SiLEA14, a novel atypical LEA protein, confers abiotic stress resistance in foxtail millet

Background Late embryogenesis abundant (LEA) proteins are involved in protecting higher plants from damage caused by environmental stresses. Foxtail millet (Setaria italica) is an important cereal crop for food and feed in semi-arid areas. However, the molecular mechanisms underlying tolerance to these conditions are not well defined. Results Here, we characterized a novel atypical LEA gene named SiLEA14 from foxtail millet. It contains two exons separated by one intron. SiLEA14 was expressed in roots, stems, leaves, inflorescences and seeds at different levels under normal growth conditions. In addition, SiLEA14 was dramatically induced by osmotic stress, NaCl and exogenous abscisic acid. The SiLEA14 protein was localized in the nucleus and the cytoplasm. Overexpression of SiLEA14 improved Escherichia coli growth performance compared with the control under salt stress. To further assess the function of SiLEA14 in plants, transgenic Arabidopsis and foxtail millet plants that overexpressed SiLEA14 were obtained. The transgenic Arabidopsis seedlings showed higher tolerance to salt and osmotic stress than the wild type (WT). Similarly, the transgenic foxtail millet showed improved growth under salt and drought stresses compared with the WT. Taken together, our results indicated that SiLEA14 is a novel atypical LEA protein and plays important roles in resistance to abiotic stresses in plants. Conclusion We characterized a novel atypical LEA gene SiLEA14 from foxtail millet, which plays important roles in plant abiotic stress resistance. Modification of SiLEA14 expression may improve abiotic stress resistance in agricultural crops. Electronic supplementary material The online version of this article (doi:10.1186/s12870-014-0290-7) contains supplementary material, which is available to authorized users.

more β-sheets than α-helixes [5,17], which is different from subgroups 5A and 5B LEA proteins that are intrinsically unstructured [17][18][19]. For example, Arabidopsis LEA14 has an αβ-fold consisting of one α-helix and seven β-strands that form two antiparallel β-sheets as determined by nuclear magnetic resonance spectroscopy [16]. Moreover, subgroup 5C LEA proteins have other outstanding characteristics, such as lower instability index, narrower range of GRAVY values, and lower proportion of polar (hydrophilic) and small residues, but higher proportion of non-polar residues than subgroup 5A and 5B members [5]. All of these differences indicate that subgroup 5C LEA proteins may function differently from other LEA proteins in plants.
At present, only a small number of subgroup 5C LEA genes have been characterized. Their transcripts can be upregulated in response to diverse stresses, as reported for cotton LEA14-A [15], Craterostigma plantagineum PcC27-45 [20], soybean D95-4 [21], tomato ER5 [22], hot pepper CaLEA6 [23], Arabidopsis LEA14 [24] and At2g44060 [17], and sweetpotato IbLEA14 [25]. Overexpression of CaLEA6 in tobacco improves tolerance to dehydration and NaCl but not to low temperature [23]. Transgenic sweetpotato non-embryogenic calli that overexpress IbLEA14 show increased tolerance to drought and salt stress by enhancing lignification [25]. Recently, rice OsLEA5 has been reported to enhance resistance against diverse abiotic stresses in recombinant Escherichia coli cells. In vitro analysis showed that OsLEA5 was able to protect lactate dehydrogenase from aggregation under different abiotic stresses [5]. All these results suggest that subgroup 5C LEA proteins are closely associated with resistance to multiple abiotic stresses.
Foxtail millet (Setaria italica (L.) Beauv.), a member of the Poaceae family, has a long history in cultivation of about 7000 years. It has been widely planted in northern China and other Asian countries. Recently, a draft genome sequence for foxtail millet has been completed [26], which enables foxtail millet to be a tractable experimental grass model [27]. As a diploid C 4 panicoid crop species, foxtail millet is well known for its remarkable drought resistance. However, the molecular mechanisms underlying this tolerance are not well defined.
In this study, we isolated and functionally characterized a novel member of the atypical subgroup 5C LEA gene, SiLEA14, from foxtail millet. The expression of SiLEA14 was induced by ABA, polyethylene glycol (PEG) and NaCl. Overexpression of SiLEA14 resulted in enhanced resistance to abiotic stresses in E. coli, Arabidopsis and foxtail millet. The SiLEA14 promoter mediated remarkable induction of β-glucuronidase (GUS) expression in transgenic Arabidopsis under various stresses. Cis-acting regulatory elements in the SiLEA14 promoter were also predicted. These data reveal the potential application of SiLEA14 in the genetic engineering of other crops.

