Transcriptomic profiling of the high-vigour maize (Zea mays L.) hybrid variety response to cold and drought stresses during seed germination

Abiotic stresses, including cold and drought, negatively affect maize (Zea mays L.) seed field emergence and later yield and quality. In order to reveal the molecular mechanism of maize seed resistance to abiotic stress at seed germination, the global transcriptome of high- vigour variety Zhongdi175 exposed to cold- and drought- stress was analyzed by RNA-seq. In the comparison between the control and different stressed sample, 12,299 differentially expressed genes (DEGs) were detected, of which 9605 and 7837 DEGs were identified under cold- and drought- stress, respectively. Functional annotation analysis suggested that stress response mediated by the pathways involving ribosome, phenylpropanoid biosynthesis and biosynthesis of secondary metabolites, among others. Of the obtained DEGs (12,299), 5,143 genes are common to cold- and drought- stress, at least 2248 TFs in 56 TF families were identified that are involved in cold and/or drought treatments during seed germination, including bHLH, NAC, MYB and WRKY families, which suggested that common mechanisms may be originated during maize seed germination in response to different abiotic stresses. This study will provide a better understanding of the molecular mechanism of response to abiotic stress during maize seed germination, and could be useful for cultivar improvement and breeding of high vigour maize cultivars.


Materials and methods
Plant materials and stress treatments. Zhongdi175 (H446 × Q317) is a common maize cultivar in Huang-Huai-Hai summer maize area in China, which has developed a widespread reputation for outstanding the characteristics of strong stress resistance, high and stable yield, wide adaptability and so on (Fig. 1). H446 as its female parent is form elite inbred line Zheng58 hybridization with foreign hybrid X1132 then continuous self-cross from seven generations of breeding under alternate between water and drought selecting environmental with many characteristics, ie stress tolerance, flourishing root system, dry resistance, and high yield's seed production. Q317 as Zhongdi175's male parent is form (H21 × 92-8) × 598 then continuous self-cross multigeneration of breeding with the characteristics of disease-resistant, plant-type compact, high quality and high combining ability. The paternal and maternal traits of Zhongdi175 are complementary and have obvious advantages, especially the previous research indicated that Zhongdi175 is also a representative cultivar of high vigour maize in China (Fig. 1).
In this study, the seeds of Zhongdi175 were provided by Beijing Zhongdi Seed Technology Co. Ltd. and stored in a seed bank (Beijing Kulan Technology Co., Ltd., Beijing, China) at 10 °C, 40% relative humidity (RH). Enough seeds with uniformity and good health were selected for surface sterilization in 1% NaClO (w/v, Beijing Chemical Reagent Company, China) for 10 min, taking out and washing the soaked seeds with distilled water for three times. Two pieces of germination paper which were supplied by Anchor Paper Co., USA, were stacked, thoroughly moistened with distilled water, and then remove the excess water from the paper with a towel. The sterilized seeds were alternately placed in a loose roll of vertical germinating paper (Anchor Paper Co., USA) and incubated in an artificial climate incubator (GXZ-380C) at 25 ± 0.5 °C (12-h light/12-h dark), with nine replicates of 100 seeds per replicate. After four days of seed germination, different treatments were carried out, and untreated seeds/replicates were used as control. The treatments are as follows: (1) cold treatment: three replicates were germinated at 4 ± 0.5 °C (12-h light/12-h dark) for 48 h; (2) drought treatment: seeds of three replicates were transferred to a paper bed containing PEG6000 (20%) solution, then were germinated at 25 ± 0.5 °C (12-h light/12-h dark) for 48 h; (3) control: three replicates continued to germinate for 48 h at 25 ± 0.5 °C (12-h light/12-h dark). For each replicate, ten seedlings with uniformity were randomly selected for RNA-seq analysis, and the other seedlings were used for physiological indices measurement. All the samples were stored in ultralow temperature refrigerator (DW-86L 388J) at − 80 °C.All applicable international, national, and institutional guidelines for the use of plants in the present study were followed.
