Skip to main content
Download PDF

Molecular mechanisms regulating glucose metabolism in quinoa (Chenopodium quinoa Willd.) seeds under drought stress

BMC Plant Biology volume 24, Article number: 796 (2024) Cite this article

Abstract

Background

Abiotic stress seriously affects the growth and yield of crops. It is necessary to search and utilize novel abiotic stress resistant genes for 2.0 breeding programme in quinoa. In this study, the impact of drought stress on glucose metabolism were investigated through transcriptomic and metabolomic analyses in quinoa seeds. Candidate drought tolerance genes on glucose metabolism pathway were verified by qRT-PCR combined with yeast expression system.

Results

From 70 quinoa germplasms, drought tolerant material M059 and drought sensitive material M024 were selected by comprehensive evaluation of drought resistance. 7042 differentially expressed genes (DEGs) were indentified through transcriptomic analyses. Gene Ontology (GO) analysis revealed that these DEGs were closely related to carbohydrate metabolic process, phosphorus-containing groups, and intracellular membrane-bounded organelles. The Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis detected that DEGs were related to pathways involving carbohydrate metabolisms, glycolysis and gluconeogenesis. Twelve key differentially accumulated metabolites (DAMs), (D-galactose, UDP-glucose, succinate, inositol, D-galactose, D-fructose-6-phosphate, D-glucose-6-phosphate, D-glucose-1-phosphate, dihydroxyacetone phosphate, ribulose-5-phosphate, citric acid and L-malate), and ten key candidate DEGs (CqAGAL2, CqINV, CqFrK7, CqCELB, Cqbg1x, CqFBP, CqALDO, CqPGM, CqIDH3, and CqSDH) involved in drought response were identified. CqSDH, CqAGAL2, and Cqβ-GAL13 were candidate genes that have been validated in both transcriptomics and yeast expression screen system.

Conclusion

These findings provide a foundation for elucidating the molecular regulatory mechanisms governing glucose metabolism in quinoa seeds under drought stress, providing insights for future research exploring responses to drought stress in quinoa.

Peer Review reports

Background

Quinoa (Chenopodium quinoa Willd., 2n = 4x = 36) is an annual dicotyledonous and often cross-pollinating plant of the Amaranth family that has been cultivated for over 5000 years. It is the traditional food of the Inca indigenous humans, known as the “golden grain” and the “mother of food” [1,2]. Quinoa seeds contain unique trace and macronutrients, such as carbohydrates, and their secondary metabolites have profound health benefits [3]. Quinoa contains 58.1–64.2% starch and has a low glycemic index of 35. The starch consists of amylose and amylopectin, and has smaller particles than ordinary grains [4]. Quinoa exhibits rich genetic diversity and extensive ecological adaptability, tolerating drought, salinity, and low temperatures [5]. Inadequate water table is a global problem that threatens crop cultivation [6]. Arid and semi-arid areas cover about one-third of the global land area, with many regions facing periodic climatic droughts [7]. Water stress could impact metabolic activities like glucose metabolism in plants, as well as cell growth and elongation [8]. The primary response of crops to drought stress is osmotic regulation [9]. This regulation involves the synthesis and accumulation of osmotic substances, such as amino acids, betaine, and carbohydrates [10]. Water-soluble carbohydrates are crucial in osmotic protection, signal transduction, and carbon storage [11]. Under drought stress, soluble carbohydrates serve as a signal molecule, regulating the expression of many key genes in plant defense and metabolic processes, thereby controlling plant resistance and growth [12]. For instance, under drought conditions, drought-tolerant genotype had higher osmotic adjustment ability and stronger root activity than sensitive genotype in quinoa [13]. Besides, drought stress significantly enhanced total soluble sugars content in all quinoa genotypes [14]. Genome-wide identification and analysis of the tobacco (Nicotiana tabacum L.) invertase gene family revealed that NtNINV10 played a role in glucose metabolism, and inhibiting its expression reduces glucose and fructose content in leaves [15]. Accumulation of osmolytes in the leaves of drought-stressed mainly included inositol, glucose, fructose and sucrose in sunflower [16]. CsHXKl was apparently up regulated in the leaves and roots under drought stress in Camellia sinensis [17]. Some key metabolites, including galactose, glucose, fructose, sucrose, succinic acid, and malic acid, were identified under drought stress in Thymus mongolicus L. [18]. The TCA cycle is one of the most important metabolic cycles that provide cellular energy [19]. It has been reported that the fertilization of citric acid could improve drought resistance in Nicotiana tabacum L. [20]. Drought stress increased the expression of isocitrate dehydrogenase (IDH) in Pisum sativum L. [21]. NADP(+)-dependent isocitrate dehydrogenase (ICDH; EC 1.1.1.42) was the only enzyme whose activity increased as a result of water deficit in pea [22]. In Arachis hypogaea L., the WRKY transcription family could improve drought resistance by up-regulating AdWRKY40 and down-regulating AdWRKY42, AdWRKY56, and AdWRKY64 expression [23]. Drought stress upregulated SlbHLH96, while overexpressed SlbHLH96 showed stronger osmotic stress resistance in Solanum lycopersicum L. [24]. Selection of genes related to abiotic stress resistance at the genome scale is a necessary and important condition for breeding resistant plants [25]. Current studies have identified genes involved in abiotic stress tolerance on a large scale, based on gene and metabolite expression [26]. RNA-seq and metabolomics analyses are often used to screen abiotic stress-related genes. Huan et al. [27] used transcriptomics and targeted metabolomics to study the leaves of drought-tolerant quinoa seedlings after drought and rehydration treatment. DEGs and DAMs involved in starch and sucrose metabolism and flavonoid biosynthesis were crucial to improve the drought tolerance in quinoa. LOC110713661 and LOC110738152 may be the key genes of drought tolerance in quinoa. However, omics selection of drought stress response genes is not a direct method for identifying drought resistance genes, as some differentially expressed genes may not show drought resistance or drought sensitivity in drought tolerance tests [28]. Previous studies have shown that using yeast expression systems to select genes is a way to quickly identify genes with specific functions [29]. Saccharomyces cerevisiae is a single-celled organism with the best research background. Compared with Arabidopsis Thaliana, it is simpler and easier to use yeast for functional screening of genes related to stress tolerance [30]. Yeast expression systems have been shown to be an effective method of screening for resistance genes in plants [31]. Large-scale yeast functional screening system was used to identify salt-tolerant genes in Tamarix hispida [32]. Yeast expression system combined with high-throughput sequencing technology was used to identify cold-tolerance genes, 60 of which belonged to DEGs and were significantly induced or inhibited under low temperature stress in peanut [33]. With advances in sequencing technology, researchers were currently experimenting with yeast expression systems combined with high-throughput sequencing techniques to quickly screen different types of functional genes at the genomic level [33]. Currently, little is known about glucose metabolism in quinoa, particularly under drought stress. The DEGs and DAMs related to glucose metabolism pathway were explored by transcriptome-metabolomics analysis in this study, and candidate drought response genes were verified by qRT-PCR combined with yeast expression system. This results could provide reference for further analysis of the molecular regulatory mechanism under drought stress in quinoa.

Results

Differences in phenotype, anatomical structure, and glucose components of quinoa seeds

We used three different concentrations (5%, 15%, and 25%) of PEG to select an optimal PEG concentration to induce drought stress in quinoa seeds. At 5% PEG treatment, no radicle inhibitory effects were observed. The mean radicle length increased by 1.12%, 1.61%, and 4.57% in the 5% PEG treatment group for 24, 48, and 72 h, respectively, compared with the normal growth. Treatment with 25% PEG resulted in non-germination of quinoa seeds. However, treatment with 15% PEG for 24, 48, and 72 h hindered the germination of quinoa seeds (Fig. 1a). The drought tolerance of 70 quinoa materials was analyzed by simulating drought with 15% PEG, and 6 different germination indexes were statistically analyzed. Among them, the activity index was the most inhibited, with a change rate of -0.79, which indicates that the index was vulnerable to drought. In addition, the coefficient of difference for each indicator ranged from 6.60 to 48.80% under control conditions to 39.30 to 73.20% under stress, indicating that the concentration effect of 15% PEG was significant (Supplementary Table 1). Using fuzzy membership function method, the drought tolerance of 70 quinoa materials under 15% PEG was ranked based on average total membership value (ATMV) of 6 indexes (Supplementary Table 2). Among the 70 quinoa materials, M059 and M024 exhibited the highest and lowest levels of drought resistance, with average total membership values (ATMV) of 0.8729 and 0.0749, respectively. M021, M062 and M066 could not germinate at all under 15% PEG treatment. At 24 h, the seeds of C1, C2, D1 and D2 all germinated, and the radicle length showed the significant difference firstly (Fig. 1b). Therefore, we selected a 15% PEG treatment for 24 h as the drought stress concentration for further experiments.

The radicle anatomical structures of M059 and M024 were observed (Fig. 1c). The radicle lengths of C1, C2, D1, and D2 were 9.25 ± 0.03, 3.15 ± 0.01, 2.90 ± 0.01, and 0.90 ± 0.01 mm, respectively. Under 15% PEG treatment for 24 h, the radicle lengths of M059 and M024 decreased by 68.65% and 71.43%, respectively, compared with the normal growth for 24 h. Under normal growth for 24 h, the root tip of M059 was full. Under 15% PEG treatment for 24 h, the root tip became wrinkled, and the cells turned small and tightly arranged. Under normal growth for 24 h, the root tip of M024 was full, M024 cells became smaller.

Additionally, the contents of sucrose, glucose, fructose, and total soluble sugar (TSS) were significantly different among C1, C2, D1, and D2 samples (Fig. 1d). The contents of total soluble sugar (TSS), sucrose, glucose, and fructose in M059 were 18.58%, 97.84%, 70.54%, and 32.77% higher than those in M024 under normal growth for 24 h. However, under 15% PEG treatment for 24 h, the sucrose content of M024 became 23.01% higher than that of M059. The TSS and glucose contents of M059 were 7.26% and 25.00% higher than those in M024, while the fructose content showed no significant difference. The TSS content of M059 treated with 15% PEG for 24 h was 8.60% higher than that treated with normal growth for 24 h (p < 0.05). The contents of TSS and sucrose in M024 treated with 15% PEG for 24 h were 20.09% and 142.70% higher than those treated with normal growth for 24 h (p < 0.001).

