Skip to main content
NCBI home page
Food Chemistry: X logoLink to Food Chemistry: X
. 2025 Dec 19;33:103409. doi: 10.1016/j.fochx.2025.103409

Ecological and genetic regulation of nutritional composition, pasting properties and metabolomic profiles in foxtail millet

Xin Zhao a,b,c, Yueyue Wang a, Ziyi Deng a, Zhongxiang Li a, Meng Yue a, Hongying Li a,⁎, Yuanhuai Han a,⁎
PMCID: PMC12807849  PMID: 41551807

Abstract

The chemical composition and functional properties of foxtail millet is jointly regulated by genetic and ecological factors. In this study, field trials and UPLC-ESI-MS-based widely-targeted metabolomics were combined to evaluate the nutritional quality, pasting properties, and metabolomic profiles of three cultivars grown in three distinct ecological regions. Nutritional composition was mainly influenced by altitude, effective accumulated temperature, and available potassium, whereas pasting properties were mainly influenced by precipitation and altitude. Metabolomic analysis identified lipids, flavonoids, and alkaloids as the predominant classes, with significant variations observed across different regions. High-altitude conditions promoted the accumulation of L-malic acid (an organic acid) and flavonoid glycosides, supported by enhanced carbon and linoleic acid metabolism. In contrast, low-altitude regions favored phenolamine alkaloids and amino acid derivatives. Cultivar-specific metabolic patterns were also identified. The traditional cultivar accumulated alkaloids and phenolic acids with potential health-promoting properties, whereas the modern cultivar was characterized by a distinct abundance of flavonoids with strong antioxidant activity. These results clarify the metabolic foundation of foxtail millet quality and provide a strategy for cultivar selection and breeding to enhance its value for functional food applications.

Keywords: Setaria italica, Cultivar, Environment, Quality, Ultra performance liquid chromatography-electrospray ionization-mass spectrometry (UPLC-ESI-MS)

Highlights

  • •

    Altitude and precipitation shape nutritional and pasting properties of foxtail millet.

  • •

    High-altitude promotes L-malic acid and flavonoid glycosides; low-altitude favors alkaloids and amino acid derivatives.

  • •

    Cultivar-specific metabolic signatures support precision breeding.

  • •

    Ecology and genetics distinctly regulate lipid and flavonoid pathways.

1. Introduction

Foxtail millet (Setaria italica L.) is crucial for agriculture in arid and semi-arid regions due to its exceptional drought tolerance and adaptability (Liang & Han, 2024; Yang et al., 2020; Zhao, Huang, Huang, & Liu, 2023). It also possesses a unique nutritional profile with various health benefits, making it a promising raw material for functional foods (Muthamilarasan and Prasad, 2021 and Prasad, 2021; Sachdev, Goomer, & Singh, 2021). Foxtail millet is rich in proteins, lipids, vitamins, and minerals, along with diverse bioactive compounds such as flavonoids, alkaloids, and phenolic acids (He, Zhang, Wang, Li, & Han, 2015). These compounds reported in literature to have antioxidant, anti-inflammatory, and hypoglycemic effects, indicating potential uses in health-promoting foods (Arora, Aggarwal, Dhaliwal, Gupta, & Kaushik, 2023; Zhang et al., 2021).

The nutritional and functional properties of foxtail millet are shaped by genetic background and ecological conditions (Hou et al., 2023; Yang et al., 2021). Environmental factors, including temperature, solar radiation, precipitation, and soil nutrient availability, regulate the nutritional and functional traits of foxtail millet by influencing its physiological processes and secondary metabolite synthesis (Li et al., 2022). The ecological effects on quality are complex. High altitude, sufficient precipitation, and large diurnal temperature fluctuations generally enhance nutritional value, while excessive accumulated temperatures may reduce it (Ning et al., 2017). Notably, lower UV-B radiation has been shown to decrease carotenoid content (Huo et al., 2024). Meteorological factors mainly affect commercial quality, whereas soil organic matter has a stronger influence on cooking properties (Ma et al., 2023). Furthermore, the genetic background plays a crucial role, as evidenced by the fact that the accumulation of carotenoids and flavonoids is closely associated with allelic variations of the PSY1 gene (He et al., 2022).

Metabolomics is a high-throughput approach that offers comprehensive profiling of crop metabolic composition, revealing the biochemical mechanisms underlying trait variation and environmental responses (Mushtaq, Choi, Verpoorte, & Wilson, 2014; Qian et al., 2023; Yun, Chin, Hong, & Lee, 2025). In foxtail millet, metabolomic studies have revealed distinct profiles among cultivars and identified key metabolic pathways, such as linoleic acid metabolism and phenylpropanoid biosynthesis, which correlate with sensory and functional properties (Li et al., 2022; Yu, Zong, Zhang, & Zhang, 2021; Zhang et al., 2024). Environmental factors also profoundly shape metabolic networks, altering pathways such as carbohydrate metabolism, the TCA (tricarboxylic acid) cycle, and the synthesis of phenolics and lipids, ultimately modulating nutritional value and bioactivity (Wang et al., 2023; Wang et al., 2024). When applied to food crops, this food metabolomics approach becomes essential for connecting genetic and environmental factors with the dynamic chemical makeup that governs food quality, functionality, and biological activity.

Although foxtail millet is recognized for its nutritional benefits, the combined effects of genotype-by-environment (G × E) interactions on its quality are not fully understood. It was hypothesized that the foxtail millet metabolome is distinctly shaped by genetic and ecological factors, leading to differentiated quality traits. To test this, an approach integrating multi-location field trials with widely-targeted metabolomics was applied to three cultivars across three ecological regions. This design advances previous metabolomic studies by explicitly resolving genotype-by-environment (G × E) interactions and linking specific environmental variables to functional metabolite accumulation. The study had three main objectives: (1) to discover cultivar-specific and region-specific metabolites with potential for functional foods; (2) to conduct a comprehensive metabolomic analysis to pinpoint functionally important metabolites; and (3) to assess genetic and geographic variations in nutritional and RVA (Rapid Visco Analyzer) properties and identify their key regulatory factors. These findings establish a metabolic basis for precision breeding and quality improvement in foxtail millet.

2. Materials and methods

2.1. Experimental materials

Three representative high-quality foxtail millet cultivars from China were used for this study: Jingu 21 (JG21, traditional elite cultivar), Qinhuang 2 (QH2, derived from a superior landrace), and Zhonggu 19 (ZG19, newly bred elite cultivar). These cultivars were cultivated in three representative locations in Shanxi Province, China: Wutai (WT), Fenyang (FY), and Changzhi (CZ) (Fig. 1A). Each location had distinct environmental conditions (Fig. 1B and C).

Fig. 1.

Fig. 1

Open in a new tab

A: Geographic information of regions. B: Meteorological characteristics across regions (from May to September). C: Soil chemical properties across regions. D-E: Correlation analysis between nutritional composition (D) and RVA profile (E) with ecological factors. ALT: Altitude (m); ADT: daily average temperature (°C); DTR: diurnal temperature range (°C); EAT: effective accumulated temperature (°C); PRE: precipitation (mm); SSD: sunshine duration (h); SOM: soil organic matter (g/kg), TN: total nitrogen (g/kg); AP: available phosphorus (mg/kg); AK: available potassium (mg/kg); Ca: available calcium (g/kg); PV: peak viscosity (cP); BD: breakdown (cP); SB: setback (cP); GT: gelatinization temperature (°C).