RNA-seq analysis and sequence assembly. The Illumina HiSeq™ 2500 plant form was applied for RNAseq analysis by Gene Denovo Biotechnology Co. (Guangzhou, China). To obtain a comprehensive overview of Zhongdi175 transcriptome, nine libraries were constructed and pariedend sequencing was carried out. In brief, the mRNA was enriched by Oligo (dT) magnetic beads and the cDNA fragments were purified by QIAquick www.nature.com/scientificreports/   25 . The reconstruction of transcripts was carried out with software Cufflinks 26 , which together with TopHat2, were used to identify new genes and known genes. All the reconstructed transcripts were aligned to the reference annotation by using Cuffcompare. Transcripts with nucleotide more than 200 bp and exon numbers more than two were defined as novel genes. The flow chart of bioinformatics analysis see Supplementary Figure S1.
Functional annotation and classification of transcripts. BLASTX analysis was used to analyze gene function annotation with the non-redundant (Nr) protein sequence database at GeneBank (http:// www. ncbi. nlm. nih. gov), Swiss-Prot (http:// www. expasy. ch/ sprot) and KEGG (http:// www. genome. jp/ kegg). The significant threshold for E-value was set to ≤ 10 −6 . Gene Ontology (http:// www. geneo ntolo gy. org) was employed to get GO enrichment based on biological process, cellular component features, and molecular function, with FDR ≤ 0.05 as a threshold. All genes were blasted against the plantTFDB (http:// plant tfdb. cbi. edu. cn) with a cutoff E-value ≤ 1e −5 to identify putative transcription factors (TFs).

DEGs identification.
To identify DEGs across treatments, the edgeR package (http:// www.r-proje ct. org/) was used. Cleaning reads of each library were rearranged with reference genes using Bowtie software, and the number of mapped reads was calculated by RSEM 27 . The gene expression level was normalized by using fragments per kilobase million (FPKM) method, and the effect of different gene lengths and sequencing data amount on the calculation of gene expression were eliminated 26 . The genes from at least one FPKM ≥ 1 treatment were used for further analysis. The fold change of each gene under cold or drought condition was determined by comparing the FPKM value with that of the control, and the genes with fold change ≥ 2 and false discovery rate (FDR) < 0.05 in the comparison were taken as significant DEGs by edgeR package 28 . qRT-PCR analysis. A total of eight genes with different expression patterns in our Illumina RNA-seq data were randomly chosen to further verify by qRT-PCR. First strand cDNA synthesis and qRT-PCR were performed using PrimeScript™ RT reagent kit with gDNA Eraser (Perfect Real Time) (Takara, Dalian, China) and SYBR® Premix Ex Taq™ II (Tli RNaseH Plus, Shiga, Japan) (TaKaRa), respectively. The reaction was performed on the BIO-RAD CFX96 sequence detection system. The specific primers (Supplementary Table S1) of eight genes were designed with online Integrated DNA Technologies (https:// sg. idtdna. com/ scito ols/ appli catio ns/ realt imepcr/). And ZmActin was used as an internal reference gene. Each PCR reaction (20 µL) contained 10 µL of SYBR® Premix Ex Taq™ II, 10 µM of forward and reverse primers, and 2 µL of template cDNA which diluted 10 folds with deionized water. Three independent biological replications were performed for each sample. Relative expression levels were calculated using the 2 −∆∆CT method 29 . And the regression coefficient between qRT-PCR results and RNA-seq data was analyzed using IBM SPSS Statistics for Windows Version 19.0 (IBM Corp.).