Fig. 1
figure 1

Phenotype, anatomical structure map and significant difference of sugar components of quinoa seeds. a: Average radicle length of all tested varieties treated with 4 PEG concentrations for 24, 48, and 72 h b: Phenotypic observation of four samples at 24, 48, and 72 h c: slitting scan of four samples d: Contents of total soluble sugar, glucose, sucrose and fructose, * representing p < 0.05, ** representing p < 0.01, *** representing p < 0.001, ns representing no significant difference

Full size image

Transcriptomic analysis

RNA sequencing and quality control analysis

The cDNA library of RNA samples was constructed, and 72.58 GB of Clean data was obtained. After filtering, the sample base information was as follows: Q20 ≥ 97.95%, Q30 ≥ 93.74%, N = 0.00%, and GC content was 43.17–43.48%. The number of effective reads that could be located on the genome of the sample ranged from 96.86 to 97.71%. Principal component analysis (PCA) results showed that PC1 and PC2 accounted for 71.0% and 20.4% of the variation, respectively, and the samples were clustered clearly (Fig. 2a). The correlation heat map analysis displayed that pearson correlation coefficient (PCC) was between 0.786 and 1; the PCC between the biological replicates of this experiment was > 0.8 (Fig. 2b).

Fig. 2
figure 2

Sample relationship analysis diagram of transcriptome. a: principal component analysis of sample b: correlation heat map of sample

Full size image

Identification of DEGs

A total of 54,033 genes were identified by transcriptomic analysis. In the C1vsD1 group, there were 233 upregulated and 914 downregulated genes. In the C2vsD2 group, there were 855 upregulated and 2743 downregulated genes. In the C1vsC2 group, there were 1096 upregulated and 1978 downregulated genes. In the D1vsD2 group, there were 1320 upregulated and 3065 downregulated genes (Fig. 3a). Gene overlap analysis of DEGs showed that there were 7042 DEGs in the four comparison groups C1vsD1, C2vsD2, C1vsC2, and D1vsD2 (Fig. 3b). Venn analysis showed that there were a total of 211 commonable DEGs and 132, 1270, 578 and 914 unique DEGs in the four comparison groups. The amount of DEGs in the C1vsD1 comparison group was 68.12% lower than that in the C2vsD2 comparison group, and the amount of DEGs in the C1vsC2 comparison group was 29.90% lower than that in the D1vsD2 comparison group. The amount of upregulated DEGs in C2vsD2 comparison group was 68.83% less than that of downregulated DEGs, and the amount of upregulated DEGs in D1vsD2 comparison group was 56.93% less than that of downregulated DEGs. To sum up, there were more downregulated DEGs in M024 than in M059 under drought stress.

Fig. 3
figure 3

Statistical analysis of the differentially expressed genes. a: The MA diagram represents the expressed genes among different comparison groups b: Venn diagram

Full size image

GO enrichment analysis

To analyze the response of the two materials to drought stress, GO enrichment analysis of DEGs was carried out. The results showed that all DEGs in C1vsD1, C2vsD2, C1vsC2, and D1vsD2 enriched 828, 1373, 1359, and 1589 GO terms, respectively. The top 20 GO terms with FDR < 0.05 were selected for in-depth analysis, mainly including photosystem (GO:0009521), thylakoid (GO:0009579), photosynthetic membrane (GO:0034357), thylakoid part (GO:0044436), and generation of precursor metabolites and energy (GO:0006091), etc. (Fig. 4). The process of photophosphorylation was closely enriched by GO term analysis, which indicated the photochemical reaction completed through the photosystem, and transformed light energy into chemical energy on the thylakoid membrane to form ATP and NADPH.

Fig. 4
figure 4

Bar chart of GO enrichment analysis

Full size image

KEGG enrichment analysis

In order to study the metabolic pathways in-depth, the enriched DEGs were compared with KEGG database. KEGG metabolic pathway enrichment analysis showed that 91, 113, 117, and 125 pathways were enriched in C1vsD1, C2vsD2, C1vsC2, and D1vsD2, respectively. The top 20 metabolic pathways with FDR < 0.05 have been selected out (Fig. 5). The enrichment factors of the photosynthesis-antenna proteins (ko00196) were 0.387, 0.677, 0.645 and 0.742, respectively. The enriched DEGs of the metabolic pathways (ko01100) were 159, 423, 319, and 480, respectively.

Fig. 5
figure 5

Sourcewww.kegg.jp/kegg/kegg1.html.

Bubble diagram of KEGG enrichment analysis.

Full size image

Analysis of expression profile of DEGs related to glucose metabolism

In this study, 23 DEGs associated with quinoa glucose metabolism were identified based on functional annotation of genes (Fig. 6), including 1 CqAGAL2, 1 CqGOLS2, 2 CqINV, 2 Cqbg1x, 2 CqFK, 1 CqHK2, 1 CqSUS, 1 CqCELB, 1 CqFBP, 1 CqALDO, 1 CqPGM, 3 CqIDH, 1 CqGAPDH, 1 CqMDH1, 2 CqPFKA, 1 CqPK, and 1 CqSDHA. Among them, CqFK (LOC110734349) was significantly down-regulated in D1vsD2, and the difference ratio log2FC reached to -6.43; CqIDH (LOC110740096) was significantly up-regulated in C1vsC2, and log2FC reached to 8.95.

Fig. 6
figure 6

Heat map of DEGs associated with quinoa glucose metabolism. The color represents the log2FC value, red represents up-regulated expression, blue represents down-regulated expression, and white represents no differential expression

Full size image

Enrichment analysis of DEGs associated with glucose metabolism

The 23 DEGs associated with quinoa glucose metabolism were analyzed by GO functional annotation and KEGG enrichment. Annotation analysis showed that the five GO terms were closely related to the molecular response of glucose metabolism in quinoa under drought stress (Fig. 7a), including intracellular membrane bounded organelle (GO:0043231), transferase activity, transferring phosphorus-containing groups (GO:0016772), single-organism metabolic process (GO:0044710), glucose metabolic process (GO:0006006), and carbohydrate metabolic process (GO:0005975). Under drought stress, DEGs related to glucose metabolism were mainly enriched to galactose metabolism (ko00052), starch and sucrose metabolism (ko00500), glycolysis and gluconeogenesis (ko00010), and carbon metabolism (ko01200) (Fig. 7b). In C1vsD1, C2vsD2, C1vsC2 and D1vsD2, the enriched DEGs in carbon metabolism were 11, 29, 43, and 57, respectively. Among the above four pathways, the galactose metabolic pathway (ko00052) had the highest enrichment factor, which were 0.071, 0.117, 0.134 and 0.116, respectively. These results indicated that the galactose metabolism (ko00052) and carbon metabolism (ko01200) were research-worthy KEGG metabolic pathways for the enrichment of DEGs related to glucose metabolism in quinoa seeds under drought stress.

Fig. 7
figure 7

Sourcewww.kegg.jp/kegg/kegg1.html.

GO and KEGG enrichment analysis of DEGs with glucose metabolism. a: String diagram of GO enrichment analysis. DEGs with different colors of different comparison groups were indicated by gene IDs. The gene alignation on the circle were arranged according to the log2 (FC) value. Up-regulated and downregulated expression were represented with pink and geen rectangle blocks. b: Bubble diagram of KEGG enrichment analysis. The vertical axis represented the metabolic pathways obtained by KEGG enrichment analysis, and the horizontal axis represented enrichment factors.

Full size image

Metabolomics analysis

Qualitative and quantitative analysis of metabolites

All metabolites in quinoa seeds under drought stress were analyzed by a broadly targeted metabolomics method (Supplementary Table 3). The PCA map showed that PC1 and PC2 explained 32% and 20.1% of the variation, respectively (Supplementary Fig. 1), indicating significant changes in metabolites in samples under different treatment conditions. The results of PLS-DA showed that Q2 values of the C1vsD1, C2vsD2, C1vsC2, and D1vsD2 groups were 0.954, 0.98, 0.986, and 0.994 (Q2 > 0.9) (Supplementary Table 4), respectively, indicating the reliability of the model.

Identification and expression analysis of DAMs

We identified twelve DAMs with significant differences that were involved in the glucose metabolism of quinoa seeds under drought stress (Fig. 8). These were inositol and D-galactose from galactose metabolism (ko00052), UDP-glucose, D-glucose, D-fructose-6-phosphate, D-glucose-6-phosphate, and D-glucose-1-phosphate from starch and sucrose metabolism (ko00500), dihydroxyacetone phosphate and ribulose-5-phosphate from glycolysis and gluconeogenesis (ko00010), and citric acid, succinate and L-malate from TCA cycle (ko00020). The color of the heat map changed from red to blue, indicating that the expression abundance of DAMs changes from high to low. The more important of the DAMs in the corresponding comparison group, the higher the VIP value would be. There were eight DAMs in C1vsD1 group, the highest VIP value of UDP-glucose was 2.92. There were twelve DAMs in C2vsD2 group and five DAMS in C1vsC2 group, and the highest VIP value of succinic acid was 3.25 and 3.70. There were eight DAMs in D1vsD2 group, the highest VIP value of L-malate was 2.91. Notably, D-glucose, UDP-glucose and dihydroxyacetone phosphate were present in all comparison groups.

Fig. 8
figure 8

Expresses abundance heat map and VIP map of DAMs. Note The vertical coordinate of the VIP diagram represents the DAMs, and the horizontal coordinate represents the VIP value

Full size image

Integrated transcriptome-metabolome analysis

In order to obtain the key pathways that respond to drought stress, we analyzed the correlation between DAMs and DEGs. An integrated analysis of DEGs and DAMs showed that starch and sucrose metabolism (ko00500) was the joint KEGG pathway. UDP-glucose, D-glucose, D-fructose-6-phosphate, D-glucose-6-phosphate, and D-glucose-1-phosphate were positively correlated with CqINV (LOC110717458), CqFK4 (LOC110718125), CqCELB (LOC29490), Cqbg1x (LOC110695967, LOC110685748), CqSUS (LOC110689796, LOC110727927), and CqHK2 (LOC110729879), respectively (PCC ≥ 0.6). Fructose-6-phosphate, D-glucose-6-phosphate, and D-glucose-1-phosphate were highly positively correlated with CqINV (LOC110717458), and PCCs were 0.98, 0.91, 0.93, respectively. UDP-glucose and D-glucose-1-phosphate were highly positively correlated with CqSUS (LOC110689796), PCCs were 0.91 and 0.91, respectively. D-glucose was highly positively correlated with CqSUS (LOC110689796, LOC110727927), and PCCs were 0.97 and 0.90, respectively (Fig. 9).

