Search
2023 Volume 2
Article Contents
ARTICLE   Open Access    

Dynamic transcriptome landscape of foxtail millet grain development

More Information
  • Received: 15 July 2023
    Accepted: 24 October 2023
    Published online: 24 November 2023
    Seed Biology  2 Article number: 19 (2023)  |  Cite this article
  • Grain development of foxtail millet (Setaria italica L.) is essential for yield and quality. However, its transcriptional dynamics molecular mechanisms and morphological analyses remain scarcely described. Thus, we conducted detailed daily morphological analyses of foxtail millet grain development throughout the 30 d post-fertilization development period. On the basis of the morphological analyses, we used RNA-sequencing (RNA-seq) to examine the transcript dynamics involved in foxtail millet grain development at four stages. These genes included those associated with transcriptional regulation, hormone signaling, sucrose and starch metabolism, zein family members, amino acid metabolism, carotenoid metabolism, flavonoid biosynthesis, and folate synthesis. We have validated the accuracy of the transcriptome data by means of reverse transcription quantitative polymerase chain reaction (RT-qPCR). This study provides precious genetic resources for understanding grain developmental process in the future. These results expand our understanding of the molecular mechanisms of grain development in foxtail millet and contribute to the functional studies of genes related to grain development in the future.
  • 加载中
  • Supplemental Fig. S1 Dynamics of foxtail millet grain development: from 1 to 30 days after pollination (DAP). (a) Changes in the grains area of during foxtail millet grains development; (b) Changes in the perimeter of during grains development; (c) The growth of the foxtail millet grains in the length and width; (d) The change in the length/width ratio during grains development. Values for (a) , (b) , (c) and (d) are means ± s.d. (from 1 to 15 days after pollination n = 40, from 16 to 30 days after pollination n = 20).
    Supplemental Fig. S2 Cell morphological analysis of the developing grains in foxtail millet: Paraffin section analysis. (a) and (e) Morphology of mature ovule cells before pollination; (b) and (f) Milk stage of grain development at 6-11 DAP; (c) and (g) Dough stage of grain development at 12-22 DAP; (d) and (h) Maturity stage of grain development at 23-30 DAP. DAP: days after pollination. (a-d) stained with 0.1% Toluidine Blue-O to observe. (e-f) stained with Safranin O and Fast Green to observe. Scale bars =10 μm.
    Supplemental Fig. S3 Analysis of different samples global gene expression. (a) The principal component analysis (PCA) of the RNA-seq data of the 12 samples; (b) Total number of genes expressed at each developmental stage; (c) Percentage of gene numbers in different categories according to their expression levels in each tissue, based on FPKM values; (d) Gene ontology (GO) enrichment analysis of the all expression gene in four developmental stages.
    Supplemental Fig. S4 Hierarchical clustering analysis of the relative expression levels of transcription factors during grain development. (a) Hierarchical clustering analysis of the relative expression levels of transcription factors during grain development; (b) Gene ontology (GO) enrichment analysis of the TFs in foxtail millet grain development; (c) KEGG pathway classification top 20 of the TFs in foxtail millet grain development.
    Supplemental Fig. S5 Clustering expression patterns of cytokinin signaling pathway related genes in foxtail millet grains development. (a) Diagram of the cytokinin signaling pathway; (b) Expression patterns of histidine kinase in the cytokinin signaling pathway; (c) Expression patterns of histidine containing phosphotransfer protein in the cytokinin signaling pathway; (d) Expression patterns of the response regulator in the cytokinin signaling pathway; (e-f) The dynamic transcript levels of cytokinin signaling pathway in the different development stages of the foxtail millet grain. HK4/CRE1: AHK4/CYTOKININ RESPONSE1; AHP1: Histidine-containing phosphotransfer protein 1; AHP2: Histidine-containing phosphotransfer protein 2; A-ARR: two-component response regulator ORR belong to ARR family Type-A subfamily; B-ARR: two-component response regulator ORR belong to ARR-B family.
    Supplemental Fig. S6 Dynamic transcriptome analysis of genes associated with gibberellin signaling pathway. (a) Diagram of the gibberellin signaling pathway; (b) Hierarchical clustering analysis of the relative expression levels of gibberellin signaling pathway genes during grain development; (c-d) The dynamic transcript levels of gibberellin signaling pathway in different development stages. GID1/2: GA-insensitive dwarf 1/2; DELLA: DELLA protein; TFs: Transcription factors.
    Supplemental Fig. S7 Dynamic transcriptome analysis of genes involved in abscisic acid signaling pathway of foxtail millet grains development. (a) Diagram of the abscisic acid signaling pathway; (b) Hierarchical clustering analysis of the relative expression levels of abscisic acid signaling pathway genes during grain development; (c-f) The dynamic transcript levels of abscisic acid signaling pathway DEGs in the different development stages of the foxtail millet grain. PYRs: Abscisic acid receptor PYR family; PYLs: Abscisic acid receptor PYL family; PP2C: Protein phosphatase 2 C; SnRK2: serine/threonine-protein kinase SnRK2; ABF: ABA responsive element binding factor.
    Supplemental Fig. S8 Dynamic transcriptome analysis of genes involved in brassinosteroid signal transduction of foxtail millet grains development. (a) Diagram of the brassinosteroid signal transduction pathway; (b) Hierarchical clustering analysis of the relative expression levels of brassinosteroid signal transduction pathway genes during grain development; (c-e) The dynamic transcript levels of brassinosteroid signal transduction pathway in the different development stages of the foxtail millet garins. BAK1: Brassinosteroid insensitive 1-associated receptor kinase 1; BRI1: Brassinosteroid insensitive 1; BKI1: BRI1 kinase inhibitor 1; BSK: BR-signaling kinase; BSU1: serine/threonine-protein phosphatase BSU1; BIN2: Brassinosteroid insensitive 2; BZR1/2: Brassinosteroid resistant 1/2; TCH4: xyloglucosyl transferase TCH4; CYCD3: Cyclin D 3.
    Supplemental Fig. S9 Heat map analysis of genes associated with sucrose-starch conversion during grain filling from Ovule stage to maturity stage.
    Supplemental Fig. S10 Dynamic analysis of genes involved to amino acid metabolism pathways in foxtail millet grains development. (a) Dynamic analysis of genes involved in amino acid biosynthesis during foxtail millet grains development. These genes are divided into essential amino acid biosynthesis genes, non-essential amino acid biosynthesis genes and genes synthesized by both; (b) Dynamic analysis of genes involved in amino acid metabolism and degradation during the development of foxtail millet grains. These genes can be divide into essential amino acid metabolism genes, non-essential amino acid metabolism genes and genes involved in the common metabolism of both.
    Supplemental Fig. S11 Dynamic analysis of the flavonoid biosynthesis pathway during foxtail millet grains development. (a) Hierarchical clustering analysis of the relative expression levels of the flavonoid biosynthesis pathway genes during grain development; (b) Flavonoid biosynthesis pathway during foxtail millet grain development. The rectangles represent the expression level of genes. C4H: Trans-cinnamate 4-monooxygenase; CHS: Chalcone synthase; CHI: Chalcone isomerase; F3H: Flavanone 3-hydroxylase; F3’H: flavonoid 3'-hydroxylase; F3’5’H: flavonoid 3',5'-hydroxylase; FLS: flavonol synthase; DFR: Dihydroflavonol 4-reductase; LAR: Leucoanthocyanidin dioxygenase; ANR: Anthocyanidin reductase.
    Supplemental Fig. S12 The biosynthetic pathway and expression patterns of folate synthesis pathway genes during foxtail millet grain development. (a) Hierarchical clustering analysis of the relative expression levels of folate metabolism pathways genes during seed development; (b) Folate synthesis pathway during foxtail millet seed development. The rectangles represent the expression level of genes. ADC: Aminodeoxychorismate; pABA: para aminobenzoic acid; GTP: guanosine-triphosphate; DHN: dihydroneopterin; ADCS: aminodeoxychorismate synthase; ADCL: aminodeoxychorismate lyase; HPPK: hydroxymethyldihydropterin pyrophospho kinase; DHPS: dihydropteroate synthase; DHFS: dihydrofolate Synthetase; DHFR: dihydrofolate reductase; FPGS: folylpolyglutamate-synthase; GGH: gamma glutamyl hydrolase; GTPCHI: GTP cyclohydrolase I; DHNA: dihydroneopterine aldolase.
    Supplemental Fig. S13 RT-qPCR was used for quantitatively verification of the key genes during grain development. (a-e) Relative expression levels of five key TFs (SiMADS-box21, SiMADS-box13, SiMADS-box7, SiMADS-box1, SiERF058) during grain development. (f-j) Relative expression levels of fix auxin signaling pathway related genes (SiIAA2, Si6g23990-SAUR36, Si1g25820-SAUR36, SiARF4, SiARF22) during grain development. (l-n) The relative expression levels of rate-limiting enzyme gene SiPSY3 and three zein genes.
    Supplemental Table S1 Morphological analyses of foxtail millet grain.
    Supplemental Table S2 Summary of RNA-Seq read mapping results.
    Supplemental Table S3 The classification of gene expression levels from the four tissues in this study.
    Supplemental Table S4 Multiple TFs involved in grain development in foxtai millet.
    Supplemental Table S5 Zein gene in transcriptome data.
    Supplemental Table S6 Primers used for RT-qPCR.
    Supplemental Data Set S1 The total number of genes detected in the four RNA-seq samples.
    Supplemental Data Set S2 All expressed genes with FPKM value >1 in the four RNA-seq samples.
    Supplemental Data Set S3 Many transcription factors were detected, belonging to 42 transcription factor families and other types of transcription factors.
  • [1]

