Regulatory networks of gene expression in maize (Zea mays) under drought stress and re-watering

Drought can severely limit plant growth and production. However, few studies have investigated gene expression profiles in maize during drought/re-watering. We compared drought-treated and water-sufficient maize plants by measuring their leaf relative water content, superoxide dismutase and peroxidase activities, proline content, and leaf gas exchange parameters (photosynthetic rates, stomatal conductance, and transpiration rates). We conducted RNA sequencing analyses to elucidate gene expression profiles and identify miRNAs that might be related to drought resistance. A GO enrichment analysis showed that the common DEGs (differently expressed genes) between drought-treated and control plants were involved in response to stimulus, cellular process, metabolic process, cell part, and binding and catalytic activity. Analyses of gene expression profiles revealed that 26 of the DEGs under drought encoded 10 enzymes involved in proline synthesis, suggesting that increased proline synthesis was a key part of the drought response. We also investigated cell wall-related genes and transcription factors regulating abscisic acid-dependent and -independent pathways. The expression profiles of the miRNAs miR6214-3p, miR5072-3p, zma-miR529-5p, zma-miR167e-5p, zma-miR167f-5p, and zma-miR167j-5p and their relevant targets under drought conditions were analyzed. These results provide new insights into the molecular mechanisms of drought tolerance, and may identify new targets for breeding drought-tolerant maize lines. Abbreviations leaf relative water content: RWC, superoxide dismutase activity: SOD, peroxidase activity: POD, proline content: Pro, photosynthetic rates: Pn, stomatal conductance: Cond, transpiration rates: Tr.; quantitative real-time polymerase chain reaction: qPCR; abscisic acid; ABA; polyethylene glycol :PEG; Principal component analysis :PCA; polyacrylamide gel electrophoresis :PAGE Highlight The study of physiology and molecular mechanism of maize laid a theoretical foundation for drought resistance breeding under drought stress and re-watering.

were analyzed for each gene. To investigate the effect of drought stress on maize, the plants were exposed to 96 2 3 4 hours of water deficit, and then re-watered. Samples were collected from the the CK group than in drought-stressed group in the first 96 hours (P <0.01) (Fig. 1A).

4 0
The differences in RWC between the drought-stressed and CK plants increased until 2 4 1 96 hours, then decreased after re-watering (Fig. 1A).

4 2
Drought stress leads to the accumulation of reactive oxygen species (ROS), which between drought-stressed and CK plants was at 96 h ( Fig. 1B). After re-watering, the 2 4 9 SOD activities were approximately equal in the CK and drought-stressed groups. The 2 5 0 POD activity was lower in the drought-stressed group than in the CK group in the first 2 5 1 96 hours (Fig. 1C), but higher in the drought-stressed group than in the CK group at 3 2 5 2 d after re-watering.

5 3
To evaluate drought adaptation, we measured gas exchange parameters. The corresponding values in the control. After re-watering, the stomatal conductance and 2 5 7 transpiration rate recovered rapidly so that their levels in the drought-stressed group 2 5 8 were equal to those in the CK group. The photosynthetic rate in the drought-stressed 2 5 9 also increased after re-watering, but not to the same level as that in the CK group. To study the involvement of regulatory miRNAs in the complex process of the maize Through target prediction, a total of 1028 miRNAs were predicted to regulate 11,118 2 7 0 genes with 22,677 target loci. We constructed a normalized cDNA library using a mixed pool of equal amounts of with lengths of at least 2 × 150 nucleotides (Supplementary Table S1).

7 7
The resulting reads were aligned to the Z. mays genome that was retrieved from respectively, between the CK group and the drought-treated group (Fig. 2B).

9 6
The miRNA sequencing results showed that in the first 60 hours of drought, 58 2 9 7 miRNAs were up-regulated and 163 miRNAs were down-regulated. In T96 versus 2 9 8 CK96, 106 miRNAs were up-regulated and 120 were down-regulated. In TR3d versus 2 9 9 CKR3d, 97 miRNAs were up-regulated and 118 were down-regulated (Fig. 2B). According to their functional classifications, we compared the significantly 3 0 1 regulated genes at each of the three time points (Fig. 3). On the whole, the carbohydrate synthesis and metabolism (Table 1). At 96 hours, the differentially 3 0 5 regulated genes were mainly related to phenylpropanoid biosynthesis and carbon 3 0 6 fixation in photosynthetic organisms. After re-watering, the differentially regulated 3 0 7 genes were mainly related to carbon metabolism and biosynthesis of amino acids. At pathways of carbon metabolism were re-activated. during drought, therefore, adaptation of membrane lipids to drought conditions might 3 1 8 be a key mechanism for drought tolerance. Significantly more genes involved in cell 3 1 9 wall formation were up-regulated in the first 60 hours of drought stress, but three novel miRNAs (novel-m0661-5p, novel-m0125-3p, and novel-m0283-3p) and five known miRNAs (miR9760-5p, miR5809-5p, miR5076-5p, miR6214-3p, and 3 2 8 miR5783-5p).