2.2. Soil chemical analysis

Soil samples were collected from the top layer (0–20 cm) of each experimental plot before sowing. After air-drying, the samples were homogenized and passed through a 1-mm sieve. Soil properties were analyzed using standard methods: pH was measured using potentiometry; soil organic matter (SOM) was measured using potassium dichromate oxidation-volumetric method; total nitrogen (TN) was measured using the Kjeldahl method; available phosphorus (AP) was measured using the sodium bicarbonate extraction – molybdenum antimony anti-colorimetric method; available potassium (AK) was measured using the ammonium acetate extraction - flame photometry method; available calcium (Ca) was measured using inductively coupled plasma optical emission spectrometry (ICP-OES) method. All measurements were conducted with three technical replicates.

2.3. Meteorological data

Meteorological data for the foxtail millet growing season (May to September) were obtained from the China meteorological data network (http://data.cma.cn). The parameters included daily average temperature (ADT), precipitation (PRE), diurnal temperature range (DTR), and sunshine duration (SSD). The effective accumulated temperature (EAT) was calculated as EAT = ∑(ADT - Tbase), where Tbase is the biological base temperature of foxtail millet (10 °C). Only days with ADT ≥ 10 °C were included in accumulated (unit: °C·d).

2.4. Determination of foxtail millet quality

Kernel color: kernel yellowness (b value) was measured using a non-contact X-Rite VS450 colorimeter (USA), where positive b values indicate increased yellowness and negative b values indicate enhanced blueness. Each treatment was analyzed with three biological replicates.

Nutritional composition: approximately 3 g of whole grains from each sample were scanned using a NIRSTM DS 2500 near-infrared spectrometer (Denmark) at wavelength between 1100 and 2498 nm. Fat, protein, and amylose contents were predicted based on pre-established calibration models.

Rapid Visco Analyzer Analysis (RVA Profile): Pasting properties were determined using an RVA-TecMaster analyzer according to a standard pasting profile (Supplementary Materials S1.1). Pasting profiles and key parameters (peak viscosity, breakdown, setback, and pasting temperature) were recorded, with three replicates per sample.

2.5. Widely targeted metabolomics analysis

2.5.1. Sample preparation and extraction

Biological samples were freeze-dried, ground, and extracted with 70 % aqueous methanol (containing internal standards) at a solvent-to-sample ratio of 24:1 (v/w). The extraction procedure was carried out as detailed in Supplementary Materials S1.2. The supernatant was filtered and used for UPLC-MS/MS analysis.

2.5.2. UPLC-ESI-MS/MS analysis

Metabolite separation was performed on an Agilent SB-C18 column (1.8 μm, 2.1 mm × 100 mm) with a water-acetonitrile gradient (both containing 0.1 % formic acid) at a flow rate of 0.35 mL/min. The column temperature was maintained at 40 °C. Mass spectrometry detection was conducted using a QTRAP-MS system with an electrospray ionization (ESI) source operated in both positive and negative ion modes. Specific MS parameters are provided in Supplementary Materials S1.3.

2.5.3. Metabolomic data quality control and normalization

Quality control was monitored by analyzing a pooled QC sample throughout the sequencing run. Metabolite identification was performed by matching MS/MS spectra and retention times against the MWDB database (MetWare). Raw data were processed using MultiQuant software (v3.0.3) and normalized by unit variance scaling. Detailed criteria for metabolite identification levels and QC results are described in Supplementary Materials S1.4 (Lines 25–41).

2.6. Identification of key active ingredients

Potential key bioactive ingredients were screened from the detected metabolites using the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP) (Ru et al., 2014). Metabolites meeting the criteria of oral bioavailability (OB) ≥ 5 % and drug-likeness (DL) ≥ 0.14 were considered potential bioactive compounds.

2.7. Differential accumulated metabolites analysis

Unsupervised principal component analysis (PCA) was performed using R software. Differential accumulated metabolites (DAMs) between groups were identified based on a variable importance in projection (VIP) score > 1 from orthogonal partial least squares-discriminant analysis (OPLS-DA) and an absolute log₂ fold change (|log₂FC|) ≥ 1.0.

Metabolites were annotated against the KEGG Compound database and mapped to KEGG pathways. Enriched pathways were identified using metabolite set enrichment analysis (MSEA) based on hypergeometric test p-values.

2.8. Data analysis and software

All statistical analyses were performed in R (www.r-project.org). OPLS-DA was conducted using the MetaboAnalyst R package, and heatmaps were generated using ComplexHeatmap. The RVA profile was controlled by ThermoCline for Windows software. Public databases including KEGG (http://www.kegg.jp/kegg/compound/) and TCMSP (https://www.91tcmsp.com) were used for metabolite pathway annotation and bioactivity screening, respectively.

3. Results and discussions

3.1. Nutritional composition of foxtail millet

Foxtail millet is rich in unsaturated fatty acids (>80 %), primarily oleic acid, linoleic acid, and α-linolenic acid (Zhang, 2019). ZG19 generally exhibited higher lipid content, particularly when cultivated in FY, whereas QH2 showed consistently lower levels, especially in WT (Table 1). This pattern suggests that cultivar ZG19 grown in FY showed potential for oil extraction and flavor-enhanced products, whereas the low-lipid QH2 grown in WT may be advantageous for shelf-stable foods by minimizing oxidation risk.

Table 1.

Quality of foxtail millet for different culivars across three regions.

Sample
 Fat (g/100 g)
 Protein (g/100 g)
 Amylose (g/100 g)
 b value
WT 3.70 ± 0.08 b 11.48 ± 0.07 b 12.79 ± 0.10 a 46.78 ± 0.59 b
JG21 FY 4.26 ± 0.19 a 12.64 ± 0.08 a 12.08 ± 0.13 b 48.42 ± 0.17 a
CZ 3.63 ± 0.11 b 12.51 ± 0.37 a 12.66 ± 0.21 a 46.19 ± 1.02 b
WT 3.64 ± 0.12 a 10.01 ± 0.24 c 15.02 ± 0.36 a 45.21 ± 0.09 b
QH2 FY 3.70 ± 0.11 a 11.00 ± 0.06 a 13.83 ± 0.11 b 46.29 ± 0.39 a
CZ 3.56 ± 0.24 a 10.51 ± 0.09 b 13.83 ± 0.74 b 45.20 ± 0.37 b
WT 3.99 ± 0.37 b 8.90 ± 0.24 a 14.11 ± 1.20 a 42.34 ± 0.46 c
ZG19 FY 4.64 ± 0.05 a 8.80 ± 0.09 ab 12.64 ± 0.23 b 48.08 ± 0.63 a
CZ 3.70 ± 0.05 b 8.54 ± 0.08 b 13.56 ± 0.05 ab 46.51 ± 0.89 b
JG21 3.86 ± 0.35 b 12.21 ± 0.64 a 12.51 ± 0.38 b 47.13 ± 1.16 a
QH2 3.63 ± 0.07 b 10.51 ± 0.49 b 14.23 ± 0.69 a 45.57 ± 0.63 a
ZG19 4.11 ± 0.48 a 8.75 ± 0.18 c 13.44 ± 0.74 ab 45.64 ± 2.97 a
WT 3.78 ± 0.19 b 10.13 ± 1.30 b 13.97 ± 1.12 a 44.78 ± 2.25 b
FY 4.20 ± 0.47 a 10.81 ± 1.93 a 12.85 ± 0.89 b 47.60 ± 1.14 a
CZ 3.63 ± 0.07 b 10.52 ± 1.98 ab 13.35 ± 0.61 ab 45.97 ± 0.68 b
ANOVA G 16.73 *** 853.77 *** 26.62 *** 20.24 ***
F value E 25.96 *** 33.43 *** 11.37 ** 52.17 ***
G × E 4.52 * 15.55 ** 1.31 ns 19.50 ***
Open in a new tab

Note: Lowercase letters indicate significant differences among the vertical groups (P < 0.05);*, ** and ***: significant difference at 0.05, 0.01 and 0.001, respectively；ns: not significant.