    Doust AN, Devos KM, Gadberry MD, Gale MD, Kellogg EA. 2004. Genetic control of branching in foxtail millet. Proceedings of the National Academy of Sciences 101(24):9045−50

    doi: 10.1073/pnas.0402892101

    CrossRef   Google Scholar

    [2]

    Doust AN, Kellogg EA, Devos KM, Bennetzen JL. 2009. Foxtail millet: a sequence-driven grass model system. Plant physiology 149(1):137−41

    doi: 10.1104/pp.108.129627

    CrossRef   Google Scholar

    [3]

    Diao X, Schnable J, Bennetzen JL, Li J. 2014. Initiation of Setaria as a model plant. Frontiers of Agricultural Science and Engineering 1(1):16–20. https://journal.hep.com.cn/fase/EN/10.15302/J-FASE-2014011

    [4]

    Liu Z, Bai G, Zhang D, Zhu C, Xia X, et al. 2011. Genetic diversity and population structure of elite foxtail millet [Setaria italica (L.) P. Beauv. ] germplasm in China. Crop science 51(4):1655−63

    doi: 10.2135/cropsci2010.11.0643

    CrossRef   Google Scholar

    [5]

    Diao X, Jia G. 2017. Origin and domestication of foxtail millet. In Genetics and genomics of Setaria, ed. Doust A, Diao X. Vol 19. Cham: Springer. pp. 61−72. https://doi.org/10.1007/978-3-319-45105-3_4

    [6]

    Shahidi F, Chandrasekara A. 2013. Millet grain phenolics and their role in disease risk reduction and health promotion: A review. Journal of Functional Foods 5(2):570−81

    doi: 10.1016/j.jff.2013.02.004

    CrossRef   Google Scholar

    [7]

    Kaur P, Purewal SS, Sandhu KS, Kaur M, Salar RK. 2019. Millets: A cereal grain with potent antioxidants and health benefits. Journal of Food Measurement and Characterization 13:793−806

    doi: 10.1007/s11694-018-9992-0

    CrossRef   Google Scholar

    [8]

    Liang S, Liang K. 2019. Millet grain as a candidate antioxidant food resource: a review. International Journal of Food Properties 22(1):1652−61

    doi: 10.1080/10942912.2019.1668406

    CrossRef   Google Scholar

    [9]

    Sachdev N, Goomer S, Singh LR. 2021. Foxtail millet: a potential crop to meet future demand scenario for alternative sustainable protein. Journal of the Science of Food and Agriculture 101(3):831−42

    doi: 10.1002/jsfa.10716

    CrossRef   Google Scholar

    [10]

    Sushree Shyamli P, Rana S, Suranjika S, Muthamilarasan M, Parida A, et al. 2021. Genetic determinants of micronutrient traits in graminaceous crops to combat hidden hunger. Theoretical and Applied Genetics 134:3147−65

    doi: 10.1007/s00122-021-03878-z

    CrossRef   Google Scholar

    [11]

    Amadou I, Amza T, Shi YH, Le GW. 2011. Chemical analysis and antioxidant properties of foxtail millet bran extracts. Songklanakarin Journal of Science & Technology 33(5):509−15

    Google Scholar

    [12]

    Liang S, Yang G, Ma Y. 2010. Chemical characteristics and fatty acid profile of foxtail millet bran oil. Journal of the American Oil Chemists' Society 87(1):63−67

    doi: 10.1007/s11746-009-1475-3

    CrossRef   Google Scholar

    [13]

    Sharma N, Niranjan K. 2018. Foxtail millet: Properties, processing, health benefits, and uses. Food Reviews International 34(4):329−63

    doi: 10.1080/87559129.2017.1290103

    CrossRef   Google Scholar

    [14]

    Muthamilarasan M, Prasad M. 2015. Advances in Setaria genomics for genetic improvement of cereals and bioenergy grasses. Theoretical and Applied Genetics 128:1−14

    doi: 10.1007/s00122-014-2399-3

    CrossRef   Google Scholar

    [15]

    Povilus RA, Gehring M. 2022. Maternal-filial transfer structures in endosperm: A nexus of nutritional dynamics and seed development. Current Opinion in Plant Biology 65:102121

    doi: 10.1016/j.pbi.2021.102121

    CrossRef   Google Scholar

    [16]