Results
Illumina sequencing, De novo assembly and functional annotation of the Zhongdi175 transcriptome. To comprehensively understand the effects of cold and drought stresses on gene expression of Zhongdi175 during seed germination, the total RNA of seedlings was sequenced by Illumine system. We performed transcriptomics analysis of maize seeds from control, cold and drought environments to investigate the response of plants to abiotic stresses during seed germination, with three replications per environment. Clean data of 35,743,617,000 bp (control), 38,835,975,900 bp (cold) and 35,041,415,400 bp (drought) were obtained. After the adapters, more than 10% of the unknown nucleotides and more than 50% of the low quality (Q-value ≤ 20) bases were removed, a total of 234,397,256 (control), 253,824,644 (cold) and 229,851,862 (drought) high quality clean reads (HQ Clean Reads) were used for assembly ( Table 1). The Q30 scores of all  (Table 1). Transcriptome assembly was carried out using the cufflinks software and 32,269 known and 2358 new genes were gotten, and the proportion of known genes and new genes were 93.2% and 6.8%, respectively (Supplementary Table S4). Based on the blast search against the PlantTFDB, a total of 2248 genes of 56 TF families were identified as transcription factors (Table 2 and Supplementary Table S5). Functional classification of DEGs. GO analysis was performed to determine the function of the identified DEGs. Based on FDR < 0.05, 43, 44 and 42 GO terms were overrepresented in cold, drought and between drought and cold, respectively (Supplementary Tables S10-S15). To control versus cold, the "binding", "catalytic activity", "cellular process", "cell", "cell part", "metabolic process", "single-organism process" and "organelle" were  Table S10). For control vs. drought and cold vs. drought, the mostly dominant terms of them were same to control vs. cold (Fig. 3b,c and Supplementary Tables S10-S15).

Identification and comparison of stress-specific differentially expressed genes (DEGs
In the DEGs identified between stress and control samples, 121 and 119 GO terms were enriched in the comparison of cold and drought to control, respectively. To determine the transcriptomic changes that occur in response to various abiotic stresses, the enriched GO terms under different stress conditions were compared, and all commonly enriched GOs were summarized in Fig. 4. 63 GO terms were enriched in all datasets, of which 30, 22, 11 GO terms belonged to biological process, molecular function and cellular component, respectively (Figs. 4 and 5). In addition, GO:0016491 and GO:0005840 were also commonly enriched in B73 leaves under three (salinity, drought and heat) and two (salinity and heat) types of abiotic stresses, respectively 31 . Under cold and drought stresses, many GO terms are closely related to the stress resistance of Zhongdi175 seed germination (Supplementary Tables S2,S10-S15). For example, the expression of genes encoding peroxidase (entrzID_542505), heat shock protein binding protein (entrzID_100280673) and peroxidase (entrzID_103641351, entrzID_103642599, entrzID_103644044, and entrzID_103647239) in seedlings under cold stress was down-regulated during seed germination of Zhongdi175. The expression of genes regulating heat shock protein (entrzID_103625886) and peroxidase (entrzID_103636702) were up-regulated under cold and drought stresses (Supplementary Tables S10-S15).
In order to further investigate the biological function of these DEGs, we performed pathway enrichment analysis using KEGG. The results showed that 2292 DEGs were enriched in 130 pathways of control vs. cold samples, 1866 were enriched in 128 pathways of control vs. drought samples, and 1048 were enriched in 127 pathways of cold vs. drought samples (Supplementary Tables S16-S18). Compared with drought stress, more DEGs were enriched in germination metabolic pathway of Zhongdi175 seeds under cold stress. It is worth noting that "ribosome", "phenylpropanoid biosynthesis", "biosynthesis of secondary metabolites" and "flavonoid biosynthesis" pathways were significantly enriched in the control vs. cold comparison; the "ribosome", "phenylpropanoid biosynthesis", "biosynthesis of secondary metabolites" and "metabolic pathways" pathways were significantly enriched in maize plants in response to drought stress during Zhongdi175 seeds germination. Moreover, the "biosynthesis of secondary metabolites", "phenylpropanoid biosynthesis", "metabolic pathways" and "photosynthesis-antenna proteins" pathways were significantly enriched in the cold vs. drought comparison (Fig. 5).

Identification of abiotic stress responsive transcription factors.