In addition to lipids, foxtail millet is distinguished by its superior essential amino acid composition (He et al., 2015; Verma, Srivastava, & Tiwari, 2015), which offers hypoallergenic and blood glucose regulation benefits (Fu et al., 2021). Protein content was primarily determined by cultivar, with JG21 accumulating significantly more protein than QH2 and ZG19 (P < 0.05; Table 1). Among regions, FY produced the highest overall protein content. The JG21-FY combination achived the highest protein content, highlighting its potential for higher protein applications. Elevated protein content also improves water retention and texture in millet-based foods, enhancing the quality of products such as pasta and meat analogues.

Amylose content influences the cooking and sensory quality (Tao, Yu, Prakash, & Gilbert, 2019), and was influenced by both cultivar and environment (P < 0.01; Table 1). QH2 had the highest amylose content, while JG21 had the lowest. The low to moderate amylose content in JG21 typically results in softer and cohesive gels upon cooking, making it ideal for traditional foods such as porridge. Among regions, WT showed significantly higher amylose content than FY (P < 0.05), further highlighting the potential for environment-specific cultivar selection to achieve desired properties.

Kernel color (b value) serves as a key quality trait due to its correlation with carotenoid content (Sun et al., 2021; Yang et al., 2017). Environmental factors demonstrated the predominant influence on b value (Table 1). FY exhibited the highest b value, significantly surpassing WT and CZ (P < 0.05). Among the cultivars, JG21 showed the highest b value. QH2 showed the lowest variability, while ZG19 showed greater variation. These results position JG21 as a promising candidate for premium food markets due to its superior and relatively stable color attributes.

Foxtail millet exhibits high sensitivity to ecological factors (Liu et al., 2020), and the three regions exhibited distinct environmental gradients (Fig. 1B and C). Correlation analysis revealed that protein content was positively correlated with EAT, pH, and AK, but negatively correlated with ALT (P < 0.05; Fig. 1D). This suggests that warmer climates with sufficient potassium promote protein biosynthesis, as potassium facilitates nitrogen assimilation into amino acids (Hess, Lu, Rabinowitz, & Botstein, 2006). In contrast, amylose content was negatively correlated with EAT and AK but positively correlated with ALT (P < 0.05). Cooler high-altitude conditions may enhance the activity of granule-bound starch synthase (GBSS), increasing the amylose-to-amylopectin ratio (Tao et al., 2019). These relationships support a strategy of region-specific cultivar selection: high-protein, low-amylose cultivars (e.g., JG21) are favored in warm, potassium-rich regions like FY, whereas high-amylose cultivars (e.g., QH2) perform better in cooler, high-altitude regions like WT.

3.2. RVA profile of foxtail millets

The pasting properties of foxtail millet, characterized by Rapid Visco Analyzer (RVA) profiles, provide crucial insights into starch gelatinization behavior and processing suitability (Balet, Guelpa, Fox, & Manley, 2019; Fitzgerald, Martin, Ward, Park, & Shead, 2003; Xuan et al., 2020). Genotype was identified as the primary factor influencing major RVA parameters (Table 2).Analysis of variance revealed that genotype was the primary factor influencing major RVA parameters, with significant effects also from environment and their interaction.The distinct RVA profiles observed among cultivars and regions suggest their potential suitability for different food processing applications (Table S1). Specifically, the phenotypic trends indicate that cultivars with higher peak viscosity and breakdown may be better suited for products requiring a soft texture, whereas those with higher setback and thermal stability could be more appropriate for foods that need to maintain structural integrity after cooking. Further correlation analysis revealed that PV was significantly positively correlated with PRE (P < 0.05) and negatively correlated with Ca (P < 0.05). BD showed a positive correlation with ALT, while SB exhibited a negative correlation with Ca (P < 0.05; Fig. 1E). These results reinforce the role of specific environmental factors as key regulators of foxtail millet's functional properties for processing.

Table 2.

RVA profile of foxtail millet for different culivars across three regions.

Sample
 PV(cP)
 BD(cP)
 SB(cP)
 GT(°C)
WT 1150.00 ± 19.08 a 319.00 ± 29.05 a 903.67 ± 13.32 a 82.02 ± 0.49 c
JG21 FY 982.00 ± 17.52 b 118.33 ± 8.08 c 900.67 ± 5.69 a 84.75 ± 0.05 a
CZ 832.00 ± 16.37 c 186.00 ± 7.00 b 834.00 ± 3.00 b 83.93 ± 0.03 b
WT 1309.67 ± 5.51 a 255.67 ± 52.77 a 1083.67 ± 20.21 a 81.75 ± 0.48 a
QH2 FY 1073.67 ± 41.14 c 107.33 ± 36.83 c 980.00 ± 33.78 b 82.07 ± 0.53 a
CZ 1172.67 ± 17.79 b 181.67 ± 0.58 b 1051.67 ± 21.46 a 82.05 ± 0.43 a
WT 1443.00 ± 59.27 a 451.00 ± 14.42 a 1034.00 ± 33.06 a 79.37 ± 0.51 a
ZG19 FY 1444.67 ± 37.53 a 364.33 ± 59.21 b 902.67 ± 14.50 b 77.40 ± 0.05 b
CZ 1348.00 ± 17.06 b 465.67 ± 18.15 a 831.67 ± 5.86 c 76.62 ± 0.03 c
JG21 988.00 ± 159.08 b 207.78 ± 102.09 b 879.44 ± 39.38 b 83.57 ± 1.40 a
QH2 1185.33 ± 118.51 b 181.56 ± 74.17 b 1038.44 ± 53.08 a 81.96 ± 0.18 a
ZG19 1411.89 ± 55.34 a 427.00 ± 54.76 a 922.78 ± 102.65 ab 77.79 ± 1.42 b
WT 1300.89 ± 146.70 a 341.89 ± 99.66 a 1007.11 ± 92.96 a 81.04 ± 1.46 a
FY 1166.78 ± 244.98 b 196.67 ± 145.31 b 927.78 ± 45.24 a 81.41 ± 3.72 a
CZ 1117.56 ± 262.38 b 277.78 ± 162.73 ab 905.78 ± 126.35 a 80.87 ± 3.80 a
ANOVA G 244.95 *** 160.93 *** 153.47 *** 195.52 ***
F value E 89.04 *** 46.92 ** 64.53 *** 5.06 *
G × E 25.39 *** 4.48 * 24.17 *** 42.31***
Open in a new tab

Note: Lowercase letters indicate significant differences among the vertical groups (P < 0.05);*, ** and ***: significant difference at 0.05, 0.01 and 0.001, respectively.

3.3. Comprehensive analysis of foxtail millet metabolite composition and variation

Widely-targeted metabolomic analysis identified 951 metabolites, categorized into 12 major classes. Lipids, flavonoids, and alkaloids were the most abundant groups (Fig. 2A), consistent with the reported importance of secondary metabolites in foxtail millet quality (Qiao, Li, Jia, Zhang, & Wang, 2025). Screening with the TCMSP database revealed 81 key bioactive compounds (OB ≥5 %, DL ≥0.14), predominantly flavonoids, lipids, and terpenoids. This further supports the dual medicinal and nutritional value of foxtail millet (Qian et al., 2023).