    Shen S, Ma S, Chen XM, Yi F, Li BB, et al. 2022. A transcriptional landscape underlying sugar import for grain set in maize. The Plant Journal 110(1):228−42

    doi: 10.1111/tpj.15668

    CrossRef   Google Scholar

    [17]

    Agarwal P, Kapoor S, Tyagi AK. 2011. Transcription factors regulating the progression of monocot and dicot seed development. Bioessays 33(3):189−202

    doi: 10.1002/bies.201000107

    CrossRef   Google Scholar

    [18]

    Locascio A, Roig-Villanova I, Bernardi J, Varotto S. 2014. Current perspectives on the hormonal control of seed development in Arabidopsis and maize: a focus on auxin. Frontiers in Plant Science 5:412

    doi: 10.3389/fpls.2014.00412

    CrossRef   Google Scholar

    [19]

    O'Neill JP, Colon KT, Jenik PD. 2019. The onset of embryo maturation in Arabidopsis is determined by its developmental stage and does not depend on endosperm cellularization. The Plant Journal 99(2):286−301

    doi: 10.1111/tpj.14324

    CrossRef   Google Scholar

    [20]

    Wang T, Lu Q, Song H, Hu N, Wei Y, et al. 2021. DNA methylation and RNA-sequencing analysis show epigenetic function during grain filling in foxtail millet (Setaria italica L.). Frontiers in Plant Science 12:741415

    doi: 10.3389/fpls.2021.741415

    CrossRef   Google Scholar

    [21]

    Slafer GA, Foulkes MJ, Reynolds MP, Murchie EH, Carmo-Silva E, et al. 2023. A 'wiring diagram' for sink strength traits impacting wheat yield potential. Journal of Experimental Botany 74(1):40−71

    doi: 10.1093/jxb/erac410

    CrossRef   Google Scholar

    [22]

    Liu D, Zhao H, Xiao Y, Zhang G, Cao S, et al. 2022. A cryptic inhibitor of cytokinin phosphorelay controls rice grain size. Molecular Plant 15(2):293−307

    doi: 10.1016/j.molp.2021.09.010

    CrossRef   Google Scholar

    [23]

    Zhao Y. 2018. Essential roles of local auxin biosynthesis in plant development and in adaptation to environmental changes. Annual Review of Plant Biology 69:417−35

    doi: 10.1146/annurev-arplant-042817-040226

    CrossRef   Google Scholar

    [24]

    Sosso D, Luo D, Li QB, Sasse J, Yang J, et al. 2015. Seed filling in domesticated maize and rice depends on SWEET-mediated hexose transport. Nature Genetics 47(12):1489−93

    doi: 10.1038/ng.3422

    CrossRef   Google Scholar

    [25]

    Ueda T, Waverczak W, Ward K, Sher N, Ketudat M, et al. 1992. Mutations of the 22- and 27-kD zein promoters affect transactivation by the Opaque-2 protein. The Plant Cell 4(6):701−9

    doi: 10.1105/tpc.4.6.701

    CrossRef   Google Scholar

    [26]

    Feng F, Qi W, Lv Y, Yan S, Xu L, et al. 2018. OPAQUE11 is a central hub of the regulatory network for maize endosperm development and nutrient metabolism. The Plant Cell 30(2):375−96

    doi: 10.1105/tpc.17.00616

    CrossRef   Google Scholar

    [27]

    Sun Q, Li Y, Gong D, Hu A, Zhong W, et al. 2022. A NAC-EXPANSIN module enhances maize kernel size by controlling nucellus elimination. Nature Communications 13(1):5708

    doi: 10.1038/s41467-022-33513-4

    CrossRef   Google Scholar

    [28]

    Zhang Z, Dong J, Ji C, Wu Y, Messing J. 2019. NAC-type transcription factors regulate accumulation of starch and protein in maize seeds. Proceedings of the National Academy of Sciences of the United States of America 116(23):11223−28

    doi: 10.1073/pnas.1904995116

    CrossRef   Google Scholar

    [29]

    Ishimaru T, Ida M, Hirose S, Shimamura S, Masumura T, et al. 2015. Laser microdissection-based gene expression analysis in the aleurone layer and starchy endosperm of developing rice caryopses in the early storage phase. Rice 8:22

    doi: 10.1186/s12284-015-0057-2

    CrossRef   Google Scholar

    [30]

    Liu J, Wu X, Yao X, Yu R, Larkin PJ, et al. 2018. Mutations in the DNA demethylase OsROS1 result in a thickened aleurone and improved nutritional value in rice grains. Proceedings of the National Academy of Sciences of the United States of America 115(44):11327−32

    doi: 10.1073/pnas.1806304115

    CrossRef   Google Scholar

    [31]

    Zhang D, Zhang M, Zhou Y, Wang Y, Shen J, et al. 2019. The Rice G Protein γ Subunit DEP1/qPE9–1 Positively Regulates Grain-Filling Process by Increasing Auxin and Cytokinin Content in Rice Grains. Rice 12:91

    doi: 10.1186/s12284-019-0344-4

    CrossRef   Google Scholar

    [32]

    Mishra BS, Sharma M, Laxmi A. 2022. Role of sugar and auxin crosstalk in plant growth and development. Physiologia Plantarum 174(1):e13546

    doi: 10.1111/ppl.13546

    CrossRef   Google Scholar

    [33]

    Kuanar SR, Molla KA, Chattopadhyay K, Sarkar RK, Mohapatra PK. 2019. Introgression of Sub1 (SUB1) QTL in mega rice cultivars increases ethylene production to the detriment of grain-filling under stagnant flooding. Scientific Reports 9(1):18567

    doi: 10.1038/s41598-019-54908-2

    CrossRef   Google Scholar

    [34]

    Wang Z, Xu Y, Chen T, Zhang H, Yang J, et al. 2015. Abscisic acid and the key enzymes and genes in sucrose-to-starch conversion in rice spikelets in response to soil drying during grain filling. Planta 241:1091−107

    doi: 10.1007/s00425-015-2245-0

    CrossRef   Google Scholar

    [35]

    Xiang J, Tang S, Zhi H, Jia G, Wang H, et al. 2017. Loose Panicle1 encoding a novel WRKY transcription factor, regulates panicle development, stem elongation, and seed size in foxtail millet [Setaria italica (L.) P. Beauv. ]. PLoS ONE 12(6):e0178730

    doi: 10.1371/journal.pone.0178730

    CrossRef   Google Scholar

    [36]

    Pan Y, Ma X, Liang H, Zhao Q, Zhu D, et al. 2015. Spatial and temporal activity of the foxtail millet (Setaria italica) seed-specific promoter pF128. Planta 241(1):57−67

    doi: 10.1007/s00425-014-2164-5

    CrossRef   Google Scholar

    [37]