SiLEA14 is an atypical LEA protein
The full-length sequence of SiLEA14 was determined by 5′ and 3′ rapid amplification of cDNA ends (RACE) [GenBank: KJ767551]. The sequence is 821 bp in length, with a 100 bp 5′ untranslated region (UTR) and a 208 bp 3′ UTR (with polyA tail) (Additional file 1A). SiLEA14 harbors two exons separated by an intron (Additional file 1B), and encodes an open reading frame of 170 aa with a predicted molecular mass of 18.77 kD and pI of 5.56. It is rich in Ser (10.6%), Lys (8.8%), and Ile (8.2%), but contains low quantities of Trp (1.2%), Asn (1.8%), Cys (1.8%), and Gln (1.8%). Three hydrophobic regions (I, II, and III) were identified in the SiLEA14 protein (Additional file 1C). A motif search of the SiLEA14 protein in InterProScan revealed that it contains a "LEA_2" motif (PF03186, 3.4e- 20), which was classified into subgroup 5C (D-95) according to Battaglia's classification of LEA proteins [12]. Further analysis showed that SiLEA14 contains a lower percentage of polar amino acids (47.2%) and higher percentage of non-polar amino acids (25.8%) than other group LEA proteins (Additional file 1D) [5]. The GRAVY value and the instability index of SiLEA14 is −0.155 and 34.68, respectively (Additional file 1D). All these characteristics were in accordance with those of other members of subgroup 5C [5].
Two additional SiLEA14 homologs (Si003774m and Si003233m, Additional file 2) existed in foxtail millet. They showed 57.7% and 48.24% identity to SiLEA14, respectively. Homologous sequences of SiLEA14 were also present in other plant species, such as rice OsLEA5 (77.06%), maize Lea14-A (50.88%) and Arabidopsis AtLEA14 (47.06%). The homologs shared different levels of sequence similarities, indicating different levels of evolutionary relationship among these proteins ( Figure 1A). A phylogenetic tree for SiLEA14 and its homologs was constructed from protein sequences (Additional file 2). The proteins were mainly separated into two clades comprising dicot and monocot species, respectively. The SiLEA14 protein shared the closest relationship with the pearl millet LEA-like protein, followed by OsLEA5 ( Figure 1B). Sequence logos were produced to examine the conservation level at each residue position. The sequences lacked significant signature motifs or consensus sequences, except for several conserved residues at specific positions, such as proline at positions 41 and 83 (Additional file 3). This is a common feature of subgroup 5C LEA proteins [5].

SiLEA14 expression profiles under normal and stress conditions
Quantitative reverse-transcription PCR (qRT-PCR) was carried out to reveal the temporal and spatial expression of SiLEA14 in foxtail millet ( Figure 3A). SiLEA14 was expressed at the highest level in roots, followed by stems and leaves, and at the lowest level in inflorescences. After pollination, SiLEA14 transcription was gradually upregulated with seed maturation, which indicated a potential role for SiLEA14 in the maturation and desiccation phases of foxtail millet seed development.
The expression level of SiLEA14 was also examined by qRT-PCR under various stresses. Under ABA treatment, SiLEA14 transcription abruptly reached its highest level (six-fold) after 1 h treatment, and then decreased gradually and reverted almost to the control level at 24 h ( Figure 3B). Under PEG stress, SiLEA14 transcription was rapidly induced within 0.5 h and increased to a peak level (8-fold) at 12 h ( Figure 3C). Salt treatment resulted in increased accumulation of SiLEA14 to a maximum level (11-fold) only after 3 h ( Figure 3D). These results indicated that SiLEA14 might play important roles in the responses to salt and drought stresses.

SiLEA14 enhances salt tolerance in transformed E. coli
In both prokaryotes and eukaryotes LEA proteins might have similar protective mechanisms [28]. Therefore, the up-regulation of SiLEA14 in response to NaCl prompted us to evaluate the salt stress tolerance of recombinant E. coli overexpressing SiLEA14 (Figure 4). There are no significant differences in colony number between transformed E. coli harboring SiLEA14 and the control under normal conditions, indicating that overexpression of SiLEA14 did not affect the growth of E. coli recombinants in non-stress conditions. However, when grown on Luria-Bertani (LB) plates supplemented with 600 mM NaCl, the number of  Total RNA was extracted from roots (R), stems (S), leaves (L), inflorescences (I) and seeds at 5 (S1), 15 (S2) and 25 (S3) days after pollination, respectively. Foxtail millet Actin7 was used as an internal control. (B-D) SiLEA14 transcripts accumulation in response to various treatments. Two-week-old foxtail millet seedlings were treated with 100 μM ABA (B), 20% (v/v) PEG 6000 (C) and 250 mM NaCl (D) for the indicated times, respectively. Total RNA was extracted from the aboveground parts of ten plants at the indicated times after the treatments. Data represent means and standard errors for three biological replicates. transformed cells was much higher than that of the control. A similar result was obtained on LB medium supplemented with 600 mM KCl. These results demonstrated that overexpression of SiLEA14 in E. coli significantly enhanced tolerance to salt stress.