2 9
Proline metabolism pathways in maize under drought stress and re-watering Proline accumulation has long been associated with stress tolerance in plants. Proline is assumed to aid in osmotic adjustment in response to drought, however, it 3 3 2 also plays roles in reactive oxygen species (ROS) scavenging and membrane stability. (encoding polyamine oxidase), and GRMZM2G054224 (encoding P4H), respectively. have shown that under drought stress, crops accumulate high levels of ABA,  The ABA signaling mechanism is conserved in economically important plants, residues that comprise the gate-latch-lock components are highly conserved across ABA-dependent or -independent gene regulation (Fig. 6A). An analysis of the miR5783-5p suggested that miR5783-5p plays a vital role in the regulation of the genes involved in the ABA signal transduction pathway were selected for qRT-PCR  The results of the fluorescent quantitative analysis of miR6214-3p were consistent with the sequencing results (Fig. 9). We used the Cytoscape platform to build the 3 8 7 network between the drought-responsive miRNAs and their targets. This allowed us to 3 8 8 study the regulation of miR6214-3p on multiple target genes and explore its 3 8 9 1 5 mechanism under drought stress. It was found that miR6214-3p may be involved in 3 9 0 the regulation of the cell wall (Fig. 4B) and pro-anabolic pathways (Fig.5A), it may 3 9 1 regulate DREB-related genes involved in ABA signaling (Fig.6A), and it may also 3 9 2 regulate genes encoding chlorophyll a-b binding protein, as well as the BHLH and 3 9 3 MYB transcription factors (Fig. 7). All of these genes were DEGs under drought 3 9 4 stress ( Supplementary Fig. S4), suggesting that they may play important roles in the 3 9 5 drought response in maize.

9 6
We selected the six miRNAs showing the largest differences in expression 3 9 7 between drought and CK conditions, and identified their targets. Among them, drought response and tissue development (Fig. 8). These genes encoded proteins Zma-miR529-5p, miR5072-3p, zma-miR167f-5p, ma-miR167j-5p, and  To confirm the reliability of miRNAs identified from the sequencing data, four 4 0 7 key miRNAs were selected randomly for qRT-PCR using the specific primers listed in 4 0 8 Supplementary indicating that zma-miR167f-5p, miR5072-3p -3p, zma-miR167j-5p, and 4 1 0 zma-miR167e-5p were down-regulated under drought stress. After re-watering, 4 1 1 zma-miR167f-5p, miR5072-3p, zma-miR167j-5p, and zma-miR167e-5p were 4 1 2 up-regulated (Fig. 9). Drought is one of the main factors limiting plant growth and production. In practice, speed of responses to drought, depending on their genetic and ecotypic backgrounds. However, a number of drought responsive genes are conserved across plant taxa,  levels at 3 days after re-watering (Fig. 1B, C). This is consistent with a previous 4 4 0 report, which noted that increased SOD and POD activities to quench ROS were  The primary reaction under drought stress is to modulate the aperture of stomata. results showed that the photosynthetic rate (Fig. 1D), stomatal conductance (Fig. 1E) 4 5 3 and transpiration rate (Fig. 1F) were decreased by degrees under drought stress 4 5 4 compared with CK, then recovered to normal levels after re-watering ( Fig. 1 D-F). We detected 221, 226, and 215 differentially expressed miRNAs between 4 5 7 drought-treated and CK plants at 60 h, 96 h, and 3d, respectively (Fig. 2B). Of these 4 5 8 DEGs, 478 were commonly shared among the three groups ( Fig. 2A). The GO  Water stress signaling induces cellular protection processes in leaves The cell wall is the first line of defense against abiotic stress. Many proteins that are increased rapidly as the drought treatment continued (Fig. 4A, B). The differential of hypoxia in various organisms. That is, proline hydroxylation targets HIF for rapid  oxidative stress (Zhang et al., 2012;Kotchoni et al., 2012). In this work, in addition to 5 0 6 the novel genes XLOC_020588 and XLOC_023274, the genes GRMZM2G058675, We determined the expression profiles of genes involved in proline biosynthesis in of the plant to retain water, and that ABA accumulates to higher levels in 5 2 0 drought-tolerant strains than in susceptible ones. In this study, many of the DEGs ABA-dependent or -independent gene regulation (Fig. 6A). The binding of ABA 5 2 4 molecules to their receptors leads to the inhibition of PP2C, which in turn activates 5 2 5 SNF1-related protein kinase 2 (SnRK2) (Ng et al., 2011;Soon et al., 2012). SnRK2 is 5 2 6 an important signaling molecule that phosphorylates its downstream targets, including drought-responsive manner (Yoshida et al., 2010;Bechtold et al., 2013). A previous 5 3 4 report indicated that the transcription factor MYB, the target of miR858, leads to under drought stress. Its transcript levels then decreased after re-watering (Fig. 6A).

3 9
ZmMYB31 is involved in the drought stress response via the ABA-dependent It has been reported that miRNAs are involved in biological processes such as growth maize miRNAs under stress conditions. In this study, we searched for genes related to 5 5 1 drought resistance that are controlled by miRNAs. These results will not only help us 5 5 2 to understand the mechanism of drought resistance in maize, but also identify 5 5 3 potential targets to breed new drought-resistant maize varieties. In addition, the 5 5 4 discovery of miRNAs related to drought tolerance in maize has good prospects for In a previous study, the down-regulation of miR6214-3p was associated with the 5 5 9 activation of the stress response and antioxidant system in plants (Lu et al., 2015). In 5 6 0 this work, we found that miR6214-3p was down-regulated under drought stress and 5 6 1 up-regulated at 3 days after re-watering (Fig. 9). The predicted targets of miR6214-3p 5 6 2 were involved in the metabolism and signaling pathways of the drought tolerance 5 6 3 response (Fig. 7).

6 4
The BHLH family is one of the largest families of plant transcription factors, and     Asterisks indicate statistically significant difference between groups (Student's t-test): *p<0.05, **p<0.01, ns: no significant difference.