    Liu K, Qi S, Li D, Jin C, Gao C, et al. 2017. TRANSPARENT TESTA GLABRA 1 ubiquitously regulates plant growth and development from Arabidopsis to foxtail millet (Setaria italica). Plant Science 254:60−69

    doi: 10.1016/j.plantsci.2016.10.010

    CrossRef   Google Scholar

    [38]

    Wang M, Li P, Li C, Pan Y, Jiang X, et al. 2014. SiLEA14, a novel atypical LEA protein, confers abiotic stress resistance in foxtail millet. BMC Plant Biology 14:290

    doi: 10.1186/s12870-014-0290-7

    CrossRef   Google Scholar

    [39]

    Guo J, Zhou X, Dai K, Yuan X, Guo P, et al. 2022. Comprehensive analysis of YABBY gene family in foxtail millet (Setaria italica) and functional characterization of SiDL. Journal of Integrative Agriculture 21(10):2876−87

    doi: 10.1016/j.jia.2022.07.052

    CrossRef   Google Scholar

    [40]

    Zhang B, Liu J, Cheng L, Zhang Y, Hou S, et al. 2019. Carotenoid composition and expression of biosynthetic genes in yellow and white foxtail millet [Setaria italica (L.) Beauv.]. Journal of Cereal Science 85:84−90

    doi: 10.1016/j.jcs.2018.11.005

    CrossRef   Google Scholar

    [41]

    Fan Y, Wei X, Lai D, Yang H, Feng L, et al. 2021. Genome-wide investigation of the GRAS transcription factor family in foxtail millet (Setaria italica L.). BMC Plant Biology 21:508

    doi: 10.1186/s12870-021-03277-y

    CrossRef   Google Scholar

    [42]

    Hussin SH, Wang H, Tang S, Zhi H, Tang C, et al. 2021. SiMADS34, an E-class MADS-box transcription factor, regulates inflorescence architecture and grain yield in Setaria italica. Plant Molecular Biology 105(4-5):419−34

    doi: 10.1007/s11103-020-01097-6

    CrossRef   Google Scholar

    [43]

    Tang S, Zhao Z, Liu X, Sui Y, Zhang D, et al. 2023. An E2-E3 pair contributes to seed size control in grain crops. Nature Communications 14:3091

    doi: 10.1038/s41467-023-38812-y

    CrossRef   Google Scholar

    [44]

    Wang T, Song H, Li P, Wei Y, Hu N, et al. 2020. Transcriptome Analysis Provides Insights into Grain Filling in Foxtail Millet (Setaria italica L.). International Journal of Molecular Sciences 21(14):5031

    doi: 10.3390/ijms21145031

    CrossRef   Google Scholar

    [45]

    Zhao Z, Liu D, Cui Y, Li S, Liang D, et al. 2020. Genome-wide identification and characterization of long non-coding RNAs related to grain yield in foxtail millet [Setaria italica (L. ) P. Beauv. ]. BMC genomics 21:1

    doi: 10.1186/s12864-019-6419-1

    CrossRef   Google Scholar

    [46]

    Yuan Y, Liu C, Zhao G, Gong X, Dang K, et al. 2021. Transcriptome analysis reveals the mechanism associated with dynamic changes in fatty acid and phytosterol content in foxtail millet (Setaria italica) during seed development. Food Research International 145:110429

    doi: 10.1016/j.foodres.2021.110429

    CrossRef   Google Scholar

    [47]

    Hou S, Man X, Lian B, Ma G, Sun Z, et al. 2022. Folate metabolic profiling and expression of folate metabolism-related genes during panicle development in foxtail millet (Setaria italica (L.) P. Beauv.). Journal of the Science of Food and Agriculture 102(1):268−79

    doi: 10.1002/jsfa.11355

    CrossRef   Google Scholar

    [48]

    Wang T, Xing L, Song H, Wei Y, Li P, et al. 2023. Large-scale metabolome analysis reveals dynamic changes of metabolites during foxtail millet grain filling. Food Research International 165:112516

    doi: 10.1016/j.foodres.2023.112516

    CrossRef   Google Scholar

    [49]

    Song H, Wang T, Li L, Xing L, Xie H, et al. 2022. Comparative transcriptome analysis provides insights into grain filling commonalities and differences between foxtail millet [Setaria italica (L.) P. Beauv.] varieties with different panicle types. PeerJ 10:e12968

    doi: 10.7717/peerj.12968

    CrossRef   Google Scholar

    [50]

    Moreno-Sanz P, D'Amato E, Nebish A, Costantini L, Grando MS. 2020. An optimized histological proceeding to study the female gametophyte development in grapevine. Plant Methods 16:61

    doi: 10.1186/s13007-020-00604-6

    CrossRef   Google Scholar

    [51]

    Wu X, Liu J, Li D, Liu CM. 2016. Rice caryopsis development II: Dynamic changes in the endosperm. Journal of Integrative Plant Biology 58(9):786−98

    doi: 10.1111/jipb.12488

    CrossRef   Google Scholar

    [52]

    Wu X, Liu J, Li D, Liu CM. 2016. Rice caryopsis development I: dynamic changes in different cell layers. Journal of Integrative Plant Biology 58(9):772−85

    doi: 10.1111/jipb.12440

    CrossRef   Google Scholar

    [53]

    Grimault A, Gendrot G, Chamot S, Widiez T, Rabillé H, et al. 2015. ZmZHOUPI, an endosperm-specific basic helix-loop-helix transcription factor involved in maize seed development. The Plant Journal 84(3):574−86

    doi: 10.1111/tpj.13024

    CrossRef   Google Scholar

    [54]

    Bolger AM, Lohse M, Usadel B. 2014. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30(15):2114−20

    doi: 10.1093/bioinformatics/btu170

    CrossRef   Google Scholar

    [55]

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

    doi: 10.1038/nmeth.3317

    CrossRef   Google Scholar

    [56]

    Bennetzen JL, Schmutz J, Wang H, Percifield R, Hawkins J, et al. 2012. Reference genome sequence of the model plant Setaria. Nature Biotechnology 30(6):555−61

    doi: 10.1038/nbt.2196

    CrossRef   Google Scholar

    [57]

    Anders S, Huber W. 2012. Differential expression of RNA-Seq data at the gene level–the DESeq package. Heidelberg, Germany: European Molecular Biology Laboratory (EMBL). 10:f1000 https://bioconductor.statistik.tu-dortmund.de/packages/3.8/bioc/vignettes/DESeq/inst/doc/DESeq.pdf

    [58]

    Kanehisa M, Araki M, Goto S, Hattori M, Hirakawa M, et al. 2008. KEGG for linking genomes to life and the environment. Nucleic acids research 36:D480−D884

    doi: 10.1093/nar/gkm882

    CrossRef   Google Scholar

    [59]

    Wang C, Wang Y, Cheng Z, Zhao Z, Chen J, et al. 2016. The role of OsMSH4 in male and female gamete development in rice meiosis. Journal of Experimental Botany 67(5):1447−59

    doi: 10.1093/jxb/erv540

    CrossRef   Google Scholar

    [60]