Fig. 2.

Fig. 2

Open in a new tab

A: Classification of metabolites and the proportion of various categories. B: OPLS-DA score plot of different samples. C: Heatmap of all metabolites across different samples. A1: WT-JG21; A2: WT-QH2; A3: WT-ZG19; B1: FY-JG21; B2: FY-QH2; B3: FY-ZG19; C1: CZ-JG21; C2: CZ-QH2; C3: CZ-ZG19.

Lipids constituted the most diverse class, including free fatty acids, glycerol esters, and lysophospholipids (Fig. S2). Notably, unique lipid metabolites such as 9,10-dihydroxy-12,13-epoxyoctadecanoic acid were specifically detected in samples from WT. The accumulation of these specific lipids may be associated with the high-altitude and low-temperature conditions (Chen et al., 2022). Furthermore, various lysophospholipids were identified, which are a class of metabolites reported in the literature to possess bioactivity (Alves et al., 2020).

Flavonoids constituted the second largest class of metabolites, comprising flavones, flavonols, isoflavones, and others (Fig. S2). In plants, flavonoids aid in UV protection and antioxidant defense, and they are also associated with health-promoting activities in humans, such as antioxidant and immunomodulatory effects (Perez-Vizcaino & Fraga, 2018). Their accumulation varied significantly by region, and cultivar-specific patterns were observed: JG21, QH2, and ZG19 contained 2, 3, and 4 characteristic flavonoid metabolites, respectively, indicating their potential use as markers for geographical authentication (Zhang et al., 2021).

Alkaloids constituted the third major class, including phenolamines, plumeranes, and pyridine alkaloids (Fig. S2). Phenolamine, quinoline, and pyridine alkaloids identified in this study exhibit diverse bioactivities, including potential neuroprotective and metabolic health effects, respectively, but these attributions need further validation (Rauf et al., 2019). The biosynthesis of alkaloids was also influenced by regional variations in soil and climate. Specifically, the accumulation of certain types, such as quinolines, appears sensitive to differences in soil microbial communities (Sugiyama, Takahashi, Katsube, Koyama, & Itamura, 2016).

The unique metabolite composition of foxtail millet distinguishes it from other cereals. Unlike pearl and finger millets, which are richer in tannins, foxtail millet is characterized by an exceptional abundance of lysophospholipids and phenolamine alkaloids (Li et al., 2021; Li et al., 2021; Ma et al., 2023). It also possesses greater flavonoid diversity than major staples such as rice and maize (Tang et al., 2022). Additionally, foxtail millet accumulates rare metabolites related to sugar metabolism, such as N-(1-deoxy-1-fructosyl)leucine, along with stress-response compounds like L-malic acid and specific phenolamides. Functionally, lipids contribute to cardiovascular benefits and sensory quality, while flavonoids and alkaloids provide antioxidant and anti-inflammatory potential. These bioactive components collectively influence starch digestibility and food texture, underscoring the potential of foxtail millet as a raw material for health-oriented functional foods.

PCA (principal component analysis) revealed clear metabolic differences among cultivars and regions (Fig. 2B). All WT samples formed a distinct cluster, reflecting the impact of high-altitude conditions (Zhang et al., 2023). JG21 and QH2 showed similar metabolic profiles in FY and CZ, whereas ZG19 maintained a consistent profile across both regions (Fig. 2C), indicating genotype-by-environment interactions (Chen et al., 2022). Differential accumulated metabolite (DAM) analysis further identified over 100 DAMs in most comparison groups (Fig. S3 A). Cultivar-specific analysis revealed 5, 10, and 10 region-specific DAMs for JG21, QH2, and ZG19, respectively, suggesting their utility as geographical markers. WT consistently up-regulated more DAMs across cultivars (Fig. S3B), implying that its plateau environment triggers shared metabolic mechanisms (Chen et al., 2022). Collectively, these results demonstrate that both ecological and genetic factors significantly shape the metabolome of foxtail millet. To further dissect these influences, subsequent analyses focused on region- and cultivar-specific DAMs and their related metabolic pathways.

3.4. Ecological regulation of metabolite profiles in foxtail millet

3.4.1. Region-specific metabolic profiling and functional characterization

Principal component analysis (PCA) of the total metabolites (mean values across three cultivars) revealed clear separation trends among the three regions along PC1 and PC2 (Fig. 3B), indicating distinct region-specific metabolic profiles. Differential analysis identified 256 region-specific metabolites (FC ≥ 2 or ≤ 0.5, VIP ≥ 1), predominantly flavonoids (81), lipids (42), and nucleotides and derivatives (24) (Fig. 3A and C). K-means clustering analysis further classified these DAMs into four distinct clusters (Fig. 3D; Fig. S4), revealing clear geographical patterns and confirming the substantial influence of production region on the foxtail millet metabolome.

Fig. 3.

Fig. 3

Open in a new tab

A: Heatmap of differentially accumulated metabolites across different regions. B: OPLS-DA score plot of different samples. C: Venn diagram of differential accumulated metabolites across different regions. D: Variation trends of metabolites across three regions.

Cluster 1 was identified as the characteristic metabolite profile of FY. Notably, 2-aminoethanesulfonic acid (taurine) was notably accumulated. This compound, commonly used in energy drinks, is associated with potential anti-fatigue and endurance-enhancing effects (Aamer et al., 2024). Soil analysis indicated that this regional specificity might be attributed to the relatively high soil pH (8.51), potentially up-regulating sulfur assimilation pathways in foxtail millet (Stipanuk & Ueki, 2011).

Cluster 2 was identified as the co-enriched metabolites in both the FY and CZ, contained key bioactive compounds including a flavonoid, a lipid, and a terpenoid (Table S3). The common accumulation patterns in these regions likely reflect their comparable ecological conditions. Notably, ergothioneine was identified as a characteristic functional constituent, known for its potent antioxidant and anti-aging properties, suggesting its potential application in protective formulations for skin and organ health (Ta, Buchmeier, Newton, Rawat, & Fahey, 2011).

Cluster 3 represents the characteristic metabolic profile of WT region, contained eight key bioactive compounds (Table S3). Particularly, high altitude and low effective accumulated temperature promoted the accumulation of L-malic acid and N-(1-deoxy-1-fructosyl)leucine. The latter metabolite plays significant roles in energy metabolism and flavor profile (Moller, Andrews, & Cheeseman, 1977), while L-malic acid enhances iron absorption efficiency (Li et al., 2022). The unique WT environment thus promotes a metabolome with implications for both nutritional functionality and distinctive taste.

Cluster 4 exhibited a distinct WT > CZ > FY gradient metabolic profile, comprising eleven key bioactive compounds (Table S3), primarily flavonoids and lipids. Notably, the phenolic amine alkaloid N-p-coumaroyl hydroxydehydroagmatine and the saccharide derivative D-xylonic acid were identified as significant geographical biomarkers (Table S2). The former has demonstrated antioxidant and anti-inflammatory properties in pharmacological studies (Rauf et al., 2019), while the latter is of interest for its potential prebiotic activity (Yim et al., 2017).

This study elucidated region-specific metabolic profiles in foxtail millet, establishing a scientific framework for geographical authentication and quality assessment. Nutritionally, the unique composition of WT-grown foxtail millet (L-malic acid, linoleic acid, and flavonoid glycosides) contributes to enhanced energy metabolism and antioxidant capacity, whereas foxtail millet from FY and CZ regions contain distinct alkaloids and amino acid derivatives that influence flavor and digestibility. These findings provide metabolic markers for origin traceability and highlight the potential to leverage region-specific compositions for developing functionally tailored foxtail millet products.