Additionally, the DEGs encoding the transcription factor (TF) was analyzed. A total of 2248 DEGs encoding cold and drought stress responsive transcription factors (TFs) were detected during Zhongdi175 seeds germination, and these transcription factors belonged to 56 transcription factor families respectively. Most of the identified DEGs encoded members of the bHLH, bZIP, C2H2, ERF, GRAS, MYB, NAC and WRKY related TF families ( Table 2). The MYB family was the largest TF family responding to abiotic stress, with 203 DEGs. A total of 196, 165, 132, 129, 123, 120, and 74 DEGs were identified, belonging to the bHLH, ERF, WRKY, bZIP, NAC, C2H2 and GRAS families respectively. Expression of most MYBs and bHLH was up-regulated; conversely, WRKY was typically down-regulated. For example, the expression of MYB (entrzID_100285825 and entrzID_103634904) and bHLH (entrzID_103631776) was upregulated and WRKY (entrzID_100279570, entrzID_100281558, and entrzID_100285217) was down-regulated under cold and drought stresses (Supplementary Table S5). The different expression patterns of TFs in maize seedlings under cold and drought stress indicated that Zhongdi175 had a wide range of abiotic stress resistance mechanisms during seed germination.     Tables S7-S9). During seed germination, the expression of most of these phytohormone biosynthesis related genes were up-regulated in Zhongdi175 seedlings, but their expression levels were significantly affected by stress conditions. In addition, entrzlD_100272864, entrzlD_100217270 and entrzlD_100281366 were commonly enriched in B73 leaves under abiotic stresses (red stars in Fig. 6) 31 . These results suggest that the biosynthesis of phytohormones is reprogrammed under different abiotic stresses.
Flavonoid metabolism and response to abiotic stresses. Plants produce a myriad of specialized metabolites to abiotic stresses. Flavonoids are a large class of important plant polyphenolic secondary metabolites, which have antioxidant, anti-ultraviolet, anti-plant pathogens and other physiological function 32,33 . In this study, 21 (cold stress) and 23 (drought stress) DEGs in flavonoid metabolism pathways were identified (Fig. 7), of which 15 were common genes. Among the 15 genes, 7 genes were up-regulated and 8 genes were down-regulated under cold and drought stresses. Two genes (entrzID_100381820 and entrzID_542258) encoding phenylalanine ammonia-lyase, two genes (entrzID_100274415 and entrzID_100282642) encoding chalcone synthase and one gene (entrzID_100284998) encoding were identified trans-cinnamate 4-monooxygenase, which were mainly down-regulated by cold and drought stresses. And phenylalanine ammonia-lyase (entrzID_109943525), chalcone isomerase (entrzID_100276821), flavanone 3-hydroxylase (entrzID_542712), flavonoid 3' ,5'-hydroxylase (entrzID_103639113), were up-regulated under cold and drought stresses. In addition, the expression of the gene encoding caffeoyl-Coa methyltransferase, entrzID_100273683, and the gene encoding flavonoid 3-monooxygenase, entrzID_103653707, was up-regulated and down-regulated under low temperature stress, respectively. In particular, we identified an important gene in the flavonoid pathway, entrzID_100127010, encoding leucoanthocyanidin dioxygenase, which was up-regulated under drought stress (Supplementary Tables S7-S9). Moreover, some of the genes identified have also been reported in other studies, such as entrzID_100381820 (Zm00001d017279) encoding phenylalanine ammonia-lyase, which is down-regulated in drought-tolerant maize line YE8112 34 ; entrzID_100274415 (GRMZM2G422750) encodes a chalcone synthase that is down-regulated in the ovary by drought in maize inbred line B73 to improve drought tolerance in maize 35 .
To cell-wall metabolism related genes, the expression of entrzID_542649 encoding beta-expansin precursor, was up-regulated under drought stress, and entrzID_541807, entrzID_542685 and entrzID_542139 encoding cellulose synthase, were up-regulated under both cold and drought stresses. And the expression of hypocotyl elongation protein ortholog gene (entrzID_100192868) was up-regulated in Zhongdi175 seedlings under cold stress (Supplementary Tables S7-S9). qRT-PCR verification. Eight DEGs were randomly selected for qRT-PCR analysis of control and stress samples to verify the reproducibility of the gene expression data of Zhongdi 175 seedlings obtained by RNA-seq analysis (Supplemental Table S1). The ratio of expression levels found between stressed samples and controls using qRT-PCR was compared to the ratio of expression measured by RNA-Seq. A significant correlation was observed between RNA-Seq and qRT-PCR data (r 2 = 0.819, p < 0.001, Fig. 8), which confirms the effectiveness of DEGs in this study (Fig. 8). Therefore, the comparison of qRT-PCR and RNA-Seq analysis data of Zhongdi175 seedlings fully verifies the results of our transcriptome study.