    Itoh JI, Nonomura KI, Ikeda K, Yamaki S, Inukai Y, et al. 2005. Rice Plant Development: from Zygote to Spikelet. Plant and Cell Physiology 46(1):23−47

    doi: 10.1093/pcp/pci501

    CrossRef   Google Scholar

    [61]

    Counce PA, Moldenhauer KAK. 2019. Morphology of rice seed development and its influence on grain quality. In Rice Grain Quality. Methods in Molecular Biology, ed. Sreenivasulu N. vol 1892. New York: Humana Press. pp. 57−74. https://doi.org/10.1007/978-1-4939-8914-0_3

    [62]

    Bai F, Settles AM. 2015. Imprinting in plants as a mechanism to generate seed phenotypic diversity. Frontiers in Plant Science 5:780

    doi: 10.3389/fpls.2014.00780

    CrossRef   Google Scholar

    [63]

    Chateigner-Boutin AL, Ordaz-Ortiz JJ, Alvarado C, Bouchet B, Durand S, et al. 2016. Developing pericarp of maize: A model to study arabinoxylan synthesis and feruloylation. Frontiers in Plant Science 7:1476

    doi: 10.3389/fpls.2016.01476

    CrossRef   Google Scholar

    [64]

    Ciampitti IA, Elmore RW, Lauer J. 2011. Corn growth and development. Report. Dent 5. USA: Iowa State University Extension. pp. 28−39.

    [65]

    Zadoks JC, Chang TT, Konzak CF. 1974. A decimal code for the growth stages of cereals. Weed Research 14(6):415−21

    doi: 10.1111/j.1365-3180.1974.tb01084.x

    CrossRef   Google Scholar

    [66]

    Waddington SR, Cartwright PM, Wall PC. 1983. A quantitative scale of spike initial and pistil development in barley and wheat. Annals of Botany 51(1):119−30

    doi: 10.1093/oxfordjournals.aob.a086434

    CrossRef   Google Scholar

    [67]

    Zhang S, Ghatak A, Bazargani MM, Bajaj P, Varshney RK, et al. 2021. Spatial distribution of proteins and metabolites in developing wheat grain and their differential regulatory response during the grain filling process. The Plant Journal 107(3):669−87

    doi: 10.1111/tpj.15410

    CrossRef   Google Scholar

    [68]

    Becraft PW, Yi G. 2011. Regulation of aleurone development in cereal grains. Journal of Experimental Botany 62(5):1669−75

    doi: 10.1093/jxb/erq372

    CrossRef   Google Scholar

    [69]

    Pérez-Rodríguez P, Riaño-Pachón DM, Corrêa LGG, Rensing SA, Kersten B, et al. 2010. PlnTFDB: updated content and new features of the plant transcription factor database. Nucleic Acids Research 38:D822−D827

    doi: 10.1093/nar/gkp805

    CrossRef   Google Scholar

    [70]

    Sun C, Palmqvist S, Olsson H, Borén M, Ahlandsberg S, et al. 2003. A novel WRKY transcription factor, SUSIBA2, participates in sugar signaling in barley by binding to the sugar-responsive elements of the iso1 promoter. The Plant Cell 15(9):2076−92

    doi: 10.1105/tpc.014597

    CrossRef   Google Scholar

    [71]

    Zhang CQ, Xu Y, Lu Y, Yu HX, Gu MH, et al. 2011. The WRKY transcription factor OsWRKY78 regulates stem elongation and seed development in rice. Planta 234(3):541−54

    doi: 10.1007/s00425-011-1423-y

    CrossRef   Google Scholar

    [72]

    Luo X, Bai X, Sun X, Zhu D, Liu B, et al. 2013. Expression of wild soybean WRKY20 in Arabidopsis enhances drought tolerance and regulates ABA signalling. Journal of Experimental Botany 64(8):2155−69

    doi: 10.1093/jxb/ert073

    CrossRef   Google Scholar

    [73]

    Schilling S, Pan S, Kennedy A, Melzer R. 2018. MADS-box genes and crop domestication: the jack of all traits. Journal of Experimental Botany 69(7):1447−69

    doi: 10.1093/jxb/erx479

    CrossRef   Google Scholar

    [74]

    Lai D, Yan J, He A, Xue G, Yang H, et al. 2022. Genome-wide identification, phylogenetic and expression pattern analysis of MADS-box family genes in foxtail millet (Setaria italica). Scientific Reports 12(1):4979

    doi: 10.1038/s41598-022-07103-9

    CrossRef   Google Scholar

    [75]

    Zuo ZF, Lee HY, Kang HG. 2023. Basic Helix-Loop-Helix Transcription Factors: Regulators for Plant Growth Development and Abiotic Stress Responses. International Journal of Molecular Sciences 24(2):1419

    doi: 10.3390/ijms24021419

    CrossRef   Google Scholar

    [76]

    Das AK, Hao L. 2022. Functional characterization of ZmbHLH121, a bHLH transcription factor, focusing on Zea mays kernel development. Gene Reports 28:101645

    doi: 10.1016/j.genrep.2022.101645

    CrossRef   Google Scholar

    [77]

    Luo J, Liu H, Zhou T, Gu B, Huang X, et al. 2013. An-1 encodes a basic helix-loop-helix protein that regulates awn development, grain size, and grain number in rice. The Plant Cell 25(9):3360−76

    doi: 10.1105/tpc.113.113589

    CrossRef   Google Scholar

    [78]

    Yang X, Ren Y, Cai Y, Niu M, Feng Z, et al. 2018. Overexpression of OsbHLH107, a member of the basic helix-loop-helix transcription factor family, enhances grain size in rice (Oryza sativa L.). Rice 11:41

    doi: 10.1186/s12284-018-0237-y

    CrossRef   Google Scholar

    [79]

    Li C, Qiao Z, Qi W, Wang Q, Yuan Y, et al. 2015. Genome-wide characterization of cis-acting DNA targets reveals the transcriptional regulatory framework of opaque2 in maize. The Plant Cell 27(3):532−45

    doi: 10.1105/tpc.114.134858

    CrossRef   Google Scholar

    [80]

    Cao R, Zhao S, Jiao G, Duan Y, Ma L, et al. 2022. OPAQUE3, encoding a transmembrane bZIP transcription factor, regulates endosperm storage protein and starch biosynthesis in rice. Plant Communications 3(6):100463

    doi: 10.1016/j.xplc.2022.100463

    CrossRef   Google Scholar

    [81]

    Saidi A, Hajibarat Z. 2021. Phytohormones: plant switchers in developmental and growth stages in potato. Journal, Genetic Engineering & Biotechnology 19(1):89

    doi: 10.1186/s43141-021-00192-5

    CrossRef   Google Scholar

    [82]

    Zažímalová E, Murphy AS, Yang H, Hoyerová K, Hošek P. 2010. Auxin transporters—why so many? Cold Spring Harbor perspectives in biology 2(3):a001552

    doi: 10.1101/cshperspect.a001552

    CrossRef   Google Scholar

    [83]