3.4.2. Pathway analysis and nutritional implications of different regions

KEGG pathway analysis revealed that the 54 DAMs identified in the WT vs FY comparison were mapped to 110 metabolic pathways (Fig. S5A; Fig. S6). The most significantly altered pathways were carbon metabolism, linoleic acid metabolism, and biosynthesis of alkaloids derived from terpenoid and polyketide. The up-regulation of linoleic acid metabolism particularly suggests enhanced potential for anti-inflammatory and cholesterol-lowering effects in foxtail millet from the WT region (Chen et al., 2022; Zhang, 2019). Specific metabolites in this pathway, such as 9-HODE and 13-HODE, have documented bioactivities relevant to human health (Alves et al., 2020).

In the WT vs CZ comparison, 45 DAMs were mapped to 107 pathways (Fig. S5B; Fig. S6), with linoleic acid metabolism, two-component system, and α-linolenic acid metabolism being most prominent. The latter two pathways are involved in plant environmental adaptation, potentially indicating more stressful growing conditions in WT. Notably, five DAMs were associated with human metabolic disorders, suggesting that WT foxtail millet merits further investigation for potential applications in metabolic health (Grabowska et al., 2016).

The FY vs CZ comparison showed fewer significant pathway alterations. The most notable were flavone and flavonol biosynthesis and neuroactive ligand-receptor interaction (Fig. S6). The up-regulated flavonoid metabolites in CZ suggest enhanced antioxidant activity in foxtail millet from this region. Conversely, the down-regulation of antibiotic resistance pathways may reflect inherent differences in soil microbial communities between FY and CZ.

In summary, our results demonstrate substantial ecological regulation of the foxtail millet metabolome and functional traits (Fig. 4A). Altitude, effective accumulated temperature, and soil available potassium were primary drivers of nutritional composition, while precipitation and altitude were key for pasting properties. Metabolically, the high-altitude conditions of WT consistently promoted enhanced carbon and linoleic acid metabolism, leading to the accumulation of organic acids like L-malic acid and specific flavonoid glycosides. In contrast, the warmer, potassium-rich conditions of FY favored the synthesis of different alkaloids and amino acid derivatives. This region-specific metabolic signature provides a robust scientific basis for geographical origin traceability and for tailoring cultivation regions to target specific functional or nutritional outcomes in foxtail millet products.

Fig. 4.

Fig. 4

Open in a new tab

Association of metabolites and functional attributes with ecological (A) and genetic (B) factors. EAT: effective accumulated temperature; PRE: precipitation; AK: available potassium.

3.5. Genetic regulation of metabolite profiles in foxtail millet

3.5.1. Cultivar-specific metabolic profiling and functional characterization

Principal component analysis (PCA) demonstrated high analytical reproducibility and revealed distinct metabolic fingerprints among the three cultivars (Fig. 5B), which highlights the dominant role of genetic background in determining metabolic composition. These findings align with previous reports on foxtail millet's dual medicinal-food applications (Qiao et al., 2025), while offering new perspectives on genotype-driven metabolic diversification in this crop.

Fig. 5.

Fig. 5

Open in a new tab

A: Heatmap of differentially accumulated metabolites across different cultivars. B: OPLS-DA score plot of different samples. C: Venn diagram of differential accumulated metabolites across different cultivars. D: Variation trends of metabolites across three cultivars.

Comparative metabolomic analysis identified 160 DAMs among cultivars, primarily comprising flavonoids, alkaloids, and phenolic acids (Fig. 5A and C) - a profile consistent with traditional medicinal applications (Wang et al., 2023). Particularly, 15 DAMs were identified as key bioactive components. Flavonoid composition varied distinctly by cultivar. JG21 and ZG19 predominantly accumulated flavonoid glycosides, which are associated with enhanced stability and bioavailability, whereas QH2 showed higher levels of flavonoid aglycones, typically linked to stronger biological activity. These results corroborate earlier findings on genetic regulation of nutritionally relevant metabolites (Zhang et al., 2021), while further clarifying how specific metabolite forms may affect both processing characteristics and health-related functions.

K-means clustering analysis of the 160 DAMs identified four distinct metabolic clusters (Fig. 5D; Fig. S7). Cluster 1 (QH2-enriched) was primarily involved in carbohydrate metabolism and phenylpropanoid biosynthesis. Notably, it featured a high accumulation of tampelopsisionoside, a flavonoid glycoside with potential antioxidant activity (Tanaka, Brugliera, & Chandler, 2009). Additionally, this cluster contained two key bioactive components (Table S4), indicating QH2's potential relevance in glycemic management.

Cluster 2 (co-upregulated in JG21 and ZG19) was dominated by tryptophan-derived alkaloids. Particularly, the neuroactive alkaloid 5-methoxy-N,N-dimethyltryptamine (bufotenine) was identified as a key metabolite, which is reported in literature to interact with 5-hydroxytryptamine (5-HT) receptors, a mechanism associated with potential sedative effects (Rauf et al., 2019). These results provide a metabolomic basis for the historical use of foxtail millet in traditional practices where calming effects were desired.

Cluster 3 (JG21-enriched) contained four key bioactive flavonoids. Apigenin-7-O-glucoside was particularly prominent, accumulating 3–5 times more in JG21 than in other cultivars (P < 0.01). This flavonoid glycoside is associated with potential anti-inflammatory and neuroprotective activities in pharmacological studies (Perez-Vizcaino & Fraga, 2018). The co-enrichment of stress-responsive alkaloids such as tryptamine and L-tyramine further suggests improved environmental adaptability (Ma, Wang, et al., 2023).

Cluster 4 (ZG19-enriched), the largest cluster, was characterized by a diverse set of bioactive flavonoids (Table S4), including the antioxidant kaempferol-3-O-rutinoside (Fu, 2024). ZG19 also showed significant upregulation of 1,2-O-diferuloylglycerol, a phenolic ester known to influence both antioxidant capacity and starch digestibility. As a modern bred cultivar, the abundant flavonoid profile of ZG19 likely reflects the synergistic improvement of both stress resistance and nutritional value during breeding selection.

Notably, twelve metabolites were consistently differentially accumulated across all cultivars (Fig. 5C; Table S5). Among these, L-malic acid consistently correlated with potential cancer-related pathways, revealing cultivar-specific bioactivity. Additionally, phenolic amine alkaloids (N-feruloylagmatine, sinapoylputrescine, and N-feruloylputrescine N-glucoside) emerged as signature biomarkers distinguishing cultivars. These ferulic acid-derived alkaloids are known for their antioxidant, anti-inflammatory, and neuroprotective properties (Rauf et al., 2019), potentially underpinning functional differences among foxtail millet cultivars.

3.5.2. Pathway analysis and nutritional implications of different cultivars

KEGG pathway enrichment analysis revealed distinct metabolic networks among the cultivars, reinforcing the genotype-specific metabolic specializations identified in the profiling analysis. The detailed results of all enriched pathways for each comparison are provided in Fig. S8 & Fig. S9.

The pathway findings provide a functional context for the key metabolic divergences. Most notably, the traditional cultivar JG21 showed significant enrichment in neuroactive ligand-receptor interaction pathway. This finding directly correlates with the observed accumulation of tryptamine alkaloids in JG21 and offers a mechanistic explanation for its traditional use in neuroregulation (Sugiyama et al., 2016). In contrast, the modern cultivar ZG19 exhibited pronounced enrichment in flavonoid biosynthesis pathways, consistent with its characteristic flavonoid accumulation profile. This pattern suggests enhanced antioxidant potential and supports ZG19's relevance for health-promoting food applications. Furthermore, suppression of the tricarboxylic acid (TCA) cycle in QH2 indicates distinct energy metabolism adaptations, potentially linked to its specialized growth and stress-response characteristics.