SiLEA14 improves the abiotic stress resistance of transgenic Arabidopsis
To assess the function of SiLEA14 in plants, it was overexpressed in Arabidopsis under the control of Super promoter ( Figure 5A). At least 30 transgenic plants were obtained, and two independent homozygous T 3 transgenic lines (L6 than L9) with high expression levels of SiLEA14 ( Figure 5B) were chosen for further investigation.
Under non-stress conditions, no significant differences in the fresh and dry weights were observed between the WT and transgenic plants ( Figure 5C and D). However, when grown on Murashige and Skoog (MS) medium supplemented with 125 mM NaCl, the transgenic seedlings showed significantly larger cotyledons and longer roots than those of the WT, even though growth of these organs of the transgenic and WT plants was inhibited compared with normal conditions ( Figure 5C). Consistently, the fresh and dry weights of transgenic seedlings were significantly higher than that of the WT ( Figure 5E). Similar results were obtained when seedlings were treated with 250 mM mannitol ( Figure 5C and F). These results suggest that SiLEA14 may enhance the osmotic stress resistance of transgenic Arabidopsis.

Overexpression of SiLEA14 increases the salt tolerance of transgenic foxtail millet
To further analyze the function of SiLEA14, it was transformed into foxtail millet under the control of ubiquitin promoter ( Figure 6A). The SiLEA14 integration was confirmed by genomic PCR using two pairs of primer sets specific to Hpt and 35S promoter, respectively (Additional file 4). QRT-PCR analysis showed higher expression level of SiLEA14 in transgenic lines L68, L76, and L78, which showed a 3:1 segregation ratio at T 1 generation, than that of the WT ( Figure 6B). The T 2 generation of these three lines were used for further analysis.
First, we examined the salt tolerance of transgenic foxtail millet during germination, and the results were shown in Figure 6. Compared with the WT plants, transgenic lines showed better growth performance ( Figure 6C) with enhanced shoot and root growth ( Figure 6E and G) when germinated in water for 4 days. Under salt stress for 4 days, the transgenic lines showed significantly longer shoot and root than the WT ( Figure 6E and G), even though the growth of both WT and transgenic lines were seriously suppressed ( Figure 6C). Similar results were obtained for 9 days salt stress treatment ( Figure 6D, F and H).
Further, we examined the salt tolerance of transgenic foxtail millet seedlings in soil. No striking differences in the height were observed between the transgenic and the WT plants under non-stress conditions, but wider leaves were observed in the transgenic foxtail millet compared with those of the WT ( Figure 7A). This finding is in accordance with the result that transgenic lines showed longer shoot and root lengths when germinated in water compared with the WT ( Figure 6C-H). Under 150 mM NaCl stress for 6 days, the first and second leaves of WT plants became curled, whereas there was no obvious effect on the transgenic lines. The WT plants were shorter than those of the transgenic lines. Under 250 mM NaCl stress for 6 days, the growth of WT plants was severely inhibited and leaves were curled, bleached, and senescent. However, the transgenic lines showed healthy growth except that only a few old leaves became bleached ( Figure 7A).
Electrolyte leakage always occurs following membrane damage under salinity stress [29]. Furthermore, plants accumulate several metabolites, such as amino acids (e.g., proline), sugars, and sugar alcohols (e.g., mannitol and trehalose), to prevent detrimental changes caused by severe osmotic stress [30,31]. Therefore, we measured the electrolyte leakage and change in free proline and soluble sugar contents in control and SiLEA14 transgenic lines with or without salinity stress, respectively. Under optimal or salt-stress conditions, the amount of electrolyte leakage in the transgenic lines was significantly lower than that of the WT plants ( Figure 7B). This finding is in agreement with the result that the SiLEA14 transgenic lines showed enhanced germination and growth compared with the WT under both normal  The phenotype of transgenic and WT foxtail millet under various abiotic stress treatment during the germination stage. The T 2 seeds soaked in water (as control) or in water containing 150 mM NaCl, 250 mM NaCl, 10% PEG, 20% PEG and 10 μM ABA for 1 day at 30°C and then placed on the filter paper in a Petri dish wet with the same solutions mentioned above for 4 and 9 days, respectively. This experiment had three replicates, and each experiment comprised 30-40 plants. Bar = 1 cm. (E-F) The shoot length of transgenic and WT foxtail millet germinated under above conditions for 4 and 9 days, respectively. (G-H) The root length of transgenic and WT foxtail millet germinated under above conditions for 4 and 9 days, respectively. Data in B, E-H represent means and standard errors for three biological replicates. Statistical significance was determined by Student's t-test. *P < 0.05; **P < 0.01. correlated with the higher accumulation of SiLEA14 transcripts in L68. All of these results indicated that SiLEA14 overexpressing foxtail millet is more tolerant to salt stress compared with the WT.