    Zhao Y. 2012. Auxin biosynthesis: A simple two-step pathway converts tryptophan to indole-3-acetic acid in plants. Molecular Plant 5(2):334−38

    doi: 10.1093/mp/ssr104

    CrossRef   Google Scholar

    [84]

    Chen Y, Liu B, Zhao Y, Yu W, Si W. 2021. Whole-genome duplication and purifying selection contributes to the functional redundancy of Auxin Response Factor (ARF) genes in foxtail millet (Setaria italica L.). International Journal of Genomics 2021:2590665

    doi: 10.1155/2021/2590665

    CrossRef   Google Scholar

    [85]

    Ma X, Dai S, Qin N, Zhu C, Qin J, et al. 2023. Genome-wide identification and expression analysis of the SAUR gene family in foxtail millet (Setaria italica L.). BMC Plant Biology 23(1):31

    doi: 10.1186/s12870-023-04055-8

    CrossRef   Google Scholar

    [86]

    Fu J, Yu H, Li X, Xiao J, Wang S. 2011. Rice GH3 gene family: Regulators of growth and development. Plant Signaling & Behavior 6(4):570−74

    doi: 10.4161/psb.6.4.14947

    CrossRef   Google Scholar

    [87]

    Hui S, Zhang M, Hao M, Yuan M. 2019. Rice group I GH3 gene family, positive regulators of bacterial pathogens. Plant Signaling & Behavior 14(5):e1588659

    doi: 10.1080/15592324.2019.1588659

    CrossRef   Google Scholar

    [88]

    Mok MC. 2019. Cytokinins and plant development—an overview. In Cytokinins, ed. Mok MC. Boca Raton: CRC Press. pp. 155–66. https://doi.org/10.1201/9781351071284-12

    [89]

    Wu K, Xu H, Gao X, Fu X. 2021. New insights into gibberellin signaling in regulating plant growth–metabolic coordination. Current Opinion in Plant Biology 63:102074

    doi: 10.1016/j.pbi.2021.102074

    CrossRef   Google Scholar

    [90]

    Kozaki A, Aoyanagi T. 2022. Molecular aspects of seed development controlled by gibberellins and abscisic acids. International Journal of Molecular Sciences 23(3):1876

    doi: 10.3390/ijms23031876

    CrossRef   Google Scholar

    [91]

    Hernández-García J, Briones-Moreno A, Blázquez MA. 2021. Origin and evolution of gibberellin signaling and metabolism in plants. Seminars in Cell & Developmental Biology 109:46−54

    doi: 10.1016/j.semcdb.2020.04.009

    CrossRef   Google Scholar

    [92]

    Blázquez MA, Nelson DC, Weijers D. 2020. Evolution of plant hormone response pathways. Annual Review of Plant Biology 71(1):327−53

    doi: 10.1146/annurev-arplant-050718-100309

    CrossRef   Google Scholar

    [93]

    Sano N, Marion-Poll A. 2021. ABA metabolism and homeostasis in seed dormancy and germination. International Journal of Molecular Sciences 22(10):5069

    doi: 10.3390/ijms22105069

    CrossRef   Google Scholar

    [94]

    Parwez R, Aftab T, Gill SS, Naeem M. 2022. Abscisic acid signaling and crosstalk with phytohormones in regulation of environmental stress responses. Environmental and Experimental Botany 199:104885

    doi: 10.1016/j.envexpbot.2022.104885

    CrossRef   Google Scholar

    [95]

    Tang J, Han Z, Chai J. 2016. Q&A: what are brassinosteroids and how do they act in plants? BMC Biology 14:113

    doi: 10.1186/s12915-016-0340-8

    CrossRef   Google Scholar

    [96]

    Manghwar H, Hussain A, Ali Q, Liu F. 2022. Brassinosteroids (BRs) role in plant development and coping with different stresses. International Journal of Molecular Sciences 23(3):1012

    doi: 10.3390/ijms23031012

    CrossRef   Google Scholar

    [97]

    Zhao M, Tang S, Zhang H, He M, Liu J, et al. 2020. DROOPY LEAF1 controls leaf architecture by orchestrating early brassinosteroid signaling. Proceedings of the National Academy of Sciences of the United States of America 117(35):21766−74

    doi: 10.1073/pnas.2002278117

    CrossRef   Google Scholar

    [98]

    Li C, Fu K, Guo W, Zhang X, Li C, et al. 2023. Starch and sugar metabolism response to post-anthesis drought stress during critical periods of elite wheat (Triticum aestivum L.) endosperm development. Journal of Plant Growth Regulation 42:5476−94

    doi: 10.1007/s00344-023-10930-3

    CrossRef   Google Scholar

    [99]

    Li K, Zhang T, Sui Z, Narayanamoorthy S, Jin C, et al. 2019. Genetic variation in starch physicochemical properties of Chinese foxtail millet (Setaria italica Beauv.). International Journal of Biological Macromolecules 133:337−45

    doi: 10.1016/j.ijbiomac.2019.04.022

    CrossRef   Google Scholar

    [100]

    Woo YM, Hu DWN, Larkins BA, Jung R. 2001. Genomics analysis of genes expressed in maize endosperm identifies novel seed proteins and clarifies patterns of zein gene expression. The Plant Cell 13:2297−317

    doi: 10.1105/tpc.010240

    CrossRef   Google Scholar

    [101]

    Huang Y, Wang H, Zhu Y, Huang X, Li S, et al. 2022. THP9 enhances seed protein content and nitrogen-use efficiency in maize. Nature 612:292−300

    doi: 10.1038/s41586-022-05441-2

    CrossRef   Google Scholar

    [102]

    Qiao Z, Qi W, Wang Q, Feng YN, Yang Q, et al. 2016. ZmMADS47 Regulates Zein Gene Transcription through Interaction with Opaque2. PLoS Genetics 12(4):e1005991

    doi: 10.1371/journal.pgen.1005991

    CrossRef   Google Scholar

    [103]

    Hou S, Men Y, Wei M, Zhang Y, Li H, et al. 2022. Total Protein Content, Amino Acid Composition and Eating-Quality Evaluation of Foxtail Millet (Setaria italica (L. ) P. Beauv). Foods 12(1):31

    doi: 10.3390/foods12010031

    CrossRef   Google Scholar

    [104]

    Li X, Gao J, Song J, Guo K, Hou S, et al. 2022. Multi-omics analyses of 398 foxtail millet accessions reveal genomic regions associated with domestication, metabolite traits, and anti-inflammatory effects. Molecular Plant 15(8):1367−83

    doi: 10.1016/j.molp.2022.07.003

    CrossRef   Google Scholar

    [105]

    Zhou X, Rao S, Wrightstone E, Sun T, Lui ACW, et al. 2022. Phytoene Synthase: The Key Rate-Limiting Enzyme of Carotenoid Biosynthesis in Plants. Frontiers in Plant Science 13:884720

    doi: 10.3389/fpls.2022.884720

    CrossRef   Google Scholar

    [106]