Collectively, our metabolomic findings underscore the pivotal role of genetic background in shaping the chemical diversity and functional properties of foxtail millet (Fig. 4B). The three cultivars displayed clearly differentiated metabolic specializations. The traditional cultivar JG21 served as a rich source of neuroactive tryptamine alkaloids, aligning with its historical use. The modern cultivar ZG19 was distinctly abundant in diverse flavonoids, highlighting breeding efforts to enhance antioxidant capacity and stress resilience. QH2 displayed a unique metabolic network with alterations in energy metabolism. These cultivar-specific metabolic blueprints are crucial for precision breeding, enabling the selection of parental lines based on their inherent capacity to accumulate desired bioactive compounds, irrespective of the environment to a certain extent.

3.6. Integrated effects of genotype and environment and future perspectives

This study demonstrates that foxtail millet quality is co-regulated by genetic and environmental factors (Fig. 4). Ecological conditions primarily drive geographical variations in metabolite profiles and quality traits, while genetic background determines cultivar-specific metabolic patterns. The genotype-environment interaction was clearly demonstrated by the cultivar-dependent accumulation of L-malic acid under high-altitude conditions. Notably, foxtail millet's unique metabolic profile—characterized by abundant lysophospholipids, uncommon phenolamine alkaloids, and superior flavonoid diversity—distinguishes it nutritionally from other major cereals like sorghum, pearl millet, rice, and wheat.

These findings provide a strategic framework for foxtail millet improvement. For precision breeding, cultivars such as ZG19 and JG21 represent valuable germplasm resources for enhancing antioxidant and neuroactive compounds, respectively. A genotype-environment matching strategy is proposed, where warmer regions are recommended for cultivating high-protein cultivars, while high-altitude regions favor production of grains rich in linoleic acid and flavonoid glycosides. This approach leverages foxtail millet's unique metabolic characteristics to address specific nutritional gaps in conventional cereal-based diets, positioning it as a complementary grain with distinct health benefits rather than merely an alternative crop. Future research should prioritize validation of the identified metabolites' health benefits through in vivo studies and integration of multi-omics approaches to elucidate underlying regulatory mechanisms. These directions will enhance the molecular foundation for developing functionally optimized foxtail millet varieties.

4. Conclusion

This study demonstrates that both environmental and genetic factors significantly shape the nutritional and metabolic profiles of foxtail millet. Key environmental drivers, including altitude, effective accumulated temperature, and soil available potassium, predominantly influenced nutritional composition and pasting properties. Concurrently, cultivar-specific genetic backgrounds shaped distinct metabolite profiles, notably the accumulation of neuroactive alkaloids in traditional cultivars and flavonoids in modern varieties. These insights establish a metabolic foundation for precision breeding and region-specific cultivation, enabling the design of foxtail millet with tailored functionality for the food industry. Future work should focus on validating the health benefits of identified metabolites and integrating multi-omics data to support targeted crop improvement.

4.1. Chemical compounds

The following key chemical compounds studied in this article are listed with their PubChem CID (Compound ID) for reference: L-Malic acid (CID: 222656); Linoleic acid (CID: 5280450); Apigenin-7-O-glucoside (CID: 12304093); Kaempferol-3-O-rutinoside (CID: 5318767); Ricinoleic acid (CID: 643684); Tectoridin (CID: 5281810); N-Feruloylagmatine (CID: 132280331); Tryptamine (CID: 1150); 9-Hydroxy-10,12,15-octadecatrienoic acid (CID: 6439873); Raffinose (CID: 439242).

CRediT authorship contribution statement

Xin Zhao: Writing – original draft, Methodology, Formal analysis. Yueyue Wang: Methodology, Investigation. Ziyi Deng: Investigation, Formal analysis. Zhongxiang Li: Software, Methodology, Investigation. Meng Yue: Resources, Project administration. Hongying Li: Writing – review & editing, Supervision, Conceptualization. Yuanhuai Han: Writing – review & editing, Supervision, Funding acquisition.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was supported by National Natural Science Foundation of China (U21A20216); Project of Shanxi Province Key Lab Construction (Z135050009017-2-16); Shanxi Key Research Project (2022ZDYF108); Central Government Guided Local Science and Technology Development Fund (YDZJSX2022B007); Shanxi Institute of Organic Dryland Farming Talent Support and Self-Research Initiative Project (YRCZY202502). Special thanks to Longlong Liu, Kai Zhao, Junjie Wang, Erhu Guo, Chengyuan Yang, and Yuemei Ren from Shanxi Agricultural University for their contributions in providing foxtail millet cultivars and coordinating the regional field trials.

Footnotes

Appendix A

Supplementary data to this article can be found online at https://doi.org/10.1016/j.fochx.2025.103409.

Contributor Information

Hongying Li, Email: hongying1964@163.com.

Yuanhuai Han, Email: hanyuanhuai@sxau.edu.cn.

Appendix A. Supplementary data

Supplementary material

mmc1.docx (5.8MB, docx)

Data availability

Data will be made available on request.