Overexpression of SiLEA14 increases drought resistance of transgenic foxtail millet
We also examined the drought resistance of transgenic foxtail millet during germination. When germinated in water for 4 days, the transgenic lines showed better growth performance than WT as described above (Figure 6C, E and G). Compared with in water, no obvious suppression under 10% PEG stress but serious suppression under 20% PEG stress for 4 days were observed ( Figure 6C). The shoot and root length of transgenic lines was longer than that of the WT no matter germinated in water, 10% or 20% PEG ( Figure 6E and G). When germinated in water or PEG for 9 days, similar results were obtained ( Figure 6D, F and H).
The observation that SiLEA14 improved osmotic stress resistance during germination prompted us to investigate the drought resistance of transgenic foxtail millet seedlings in soil. As it is shown in Figure 8A, no significant differences in the height were observed between the transgenic lines and the WT when watered well, except that the leaves of the transgenic lines were wider. However, under drought stress for 7 days, more than 90% of WT leaves, but less than 20% of transgenic plant lines leaves, became curled and wilted. After rewatering, 74- 100% plants of trangenic lines survived in contrast to only 14% of WT ( Figure 8B).
The change in free proline and soluble sugar contents in control and SiLEA14 transgenic lines with or without drought stress was examined. No significant differences in free proline and soluble sugar contents were observed between the WT and the transgenic lines under nonstress conditions. Under drought stress for 3 days, the free proline and soluble sugar contents increased in all WT and transgenic plants. However, a significantly higher increase in free proline content was observed in the transgenic lines L68 and L78 when compared to the WT. In addition, compared with the WT, the soluble sugar content increased significantly in the transgenic lines L68 and L76 ( Figure 8C, D). Taken together these results indicate that SiLEA14 overexpressing foxtail millet showed improved drought resistance.

GUS activity is induced by various stresses in proSiLEA14:: GUS transgenic Arabidopsis
The putative SiLEA14 promoter region was isolated from the foxtail millet genome using PCR. This fragment consisted in nt −1273 to +96 (transcription start site is +1) upstream of SiLEA14 coding sequence. The SiLEA14 promoter was fused to the GUS gene and transformed into Arabidopsis. Then, qRT-PCR was carried out to examine the GUS expression level under various stresses. As it is shown in Figure 9, GUS transcription was up-regulated to its highest level (about 3-and 8-fold increases) after 6 h treatment with ABA and NaCl, respectively. For PEG stress, GUS expression gradually increased and peaked (an almost 12-fold increase) after 18 h treatment. To further confirm these results, a histochemical GUS-staining assay was performed. No visible GUS activity was noted for the control. However, induction of GUS activity was observed with ABA, PEG or NaCl application for 3 to 18 h, although the GUS signals were not particularly strong (Additional file 5). Nevertheless, collectively these results suggest that the selected SiLEA14 promoter sequence is enough to regulate ABA, PEG and NaCl induction of the gene.

SiLEA14 promoter contains stress-associated cis-elements
To characterize the mechanism of SiLEA14 function, a 1273 bp SiLEA14 promoter fragment (without the 96 bp 5′UTR) was subjected to a putative cis-acting regulatory element search using the Plant Cis-acting Regulatory DNA Elements (PLACE) database [32]. Cis-acting elements reported to be involved in ABA-and dehydrationmediated gene expression were identified (Table 1). Abscisic acid-responsive elements (ABREs) are common motifs for ABA-and dehydration-mediated regulation of transcription [33][34][35]. In the SiLEA14 promoter, six ABRE-like sequences and four ABA-and dehydration-responsive ACGT-box motifs [34,36] were identified. A dehydrationresponsive element (DRE)-like site, as a cis-acting element involved in ABA induction, drought, salinity, or cold stresses [37,38], was also identified at position −158. MYC and MYB binding sites have been identified as regulatory elements associated with ABA, drought, or cold responsiveness in plants [34,[39][40][41][42][43]. In the SiLEA14 promoter, as many as 10 and three MYC-and MYB-like sequences, respectively, were observed. These results suggest that SiLEA14 may be under the regulation of DREB-, DBF-, CBF-, MYC-or MYB-like transcription factors as in Arabidopsis and rice. In addition, other regulatory cis-acting elements were identified in the SiLEA14 promoter, such as low temperature-responsive, pathogen-, defense-and wounding-associated elements. However, whether SiLEA14 participates in these abiotic or biotic stresses awaits further investigation.