    Dhaka A, Muthamilarasan M, Prasad M. 2021. A comprehensive study on core enzymes involved in starch metabolism in the model nutricereal, foxtail millet (Setaria italica L.). Journal of Cereal Science 97:103153

    doi: 10.1016/j.jcs.2020.103153

    CrossRef   Google Scholar

    [107]

    Zhang Y, Gao J, Qie Q, Yang Y, Hou S, et al. 2021. Comparative analysis of flavonoid metabolites in foxtail millet (Setaria italica) with different eating quality. Life 11(6):578

    doi: 10.3390/life11060578

    CrossRef   Google Scholar

    [108]

    Chowdhary AA, Mishra S, Mehrotra S, Upadhyay SK, Bagal D, et al. 2023. Plant transcription factors: an overview of their role in plant life. In Plant Transcription Factors, eds. Srivastava V, Mishra S, Mehrotra S, Upadhyay SK. Netherlands: Elsevier. pp. 3–20. https://doi.org/10.1016/b978-0-323-90613-5.00003-0

    [109]

    Srivastava V, Mishra S, Mehrotra S, Upadhyay SK. 2022. Plant Transcription Factors: Contribution in Development, Metabolism, and Environmental Stress. Netherlands: Elsevier. https://doi.org/10.1016/C2020-0-04071-5

    [110]

    Dai D, Ma Z, Song R. 2021. Maize endosperm development. Journal of Integrative Plant Biology 63(4):613−27

    doi: 10.1111/jipb.13069

    CrossRef   Google Scholar

    [111]

    Feng F, Song R. 2018. O11 is multi-functional regulator in maize endosperm. Plant Signaling & Behavior 13(4):e1451709

    doi: 10.1080/15592324.2018.1451709

    CrossRef   Google Scholar

    [112]

    Jain R, Dhaka N, Yadav P, Sharma R. 2023. Role of phytohormones in regulating agronomically important seed traits in crop plants. In Plant Hormones in Crop Improvement, eds. Khan MIR, Singh A, Poór P. Netherlands: Elsevier. pp. 65–88. https://doi.org/10.1016/b978-0-323-91886-2.00002-1

    [113]

    Liu J, Shi X, Chang Z, Ding Y, Ding C. 2022. Auxin efflux transporters OsPIN1c and OsPIN1d function redundantly in regulating rice (Oryza sativa L.) panicle development. Plant and Cell Physiology 63(3):305−16

    doi: 10.1093/pcp/pcab172

    CrossRef   Google Scholar

    [114]

    Zhao Z, Tang S, Li W, Yang X, Wang R, et al. 2021. Overexpression of a BRASSINAZOLE RESISTANT 1 homolog attenuates drought tolerance by suppressing the expression of PLETHORA-LIKE 1 in Setaria italica. The Crop Journal 9(5):1208−1213

    doi: 10.1016/j.cj.2021.02.006

    CrossRef   Google Scholar

    [115]

    Yu G, Gaoyang Y, Liu L, Shoaib N, Deng Y, et al. 2022. The structure, function, and regulation of starch synthesis enzymes SSIII with emphasis on maize. Agronomy 12(6):1359

    doi: 10.3390/agronomy12061359

    CrossRef   Google Scholar

    [116]

    Jeon JS, Ryoo N, Hahn TR, Walia H, Nakamura Y. 2010. Starch biosynthesis in cereal endosperm. Plant Physiology and Biochemistry 48(6):383−92

    doi: 10.1016/j.plaphy.2010.03.006

    CrossRef   Google Scholar

    [117]

    Smith AM. 2012. Starch in the Arabidopsis plant. Starch 64(6):421−34

    doi: 10.1002/star.201100163

    CrossRef   Google Scholar

    [118]

    Fukunaga K, Kawase M, Kato K. 2002. Structural variation in the Waxy gene and differentiation in foxtail millet [Setaria italica (L.) P. Beauv.]: implications for multiple origins of the waxy phenotype. Molecular Genetics and Genomics 268(2):214−22

    doi: 10.1007/s00438-002-0728-8

    CrossRef   Google Scholar

    [119]

    Van K, Onoda S, Kim MY, Kim KD, Lee SH. 2008. Allelic variation of the Waxy gene in foxtail millet [Setaria italica (L.) P. Beauv.] by single nucleotide polymorphisms. Molecular Genetics and Genomics 279(3):255−66

    doi: 10.1007/s00438-007-0310-5

    CrossRef   Google Scholar

    [120]

    Yang Q, Yuan Y, Liu J, Han M, Li J, et al. 2023. Transcriptome analysis reveals new insights in the starch biosynthesis of non-waxy and waxy broomcorn millet (Panicum miliaceum L.). International Journal of Biological Macromolecules 230:123155

    doi: 10.1016/j.ijbiomac.2023.123155

    CrossRef   Google Scholar

    [121]

    Wu Y, Holding DR, Messing J. 2010. γ-Zeins are essential for endosperm modification in quality protein maize. Proceedings of the National Academy of Sciences of the United States of America 107:12810−15

    doi: 10.1073/pnas.1004721107

    CrossRef   Google Scholar

    [122]

    Zhang Z, Yang J, Wu Y. 2015. Transcriptional regulation of zein gene expression in maize through the additive and synergistic action of opaque2, Prolamine-box binding factor, and O2 heterodimerizing proteins. The Plant Cell 27(4):1162−72

    doi: 10.1105/tpc.15.00035

    CrossRef   Google Scholar

    [123]

    Liu CN, Rubenstein I. 1993. Transcriptional characterization of an α-zein gene cluster in maize. Plant Molecular Biology 22(2):323−36

    doi: 10.1007/BF00014939

    CrossRef   Google Scholar

    [124]

    Song R, Segal G, Messing J. 2004. Expression of the sorghum 10-member kafirin gene cluster in maize endosperm. Nucleic Acids Research 32(22):e189

    doi: 10.1093/nar/gnh183

    CrossRef   Google Scholar

    [125]

    He L, Cheng L, Wang J, Liu J, Cheng J, et al. 2022. Carotenoid Cleavage Dioxygenase 1 catalyzes lutein degradation to influence carotenoid accumulation and color development in foxtail millet grains. Journal of Agricultural and Food Chemistry 70(30):9283−94

    doi: 10.1021/acs.jafc.2c01951

    CrossRef   Google Scholar

  • Cite this article

    Wang D, Su M, Hao JH, Li ZD, Dong S, et al. 2023. Dynamic transcriptome landscape of foxtail millet grain development. Seed Biology 2:19 doi: 10.48130/SeedBio-2023-0019
    Wang D, Su M, Hao JH, Li ZD, Dong S, et al. 2023. Dynamic transcriptome landscape of foxtail millet grain development. Seed Biology 2:19 doi: 10.48130/SeedBio-2023-0019

