fig 3). In pea leaves, PsPLC  transcription was also stimulated after 30 min induction by ABA [40]. It has been reported that PLC plays a role in the events associated with the inhibition of stomatal opening by ABA [41]. This implies that PLC takes part in the regulation of drought stress in plants. On the other hand, the expression levels of GmPLC9 were induced by PEG stress, specifically. Transcript levels of GmPLC10 were increased after alkali treatment. After induction by salt, GmPLC12 transcript levels were more than 3-fold greater than in the control (Fig 3). Recently, TaPLC1 was characterized to have biological functions in regulating seedling growth and responses to drought and salinity stress [42]. The expression of TaPLC1 and AtPLC5 could be induced by salt and PEG treatments [15,42]. The overexpression of AtPLC9 improved thermotolerance in Arabidopsis [43]. These results show that plant PLCs play a positive role in abiotic stress responses. However, the expression levels of GmPLC3 and GmPLC9 decreased under ABA treatment, and GmPLC6 transcripts were decreased under PEG treatment (Fig 3). This phenomenon was also observed in rice, in which OsPLC4 transcripts decreased during salt, cold and drought stresses [24]. It was also shown that one family member had dominant negative effects on the function of the others in Arabidopsis [44]. It is essential to investigate the mechanism by which PLCs function in the regulation of plant abiotic stress responses.
PA reportedly binds to a number of proteins that play a role during water limiting conditions, such as drought and salinity stress, and has been shown to play an important role in maintaining root system architecture [45]. In addition, DAG, which is a product of PLC, can be subsequently phosphorylated to PA by DGK. Therefore, the expression profiles of the soybean PLC genes were also analyzed in roots. The expression of GmPLC1 decreased under salt, saline-alkali, alkali and PEG treatments in roots. GmPLC3, GmPLC5, GmPLC6, GmPLC10, GmPLC11 and GmPLC12 transcripts were induced by salt stress (Fig 4). The expression levels of GmPLC3, GmPLC5, GmPLC6, GmPLC10, GmPLC11 and GmPLC12 were increased under alkali treatment. GmPLC3, GmPLC5, GmPLC6, GmPLC10 and GmPLC11 were expressed under saline-alkaline treatment. ABA treatment stimulated the transcription of GmPLC5, GmPLC6, GmPLC7, GmPLC10 and GmPLC12, especially GmPLC7, GmPLC10 and GmPLC12 (Fig 4). Although the expression profiles of PLCs in roots have not been analyzed before, PLC signaling has been characterized in the activation of plasmolysis in root cells [46]. Moreover, most of the soybean PLC genes were induced by abiotic stress, especially ABA treatment. This suggests that soybean PLC genes are involved in various stress pathways.

Expression analysis of GmPLCs in various organs
Plant PLC members have been characterized in various cellular processes and signaling networks, and are triggered in response to developmental events in different plant species [47]. To investigate the steady-state expression patterns of the PLC genes in soybean, qRT-PCR was used to detect transcript levels in various organs. High transcript levels of GmPLC7 were detected in all organs, especially stems and early pods. GmPLC12 expression was found in early pods specifically, although low transcript levels were found in other organs (Fig 5). In Arabidopsis, AtPLC2 was highly expressed in leaves, stems, roots and flowers [33]. Recently, microarray analysis of the Arabidopsis PI-PLC gene family was carried out and confirmed that AtPLC2 transcripts were highly expressed in all organs [48]. GmPLC3, GmPLC5, GmPLC10 and GmPLC11 had medium transcript levels in all organs. The expression levels of GmPLC1, GmPLC6, GmPLC8 and GmPLC9 were lower than the others (Fig 5). In rice, OsPLC1 and OsPLC3 were highly expressed in all organs, but lower expression of OsPLC2 was detected in various organs [24]. The varied expression patterns of the soybean PLC gene family imply that GmPLCs are involved in different stages during soybean development.  Cis-regulatory elements in the promoters of soybean PLC genes Gene expression can be regulated at several levels, such as transcription and post-translational modification. The promoter is the region of DNA that initiates the transcription of a particular gene. Cis-regulatory elements are functional DNA sequences that precisely control the temporal and spatial expression patterns of the genes expressed in higher plants [49]. In this study, soybean PLC genes were mapped on the genome and their promoter sequences were obtained. The cis-regulatory elements, such as MBS (MYB binding site), ABRE (abscisic acid responsive element), TC-rich, Skn-1 and GCN4, were analyzed. These cis-regulatory elements are known to regulate various stress responses and plant development [50][51][52][53]. Each soybean PLC gene had 1-6 cis-regulatory elements in its promoter region (Fig 6). The promoter of GmPLC10, The Phospholipase C Family in Soybean which was highly expressed in all organs and induced strongly by abiotic stresses, contained one TC-rich and Skn-1, and two ABRE motifs. GmPLC3, GmPLC11 and GmPLC12 had more cis-regulatory elements, which might account for their responsiveness to abiotic stresses and development (Fig 6, S3 Table). Surprisingly, six cis-regulatory elements were found in the promoter of GmPLC9, but GmPLC9 transcripts were low in all organs and insensitive to nearly all abiotic stresses. This is in agreement with previous data on Glu-A1-1, -B1-1 and -B1-2 [54]. This result implies that other mechanisms modulate the expression of GmPLC9.

Subcellular localization of GmPLC10
Previously, it was reported that the substrates of PLC, PI4P and PI(4,5)P 2 , are located in the plasma membrane [55]. To analyze the location of soybean PLC proteins, GmPLC10 was selected for subcellular localization analysis. GmPLC10 was fused to the N-terminus of GFP and transformed into N. benthamiana transiently. In confocal observation, GmPLC10-GFP fluorescence was located in the plasma membrane, whereas fluorescence of the vector control was observed everywhere (Fig 7). This result is consistent with research on AtPLC9 [43]. However, Singh et al. [24] reported that OsPLC1 and OsPLC4 were distributed throughout the cytoplasm and nucleus. The diverse localization of plant PLCs implies that they might take part in different cellular processes during development and abiotic stress.

Conclusions
In conclusion, we identified 12 GmPLC genes from the soybean genome in this study. Conversed domain and phylogenetic analyses were carried out and verified that all soybean PLC genes belong to the PI-PLC subfamily. The expression profiles of the soybean PLC genes were analyzed during development and various stresses. Together with cis-regulatory elements analysis, our results suggest that soybean PLCs are involved in various cellular processes. Moreover, subcellular localization analysis confirmed that GmPLC10 was located in the plasma membrane. Taken together, these results suggest that soybean PLCs is important for plant development and adaptation to environmental stress conditions. Supporting Information S1