Discussion
Typical LEA proteins can retain water molecules and protect other proteins from aggregation or desiccation because of their highly hydrophilic properties [44]. Conversely, atypical LEA proteins have higher content of hydrophobic residues than typical LEAs. The latter have been speculated to be involved in diverse stress tolerances, although few studies have been carried out to characterize their functions [5,23,25]. Here, we reported the identification and characterization of SiLEA14, a novel atypical LEA member, in foxtail millet.
As a key phytohormone, ABA plays an important role in plant stress responses. Osmotic stress-regulated genes can be activated through both ABA-dependent and ABA-independent pathways [45]. However, it is considered that stress-signalling pathways for the activation of One-week-old seedlings were treated with 100 μM ABA, 250 mM NaCl and 20% (m/v) PEG 6000 solution for 0, 3, 6 and 18 h, respectively. Total RNA was extracted from at least 30 whole seedlings per treatment. The expression of GUS in proSiLEA14::GUS transgenic Arabidopsis was determined by qRT-PCR using GUS-specific primers (Additional file 4). Actin2 was used to normalize the gene expression. Data represent means and standard errors for three biological replicates.
LEA-like genes completely independent of ABA may not exist [1]. In our study, SiLEA14 accumulation was remarkably induced by ABA and peaked rapidly after 1 h treatment ( Figure 3). In addition, SiLEA14 promoter-driven GUS activity was distinctly stimulated by ABA (Figure 9). Compared with the WT, transgenic foxtail millet seeds showed better germination in ABA solution ( Figure 6C-H). All of these results indicate that activation of SiLEA14 under salt and drought stresses may be dependent on ABA.
Yeast and E. coli heterologous systems have been widely used to investigate LEA gene functions [5,6,46]. In the present study, overexpression of SiLEA14 protected E. coli cells from damage caused by salt stress (Figure 4). Overexpression of SiLEA14 in Arabidopsis imparted increased tolerance to salt and mannitol stresses ( Figure 5). This result suggests that SiLEA14 from foxtail millet, a monocot, can function properly in the dicot Arabidopsis. Interestingly, transgenic foxtail millet overexpressing SiLEA14 exhibited superior germination and subsequent growth in soil compared with the WT even under normal conditions. Under salt and drought stresses, these differences were more remarkable, which were indicative of the key roles of SiLEA14 in foxtail millet (Figures 6, 7 and 8). Consistent with these findings, both SiLEA14 transcription and SiLEA14 promoter-driven GUS activity were remarkably induced by NaCl and PEG (Figures 3, 9 and Additional file 5). However, it should be noted that some discrepancies were observed between the expression pattern of the endogenous SiLEA14 upon abiotic stress and ABA treatment ( Figure 3) and the SiLEA14 promoter-driven GUS transcript accumulation responsive to abiotic stress and ABA driven by SiLEA14 promoter region in transgenic Arabidopsis (Figure 9). For example, the SiLEA14 transcripts accumulated to the highest level after 250 mM NaCl treatment for 3 h (Figure 3), whereas the GUS transcripts reached the peak after 6 h treatment ( Figure 9). Additionally, the histochemical analysis showed strong GUS signals in the petioles of transgenic Arabidopsis (Additional file 5), whereas the endogenous SiLEA14 was highly expressed in roots ( Figure 3). This is probably due to that SiLEA14 is from the monocot foxtail millet and the GUS expression driven by SiLEA14 promoter is performed in the dicot Arabidopsis.
So far, there is only one report on the functional mechanism of the atypical LEA protein in plants [25]. Overexpression of IbLEA14, a homologous gene of SiLEA14, enhanced tolerance to drought and salt stress in the sweetpotato calli. The contents of lignin in the IbLEA14-overexpressing calli were increased under normal conditions. The authors inferred that IbLEA14 may be involved in these functions as a consequence of regulating increased lignin production. In our study, we found that overexpression of SiLEA14 in Arabidopsis and foxtail millet obviously improved the osmotic stress resistance of transgenic plants. Meanwhile, the proline and sugar accumulated at a higher level in transgenic lines, especially in L68 which accumulated higher SiLEA14 transcripts, than WT after osmotic  and NaCl stress. These results implied that overexpression of SiLEA14 might up-regulate these metabolites. However, further study is needed. The identification of cis-acting elements in the SiLEA14 promoter may help to provide insight into the molecular mechanism of SiLEA14 function. As a major cis-acting element, ABRE has been identified in the promoters of many ABA-inducible genes of plants such as the cotton LEA gene D-113 [47]. The bZIP transcription factors AREB/ABF can bind to ABRE and activate ABA-inducible gene expression [48][49][50][51]. In the present study, ten cis-elements containing the ABRE ACGT-core were identified in the −793 to −77 bp region of the SiLEA14 promoter. The DRE element was first identified in the promoter of rd29A, a gene responsive to dehydration stress in Arabidopsis [52]. In plants, the AP2 transcription factors DREB/CBF specifically bind to the DRE element to regulate the expression of the downstream stress-responsive genes [53]. There are interactions between ABRE and DRE in Arabidopsis rd29A gene expression in response to dehydration and highsalinity stresses [54]. One DRE-like element was predicted in the SiLEA14 promoter. In Arabidopsis, an MYC transcription factor, AtMYC2 and an MYB transcription factor, AtMYB2, have been shown to bind the cis-elements CANNTG and C/TAACNA/G, respectively, to regulate expression of the dehydration-responsive gene RD22 [41,55]. Ten MYC and three MYB-like sequences were identified in the SiLEA14 promoter. Proteins that bind to these elements remain to be isolated. It is necessary to further elucidate the detailed functions of these putative regulatory cis-elements. In addition, pathogen-related elements were also identified. However, whether SiLEA14 participates in biotic stress responses needs to be investigated.