Fig. 9
figure 9

Correlation network diagram of DAMs and DEGs. Note circle represented DEGs and triangle represented DAMs. a: Overall correlation analysis of DAMs and DEGs b: Correlation network diagram of DAMs and DEGs (PCC ≥ 0.8)

Full size image

Transcription factor analysis

There were 15, 156, 67, and 158 significantly differentially expressed TFs (DETFs) in the C1vsD1, C2vsD2, C1vsC2, and D1vsD2 groups, respectively (Fig. 10a). In the C1vsD1 group, there were 2 upregulated and 13 downregulated DETFs. In the C2vsD2 group, there were 44 upregulated and 112 downregulated DETFs. In the C1vsC2 group, there were 24 upregulated and 43 downregulated DETFs. In the D1vsD2 group, there were 40 upregulated and 118 downregulated DETFs. In this study, 42 DETFs from WRKY family and bHLH (basic Helix Loop Helix) family were identified (Supplementary Table 5). The FPKM (fragments per kilobase per million) value of DETFs showed that LOC110720931_WRKY6, LOC110731563_WRKY54, LOC110735049_WRKY4, and LOC110719592_WRKY56 were downregulated. LOC110725693_SPCH, LOC110681689_HEC3 and LOC110706295_bHLH25 were upregulated. A biomic correlation analysis was performed between seven DETFs and glucose metabolism related DEGs (Fig. 10b). There was a significant positive correlation between LOC110720931_WRKY6 and LOC110716244_FBP (p < 0.01). There was a significant positive correlation between LOC110719592_WRKY56 and LOC110729879_HK2 (p < 0.01). LOC110681689_HEC3 was found to be negatively correlated with LOC110700602_MDH1 (p < 0.01). There was a significant positive correlation between LOC110681689_HEC3 and LOC14088_IDH1 (p < 0.01).

Fig. 10
figure 10

Statistical analysis of DETFs and correlation heat maps between DETFs and DEG. a: Statistical analysis of DETFs b: Heat map of correlation between DETFs and DEGs, red indicated positive correlation, blue indicated negative correlation, * indicated correlation (p < 0.05); ** indicated significant association (p < 0.01)

Full size image

qRT-PCR validation of key DEGs

Ten key DEGs in different metabolic pathways were subjected to qRT-PCR validation (Fig. 11). The results showed that 4, 8, 5 and 7 DEGs in C1vsD1, C2vsD2, C1vsC2 and D1vsD2 were verified by qRT-PCR, respectively. The expression trend of qRT-PCR of ten key DEGs and the difference ratio of transcriptome detection results were consistent, and the transcriptome data were reliable.

In the entire glucose metabolism pathways during seed germination in quinoa, starch and sucrose metabolism (ko00500) served as the upstream pathway, while glycolysis and gluconeogenesis (ko00010) as the intermediate pathway, and the TCA cycle (ko00020) acted as the downstream pathway. Amongst the ten key DEGs, upstream DEGs including LOC110702784_AGAL2, LOC110734349_FK7, LOC110689372_INV, LOC29490_CELB, and LOC110695967_bg1x were verified by qRT-PCR. The identified DEGs in the glycolysis and gluconeogenesis pathway were LOC110716244_FBP, LOC110736667_ALDO and LOC110682269_PGM. The verified downstream DEGs were LOC110740096_IDH3 and LOC110717784_SDH, which expression trends were consistent with the transcriptomic results.

Fig. 11
figure 11

qRT-PCR validation of 10 key DEGs. The left Y-axis representrd the FPKM value in transcriptomics, and the right Y-axis represented the relative expression of samples in qRT-PCR. The column represented the FPKM value of each component in each comparison group, corresponding to the left Y-axis. The red circle represented the relative qRT-PCR expression of each component in each comparison group, corresponding to the right Y-axis

Full size image

Drought tolerance genes expressed in yeast expression system

Functional analysis of expressed genes involved in glucose metabolism in yeast system

196 drought tolerance genes were verified by functional screening in yeast (Supplementary Table 6), among which five genes were involved in glucose metabolism pathways (Table 1), namely CqCS (LOC110738229) and CqSDHF2 (LOC110698560) from TCA cycle (ko00020), CqTIM (LOC110739768) and CqRBCS (LOC110728202, LOC110720405) from glycolysis and gluconeogenesis (ko00010), and Cqβ-GAL13 (LOC110694437) from galactose metabolism (ko00052).

Table 1 Function of genes related to glucose metabolism
Full size table

Joint analysis of omics technique and yeast expression system

Ten key DEGs (CqAGAL2, CqINV, CqFrK7, CqCELB, Cqbg1x, CqFBP, CqALDO, CqPGM, CqIDH3, and CqSDH) were verified by qRT-PCR. In the TCA cycle (ko00020), CqIDH3 was the downstream gene of CqCS. CqSDH was a drought tolerance candidate gene that has been validated in both transcriptomic and yeast systems. In glycolysis and gluconeogenesis (ko00010), the downstream genes of CqALDO and CqRBCS were CqTIM and CqPGM, respectively. In galactose metabolism (ko00052), CqAGAL2 and Cqβ-GAL13 were drought tolerance candidate genes that have been validated both in transcriptomics and yeast systems, respectively (Fig. 12).

Fig. 12
figure 12

Molecular mechanisms of glucose metabolism in quinoa seeds under drought stress. Each graph represents the normalized intensity of the corresponding differentially expressed gene or differentially accumulated metabolite for in two treatments from two genotypes

Full size image

Discussion

Quinoa not only has a high nutritional value but also a good tolerance for abiotic stress [34]. Carbohydrates are key nutrients in plants, and signaling molecules to adversity responses [35]. In this study, drought-tolerant material M059 and drought-sensitive material M024 were selected for combined transcriptome and metabolomics analysis. The number of DEGs in M024 was significantly higher than that in M059, the number of downregulated DEGs in M024 was significantly higher than that of upregulated DEGs. The DAMs of M059 were derived from starch and sucrose metabolism (ko00500). The DAMs of M024 were derived from starch and sucrose metabolism (ko00500), galactose metabolism (ko00052), and carbon metabolism (ko01200). There were more downregulated DEGs and more DAMs changes in M024 than M059 under drought stress. The type and quantity of plant metabolites can reflect environmental adaptability [36]. The results showed that drought tolerant genotype M059 was less affected by water deficit at transcriptional and metabolic levels. The candidate genes and key metabolic pathways involved in drought response in quinoa were identified by integrating transcriptomic and metabolomic analyses.

Phenotypic analysis of quinoa seeds

By observing the phenotype and comparing the sugar contents, under normal conditions, the concentration of TSS, sucrose, glucose and fructose in M059 was 18.58%, 97.84%, 70.54% and 32.77%, which was much higher than that of M024. The content change of these carbohydrate substances is one of the important reasons for the root development speed during seed germination. The crux of the matter was that M059 and M024 have different contents of total soluble sugar in the original cells, which was the main reason that M059 was more drought tolerant than M024. Of course, in order to more clearly represented the changes in glucose metabolism, this study also carried out a specific analysis of sucrose, glucose and fructose. Total soluble sugars included not only monosaccharides and partial disaccharides, but also sugars that were soluble in water, as well as oligosaccharides and polysaccharides that could be hydrolyzed into reducing sugars under determined conditions. Among them, soluble sugars included sucrose, glucose, fructose, maltose, trehalose, raffinose and so on. The contents of other soluble sugars (maltose, trehalose, raffinose, etc.) may show an increasing trend resulting in different contents of total soluble sugars. In this study, the accumulation of glucose and fructose was reduced under drought stress in both materials. Consistent with previous studies, under drought stress, the content of glucose and fructose of roots firstly increased and then decreased with the extension of stress time in six wheat varieties [37]. Sucrose plays an active role in glucose metabolism, and sucrose synthase and sucrose invertase break down sucrose to form UDP-glucose and fructose. The catalytic reaction of sucrose synthase was reversible, so sucrose invertase played a major role in the decomposition of sucrose into glucose and fructose in the cells (Fig. 13).

Fig. 13
figure 13

sucrose decomposition model graph in quinoa seed

Full size image

Regulation of key genes and metabolites

Galactosidase (GAL) referred to a class of enzymes that hydrolyzed substances containing galactosidase, mainly divided into α-Galactosidase (AGAL) and β-Galactosidase (β-GAL) [38]. The catalytic action of AGAL was to remove the α-linked non-reducing D-galactose at the end of the substrate [39]. The relationship between AGAL activity and seed dehydration tolerance in early corn germination found that the AGAL activity of four corn varieties peaked at 24 h of imbibition, with drought-resistant varieties showing higher AGAL activity than drought-sensitive ones. Overexpression of ZmAGA1 in Arabidopsis thaliana leaded to higher seed germination rates [40]. Masakazu et al. founded that the transcription level of alkaline AGAL increased significantly when the spinach (Spinacia oleracea L.) was subjected to drought stress [41]. In this study, there was no differential expression of CqAGAL2 in drought tolerant material M059 under drought stress. The expression of CqAGAL2 in drought sensitive material M024 was down-regulated. The expression of D-galactose and inositol were down-regulated. β-GAL involved in cell wall polysaccharide metabolism can catalyze the decomposition of lactose into glucose and galactose [42]. β-GAL was involved in the growth and differentiation of the embryo in chickpea [43]. In this study, Cqβ-GAL13 was also detected in the germinant seeds of quinoa. The expressing of Cqβ-GAL13 in transgenic yeast showed stronger resistance to PEG simulated drought stress than empty carrier yeast. Cqβ-GAL13 decomposed cell wall polysaccharides, and the accumulation of sugars not only maintained energy homeostasis, but also might promote cell viability through osmoregulation. This result demonstrateded that CqAGAL2 and Cqβ-GAL13 regulated the seed growth under drought stress in quinoa.