Discussion
Abiotic stresses (such as cold, drought, salt, etc.) seriously affect crop growth, development and yield. Worldwide, abiotic stresses cause major crop yields to drop by more than 50% 4 . Especially, spring sowing maize is often affected by low temperature, drought and other harsh environments. Zhongdi175 is the representative of spring sowing maize varieties with low temperature tolerance and drought resistance in arid and semi-arid areas (Fig. 1). Understanding of the molecular mechanism of Zhongdi175 in response to abiotic stress would be helpful for the breeding of high vigour cultivars. With the development and maturity of high-throughput transcriptome sequencing technology, RNA-seq technology has been successfully applied to transcriptome or genome-wide analysis of rice 37 , maize 38,39 , wheat 40 and other crops. In particular, this study may help us to better understand the molecular basis of maize seed germination response to abiotic stress.
Overview of Zhongdi175 transcriptome. Seed germination (generalized) includes four stages: imbibition, protrusion, germination and seedling establishment 30 . In this study, we selected the seedling establishment of seed germination, because the germination stage of maize is particularly sensitive to abiotic stress 15 . A total of 32,269 known and 2358 new genes were identified by transcriptomic analysis in maize seedling during seed germination under different conditions (Supplementary Table S4). Under cold stress (4 °C), 5300 and 4305 DEGs were up-regulated and down-regulated, respectively (Supplementary Table S7). And the changes of gene expression at the whole transcriptome level indicated that the gene expression was up-regulated (5239) and downregulated (2598) under drought stress (Supplementary Table S8). Here we performed comparative transcriptome profiling of Zhongdi175 in response to cold and drought stresses, and found that an activation of gene network was involved during seed germination. In particular, 5143 co-regulated DEGs by two stresses, 4462 specific to   2017), more DEGs were found in this study, which may be caused by different maize varieties and different growth stages. The co-regulated DEGs belongs to a subset of key genes for cold and drought resistance in maize seed germination, which may play an important role in the adaptation to stress during maize seed germination. Based on the functional annotation of DEGs, the responses of maize seed germination to abiotic stresses were revealed to involve multiple biological pathways, such as hormone metabolism and signal transduction, transcriptional regulation and flavonoid metabolism.

DEGs encoding transcription factors. Transcription factors (TFS) play an important role in stress signal
transduction pathways, which regulate the expression of specific stress-responsive genes in plants 22,45 . Major plant TF families, including MYB, NAC, and WRKY, as important regulatory factors in plant response to various abiotic and biotic stresses, have attracted much attention 12,46,47 . In this study, at least 2248 TFs in 56 TF families were identified that were involved in drought and/or cold treatments during seed germination. According to the number of genes in each TF family identified, the top eight TF families in order were MYB, bHLH, ERF, WRKY, bZIP, NAC, C2H2, and GRAS (Table 2). Different members of these families were up-regulated or down-regulated under cold or/and drought stress, such as ARF, bHLH, bZIP, GRAS, MYB, NAC, WRKY. The MYB TF family was the largest class in response to abiotic stress in Zhongdi175 seed germination. MYBs involve in regulating secondary metabolism and cell morphology, and are response to biotic and abiotic stresses 46,48 , some of which regulate flavonoid biosynthesis and involve in abiotic stress responsiveness in tobacco 49,50 55 . The bZIP and GRAS families are another group of TFs involved in plant tolerance to abiotic stresses, in particular, to cold and drought stresses 12 . And the transcription factor OsABF1 (bZIP), as a binging factor of ABA response element, can enhance abiotic stress signal transduction in rice 56 . SlGRAS40 enhances tomato tolerance to abiotic stress by regulating auxin and gibberellin signal transduction 57 . These results indicated that TFS plays an important role in improving the ability of maize and other crops to resist abiotic stress during growth and development.
DEGs involved in hormone metabolism and signal transduction. Plant hormones are important growth regulators that regulate plant growth, development, nutrient allocation and source-sink conversion to adapt to stress environment. To date, many studies have shown that different plant hormones, such as ABA, GAs, IAA, CKS, JA, BRs and ETH, control many physiological functions and biochemical processes in plants, including seed germination 18,30,58 .