Conclusions
In conclusion, this study characterizes a novel atypical LEA gene SiLEA14 from foxtail millet. SiLEA14 is responsive to ABA, PEG and NaCl and the SiLEA14 is localized in the cytosol. SiLEA14 improves the salt tolerance of E. coli transformant and transgenic Arabidopsis. Furthermore, overexpression of SiLEA14 significantly enhances the salt and drought tolerances of transgenic foxtail millet. SiLEA14 plays important roles in plant abiotic stress resistance and could be used in crops genetic engineering with the aim of improving stress tolerance.

Plant materials and growth conditions
Seeds of foxtail millet (Setaria italica cv. Jigu11), kindly provided by Prof. Xianmin Diao of the Institute of Crop Science, Chinese Academy of Agricultural Sciences, China, were geminated on moist gauze for 24 h at 30°C, and then grown in pots filled with nutrient soil and vermiculite mixed at 1:1 (v/v) in the controlled chamber (25-26°C, humidity 60-70%, under 16-h light/8-h darkness). Foxtail millet leaves, stems, roots, inflorescences and seeds at 5, 15 and 25 days after pollination were harvested and stored at −80°C after frozen in liquid nitrogen.
Seeds of Arabidopsis thaliana (Col-0), after stratification, were plated on ½MS medium with 2% sucrose and 0.7% agar for three weeks at 21-22°C with a 16 h/8 h (day/night) photoperiod and 60-70% relative humidity. Then, the young seedlings were planted on fertilized soil and grown in the same conditions.

Stress treatments
For SiLEA14 expression profile in response to ABA, NaCl and PEG, two-week-old foxtail millet seedlings were carefully removed from soil and washed. The cleaned plants were fixed in plastic foam, and grown hydroponically in water for 1 d. Then, the seedling roots were immersed separately in water containing 100 μM ABA, 250 mM NaCl and 20% (m/v) PEG 6000 for the indicated time in Figure 3, respectively. Six seedlings were used in each treatment. After drying on the filter paper, the seedlings were harvested, and then immediately frozen in liquid nitrogen and stored at −80°C until use.
For GUS expression in response to ABA, NaCl and PEG, about 20 one-week-old proSiLEA14::GUS transgenic Arabidopsis plants were carefully removed from the plates and immersed in water containing 100 μM ABA, 250 mM NaCl and 20% (m/v) PEG 6000 for 0, 3, 6 and 18 h, respectively. For histochemical GUS staining, three-weekold proSiLEA14::GUS transgenic Arabidopsis plants were used.