Fructokinase (FK) was a major kinase in fructose metabolism, phosphorylating fructose to produce D-fructose-6-phosphate into cytoplasmic glycolysis pathway [44]. UDP-glucose was a substrate for glycosylation [45]. UDP-glucose combined with D-fructose-6-phosphate to form sucrose phosphate, which in turn dephosphorylated to produce sucrose, which could serve as an initial substrate for various metabolic pathways [46]. In Helianthus annuus L., FK expression and related proteins were co-upregulated due to drought stress, increasing the root-shoot ratio in stressed plants compared to controls [47]. In maize, short-term salt stress upregulated FK2 expression, serving as an early marker of salt stress [48]. In this study, under drought stress, CqFK7 expression was upregulated in drought tolerant material M059, while CqFK7 was downregulated in drought sensitive material M024. Although D-fructose-6-phosphate, a downstream metabolite of CqFK7, was downregulated in both M059 and M024, the Log2FC of D-fructose-6-phosphate in M024 and M059 were − 0.94 and − 0.48, respectively. The results showed that CqFK7 was a candidate gene involved in drought response in quinoa.

β-glucosidase (bg1x) was a member of Glycosyl hydrolase family I, involved in plant metabolism and regulating plant growth and development [49]. Overexpression of AtBG1 enhanced drought resistance in creeping bentgrass (Agrostis stolonifera L.) [50]. Under drought stress, the abundance of bg1x increased in both leaves and roots in soybean [51]. In this study, no expression difference in Cqbg1x was observed in drought tolerant material M059 under drought treatment, while it was downregulated in drought sensitive material M024. The downstream metabolite of Cqbg1x was D-glucose, which was downregulated in both M059 and M024. The Log2FC of D-glucose in M024 and M059 were − 0.94 and − 0.25, respectively. Under drought stress, the glucose concentration of M059 was 25.00% higher than that of M024 (p < 0.001). The difference trends of glucose concentration and D-glucose metabolic abundance were consistent with that of M059 and M024. The results suggested that Cqbg1x was a candidate gene involved in drought response in quinoa.

D-glucose was underwent a four-step reaction to produce two molecules of three-carbon sugars, glyceraldehyde 3-phosphate and dihydroxyacetone phosphate (DHAP). The two three-carbon sugars were isomers that could be converted into each other by the action of triose phosphate isomerase (TIM) [52]. In this study, the expressing of CqTIM in transgenic yeast showed stronger resistance to PEG simulated drought stress than empty carrier yeast. The upstream gene of CqTIM was CqALDO. Under drought stress, CqALDO expression was down-regulated both in M059 and M024. The Log2FC of CqALDO in M059 and M024 were − 1.00 and − 1.68, respectively. The downstream metabolite of CqALDO was DHAP, and the metabolic abundance of DHAP was down-regulated in C1vsD1, C2vsD2, C1vsC2 and D1vsD2. DHAP, as an intermediate product of glycolysis, was also a raw material for synthesis of glycerol [53]. We speculated that under drought stress, CqALDO and DHAP were down-regulated to reduce energy consumption of other metabolic pathways and retain more ATP for glycolysis pathways. One molecule of glucose underwent glycolysis, producing a two molecules of ATP and two molecules of NADPH + H+. Ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) was a key enzyme in carbon assimilation in photosynthesis. Rubisco catalyzed the covalently binding of free CO2 to Ribulose 1, 5-diphosphate (RuBP) to produce two-molecule 3-phosphoglyceric acid [54]. The whole enzyme of Rubisco was composed of eight large subunits (RBCL) and eight small subunits (RBCS). RBCS was encoded by nuclear genes and had the function of regulating the whole enzyme activity of Rubisco [55]. Under drought stress, the protein abundance of RBCS significantly increased during grout in wheat [56]. Jurczyk et al. founded that ryegrass (Lolium perenne L.) could enhance the expression of RBCS at low temperature, thereby enhancing the vitality of Rubisco to maintain the rate of photosynthetic carbon assimilation to resist cold stress [57]. In this study, the expressing of CqRBCS in transgenic yeast showed stronger resistance to PEG simulated drought stress than empty carrier yeast. The upstream metabolite of RuBP was ribulose-5-phosphate. Under drought stress, the metabolic abundance of ribulose-5-phosphate was accumulated in drought-tolerant material M059, but down-regulated in drought-sensitive material M024. We hypothesized that the metabolic abundance of ribulose-5-phosphate was upregulated, increasing the conversion of inorganic carbon to organic carbon and generating more substrates in response to drought stress.

Citrate synthase (CS) was a key enzyme in the TCA cycle. CS catalyzed the synthesis of citric acid from acetyl CoA and oxaloacetic acid. Under high or low temperature, salt stress and drought stress, CS was widely involved in the metabolic regulation and environmental response in rape [58]. CS increased in response to drought stress in Sinapis alba L. [59]. In this study, CqCS-transgenic yeast could grow on PEG medium, but empty carrier yeast could not as the former yeast. Under drought stress, the metabolic abundance of citric acid was accumulated in drought-tolerant material M059, but down-regulated in drought-sensitive material M024. CqCS were involved in metabolic regulation and drought response of quinoa seeds. Isocitrate dehydrogenase (IDH) was the second regulatory enzyme in the TCA cycle, which catalyzed the oxidative decarboxylation of isocitric acid to form α-ketoglutaric acid, and decreased oxidized NAD+ to reduced NADH + H[+ [60]]. The activity of IDH increased due to water shortage in pea (Pisum sativum L.) [61]. In this study, CqIDH3 was upregulated under drought stress. Succinate was converted to fumaric acid by succinate dehydrogenase (SDH). Fumaric acid, as an antioxidant, had the function of trapping and neutralizing free radicals and reducing the damage to organisms [62]. The increase of SDH activity during seed germination promoted the efficiency of carbohydrate assimilation and metabolism, and enhanced the drought tolerance in alfalfa (Medicago sativa L.) [63]. Under drought stress in rapeseed, the SDH expression could resist the stress by producing non-enzymatic antioxidants [64]. In this study, CqSDH was not differentially expressed in M059, but up-regulated in M024 under drought stress. The abundations of DAMs succinate and L-malate were down-regulated. CqSDHF2 was also identified in the yeast expression system, and the CqSDHF2 transgenic yeast expressed stronger resistance to PEG simulated drought stress than that of empty vector yeast. CqSDH was up-regulated in sensitive material M024 in response to drought stress. However, crop drought resistance is a complex quantitative trait controlled by multiple genes [65]. Compared with M059, M024 has more down-regulated DEGs and DAMs, and is more easy affected by drought stress in both of transcriptional and metabolic levels. Under drought stress, the TCA cycle became more active and catabolized soluble sugars to ensure that there was sufficient ATP during seed germination.

Transcription factor analysis

It is well known that bHLH TFs play crucial roles in promoting abiotic stress tolerance in plants [66]. Arabidopsis thaliana overexpressing AtbHLH146 exhibited stronger drought tolerance [67]. In this study, we found that bHLH HEC3 correlated with the CqIDH1 and CqMDH1. There was a significant positive correlation between HEC3 and CqIDH1 (p < 0.01). A negative correlation existed between HEC3 and CqMDH1 (p < 0.01).

WRKY TFs were known to regulate plant defense and abiotic stress responses [68]. In Arabidopsis thaliana, AtWRKY46/54/70 negatively regulated drought tolerance [69]. In this study, WRKY56 was positively with CqHXK2 and CqFBP (p < 0.01). This aligned with sucrose metabolism gene studies in Hylocereus undulatus Britt. [70]. When TFs interacted with cis-responsive elements in the promoters of stress-inducing genes, it activates the signal transduction cascade. The regulatory mechanism of TFs was a complex phenomenon, and the specific regulatory mechanisms of HEC3, WRKY56 and WRKY6 require verification in overexpressed transgenic plants.

Conclusions

The glycometabolic synthesis pathway has been identified as one of the key metabolic pathways during quinoa seed germination through the combined analysis of transcriptomics-metabolomics. Transcriptomic-metabolomic analysis and large-scale yeast function screening system were performed for two different drought-response genotypes of quinoa. The results showed that galactose metabolism (ko00052), starch and sucrose metabolism (ko00500), glycolysis and gluconogenesis (ko00010) and TCA cycle (ko00020) were closely related to glucose metabolism under drought stress. CqAGAL2, Cqβ-GAL13 and CqSDH were identified to be the key genes of drought tolerance in quinoa. The results of this study confirmed that glucose metabolism plays an important role in the drought stress response of quinoa seeds.

Materials and methods

Test materials and sampling

In this study, seventy quinoa germplasm materials were planted at the test plot of the Quinoa Industrial Technology Research Institute of Hebei Province, Zhang Jiakou, P. R. China (Table 2). These materials were provide by the S&T Innovation Service Center of Hebei Province, Shijiazhuang, P. R. China and the formal identifcations were carried out by Researcher Wei Lv. No voucher specimen have been deposited in a genebank and no special permissions are needed to study these species. Two germplasm materials, drought tolerant material M059 (BO-60, Peru) and drought sensitive material M024 (D-12290, Peru), were selected by the hyperosmotic method PEG-6000 [71] simulation of drought stress in this experiment. The samples were cultured on paper in an artificial climate chamber. In the first stage, conditions were set at 25.0℃, 10,000 lx, 70% relative humidity (RH), and incubated for 12 h; the second stage was cultured at 22.0℃, 0 lx, 70% RH for 12 h. the seeds of normal growing M059 and M024 (C1 and C2), 15% PEG-treated M059 and M024 (D1 and D2) were collected, 3 g each, with three biological replicates. The samples were placed in a frozen tube, quickly frozen with liquid nitrogen, and stored in an ultra-low-temperature refrigerator at -80℃.

