LcMYB2 expression pattern analysis
Based on 454 high throughput sequencing and expression profile analyses of sheepgrass under drought stress, we found 15 MYB and MYB-related transcription factors that were responsive to changes of water content in plant tissues [32, 35]. Contig41859, which was up-regulated by drought stress and named LcMYB2, was a MYB-related transcription factor with unknown function that attracted our attention (Additional file S1).
LcMYB2 is highly induced by 300mM mannitol at the 8th hour after treatment (Fig. 1a), whereas it is relatively slower responding to salt and cold stress (24 h; Fig. 1b,c). However, it is quickly upregulated by ABA treatment (Fig. 1d). The maximal level of mRNA accumulation under mannitol treatment is higher than under salt, cold and ABA treatments, indicating that LcMYB2 mainly functions in response to osmotic stress. Furthermore, the expression level of LcMYB2 in different organs is also detected under normal growth conditions. The results show that LcMYB2 has the highest transcript level in roots (Fig. 1e). Based on these combined results, we predict that LcMYB2 is mainly responsible for the osmotic stress response in roots, which may benefit plants under drought stress.
Isolation and sequence analysis of the LcMYB2
Putative full-length LcMYB2 was isolated from sheepgrass by Rapid-Amplification of cDNA Ends-PCR (RACE-PCR) and classical PCR based on the 454 high-throughput data (SRA065691; Additional file S2). The length of LcMYB2 Open reading frame (ORF), region is 1092 bp, encoding 363 amino acids (GenBank: KY316376). The molecular mass of the putative protein is approximately 38.4 kDa, and its theoretical isoelectric point (pI) is 8.3 (predicted by DNAMAN 7.0). Multiple sequence alignment of LcMYB2 with its homologs shows that a conserved domain exists among these sequences (amino acids 101-171; Fig. 2a). Sequence similarity and phylogenetic analysis show that LcMYB2 forms a clade with BAK02871 (Hordeum vulgare), CDM81700 (Triticum aestivum) and EMT01615 (Aegilops tauschii) by a high nodal support values (Fig. 2b, c). However, the functions of these LcMYB2 homologs have not been reported thus far. The analysis of LcMYB2 biological function is of great significance for Leymus chinensis and closely related species Triticum aestivum and Aegilops tauschii homologs.
Subcellular localization and transcription activity assay of LcMYB2
To determine the subcellular localization of LcMYB2, the ORF of LcMYB2 (without the TGA stop codon) was fused to a GFP reporter gene under the control of the CaMV 35S promoter (Fig. 3a). Recombinant CaMV35S::LcMYB2-GFP and CaMV35S::GFP were transformed into Arabidopsis separately by the floral dip method. Confocal microscopy showed that the GFP protein was localized throughout the whole cell, whereas the LcMYB2-GFP fusion protein was present only in the nucleus (Fig. 3b), suggesting that LcMYB2 is a nuclear-localized protein.
The transcriptional activation of LcMYB2 was tested using a yeast one-hybrid assay system. The LcMYB2 ORF was inserted at the 3’-end of GAL-BD under the control of PADH1 to form a BD-LcMYB2 fusion gene (Fig. 3c). The yeast strain AH109, harboring BD-LcMYB2 or BD-WRKY15 (positive controls), grew normally on SD/-His-Trp medium, whereas AH109 harboring only BD (negative control) did not grow. In β-galactosidase activity assays on Whatman filter paper, blue signal appeared in the regions where BD-LcMYB2 or BD-WRKY15-containing yeast were growing (Fig. 3d). Therefore, we suggest that LcMYB2 serves as a transcription activator and functions in the nucleus.
Performance of transgenic plants under osmotic stress, ABA treatment and natural drought treatment
First, we probed the biological functions of LcMYB2 at the seed germination stage under osmotic or ABA treatment. Under normal conditions (Murashige-Skoog medium ,MS), there were no significant differences between transgenic and wild-type seeds in germination rate, cotyledon greening rate or root length (Fig. 4a,d,g,c,f,i; Fig. 5a,d,g,h,i; Additional file S5). Under treatment with 300 mmol/L mannitol, there were significant differences in germination rate (p < 0.01; Fig. 4b,c), and the cotyledon greening rate and root length had very significant differences (p < 0.001) between transgenic and wild-type seeds (Fig. 4e,f,h,i; Additional file S5). Under treatment with 0.25 µmol/L ABA, the germination rate (p < 0.01), cotyledon greening rate (p < 0.001) and root length (p < 0.001) were significantly different between transgenic and wild-type seeds, and similar results were obtained with 0.5 µmol/L ABA treatment (Fig. 5b,c,e,f,g,h,i). Taken together, these data indicate that LcMYB2 can promote seed germination and root growth under osmotic stress and possibly via the ABA signaling pathway. In addition, the transgenic plants maintained green leaves longer under natural drought stress conditions and had a higher refresh rate after rewatering than did wild-type (Fig. 6).
To investigate the physiological responses of transgenic and wild-type A. thaliana under osmotic stress, we irrigated 4-week-old seedlings with 300 mmol/L mannitol. Two days later, the malondialdehyde (MDA), Superoxide dismutase (SOD), soluble sugars and proline contents were measured. The results showed that the two transgenic lines overexpressing LcMYB2 accumulated greater amounts of SOD (p < 0.01), soluble sugars (p < 0.05/0.01) and proline (p < 0.001) than wild-type lines under 300 mmol/L mannitol treatment, whereas they had lower MDA content (Fig. 7a,b,c,d). The greater accumulation of proline and soluble sugars in the transgenic lines might provide extra protection to the cells under drought stress. The lower MDA and higher SOD content indicates that less damage occurred in the cells of transgenic plants. These results together suggest that LcMYB2 promotes osmotic stress resistance. The gene expression levels of AtDREB2A, AtP5CS1, and AtLEA14 were measured by Quantitative Real Time PCR (qPCR) on the 9th hour after treatment with 300 mmol/L mannitol. The expression levels of these genes were higher in transgenic plants than in wild-type plants both in the control check (CK) and in the treatment group (M9; Fig. 7e).
CHIP analysis
It has been shown that MYB proteins can recognize the motifs A/TAACCA and C/TAACG/TG [43]. Therefore, we analyzed the promoter sequences of AtLEA14, AtP5CS1, AtDREB2A and LcDREB2 using the sequences ~1500-2000 bp upstream of the predicted transcription start sites (TSSs), and several possible motifs were found in the putative promoter regions (Additional file S3). We further confirmed our prediction with CHIP experiments, and the signals were detected in qPCR and universal PCR reactions with DNAs, released from proteins LICHIP, L2CHIP or LcCHIP, as templates. These results indicated that the LcMYB2 (or together with its interaction proteins) regulates the expression of AtLEA14, AtP5CS1, AtDREB2A and LcDREB2 (Fig. 8).