References

  1. Aamer H., Tariq M., Yawar M., Shaheryar M., Khan A.A., Iftikhar S.…Khan A. Exploring taurine’s potential in Alzheimer’s treatment: A comprehensive review. Cureus. 2024;16(5) doi: 10.7759/cureus.60997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alves E., Dias M., Lopes D., Almeida A., Domingues M., Rey F. Antimicrobial lipids from plants and marine organisms: An overview of the current state-of-the-art and future prospects. Antibiotics (Basel) 2020;9:441. doi: 10.3390/antibiotics9080441. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Arora L., Aggarwal R., Dhaliwal I., Gupta O.P., Kaushik P. Assessment of sensory and nutritional attributes of foxtail millet-based food products. Frontiers in Nutrition. 2023;10 doi: 10.3389/fnut.2023.1146545. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Balet S., Guelpa A., Fox G., Manley M. Rapid Visco analyser (RVA) as a tool for measuring starch-related physiochemical properties in cereals: A review. Food Analytical Methods. 2019;12(10):2344–2360. doi: 10.1007/s12161-019-01581-w. [DOI] [Google Scholar]
  5. Chen M., Huang W., Yin Z., Zhang W., Kong Q., Wu S.…Yan S. Environmentally-driven metabolite and lipid variations correspond to altered bioactivities of black wolfberry fruit. Food Chemistry. 2022;372 doi: 10.1016/j.foodchem.2021.131342. [DOI] [PubMed] [Google Scholar]
  6. Fitzgerald M.A., Martin M., Ward R.M., Park W.D., Shead H.J. Viscosity of rice flour: A rheological and biological study. Journal of Agricultural and Food Chemistry. 2003;51:2295–2299. doi: 10.1021/jf020574i. [DOI] [PubMed] [Google Scholar]
  7. Fu Y. Kaempferol 3-o-rutinoside, a flavone derived from tetrastigma hemsleyanum diels et gilg, reduces body temperature through accelerating the elimination of il-6 and tnf-α in a mouse fever model. Molecules. 2024;29 doi: 10.3390/molecules29071641. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Fu Y.X., Yin R.Y., Guo E.H., Cheng R.H., Diao X.M., Xue Y., Shen Q. Protein isolates from raw and cooked foxtail millet attenuate development of type 2 diabetes in streptozotocin-induced diabetic mice. Molecular Nutrition & Food Research. 2021;22 doi: 10.1002/mnfr.202000365. [DOI] [PubMed] [Google Scholar]
  9. Grabowska K., Podolak I., Galanty A., Załuski D., Makowska-Wąs J., Sobolewska D., Zmudzki P. In vitro anti-denaturation and anti-hyaluronidase activities of extracts and galactolipids from leaves of impatiens parviflora DC. Natural Product Research. 2016;30(10):1219–1223. doi: 10.1080/14786419.2015.1049175. [DOI] [PubMed] [Google Scholar]
  10. He L., Cheng L., Wang J.J., Liu J., Cheng J.J., Yang Z.R., Cao R., Han Y.H., Li H.Y., Zhang B. Carotenoid cleavage dioxygenase 1 catalyzes lutein degradation to influence carotenoid accumulation and color development in foxtail millet grains. Journal of Agricultural and Food Chemistry. 2022;70(30):9283–9294. doi: 10.1021/acs.jafc.2c01951. [DOI] [PubMed] [Google Scholar]
  11. He L., Zhang B., Wang X.C., Li H.Y., Han Y.H. Foxtail millet: Nutritional and sensory quality, and prospects for genetic improvement. Frontiers of Agricultural Science and Engineering. 2015;2:124–133. doi: 10.15302/J-FASE-2015054. [DOI] [Google Scholar]
  12. Hess D.C., Lu W., Rabinowitz J.D., Botstein D. Ammonium toxicity and potassium limitation in yeast. PLoS Biology. 2006;4(11) doi: 10.1371/journal.pbio.0040351. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Hou S.Y., Men Y.H., Wei M., Zhang Y.J., Li H.Y., Sun Z.X., Han Y.H. Total protein content, amino acid composition and eating-quality evaluation of foxtail miillet. Foods. 2023;12(1):31. doi: 10.3390/foods12010031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Huo Y.Q., Pang X.D., Feng H.X., Hao R.L., Zhang R.J., Wang Y.S.…Han Y.H. Effect of ultraviolet B radiation on the biosynthesis of carotenoids in foxtail millet grains. Agriculture. 2024;14(2):289. doi: 10.3390/agriculture14020289. [DOI] [Google Scholar]
  15. Li H., Lv Q., Liu A., Wang J., Sun X., Deng J., Wu Q. Comparative metabolomics study of Tartary (Fagopyrum tataricum (L.) Gaertn) and common (Fagopyrum esculentum Moench) buckwheat seeds. Food Chemistry. 2021 doi: 10.1016/j.foodchem.2021.131125. [DOI] [PubMed] [Google Scholar]
  16. Li W., Wen L., Chen Z., Zhang Z., Pang X., Deng Z., Liu T., Guo Y. Study on metabolic variation in whole grains of four proso millet varieties reveals metabolites important for antioxidant properties and quality traits. Food Chemistry. 2021;345 doi: 10.1016/j.foodchem.2021.129791. [DOI] [PubMed] [Google Scholar]
  17. Li X.K., Gao J.H., Song J.Y., Guo K., Hou S.Y., Wang X.C.…Han Y.H. Multi-omics analyses of 398 foxtail millet accessions reveal genomic regions associated with domestication, metabolite traits, and anti-inflammatory effects. Molecular Plant. 2022;15(8):1367–1383. doi: 10.1016/j.molp.2022.07.003. [DOI] [PubMed] [Google Scholar]
  18. Liang Y.P., Han Y.H. Pan-genome brings opportunities to revitalize the ancient crop foxtail millet. Plant Communications. 2024;5(1) doi: 10.1016/j.xplc.2023.100735. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Liu S.C., Cao X.N., Wen Q.F., Wang H.G., Tian X., Wang J.J.…Qiao Z.J. Comprehensive evaluation of agronomic traits and quality traits of foxtail millet landrace in Shanxi. Scientia Agricultura Sinica. 2020;53(11):2137–2148. doi: 10.3864/j.issn.0578-1752.2020.11.001. [DOI] [Google Scholar]
  20. Ma K., Zhao L.R., Zhao X.T., Li X.R., Dong S.Q., Zhang L.G.…Diao X.M. The relationship between ecological factors and commercial quality of high-quality foxtail millet “Jingu 21”. Food Research International. 2023;163 doi: 10.1016/j.foodres.2022.112225. [DOI] [PubMed] [Google Scholar]
  21. Ma Q., Wang H., Wu E., Zhang H., Feng Y., Feng B. Widely targeted metabolomic analysis revealed the effects of alkaline stress on nonvolatile and volatile metabolites in broomcorn millet grains. Food Research International. 2023 doi: 10.1016/j.foodres.2023.113066. [DOI] [PubMed] [Google Scholar]
  22. Moller A.B., Andrews A.T., Cheeseman G.C. Chemical changes in ultra-heat-treated milk during storage: II. Lactuloselysine and fructoselysine formation by the Maillard reaction. Journal of Dairy Research. 1977;44(2):267–276. doi: 10.1017/S0022029900019634. [DOI] [Google Scholar]
  23. Mushtaq M.Y., Choi Y.H., Verpoorte R., Wilson E.G. Extraction for metabolomics: Access to the metabolome. Phytochemical Analysis. 2014;25(4):291–306. doi: 10.1002/pca.2505. [DOI] [PubMed] [Google Scholar]
  24. Muthamilarasan M., Prasad M. Small millets for enduring food security amidst pandemics. Trends in Plant Science. 2021;26(4):189–202. doi: 10.1016/j.tplants.2020.12.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Ning N., Yang Y.J., Hong J.P., Yuan X.Y., Song X.E., Wang H.F., Guo P.Y. Correlation between grain quality of foxtail millet (Setaria italica [L.] P. Beauv.) and environmental factors on multivariate statistical analysis. Chilean Journal of Agricultural Research. 2017;77(4):303–310. doi: 10.4067/S0718-58392017000400303. [DOI] [Google Scholar]
  26. Perez-Vizcaino F., Fraga C.G. Research trends in flavonoids and health. Archives of Biochemistry and Biophysics. 2018;646:107–112. doi: 10.1016/j.abb.2018.03.022. [DOI] [PubMed] [Google Scholar]