ABA is the key phytohormone of plants, which regulates gene expression, protein synthesis, signal transduction, and ion transport in response to abiotic stresses such as drought and photoinhibition 59 . Hyperosmotic stress caused by drought or salt stress leads to the accumulation of ABA, and then quickly triggers the downstream response of plants 18,22 . As an example, de Zelicourt et al. (2016) found that cold stress could induce the biosynthesis of ABA, and exogenous ABA can improve the cold tolerance of plants to a certain extent. And previous studies have shown that the bZIP transcription factor OsABF1 is involved in ABA signaling transduction in rice under abiotic stress 56 , NAC family TFs (such as OsNAP, OsNAC54) mediates rice response to abiotic stress through ABA pathway 60 . In the current study, XLOC_046985, entrzID_100501444 (up-regulated under cold stress), entrzID_100501454 (up-regulated under drought stress) and entrzID_732819 (up-regulated under cold and drought stresses) encoding NCED, was identified during Zhongdi175 seed germination (Fig. 6), which is consistent with the results of NCED1 (c63716_g5, up-regulated by drought) reported previously in Zea mays ssp. mexicana L. 44 and the results of GRMZM2G407181 (entrzID_100501444) and GRMZM2G110192 (entrzID_100501454) encoding NCED up-regulated under drought stress reported by Kakumanu et al. (2012), suggesting conserved regulation of this drought induced response and ABA-dependent pathways play a role in maize response to drought. Therefore, bZIP/NAC-ABA pathways play critical roles in the response to cold or/and drought stresses during seed germination of Zhongdi175. Our results further indicated that ABA biosynthesis played an important role in plant response to environmental stresses. Gibberellins (GAs) play an important role in regulating plant growth and development, including seed germination and stem elongation 61 . GAs are also involved in plant tolerance to abiotic stresses, e.g. the effect of exogenous gibberellin (GA4 + 7) on soybean under flooding stress, and found that GAs could improve plant stress resistance during short-term flooding 62 . And GRAS is involved in plant tolerance to abiotic stress and affects auxin and gibberellin signal transduction 57 . Here, two gibberellin 2-beta-dioxygenase genes (entrzID_100273040 and entrzID_100280480), one ent-kaurene oxidase gene (entrzID_103638615), and two gibberellin 20-oxidase genes (entrzID_100283148 and entrzID_100284800) were regulated by cold or/and drought (Fig. 6), which may play a role in the catabolic pathway of gibberellin by 2β-hydroxylation 63 . And a previous study found that, gene expression of GA 2-beta-dioxygenase (c57017_g1, c33506_g1, and c57017_g2), precursor of GAs receptor GID1L2 (c47232_g1) and GA 20-oxidase (c63822_g4, c49254_g1, and c56760_g2) were up-regulated under cold stress 44 , of which entrzID_100284800 (c63822_g4) encoding gibberellin 20-oxidase gene was up-regulated by drought stress in this study. Therefore, GRAS-GA/auxin pathways may play a key role in the respond of Zhongdi175 seed germination to cold or/and drought stresses.