Table 2 Sample number, origin identification name and origin of all tested quinoa strains
Full size table

Section staining and observation of quinoa seeds

The tissue samples were cut longitudinally to a thickness of 8 μm and then stained with safranin. The tissue paraffin sections of C1, D1, C2, and D2 were fixed with the formalin-acetic acid-alcohol (FAA) solution, following the procedure of fixing, dehydrating, clearing, dipping, embedding, slicing, drying, dyeing, and sealing, as reported previously [72]. Samples were sliced using a pathological slicing instrument (Shanghai Leica Instruments Co., Ltd., China P.R.). Slices were observed and photographed through an upright optical microscope (Nikon Corporation, Japan). Digital scanning and imaging were performed using a slice scanner (Pannoramic DESK, 3DHISTECH, Hungary). The sliced images were collected and analyzed through Caseviewer C.V 2.4.

Determination of sugar content: sucrose, fructose, and glucose

Use 1 mL pipette to take the standard sugar working liquid in a dry and clean test tube according to Table 3, and adjust the volume to 2.00 mL with distilled water respectively. 6.00 mL anthrone reagent was added to each tube. Shake well and immediately place in a boiling water bath for 3.5 min. Remove immediately and quickly shake in cold or ice water to cool to stop the reaction. Repeat with 2.00 mL of distilled water as blank. The absorbance was measured at 640 nm wavelength, and the standard curve was drawn with absorbance as horizontal coordinate and sugar content as vertical coordinate, and the standard linear equation was obtained (Fig. 14).

Table 3 Standard curve solution of glucose, fructose and sucrose
Full size table
Fig. 14
figure 14

Standard curve for colorimetric measurement of soluble total sugar, sucrose, glucose and fructose content

Full size image

Referring to the method of Zhang [73], 0.5 g of quinoa seeds was taken. The sample was ground with a mortar, then placed in a 15-mL centrifuge tube, and 10 mL of 75% ethanol was added. It was then heated in a water bath at 80℃ for 15 min. After cooling, the mixture was centrifuged, and the supernatant was transferred to a 25 mL volumetric flask. Additionally, a small amount of 75% ethanol was used to rinse the residue repeatedly, and the supernatant was then transferred into the volumetric flask to make up to 25 mL. Glucose and fructose contents were measured using the anthrone sulfuric acid colorimetric method [74], and sucrose content was measured using the KOH anthrone sulfuric acid method [75]. The experiment was conducted with three biological replicates.

Determination of total soluble sugar (TSS) content

Take several 20 mL scale test tubes, add 0, 0.2, 0.4, 0.8, 1.2, 1.6, 2.0 mL of 100 µg/mL sucrose solution, and fill with water to 2 mL, then add 0.5 mL anthrone-ethyl acetate reagent and 5 mL concentrated sulfuric acid. The absorbance was measured at 630 nm wavelength with the blank as the control. The standard curve was drawn with the absorbance as the horizontal coordinate and the sugar content as the vertical coordinate, and the standard linear equation was obtained (Fig. 13).

All samples were crushed, 0.2 g of each sample was weighed and placed into a 15 mL centrifuge tube, and then 10 mL of distilled water was added. Afterward, the samples were placed in a boiling water bath for 30 min, cooled, and centrifuged. The supernatant was then transferred to a 25 ml volumetric flask. This step was repeated. The content of total soluble sugar was determined using the anthrone colorimetric method [76], with three biological replicates. Microsoft Excel 2010 was used for data organization, SPSS Statistics 26.0 software for significance analysis, and GraphPad Prism 8.0 for plotting.

Transcriptome sequencing and DEGs analysis

RNA library construction and quality control

RNA was extracted from quinoa seeds following the previously published Trizol et al. precipitation method [77]. The purity and integrity of the RNA were detected by 1% agarose gel electrophoresis. The quality requirement for RNA Seq samples was RIN > 7.5; 1.9 < OD260/OD280 < 2.1; 28 S/18 S ≥ 1.5. After passing the sample testing, the library was constructed through five processes [78]. Samples were sequenced using the Illumina platform (http://www.illumina.com), with three biological replicates. Sequencing by synthesis (SBS) techniques were used to calculate base mass values for the obtained Raw reads. Quality control was performed using Fastp [79], resulting in high-quality Clean reads.

Identification of DEGs

Using HISAT2 software(HISAT2 2.2.1) [80], high-quality Clean reads were compared with reference genomes of tetraploid quinoa (https://www.ncbi.nlm.nih.gov/genome/?term=txid63459 [orgn]). Reads were assembled and quantified with StringTie [81] to form the initial transcript. A comparison was conducted using BLAST (Basic Local Alignment Search Tool) [82], GO (http://www.geneontology.org/) [83], and KEGG (https://www.kegg.jp/kegg/kegg1.html) [84] to obtain annotation information. StringTie was used with the maximum flow algorithm, using FPKM as an indicator of gene expression levels. The fold change (FC) and False discovery rate (FDR) of FPKM between different comparison samples were calculated Using DESeq2 [85]. Based on the criteria of log2FC ≥ 1 and FDR < 0.05, DEGs were considered upregulated at log2(FC) > 1 and downregulated at log2(FC)<-1. The volcano figures were drawn in R (http://www.r-project.org/) to display DEG statistics in each comparison group [86]. The Venn diagram was drawn using the Venny 2.1 online tools (http://jvenn.toulouse.inra.fr/app/example.html) to show unique genes and their overlap between the comparison groups.

GO and KEGG enrichment analysis

DEGs were subjected to GO enrichment analysis using GO databases. GO has three ontologies, which describe the molecular function (MF), cell components (CC), and biological process (BP). DEGs were mapped to GO terms, and the number of genes was calculated to obtain a list of genes with certain GO functions in each term [84]. Hypergeometric tests were applied to identify DEG-associated enriched GO terms compared to the whole genomic background.

The KEGG is a major public database related to pathways. Based on KEGG Pathway as a unit, a hypergeometric test was applied to find out significantly enriched DEGs associated pathways [85]. These pathways can identify the most important biochemical metabolites and signal transduction pathways linked to DEGs.

Metabolomics analysis

Qualitative and quantitative analysis of the metabolites

C1, C2, D1, and D2 samples were analyzed using a broadly targeted metabolomics approach. A total of 840 metabolites were divided into 11 categories, including amino acids and their derivatives, saccharides and organic acids (Supplementary Table 3). Referring to the method of Chen et al. [87], the specific experimental methods were as follows: (1) The freeze-dried seeds were crushed using a mixer mill (MM 400, Retsch) with a zirconia bead for 1.5 min at 30 Hz. 100 mg powder was weighted and extracted overnight at 4℃ with 1.0 ml 70% aqueous methanol. Following centrifugation at 10, 000 g for 10 min, the extracts were absorbed (CNWBONDCarbon-GCB SPE Cartridge, 250 mg, 3 ml; ANPEL, Shanghai, China, www.anpel.com.cn/cnw) and filtrated (SCAA-104, 0.22 μm pore size; ANPEL, Shanghai, China, http://www.anpel.com.cn/) before LC-MS analysis. (2) The sample extracts were analyzed using an LC-ESI-MS/MS system (HPLC, Shim-pack UFLC SHIMADZU CBM30A system, www.shimadzu.com.cn/; MS, Applied Biosystems 6500 Q TRAP, www.appliedbiosystems.com.cn/). The analytical conditions were as follows: HPLC: column, Waters ACQUITY UPLC HSS T3 C18 (1.8 μm, 2.1 mm*100 mm); solvent system: water (0.04% acetic acid): acetonitrile (0.04% acetic acid); gradient program: 95:5 V/V at 0 min, 5:95 V/V at 11.0 min, 5:95 V/V at 12.0 min, 95:5 V/V at 12.1 min, 95:5 V/V at 15.0 min; flow rate, 0.40 ml/min; temperature: 40 °C; injection volume: 2 µl. (3) The effluent was alternatively connected to an ESI-triple quadrupole-linear ion trap (Q TRAP)-MS. LIT and triple quadrupole (QQQ) scans were acquired on a triple quadrupole-linear ion trap mass spectrometer (Q TRAP), API 6500 Q TRAP LC/MS/MS System, equipped with an ESI Turbo Ion-Spray interface, operating in a positive ion mode and controlled by Analyst 1.6.3 software (AB Sciex). (4) The ESI source operation parameters were as follows: ion source, turbo spray; source temperature 500 °C; ion spray voltage (IS) 5500 V; ion source gas I (GSI), gas II(GSII), curtain gas (CUR) were set at 55, 60, and 25.0 psi, respectively; instrument tuning and mass calibration were performed with 10 and 100 µmol/L polypropylene glycol solutions in QQQ and LIT modes, respectively. QQQ scans were acquired as MRM experiments with collision gas (nitrogen) set to 5 psi. DP and CE for individual MRM transitions were adopted with further DP and CE optimization. A specific set of MRM transitions were monitored for each period according to the metabolites eluted within this period.

Identification of DAMs

Metabolites were annotated in the MetWare database (https://sourceforge.net/p/metware/support-requests/1/), and MetaboAnalyst version 5.0 software was used to process the metabolome data [88]. Unsupervised dimension reduction and principal component analysis were conducted using the R language. Cross-validation and arrangement testing of metabolomics data were verified by the OPLS-DA model. Score and arrangement maps for each group were drawn to visually show group differences. Multiquant software V2.0 [89] was used to integrate and correct the chromatographic peaks of mass spectrometry results, obtaining the final metabolite data. Variable importance in the projection (VIP) values denotes each variable’s contribution to the model; VIP > 1 was set as a common threshold. DAMs were screened with the criteria VIP > 1 and p < 0.05.

Combined transcriptome-metabolome analysis

Transcriptome and metabolome data were examined to identify DEGs and DAMs. We used the PCC method to find correlations between DEGs and DAMs, taking a PCC threshold ≥ 0.6. |PCC|≥0.6 indicates a correlation between DEGs and DAMs, |PCC|≥0.8 indicates a strong correlation between DEGs and DAMs. Cytoscape software (Cytoscape_v3.8.2) [90] was used to map the results of the combined analysis, and the results were annotated into the KEGG database to obtain the common KEGG metabolic pathway of DEGs and DAMs.