Cloning and sequence analysis of SiLEA14
The full-length SiLEA14 was amplified by RACE in accordance with the manufacturer's protocol (GeneRacer™ Kit, Invitrogen, Carlsbad, CA, USA). The products were cloned into the pMD19-T vector (Takara, Shiga, Japan) and sequenced. Primer sets used are listed in Additional file 4. The isoelectric point and molecular mass predictions were estimated using the compute pI/Mw tool (http://expasy. org/tools/pi_tool.html). Analysis of protein hydropathy was done by constructing hydropathy plots with the Kyte and Doolittle algorithm (http://ipsort.hgc.jp/) [56]. Motif analysis was performed using the Pfam program (http://www. ebi.ac.uk/Tools/InterProScan/). The grand average of hydropathy (GRAVY) and instability index of deduced proteins were predicted using the ProtParam program (http:// au.expasy.org/tools/protparam.html). Sequence similarities were determined using the BLAST program and the GenBank database on the NCBI web server. The complete amino acid sequences of subgroup 5C LEA proteins were used to construct a phylogenetic tree. Sequence alignment was performed with ClustalW and adjusted manually. A phylogenetic tree was constructed with the neighbor-joining method using the MEGA4.0 program [57]. Sequence logos for subgroup 5C LEA14 were obtained with the WebLogo website http://weblogo.berkeley.edu/logo.cgi [58].
RNA extraction, semi-quantitative PCR and qRT-PCR Total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) and first-strand cDNA was prepared with SuperScript III Reverse Transcriptase (Invitrogen, Carlsbad, CA, USA) after digestion with RNase-free DNase I (Takara, Shiga, Japan). The semi-quantitative RT-PCR were conducted as follows: 95°C for 3 min, then 25 cycles of 95°C for 30 s, 54°C for 30 s, and 72°C for 30 s for both SiLEA14 and Actin2. For qRT-PCR, 100 ng of cDNAs were used as template in a 20 μL reaction system, containing 10 μL 2× SYBR Premix Ex Taq II (TaKaRa, Shiga, Japan), and 0.5 μM each specific forward and reverse primer (Additional file 4). Amplification was performed using the Bio-Rad CFX96 Real-Time PCR System C1000 Thermal Cycler (Bio-Rad, USA) as follows: 95°C for 30 s, 35 cycles of 95°C for 10 s, 60°C for 10 s, and 72°C for 10 s. Arabidopsis Actin2 (accession number: NM_180280) and foxtail millet actin7 (accession number: NM_001280818) were used as the endogenous references. Primers used were listed in Additional file 4.

Subcellular localization of SiLEA14
The coding sequence of SiLEA14 without the terminating codon was amplified and inserted into the XbaI/SmaI sites of pROK219-GFP to generate the construct pROK219-SiLEA14-GFP. Onion epidermal cells were bombarded with the constructs pROK219-GFP and pROK219-SiLEA14-GFP, which were validated by sequencing, using a particle gun-mediated system PDS-1000/He (Bio-Rad, Hercules, CA, USA). Foxtail millet protoplast isolation and transfection were carried out according to the procedure described by Zhai et al. [59]. Root tissues from 7-day-old seedlings were sliced and then incubated in a solution containing 1.5% Cellulase RS, 0.75% Macerozyme R10, 0.6 M mannitol, 10 mM MES, 0.1% BSA and 1 mM CaCl 2 for 4-5 h at 28°C in the dark with gentle swirling (50 rpm). The constructs pROK219-SiLEA14-GFP and pROK219-GFP were incubated with protoplasts and 40% PEG 4000 for 20 min at room temperature for transient transformation, respectively. GFP signals were observed with a confocal laser scanning microscopy (LSM 510, Carl Zeiss MicroImaging GmbH, Jena, Germany).

Assay for salt-stress tolerance of E. coli transformants
The coding sequence of SiLEA14 without the stop codon was amplified and cloned into the EcoRV/XhoI sites of pET30a(+) to construct the expression vector pET30a-SiLEA14, which was then transformed into E. coli host strain BL21. The pET30a(+) empty vector was used as the control. The expression of SiLEA14 in the recombinant cells was confirmed by SDS-PAGE analysis (Additional file 6). Transformed E. coli BL21 cells carrying pET30a-SiLEA14 or pET-30a (+) were grown in LB liquid medium supplemented with 100 μg/ml ampicillin overnight at 37°C, respectively. The bacterial cultures were diluted 100-fold using fresh liquid LB, and allowed to incubate for 2-3 h at 37°C until OD 600 = 0.5-0.6. Isopropylthio-β-D-galactoside was then added to the cultures to a final concentration of 1 mM, and the bacteria were cultured for a further 4 h at 30°C to induce expression of the inserted gene. All induced cultures were adjusted to OD 600 = 0.6 using fresh liquid LB medium with 100 μg/ml ampicillin. To measure responses to salt stress, the samples were diluted by 200-, 500-, 1000-, 2000-and 4000-fold with fresh LB medium supplemented with 100 μg/ml ampicillin. Five microliters of each diluted sample were plated on LB agar plates, LB agar plates supplemented with 600 mM KCl and 600 mM NaCl, respectively. After incubation for 12 h on LB agar plates or 24 h on LB agar plates supplemented with 600 mM KCl and 600 mM NaCl at 37°C respectively, the numbers of colonies were calculated. Growth was measured at least three times.