Auxin is another phytohormone, which is the early detection of plant hormones, widely involved in plant growth and development, including plant responses to abiotic stress 64 . Here, five auxin-responsive Aux/IAA genes (entrzID_100193444, entrzID_100274580, entrzID_100283579, entrzID_100285630, and entrzID_100286028) and five auxin-related genes, indole-3-acetaldehyde oxidase (entrzID_542228 and entrzID_542229), 3-hydroxyindolin-2-one monooxygenase (entrzID_100382554), indole-2-monooxygenase (entrzID_100192631), indole-3-glycerol phosphate synthase (entrzID_100286258) and indole-3-glycerol phosphate lyase (entrzID_542117) which may play an important role in stress resistance during Zhongdi175 seed germination. The genes of entrzID_542117, entrzID_100192631 and entrzID_100382554 were up-regulated under cold and drought stresses; the expression of entrzID_100274580 was down-regulated under cold stress; the expression of entrzID_100193444 and entrzID_100285630 was down-regulated under cold and drought stresses; the expression of entrzID_100286028, entrzID_100283579, entrzID_542228, and entrzID_542229 was up-regulated under cold stress; entrzID_100286258 was up-regulated under drought stress ( Fig. 6 and Supplementary Tables S7-S9). Auxin/indole-3-acetic acid proteins are widely involved in plant growth and development through auxin signaling pathway 65 . The stress pathway interacts with the auxin gene regulatory network through the transcription of the Aux/IAA genes, which acts as the hub of integrating genetic and environmental information to achieve plant stress resistance 66 . ARF family is a well-known family in the auxin signal transduction pathway, which regulates the transcription of auxin-induced genes. Recent evidence suggests that the ARF family also functions in plant www.nature.com/scientificreports/ responses to abiotic stresses [67][68][69] . Thus, the ARF-Aux/IAA pathway may play an important role in response to cold and/or drought stresses in Zhongdi175 seed germination. In grasses, indoles are converted to indolin-2-ones by the P450 enzyme BX2 67 . In this study, genes encoding cis-zeatin-O-glucosyltransferase (entrzID_541881 and entrzID_103636476) and cytochrome P450 (entrzID_103630138, entrzID_103635519, entrzID_103640646, and entrzID_103647886) were identified, which were involved in the tolerance of Zhongdi175 to cold and/or drought stresses ( Fig. 6 and Supplementary Tables S7-S9) 73 . In the present study, 12 putative flavonoid metabolism-related genes might be important roles in responding to cold or/and drought stresses were identified (Fig. 7). The genes entrzID_100282642 and entrzID_100274415 encoding CHS, were down-regulated in Zhongdi175 seedlings under cold and drought stresses; entrzID_100273683 was up-regulated under cold stress, encoding caffeoyl-CoA O-methyltransferase ( Fig. 7 and Supplementary Tables S7-S9), which are two key enzymes of the flavonoid/isoflavonoid biosynthesis pathway 74,75 . In particular, the expression of CHS gene is induced in plants by stresses such as UV light, bacterial or fungal infection, and the like 74 . Other nine genes (entrzID_100381820, entrzID_542258, entrzID_109943525, entrzID_100284998, entrzID_100276821, entrzID_100127010, entrzID_542712, entrzID_103653707, and entrzID_103639113), encode PAL, C4H, CHI, LDOX, F3H, F3′H, F3′5′H, respectively (Fig. 7), of which entrzID_100381820 (Zm00001d017279) and entrzID_100274415 (GRMZM2G422750) down-regulated by drought in maize were reported 34,35 . Besides, MYB and/or bHLH TFs always involves regulated the phenylpropanoid or flavonoid biosynthetic pathways 76 . Therefore, MYB-flavonoid or MYB-bHLH-flavonoid biosynthetic pathways may also be important for Zhongdi175 in response to cold or/and drought stresses during seed germination. This study will provide better understanding of the molecular regulation mechanism of flavonoid metabolism in maize seed germination resistance to cold and drought stresses.
DEGs involved in sucrose metabolism and cell growth promotion. Genes encoding enzymes related to sucrose metabolism, such as galactinol synthase, sucrose synthase, trehalose phosphate phosphatase and trehalose phosphate synthase, were induced under abiotic stresses 18,22 . For example, galactinol synthase genes (AtGolS1, 2 and 3), AtGolS1 and 2 were induced by drought and high-salinity stresses, while AtGolS3 was induced by cold stress in Arabidopsis 36 ; sucrose synthase genes (GRMZM2G089713 and GRMZM2G318780) were up-regulated in ovary of maize inbred line B73 under drought stress 35 ; overexpression of trehalose-6-phosphate phosphatase gene OsTPP1 improved salt and cold tolerance of rice 77 ; overexpression of trehalose-6-phosphate synthase gene OsTPS1 enhanced the tolerance of rice seedlings to abiotic stresses 78 .
Genes encoding fructokinase (entrzID_542108) and hexokinase (entrzID_100170246), which play a role in sugar signaling 79 and enzymatic function, were regulated under cold and drought stresses in Zhongdi175 seedlings. These genes were also induced by drought stress in maize inbred line B73 ovary 35 . The results show sucrose metabolism is vital for the adaptation of Zhongdi175 to abiotic stresses during seed germination.