  27. Qian G., Li X., Zhang H., Zhang H., Zhou J., Ma X., Sun W., Yang W., He R., Wahab A.-T., Wan H., Li L. Metabolomics analysis reveals the accumulation patterns of flavonoids and phenolic acids in quinoa (Chenopodium quinoa Willd.) grains of different colors. Food Chemistry. 2023;X, 17 doi: 10.1016/j.fochx.2023.100594. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Qiao H., Li X., Jia Z., Zhang W., Wang Y. Metabolite profiling of foxtail millet varieties from Qingyang using GC-MS and LC-MS analysis. Food Chemistry: X. 2025;27 doi: 10.1016/j.fochx.2025.102477. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Rauf A., Imran M., Abu-Izneid T., Patel S., Pan X., Naz S. Proanthocyanidins: A comprehensive review. Biomedicine & Pharmacotherapy. 2019;116 doi: 10.1016/j.biopha.2019.108999. [DOI] [PubMed] [Google Scholar]
  30. Ru J., Li P., Wang J., Zhou W., Li B., Huang C., Li P., Guo Z., Tao W., Yang Y., Xu X., Li Y., Wang Y., Yang L. TCMSP: A database of systems pharmacology for drug discovery from herbal medicines. Journal of Cheminformatics. 2014;6(1) doi: 10.1186/1758-2946-6-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Sachdev N., Goomer S., Singh L.R. Foxtail millet: A potential crop to meet future demand scenario for alternative sustainable protein. Journal of the Science of Food and Agriculture. 2021;101(8):3045–3055. doi: 10.1002/jsfa.10716. [DOI] [PubMed] [Google Scholar]
  32. Stipanuk M.H., Ueki L. Dealing with methionine/homocysteine sulfur: Cysteine metabolism to taurine and inorganic sulfur. Journal of Inherited Metabolic Disease. 2011;34(1):17–32. doi: 10.1007/s10545-009-9006-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Sugiyama M., Takahashi M., Katsube T., Koyama A., Itamura H. Effects of applied nitrogen amounts on the functional components of mulberry (Morus alba L.) leaves. Journal of Agricultural and Food Chemistry. 2016;64(37):6923–6929. doi: 10.1021/acs.jafc.6b01922. [DOI] [PubMed] [Google Scholar]
  34. Sun M., Kang X., Wang T., Fan L., Wang H., Pan H., Yang Q., Liu H., Lou Y., Zhuge Y. Genotypic diversity of quality traits in Chinese foxtail millet (Setaria italica L.) and the establishment of a quality evaluation system. Food Chemistry. 2021;353 doi: 10.1016/j.foodchem.2021.129421. [DOI] [PubMed] [Google Scholar]
  35. Ta P., Buchmeier N., Newton G.L., Rawat M., Fahey R.C. Organic hydroperoxide resistance protein and ergothioneine compensate for loss of mycothiol in mycobacterium smegmatis mutants. Journal of Bacteriology. 2011;193(8):1982–1990. doi: 10.1128/JB.01402-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Tanaka Y., Brugliera F., Chandler S. Recent progress of flower colour modification by biotechnology. International Journal of Molecular Sciences. 2009;10(12):5350–5369. doi: 10.3390/ijms10125350. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Tang J.Y., Li X.K., Zhang Y.K., Yang Y.L., Sun R., Li Y.J.…Han Y.H. Differential flavonoids and carotenoids profiles in grains of six poaceae crops. Foods. 2022;11(14):2068. doi: 10.3390/foods11142068. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Tao K., Yu W., Prakash S., Gilbert R.G. High-amylose rice: Starch molecular structural features controlling cooked rice texture and preference. Carbohydrate Polymers. 2019;219:251–260. doi: 10.1016/j.carbpol.2019.05.031. [DOI] [PubMed] [Google Scholar]
  39. Verma S., Srivastava S., Tiwari N. Comparative study on nutritional and sensory quality of barnyard and foxtail millet food products with traditional rice products. Journal of Food Science and Technology. 2015;52:5147–5155. doi: 10.1007/s13197-014-1617-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Wang G., Shen X., Bai C., Zhuang Z., Jiang H., Yang M., Wei X., Wu Z. Metabolomic study on the quality differences and physiological characteristics between rice cultivated in drought and flood conditions. Food Chemistry. 2023;425 doi: 10.1016/j.foodchem.2023.135946. [DOI] [PubMed] [Google Scholar]
  41. Wang H., Zhao M., Wu Z.Z., Qin N., Fu Y., Guo S. Nutrient composition and functional constituents of daylily from different producing areas based on widely targeted metabolomics. Food Chemistry: X. 2024;21 doi: 10.1016/j.fochx.2024.101239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Xuan Y., Yi Y., Liang H., Wei S., Chen N., Jiang L., Ali I., Ullah S., Wu X., Cao T., Zhao Q., Li T. Amylose content and RVA profile characteristics of noodle rice under different conditions. Agronomy Journal. 2020;112:117–129. doi: 10.1002/agj2.20079. [DOI] [Google Scholar]
  43. Yang Y.B., Jia G.Q., Deng L.G., Qin L., Chen E.Y., Cong X.J.…Yin Y.P. Genetic variation of yellow pigment and its components in foxtail millet (Setaria italica (L.) P. Beauv.) from different eco-regions in China. Journal of Integrative Agriculture. 2017;16(11):2459–2469. doi: 10.1016/S2095-3119(16)61598-8. [DOI] [Google Scholar]
  44. Yang Y.B., Zhang H.D., Qin L., Chen E.Y., Li F.F., Guan Y.A. Effect of planting location and processing fineness on nutrition and safety quality of millet. Journal of the Chinese Cereals and Oils Association. 2021;36(9):54–59. http://zglyxb.ijournals.cn/ch/reader/view_abstract.aspx?file_no=20200461&flag=1 [Google Scholar]
  45. Yang Z.R., Zhang H.S., Li X.K., Shen H.M., Gao J.H., Hou S.Y.…Wang X.C. A mini foxtail millet with an Arabidopsis-like life cycle as a C4 model system. Nature Plants. 2020;6(9):1167–1178. doi: 10.1038/s41477-020-0747-7. [DOI] [PubMed] [Google Scholar]
  46. Yim S.S., Choi J.W., Lee S.H., Jeon E.J., Chung W.J., Jeong K.J. Engineering of corynebacterium glutamicum for consolidated conversion of hemicellulosic biomass into xylonic acid. Biotechnology Journal. 2017 doi: 10.1002/biot.201700040. [DOI] [PubMed] [Google Scholar]
  47. Yu R., Zong E., Zhang L., Zhang D. Analysis of metabolites and metabolic pathways in different varieties of foxtail millets (Setaria italica L.) based on gas chromatography-mass spectrometry. Food Science. 2021;42(24):199–205. doi: 10.7506/spkx1002-6630-20210120-219. [DOI] [Google Scholar]
  48. Yun J., Chin Y.-W., Hong S.P., Lee J.-E. Discovery of doenjang aging marker and metabolites alteration during long-term aging period: Targeted and untargeted metabolomic study. Food Chemistry. 2025 doi: 10.1016/j.foodchem.2025.143719. [DOI] [PubMed] [Google Scholar]
  49. Zhang Q., Dong X., Lin X., Peng J., Luo D., Huang S., Zhu S. Ultra-high performance liquid chromatography-tandem mass spectrometry-based metabolomic analysis of primary metabolites in 'Fengtang' plum fruits from different geographical origins. Food Science. 2023;44(8):284–292. doi: 10.7506/spkx1002-6630-20220401-014. [DOI] [Google Scholar]
  50. Zhang Y., Yao X., Tian P., Zhang C., Liu X., Li H., Han Y. Comparative metabolome and transcriptome analysis of foxtail millet cultivars with high and low grain quality. Food Chemistry: Molecular Sciences. 2024;5 doi: 10.1016/j.fochms.2025.100259. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Zhang Y.K., Gao J.H., Qie Q.R., Yang Y.L., Hou S.Y., Wang X.C.…Han Y.H. Comparative analysis of flavonoid metabolites in foxtail millet (Setaria italica) with different sensory quality. Life. 2021;11(6):578. doi: 10.3390/life11060578. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Zhang Y.R. Shanxi Agricultural University; 2019. Study on the cooking aroma, lipid change patterns and metabolite accumulation characteristics during grain filling based on metabolomics in foxtail millet. (Doctor’s Thesis) [DOI] [Google Scholar]
  53. Zhao X., Huang M.J., Huang X.F., Liu E.K. Evaluation of drought resistance and index screening of foxtail millet cultivars. Journal of Water and Climate Change. 2023;14(7):2384–2396. doi: 10.2166/wcc.2023.086. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary material

mmc1.docx (5.8MB, docx)

Data Availability Statement

Data will be made available on request.

Articles from Food Chemistry: X are provided here courtesy of Elsevier

RESOURCES