Generation of transgenic plants
For SiLEA14 overexpression in Arabidopsis, the coding region of SiLEA14 was amplified and ligated into the HindIII/SpeI sites of pSB1300, which was kindly provided by Prof. Shuhua Yang of the College of Biological Sciences, China Agricultural University, China, to generate the construct pSB1300-SiLEA14 in which SiLEA14 was controlled by Super promoter. For SiLEA14 promoter assay, the putative SiLEA14 promoter was isolated from the foxtail millet genome using PCR and cloned into the SalI/EcoRI sites of pCAMBIA1391-GUS to generate the construct pCAMBIA1391-proSiLEA14-GUS. After validation by sequencing, the constructs were introduced into Agrobacterium tumefaciens strain GV1301, and transformed into Arabidopsis by the floral dip method [60]. Seeds were obtained following self-pollination.
For SiLEA14 overexpression in foxtail millet, the coding region of SiLEA14 fused with a flag tag at the 3′end was amplified and ligated into the SacI/KpnI sites of the binary vector pCOU [61], which includes the ubiquitin promoter to drive transgene expression. After confirmation by sequencing, the recombinant plasmid pCOU-SiLEA14-flag was introduced into A. tumefaciens strain LBA4404. Transgenic foxtail millet plants were obtained by Agrobacterium-mediated transformation as described previously [62,63]. Seeds were obtained following self-pollination.
Abiotic stress-tolerance assay of SiLEA14 transgenic Arabidopsis Seeds of WT and T 3 transgenic Arabidopsis plants were surface sterilized by the vapor-phase method [60], and sown on MS, MS + 125 mM NaCl and MS + 250 mM mannitol media, respectively. After 3 days vernalization at 4°C, they were cultured in a controlled chamber (22°C, humidity 40-50%, 120-150 μmol/m −2 s −1 under 16-h light/8-h darkness). Photographs were taken after 7 days culture. Fresh and dry weights of each sample were calculated based on the average weight of 20 individual plants.
Abiotic stress tolerance assay of SiLEA14 transgenic foxtail millet For analysis in the germination stage, 30-40 seeds of both WT and T 2 transgenic millet were geminated on the filter paper in a Petri dish wet with water (as control) or water containing 150 mM NaCl, 250 mM NaCl, 10% PEG, 20% PEG and10 μM ABA for 1 day at 30°C, respectively. Photographs were taken after 4 and 9 days, respectively. Shoot and root lengths were measured.
For salt tolerance assay in soil, two-week-old seedlings were irrigated with water, 150 and 250 mM NaCl solution every 3 days, respectively. After 6 days, the phenotypes of the transgenic lines and WT were investigated. Six to eight plants grown in one plot were used in each experiment. Two fully expanded young leaves from each foxtail millet plant per plot were harvested and cut into 1 cm segment for electrolyte leakage measurement.
For drought tolerance assay in soil, two-week-old seedlings were deprived of water for 7 days. Subsequently, the plants were irrigated with water and grown for 3 days. Six to ten plants grown in one plot were used in each experiment. The survived plants were counted. Above-ground parts of treated seedlings were collected and used to measure proline and soluble sugar contents.

Electrolyte leakage assay
Leaf tissue (0.1 g) from each sample was washed and immersed in 20 mL deionized water with 150 rpm shaking for 16 h. The initial electrical conductivity (L1) of the sample was detected using a FE-30 conductivity meter (Mettler-Toledo, Columbus, OH, USA). Then, the samples were autoclaved at 121°C for 10 min and cooled to room temperature. The ultimate conductivity (L2, maximum conductivity of tissues) was measured. Relative electrical conductivity (L) was calculated as the ratio of L1/L2.

Proline content measurement
Free proline content of foxtail millet plants was measured using the method described by Bates et al. [64]. Leaf tissue (0.1 g, dry weight) was used to extract free proline in 3% sulphosalicylic acid at 95°C for 15 min. Then, 2 mL of supernatant was transferred to a new tube and reacted with 2 mL acetic acid and 2 mL acidified ninhydrin reagent for 30 min at 95°C. Next, 5 ml of toluene was added to the tube with full shaking. The absorbance of the toluene layer was determined at 520 nm.

Soluble sugar content measurement in foxtail millet leaves
Soluble sugar content of foxtail millet plants was examined using the method of Yemm and Willis [65]. Leaf tissue (0.1 g, dry weight) was used to extract soluble sugar in 7 ml of 80% ethanol with constant stirring at 80°C for 2 h. Ethanol was evaporated in the boiled water bath. Then, 1 ml water and 5 ml of 0.15% anthrone solution was added. After incubation at 95°C for 15 min and cooling to room temperature, the absorbance of the reaction solution was determined at 620 nm. Glucose was used as a standard.

Histochemical GUS staining
Histochemical GUS staining was performed as described by Jefferson et al. [66]. After the GUS staining, Arabidopsis seedlings were treated with 70% ethanol to remove chlorophyll from the GUS-stained tissue.

Statistical analysis
The survival rate, fresh/dry weights, relative electrolyte leakage rate, proline content and soluble sugar content data were subjected to Student's t-test analyses using GraphPad Prism 5. All of the experiments were repeated three times.