Cell-wall metabolism related genes were in response to cold and drought stresses in Zhongdi175 seedlings. For example, beta-expansin 7 (entrzID_542649, Zm00001d029906) was induced in sensitive drought and tolerant drought maize 34 . It is well known that β-expansin is a key regulator of cell wall modification during tissue elongation 80,81 . Overexpression of wheat beta-expansin gene TaEXPB23 enhances tobacco root growth and water stress tolerance 82 . The cellulose in the primary cell wall, which is laid down during plant growth, determines the shape of the plant 83,84 . Chen et al. (2005) found that disruption of the cellulose synthase gene AtCesA8/IRX1 was found to enhance the tolerance of Arabidopsis to drought and osmotic stresses. In additon, the expression of entrzID_100192868 was up-regulated under cold stress in this study (Supplementary Tables S7-S9), which was consistent with that the expression of entrzID_100192868 was up-regulated in drought-sensitive and droughttolerant maize 34 . Therefore, sucrose metabolism could help plants to maintain normal growth under stress conditions, and cell wall remodeling can help the cells conserve water, thus contributing to better adaptation to abiotic stresses during maize seed germination.
Common and unique molecular mechanisms of response to different abiotic stresses in maize. Plants often simultaneously adapt to different adverse environmental conditions and evolve common mechanisms to respond to various abiotic stresses 85 . In our dataset, many up-regulated DEGs were found in both stress samples (cold and drought), however, the functions of most of the down-regulated genes remain  Tables S7-S9). Here, we screen and obtained some important DEGs related to transcriptome factors (Table 2), hormone metabolism and signal transduction, flavonoid metabolism, sucrose metabolism and cell growth promotion and so on. This study laid an important foundation for us to better understand the mechanism of maize seed germination in response to cold and drought stresses. In agricultural production, many kinds of abiotic stresses occur simultaneously 85 . At present, there are some report on that transcriptome of maize plants in response to abiotic stresses. Transcriptome comparative analysis based on different maize materials, different growth stages, different tissues and different adverse environments may play an important role in clarifying some common molecular mechanisms of plant abiotic stress response. For example, Li et al. analyzed the transcriptome of B73 seedling leaves under abiotic stresses, we compared the relevant results of this study with them, and found that GO:0016491 and GO:0005840 were also enrich in B73 leaves under multiple stress conditions (Fig. 4b) and entrzlD_100272864, entrzlD_100217270 and entrzlD_100281366 were also commonly enriched in B73 leaves under abiotic stresses, which involved in abscisic acid, ethylene and jasmonic acid biosynthesis and signal transduction pathways, respectively (Fig. 6). Relevant work needs to be further carried out in the later stage.

Conclusions
In this study, RNA-Seq was used to detect the whole transcriptional changes of maize seedlings under abiotic stresses during seed germination. Totally, 12,299 DEGs were obtained, with 9605 and 7837 DEGs responding to cold and drought stresses, respectively. Among them, 5143 DEGs were regulated by both stresses, indicating that there were many common and unique molecular mechanisms in the resistance of Zhongdi175 seed germination to different abiotic stresses. Genes related to TFs, hormone metabolism and signaling, flavonoid metabolism, sucrose metabolism and cell growth promotion were found to involve in the resistance to cold and/or drought stresses in Zhongdi175 seedlings during seed germination. Importantly, a total of 2248 transcription factor (TF) genes from 56 TF families were identified; the identified DEGs mainly enriched in ribosome, phenylpropanoid biosynthesis and secondary metabolites biosynthesis pathways under and/or drought stresses. Furthermore, some important genes and pathways have been found, such as sucrose metabolism and cell growth promotion genes, MYB-flavonoid or MYB-bHLH-flavonoid biosynthetic pathways, ARF-Aux/IAA, bZIP/NAC-ABA and GRAS-GA/auxin pathways, etc. This research expanded the understanding of the molecular mechanism of high vigour maize seed germination (seedling establishment) stress resistance, and provided the basis for the selection of major candidate genes and molecular markers of maize stress resistance and the breeding of high vigour maize cultivars.