Analysis of differentially expressed transcription factors

The selected DEGs were BLAST compared with the quinoa genome database and combined with Plant TFDB (Plant transcription factor database 3.0, http://planttfdb.cbi.pku.edu.cn/) [91] to screen differentially expressed TFs. The TF family and related genes associated with drought resistance in quinoa were further analyzed. The Pearson coefficient method was used to determine the correlation between key DEGs and TFs, and a heat map was drawn to analyze the expression patterns of TFs.

qRT-PCR validation

Ten key DEGs were selected for qRT-PCR validation. The sprouted seed RNA of C1, D1, C2 and D2 was extracted by Trizol precipitation method, and cDNA was synthesized by HiFiScript gDNA Removal RT Master Mix reverse transcription kit. qRT-PCR was performed using the 2×SYBR Green qPCR Master Mix kit. qRT-PCR was performed on the Quant Studio 5 real-time fluorescence quantitative PCR apparatus (Thermo Fisher, America). The reaction procedure was as follows: denaturation at 95℃ for 10 s, annealing at 60℃ for 20 s, and extension at 72℃ for 15 s, for a total of 40 cycles, with three biological repeats. CqACT-1 (LOC110715281) was used as the internal reference gene. Based on the CDS sequence, primers were designed using the Premier5.0 software [92] (Supplementary Table 7). The relative gene expression levels were calculated by the 2−△△Ct method [93].

cDNA library construction and selection of drought tolerance yeast transformants

In order to construct cDNA library, total RNA of C1, C2, D1 and D2 germinated seeds was extracted according to Trizol method [77]. cDNA was synthesized and amplified by PCR using the SMART™cDNA Library Construction Kit (Clontech, Palo Alto, CA, USA). Double-stranded cDNA was linked to pYES2 carrier by homologous recombination in vitro and quantitatively transferred to Escherichia coli by electric shock (TOP10) [94]. The quality of the library was determined by computational cloning and PCR. High purity plasmid extraction kit (Tiangen, DP116) was used to extract the library plasmid. The library was constructed by Nanjing Ruiyuan Biotechnology Co., LTD.

The pYES2 carrier and cDNA library were transformed into Saccharomyces cerevisiae INVSc1 [94]. A series of liquid medium containing galactose-substituted glucose (SG-U) and 0, 30, 60, 90, 120, 135 mM PEG 3350 were prepared to determine the appropriate concentration of inverters for screening drought-tolerant yeast (Supplementary Fig. 2). Under 120 mM PEG 3350 stress, all the drought-tolerant clones were collected. In order to identify the genes of the positive clones, the yeast colony detection kit (Nanjing Ruiyuan Biotechnology Co., LTD., P. R. China) was used. The positive clones were sequenced and compared with the sequences in GenBank database by Seqman and BLAST.

Data availability

The raw reads generated during the current study have been deposited in BioProject with the accession number of PRJNA1043090 (https://www.ncbi.nlm.nih.gov/bioproject/1043090).

Abbreviations

PEG:

Polyethylene Glycol

TSS:

Total Soluble Sugar

DEGs:

Differentially Expressed Genes

GO:

Gene Ontology

KEGG:

Kyoto Encyclopedia of Genes and Genomes

DAMs:

Differentially Accumulated Metabolites

C1:

M059 under normal growth

C2:

M024 under normal growth

D1:

M059 under 15% PEG treatment

D2:

M024 under 15% PEG treatment

PCA:

Principal Component Analysis

PCC:

Pearson Correlation Coefficient

CC:

Cellular Components

BP:

Biological Processes

MF:

Molecular Functions

DETFs:

Differentially Expressed TFs

bHLH:

basic Helix Loop Helix

FPKM:

Fragments Per Kilobase Per Million

RH:

Relative Humidity

FAA:

Formalin-Acetic Acid-Alcohol

SBS:

Sequencing By Synthesis

BLAST:

Basic Local Alignment Search Tool

FC:

Fold Change

FDR:

False Discovery Rate

LC-MS/MS:

Liquid Chromatography-Tandem Mass Spectrometry

VIP:

Variable Importance In The Projection

References

  1. Murphy KM, Bazile D, Kellogg J, Rahmanian M. Development of a Worldwide Consortium on Evolutionary participatory breeding in Quinoa. Front Plant Sci. 2016;7:608.

    Article  PubMed  PubMed Central  Google Scholar 

  2. Lin C, Liu ZJ, Dong YM, Vales M, Mao ZC. Domesticated cultivation and genetic breeding of Chenopodium quinoa. Yi Chuan. 2019;41:1009–22.

    PubMed  Google Scholar 

  3. Yao Y, Yang X, Shi Z, Ren G. Anti-inflammatory activity of saponins from quinoa (Chenopodium quinoa Willd.) Seeds in lipopolysaccharide-stimulated RAW 264.7 macrophages cells. J Food Sci. 2014;79:18–23.

    Article  Google Scholar 

  4. Vega-Gálvez A, Miranda M, Vergara J, Uribe E, Puente L, Martínez EA. Nutrition facts and functional potential of quinoa (Chenopodium quinoa willd.), an ancient Andean grain: a review. J Sci Food Agric. 2010;90:2541–7.

    Article  PubMed  Google Scholar 

  5. Afzal I, Haq MZU, Ahmed S, Hirich A, Bazile D. Challenges and perspectives for integrating Quinoa into the Agri-Food System. Plants (Basel). 2023;12:3361.

    CAS  PubMed  Google Scholar 

  6. Ren D, Yang Y, Yang Y, Richards K, Zhou X. Land-Water-Food Nexus and indications of crop adjustment for water shortage solution. Sci Total Environ. 2018;626:11–21.

    Article  CAS  PubMed  Google Scholar 

  7. Fang Y, Xiong L. General mechanisms of drought response and their application in drought resistance improvement in plants. Cell Mol Life Sci. 2015;72:673–89.

    Article  CAS  PubMed  Google Scholar 

  8. Wang X, Cai X, Xu C, Wang Q, Dai S. Drought-Responsive mechanisms in Plant leaves revealed by Proteomics. Int J Mol Sci. 2016;17:1706.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Ozturk M, Turkyilmaz Unal B, García-Caparrós P, Khursheed A, Gul A, Hasanuzzaman M. Osmoregulation and its actions during the drought stress in plants. Physiol Plant. 2021;172:1321–35.

    Article  CAS  PubMed  Google Scholar 

  10. Di Martino C, Pizzuto R, Pallotta ML, De Santis A, Passarella S. Mitochondrial transport in proline catabolism in plants: the existence of two separate translocators in mitochondria isolated from durum wheat seedlings. Planta. 2006;223:1123–33.

    Article  PubMed  Google Scholar 

  11. Zhou Q. The role of non-structural glucose metabolism pathway in response and adaptation to drought stress in Jatropha seedlings based on RNA-Seq. Yunnan Normal Univ. 2019; doi:CNKI:CDMD:2.1018.796925.

  12. Cazor A, Deborde C, Moing A, Rolin D, This H. Sucrose, glucose, and fructose extraction in aqueous carrot root extracts prepared at different temperatures by means of direct NMR measurements. J Agric Food Chem. 2006;54:4681–6.

    Article  CAS  PubMed  Google Scholar 

  13. Zhu X, Zhang M, Wang B, Song X, Wang X, Wei X. Non-targeted metabolomics analysis of metabolite changes in two quinoa genotypes under drought stress. BMC Plant Biol. 2023;23(1):503.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Iqbal H, Yaning C, Waqas M, Shareef M, Raza ST. Differential response of quinoa genotypes to drought and foliage-applied H2O2 in relation to oxidative damage, osmotic adjustment and antioxidant capacity. Ecotoxicol Environ Saf. 2018;164:344–54.

    Article  CAS  PubMed  Google Scholar 

  15. Cheng L, Jin J, He X, Luo Z, Wang Z, Yang J, Xu X. Genome-wide identification and analysis of the invertase gene family in tobacco (Nicotiana tabacum) reveals NtNINV10 participating the sugar metabolism. Front Plant Sci. 2023;14:1164296.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Fulda S, Mikkat S, Stegmann H, Horn R. Physiology and proteomics of drought stress acclimation in sunflower (Helianthus annuus L). Plant Biol (Stuttg). 2011;13(4):632–42.

    Article  CAS  PubMed  Google Scholar 

  17. Li NN, Qian WJ, Wang L, Cao HL, Hao XY, Yang YJ, Wang XC. Isolation and expression features of hexose kinase genes under various abiotic stresses in the tea plant (Camellia sinensis). J Plant Physiol. 2017;209:95–104.

    Article  CAS  PubMed  Google Scholar 

  18. Shahbazy M, Moradi P, Ertaylan G, Zahraei A, Kompany-Zareh M. FTICR mass spectrometry-based multivariate analysis to explore distinctive metabolites and metabolic pathways: a comprehensive bioanalytical strategy toward time-course metabolic profiling of Thymus vulgaris plants responding to drought stress. Plant Sci. 2020;290:110257.

    Article  CAS  PubMed  Google Scholar 

  19. Sweetlove LJ, Beard KF, Nunes-Nesi A, Fernie AR, Ratcliffe RG. Not just a circle: flux modes in the plant TCA cycle. Trends Plant Sci. 2010;15:462–70.

    Article  CAS  PubMed  Google Scholar 

  20. Xie H, Bai G, Lu P, Li H, Fei M, Xiao BG, Chen XJ, Tong ZJ, Wang ZY, Yang DH. Exogenous citric acid enhances drought tolerance in tobacco (Nicotiana tabacum). Plant Biol. 2022;24:333–43.

    Article  CAS  PubMed  Google Scholar 

  21. Leterrier M, Del Río LA, Corpas FJ. Cytosolic NADP-isocitrate dehydrogenase of pea plants: genomic clone characterization and functional analysis under abiotic stress conditions. Free Radic Res. 2007;41:191–9.

    Article  CAS  PubMed  Google Scholar 

  22. Gálvez L, González EM, Arrese-Igor C. Evidence for carbon flux shortage and strong carbon/nitrogen interactions in pea nodules at early stages of water stress. J Exp Bot. 2005;56(419):2551–61.

    Article  PubMed  Google Scholar 

  23. Zhang Y, Du P, Xiong F, Zhang X, Song H. WRKY genes improve Drought Tolerance in Arachis duranensis. Front Plant Sci. 2022;13:910408.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Liang Y, Ma F, Li B, Guo C, Hu T, Zhang M, Liang Y, Zhu J, Zhan X. A bHLH transcription factor, SlbHLH96, promotes drought tolerance in tomato. Hortic Res. 2022;9:198.

    Article  Google Scholar 

  25. Godwin J, Farrona S. Plant epigenetic stress Memory Induced by Drought: a physiological and molecular perspective. Methods Mol Biol. 2020;2093:243–59.

    Article  CAS  PubMed  Google Scholar 

  26. Wang X, Song S, Wang X, Liu J, Dong S. Transcriptomic and Metabolomic Analysis of Seedling-Stage Soybean Responses to PEG-Simulated Drought stress. Int J Mol Sci. 2022;23:6869.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Huan X, Li L, Liu Y, Kong Z, Liu Y, Wang Q, Liu J, Zhang P, Guo Y, Qin P. Integrating transcriptomics and metabolomics to analyze quinoa (Chenopodium quinoa Willd.) Responses to drought stress and rewatering. Front Plant Sci. 2022;13:988861.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Zhang L, Yu S, Zuo K, Luo L, Tang K. Identification of gene modules associated with drought response in rice by network-based analysis. PLoS ONE. 2012;7:33748.

    Article  Google Scholar 

  29. Wang L, Zheng B, Yuan Y, Xu Q, Chen P. Transcriptome profiling of Fagopyrum tataricum leaves in response to lead stress. BMC Plant Biol. 2020;20:54.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Liu M, He X, Feng T, Zhuo R, Qiu W, Han X, Qiao G, Zhang D. cDNA Library for Mining Functional genes in Sedum Alfredii Hance related to Cadmium Tolerance and characterization of the roles of a Novel SaCTP2 gene in enhancing Cadmium Hyperaccumulation. Environ Sci Technol. 2019;53:10926–40.

    Article  CAS  PubMed  Google Scholar 

  31. Zhang XH, Takagi H, Widholm JM. Expression of a novel yeast gene that detoxifies the proline analog azetidine-2-carboxylate confers resistance during tobacco seed germination, callus and shoot formation. Plant Cell Rep. 2004;22:615–22.

    Article  CAS  PubMed  Google Scholar 

  32. Nakahara Y, Sawabe S, Kainuma K, Katsuhara M, Shibasaka M, Suzuki M, Yamamoto K, Oguri S, Sakamoto H. Yeast functional screen to identify genes conferring salt stress tolerance in Salicornia europaea. Front Plant Sci. 2015;6:920.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Wang X, Liu Y, Yao R, et al. A systematic identification of cold tolerance genes in peanut using yeast functional screening system. Oil Crop Sci. 2023;8:184–90.

    Article  Google Scholar 

  34. Liu W, Yang F, Xie Z, et al. Physiological response and drought tolerance evaluation of different varieties of quinoa seedlings to drought stress. Agric Res Arid Areas. 2021;39:10–8.

    Google Scholar 

  35. Hu Y, Li WC, Xu YQ, Li GJ, Liao Y, Fu FL. Differential expression of candidate genes for lignin biosynthesis under drought stress in maize leaves. J Appl Genet. 2009;50:213–23.

    Article  CAS  PubMed  Google Scholar 

  36. Töpfer N, Kleessen S, Nikoloski Z. Integration of metabolomics data into metabolic networks. Front Plant Sci. 2015;6:49.

    PubMed  PubMed Central  Google Scholar 

  37. Gupta A, Rico-Medina A, Caño-Delgado AI. The physiology of plant responses to drought. Science. 2020;368(6488):266–9.

    Article  CAS  PubMed  Google Scholar 

  38. Thoden JB, Holden HM. The molecular architecture of galactose mutarotase/UDP-galactose 4-epimerase from Saccharomyces cerevisiae. J Biol Chem. 2005;280:21900–7.

    Article  CAS  PubMed  Google Scholar 

  39. Chen B. Study on α-galactosidase in mushroom. Shanxi Agricultural University. 2019; doi:CNKI:SUN:SYJB.0.2018-02-012.

  40. Zhang D, Guan T, Yan J, et al. Relationship between alkaline α-galactosidase activity and seed dehydration tolerance in early germination of maize. J Northwest Agricultural Sci. 2013;22:48–54.

    Google Scholar 

  41. Hara M, Tokunaga K, Kuboi T. Isolation of a drought-responsive alkaline α-galactosidase gene from New Zealand spinach. Plant Biotechnol. 2008;25:497–501.

    Article  CAS  Google Scholar 

  42. Husain Q. Beta galactosidases and their potential applications: a review. Crit Rev Biotechnol. 2010;30:41–62.

    Article  CAS  PubMed  Google Scholar 

  43. Hernández-Nistal J, Martín I, Dopico B, Labrador E. Coordinated action of β-galactosidases in the cell wall of embryonic axes during chickpea germination and seedling growth. Plant Biol. 2014;16:404–10.

    Article  PubMed  Google Scholar 

  44. Pego JV, Smeekens SC. Plant fructokinases: a sweet family get-together. Trends Plant Sci. 2000;5:531–6.

    Article  CAS  PubMed  Google Scholar 

  45. Gutmann A, Lepak A, Diricks M, Desmet T, Nidetzky B. Glycosyltransferase cascades for natural product glycosylation: use of plant instead of bacterial sucrose synthases improves the UDP-glucose recycling from sucrose and UDP. Biotechnol J. 2017. https://doi.org/10.1002/biot.201600557.

    Article  PubMed  Google Scholar 

  46. Yoon J, Cho LH, Tun W, Jeon JS, An G. Sucrose signaling in higher plants. Plant Sci. 2021;302:110703.

    Article  CAS  PubMed  Google Scholar 

  47. Fulda S, Mikkat S, Stegmann H, Horn R. Physiology and proteomics of drought stress acclimation in sunflower (Helianthus annuus L). Plant Biol. 2011;13:632–42.

    Article  CAS  PubMed  Google Scholar 

  48. Zörb C, Schmitt S, Mühling KH. Proteomic changes in maize roots after short-term adjustment to saline growth conditions. Proteomics. 2010;10:4441–9.

    Article  PubMed  Google Scholar 

  49. Wang L. Study on β-glucosidase BG3 and drought tolerance of Arabidopsis Thaliana. Henan Univ. 2014. https://doi.org/10.7666/d.D371597.

    Article  Google Scholar 

  50. Han YJ, Cho KC, Hwang OJ, Choi YS, Shin AY, Hwang I, Kim JI. Overexpression of an Arabidopsis β-glucosidase gene enhances drought resistance with dwarf phenotype in creeping bentgrass. Plant Cell Rep. 2012;31:1677–86.

    Article  CAS  PubMed  Google Scholar 

  51. Wang X, Khodadadi E, Fakheri B, Komatsu S. Organ-specific proteomics of soybean seedlings under flooding and drought stresses. J Proteom. 2017;162:62–72.

    Article  CAS  Google Scholar 

  52. Fernie AR, Carrari F, Sweetlove LJ. Respiratory metabolism: glycolysis, the TCA cycle and mitochondrial electron transport. Curr Opin Plant Biol. 2004;7:254–61.

    Article  CAS  PubMed  Google Scholar 

  53. He C, Liu L, Lu X. Analysis of glycolytis-related genes in leaves and roots of wild apple under NaCl stress based on transcriptome sequencing. J Fruit Sci. 2020;37:951–61.

    Google Scholar 

  54. Andersson I, Backlund A. Structure and function of Rubisco. Plant Physiol Biochem. 2008;46:275–91.

    Article  CAS  PubMed  Google Scholar 

  55. Cai Z, Liu G, Zhang J, Li Y. Development of an activity-directed selection system enabled significant improvement of the carboxylation efficiency of Rubisco. Protein Cell. 2014;5:552–62.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  56. Wang X, Vignjevic M, Jiang D, Jacobsen S, Wollenweber B. Improved tolerance to drought stress after anthesis due to priming before anthesis in wheat (Triticum aestivum L.) var. Vinjett J Exp Bot. 2014;65:6441–56.

    Article  CAS  PubMed  Google Scholar 

  57. Jurczyk B, Pociecha E, Grzesiak M, Kalita K, Rapacz M. Enhanced expression of Rubisco activase splicing variants differentially affects Rubisco activity during low temperature treatment in Lolium perenne. J Plant Physiol. 2016;198:49–55.

    Article  CAS  PubMed  Google Scholar 

  58. Tong J, Zhan G, Wang X, et al. Cloning and expression of citrate synthase gene in rapeseed under stress. Acta Cropologica Sinica. 2009;35:33–40.

    CAS  Google Scholar 

  59. Zhang N. Study on physiological differences and citrate synthase expression of white mustard under drought conditions. Zhengzhou Univ. 2012. https://doi.org/10.7666/d.y2102669.

    Article  Google Scholar 

  60. Yasutake Y, Watanabe S, Yao M, Takada Y, Fukunaga N, Tanaka I. Crystal structure of the monomeric isocitrate dehydrogenase in the presence of NADP+: insight into the cofactor recognition, catalysis, and evolution. J Biol Chem. 2003;278:36897–904.

    Article  CAS  PubMed  Google Scholar 

  61. Gálvez L, González EM, Arrese-Igor C. Evidence for carbon flux shortage and strong carbon/nitrogen interactions in pea nodules at early stages of water stress. J Exp Bot. 2005;56:2551–61.

    Article  PubMed  Google Scholar 

  62. Park Y, Kwon SH, Jang YL, Lee DH, Yang SO, Eo HJ, Park GH, Kwon HY. Nutritional composition and phytochemical screening in different parts of Hibiscus syriacus L. Food Sci Nutr. 2022;10:3034–42.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  63. Zhao Y. Transcriptome analysis reveals the mechanism of nitric oxide regulating alfalfa’s response to drought stress. Gansu Agricultural Univ. 2021. https://doi.org/10.27025/d.cnki.ggsnu.2020.000021.

    Article  Google Scholar 

  64. Rezayian M, Niknam V, Arabloo M. Iron nanoparticle regulate succinate dehydrogenase activity in canola plants under drought stress. Sci Rep. 2023;13:9628.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  65. Yi Y, Hassan MA, Cheng X, Li Y, Liu H, Fang W, Zhu Q, Wang S. QTL mapping and analysis for drought tolerance in rice by genome-wide association study. Front Plant Sci. 2023;14:1223782.

    Article  PubMed  PubMed Central  Google Scholar 

  66. Sun X, Wang Y, Sui N. Transcriptional regulation of bHLH during plant response to stress. Biochem Biophys Res Commun. 2018;503:397–401.

    Article  CAS  PubMed  Google Scholar 

  67. Jia Y. Functional study of Arabidopsis transcription factor bHLH146. Jilin Univ. 2022. https://doi.org/10.27162/d.cnki.gjlin.2022.006447.

    Article  Google Scholar 

  68. Banerjee A, Roychoudhury A. WRKY proteins: signaling and regulation of expression during abiotic stress responses. Sci World J. 2015;2015:807560.

    Article  Google Scholar 

  69. Chen J, Nolan TM, Ye H, Zhang M, Tong H, Xin P, Chu J, Chu C, Li Z, Yin Y. Arabidopsis WRKY46, WRKY54, and WRKY70 transcription factors are involved in Brassinosteroid-Regulated Plant Growth and Drought responses. Plant Cell. 2017;29:1425–39.

    CAS  PubMed  PubMed Central  Google Scholar 

  70. Wei W, Cheng MN, Ba LJ, Zeng RX, Luo DL, Qin YH, Liu ZL, Kuang JF, Lu WJ, Chen JY, Su XG, Shan W. Pitaya HpWRKY3 is Associated with Fruit Sugar Accumulation by transcriptionally modulating sucrose metabolic genes HpINV2 and HpSuSy1. Int J Mol Sci. 2019;20:1890.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  71. Yao Q, Qin P, Miao H, et al. Evaluation of drought resistance of quinoa during germination under PEG simulated drought stress. Xinjiang Agricultural Sci. 2019;56:1588–96.

    Google Scholar 

  72. Yao Y, Wu Y, Li M, et al. Microscopic observation and analysis of paraffin wax sections of pineapple pulp in water heart disease. J Trop Crops. 2023;44:2514–9.

    Google Scholar 

  73. Zhang Y. Glucose, fructose, sucrose and starch in fruits and vegetables were determined by anthrone spectrophotometer. Anal Chem. 1976;5:167–71.

    Google Scholar 

  74. Ren J. Genetic regularity and QTL mapping of soluble sugar content in tomato fruits. Northeast Agricultural Univ. 2019; doi:CNKI:CDMD:2.1018.135996.

  75. Zhou Z, Zhao B, Wang W, et al. Effects of different pretreatment on determination of three kinds of sugar in sweet sorghum stem juice. Pratacultural Sci. 2022;39:940–8.

    Google Scholar 

  76. Wang X. Principles and techniques of plant physiological and biochemical tests. Higher Education Press. China: Beijing; 2006. pp. 202–4.

    Google Scholar 

  77. Rio DC, Ares M Jr, Hannon GJ, Nilsen TW. Purification of RNA using TRIzol (TRI reagent). Cold Spring Harb Protoc. 2010. https://doi.org/10.1101/pdb.prot5439.

    Article  PubMed  Google Scholar 

  78. Li J. Breeding of peanut varieties with purple seed coat and high oleic acid and analysis of seed coat pigment DEGs. Hebei Agricultural Univ. 2022. https://doi.org/10.27109/d.cnki.ghbnu.2022.000305.

    Article  Google Scholar 

  79. Chen S, Zhou Y, Chen Y, Gu J. Fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics. 2018;34:884–90.

    Article  Google Scholar 

  80. Kim D, Langmead B, Salzberg SL. HISAT: a fast spliced aligner with low memory requirements. Nat Methods. 2015;12(4):357–60.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  81. Pertea M, Pertea GM, Antonescu CM, Chang TC, Mendell JT, Salzberg SL. StringTie enables improved reconstruction of a transcriptome from RNA-seq reads. Nat Biotechnol. 2015;33:290–5.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  82. Raney JA, Reynolds DJ, Elzinga DB, Page J, Udall JA, Jellen EN. Transcriptome analysis of drought induced stress in chenopodium quinoa. Am J Plant Sci. 2014;05(3):338–57.

    Article  CAS  Google Scholar 

  83. Gaudet P, Logie C, Lovering RC, Kuiper M, Lægreid A, Thomas PD. Gene Ontology representation for transcription factor functions. Biochim Biophys Acta Gene Regul Mech. 2021;1864:194752.

    Article  CAS  PubMed  Google Scholar 

  84. Kanehisa M, Furumichi M, Sato Y, Kawashima M, Ishiguro-Watanabe M. KEGG for taxonomy-based analysis of pathways and genomes. Nucleic Acids Res. 2023;51:587–92.

    Article  Google Scholar 

  85. Love MI, Huber W, Anders S. Moderated estimation of Fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15:550.

    Article  PubMed  PubMed Central  Google Scholar 

  86. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010;26:139–40.

    Article  CAS  PubMed  Google Scholar 

  87. Chen W, Gong L, Guo Z. A Novel Integrated Method for large scale detection, identification, and quantification of widely targeted metabolites: application in the study of Rice Metabolomics. Mol Plant. 2013;6:1769–80.

    Article  CAS  PubMed  Google Scholar 

  88. Tian LX, Zhang YC, Chen PL, Zhang FF, Li J, Yan F, Dong Y, Feng BL. How does the Waterlogging Regime affect crop yield? A Global Meta-Analysis. Front Plant Sci. 2021;12:634898.

    Article  PubMed  PubMed Central  Google Scholar 

  89. Ying L, Wang D, Du G. Analysis of Bioactive Components in the Fruit, roots, and leaves of Alpinia oxyphylla by UPLC-MS/MS. Evid Based Complement Alternat Med. 2021;2021:5592518.

    Article  PubMed  PubMed Central  Google Scholar 

  90. Sohrabi E, Tohidfa M, Ahmadikhah A, Ghanbari Moheb Seraj R. Data-Mining of Barley to identify salt stress hub genes, Gene expression analysis and recombinant plasmid construction. Iran J Biotechnol. 2023;21(3):e3389.

    PubMed  PubMed Central  Google Scholar 

  91. Xie H, Zhang P, Jiang C, Wang Q, Guo Y, Zhang X, Huang T, Liu J, Li L, Li H, Wang H, Qin P. Combined transcriptomic and metabolomic analyses of high temperature stress response of quinoa seedlings. BMC Plant Biol. 2023;23:292.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  92. Fernandes-Monteiro AG, Trindade GF, Yamamura AM, Moreira OC, de Paula VS, Duarte AC, Britto C, Lima SM. New approaches for the standardization and validation of a real-time qPCR assay using TaqMan probes for quantification of yellow fever virus on clinical samples with high quality parameters. Hum Vaccin Immunother. 2015;11(7):1865–71.

    Article  PubMed  PubMed Central  Google Scholar 

  93. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods. 2001;25:402–8.

    Article  CAS  PubMed  Google Scholar 

  94. Wang J, Mao X, Wang R, Li A, Zhao G, Zhao J, Jing R. Identification of wheat stress-responding genes and TaPR-1-1 function by screening a cDNA yeast library prepared following abiotic stress. Sci Rep. 2019;9:141.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

The authors thank Guangzhou Gene Denovn Biotechnology Co., LTD for RNA sequencing and metabolome analysis. We thank Shenzhen Meiji compilation information consulting Co., LTD for the polishing of this article.

Funding

This study was supported by the key research and development project of science and technology department of Hebei Province: Key Project of Technologies for Industrialization of Quinoa in Bashang Region (19227527D), and Science and Technology Innovation Team of Multi-grain and Multi-Bean Modern seed Industry (21326305D). This research was also funded by the construction subsidy of Hebei Quinoa Technology Innovation Center (227790177 H).

Author information

Authors and Affiliations

  1. North China State Key Laboratory of Crop Improvement and Regulation, Hebei Provincial Laboratory of Crop Germplasm Resources/College of Agronomy, Hebei Agricultural University, Baoding, 071000, Hebei Province, P. R. China

    Chunmei Wang, Junling Wang, Xiaoqing Liu, Xiaoxia Wang, Boxiang Wei & Guojun Mu

  2. The S&T Innovation Service Center of Hebei Province, Shijiazhuang, 050000, Hebei Province, P. R. China

    Chuan Lu, Yan Qin & Wei Lv

  3. The Quinoa Industrial Technology Research Institute of Hebei Province, Zhang Jiakou, 075000, Hebei Province, P. R. China

    Chunmei Wang, Junling Wang, Xiaoqing Liu, Xiaoxia Wang, Boxiang Wei & Guojun Mu

  4. Institute of Millet Crops, Key Laboratory of Genetic Improvement and Utilization for Featured Coarse Cereals(Co-construction by Ministry and Province), Ministry of Agriculture and Rural Affairs, The Key Research Laboratory of Minor Cereal Crops of Hebei Province, Hebei Academy of Agriculture and Forestry Sciences, Shijiazhuang, 050000, Hebei Province, P. R. China

    Zhimin Wei

  5. Shijiazhuang Fubao Ecological Technology Co., LTD, Shijiazhuang, 050000, Hebei Province, P. R. China

    Huilong Zhang

Authors
  1. Chunmei Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Chuan Lu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Junling Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Xiaoqing Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Zhimin Wei
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Yan Qin
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Huilong Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Xiaoxia Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Boxiang Wei
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Wei Lv
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Guojun Mu
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

GM and WL conceived and designed the research. CL, ZW, YQ, and HZ planted the materials and conducted field management. JW, XL, and XW does the transcriptomics data analysis. CL and BW does the metabolomics data analysis. CW wrote the manuscript. GM revised the manuscript. All authors contributed to the article and approved the submitted version.

Corresponding authors

Correspondence to Wei Lv or Guojun Mu.

Ethics declarations

Ethics approval and consent to participate

The study complies with relevant institutional, national, and international guidelines and legislations. All procedures were conducted in accordance with the guidelines.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wang, C., Lu, C., Wang, J. et al. Molecular mechanisms regulating glucose metabolism in quinoa (Chenopodium quinoa Willd.) seeds under drought stress. BMC Plant Biol 24, 796 (2024). https://doi.org/10.1186/s12870-024-05510-w

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12870-024-05510-w

Keywords

BMC Plant Biology

ISSN: 1471-2229

Contact us