Figures(9)

Article Metrics

Article views(3891) PDF downloads(563)

ARTICLE   Open Access    

Dynamic transcriptome landscape of foxtail millet grain development

Seed Biology  2 Article number: 19  (2023)  |  Cite this article

Abstract: Grain development of foxtail millet (Setaria italica L.) is essential for yield and quality. However, its transcriptional dynamics molecular mechanisms and morphological analyses remain scarcely described. Thus, we conducted detailed daily morphological analyses of foxtail millet grain development throughout the 30 d post-fertilization development period. On the basis of the morphological analyses, we used RNA-sequencing (RNA-seq) to examine the transcript dynamics involved in foxtail millet grain development at four stages. These genes included those associated with transcriptional regulation, hormone signaling, sucrose and starch metabolism, zein family members, amino acid metabolism, carotenoid metabolism, flavonoid biosynthesis, and folate synthesis. We have validated the accuracy of the transcriptome data by means of reverse transcription quantitative polymerase chain reaction (RT-qPCR). This study provides precious genetic resources for understanding grain developmental process in the future. These results expand our understanding of the molecular mechanisms of grain development in foxtail millet and contribute to the functional studies of genes related to grain development in the future.

    • Foxtail millet (Setaria italica L.), belonging to the poaceae family, was first cultivated in the green foxtail (S. viridis) about 16,000 years before present[13]. It has since become the most prominent miscellaneous cereal in northern China, especially in arid or semi-arid area[4,5]. Foxtail millet grains are rich in nutrients, including polyphenols[6], organic acids[7], vitamin E, and carotenoids[8], a variety of essential amino acids, and high-quality protein[9]. Additionally, it is a good source of trace elements such as zinc and iron[10]. Its bran is also rich in linoleic and linolenic acids[11,12], and is an excellent crude fiber source, which helps with intestinal digestion and promotes digestive health[13]. Recently, foxtail millet's small diploid genome, self-compatibility, strong stress tolerance, and low repetitive DNA content have made it a model crop for C4 monocot studies[2,3,14]. However, despite foxtail millet has many advantages, the planting area and total yield have been a declining trend, and its yield per unit area is significantly lower compared to other staple food crops such as maize, wheat, and rice. So attention should be paid to improving foxtail millet yield and quality.

      Seeds are reproductive units of flowering plants, and the ultimate goal of seeds is to successfully establishing the progeny. The development of seed is not only the growing of the embryo, but also involves close coordination of different tissues, including seed coat, embryo, and endosperm which nourishes the embryo[15]. In cereals, the endosperm contains starch, protein, and lipids, accounting for approximately 42.5% of global food calories for humans[16]. In brief, seed development can be described in three stages: cellular division, morphogenesis, and maturation[17]. Morphogenesis encompasses all processes that form and develop the different parts of mature seed, and it is during this stage that resources are allocated[18]. Maturation is a physiological process that comes to an end when the seed begins the dormant state[19]. However, the developmental process of foxtail millet grains is not clearly defined morphologically. Therefore, the morphological description of foxtail millet grain development is particularly important.

      Grain development is an important process in foxtail millet growth, related to seed setting, grain weight, yield, and quality[20]. This process is influenced by various factors including the delivery of photosynthetic products, transportation of stored substances, transport tissues and physiological activities of the grains themselves, regulation of various plant hormones, and restriction of various environmental factors[21]. The interaction of phytohormones, activity of starch biosynthesis enzymes, levels of polyamines, and synthesis and translocation of assimilates all play essential role in grain development[2224]. In recent years, numerous genes and signaling pathways have been discovered to play a role in regulating grain development. The endosperm specific bZIP transcription factor (TF) O2 can bind the O2-box and transactivates the 22-kDa α-zein protein[25]. bHLH transcription factor OPAQUE11 (O11) can directly regulate the expression of various carbohydrate metabolizing enzymes, affecting carbohydrate accumulation, amino acid metabolism, and the transcription of stress response genes[26]. ZmNAC11 and ZmNAC29 are also involved in grain development as they directly activate the expression of ZmEXPB15, which promotes early grain development in maize[27]. Additionally, ZmNAC128 and ZmNAC130 are endosperm-specific TFs that affect starch biosynthesis and 16-kDa γ-zein content[28]. The specific expression of OsSUT1 in the aleurone layer indicates a role for this gene in sucrose uptake, and the spatial distribution of AGPL2 and AGPS2b is associated with the development of starch granules in the grain[29]. TA2 encoding DNA demethylase OsROS1 in rice, restricts the aleurone layer cell number by mediating DNA demethylation, providing means to improve rice nutrition[30]. Phytohormones can also affect seed development and regulate grain filling. DEP1/qPE9-1 can promote starch accumulation and prolong the grain filling process through auxin and cytokinin[31]. Recently, a new pathway, the sugar-auxin crosstalk signaling pathway, has been discovered that is involved in sugar metabolism and carbon partitioning[32]. Ethylene and abscisic acid have been shown to regulate the activity of enzymes involved in starch synthesis, ultimately affecting the grain filling process[33,34].

      However, compared to major cereal crops, the molecular mechanisms and regulatory networks of grain development in foxtail millet are rarely reported and only a limited number of genes involved in seed development have been identified. Therefore, elucidating the regulatory networks and key regulators during foxtail millet grain development is important, which will help us to understand the molecular basis of foxtail millet grain yield and quality. LOOSE PANICLE1 (LP1) is known to encode a novel WRKY transcription factor, and the lp1 mutant can affect panicle development and seed size in foxtail millet[35]. F128 is the first seed-specific promoter gene reported in foxtail millet, encoding protein is likely a protease inhibitor / seed storage protein / lips-transfer protein. F128 was specifically expressed in immature and mature seeds and gradually decreased with seed maturity, and was no longer expressed in mature seeds of 25 DAP[36]. TRANSPARENT TESTA GLABRA1 (SiTTG1) is a WD40 repeat transcription factor, is able to induce the expression of genes associated with the accumulation of seed fatty acids and storage proteins, thereby plays an important role in seed metabolite production[37]. SiLEA14 is a late embryogenesis abundant (LEA) proteins, its transcription level was gradually increase with seed maturation, indicating that it had a potential role at the maturation and drying stages of seed development in foxtail millet[38]. SiDL is a member of the YABBY family, and the discovery that over-expression of SiDL leads to delayed flowering and reduced seed size confirms its function in foxtail millet seed development[39]. The transcriptional levels of genes SiPSY1/2/3, SiPDS, SiZDS, SiZ-ISO, SiCRTISO, SiLCYB, SiLCYE, and SiHYD, that play a role in the carotenoid pathway have been extensively studied. These studies have shown that up-regulated expression of SiPSY1 accompanied by down-regulated expression of SiCCD1 is the key to increased carotenoid accumulation in foxtail millet seeds[40]. The expression of SiGRAS41 increased gradually with the development of millet seeds, while the expression of SiGRAS01 decreased with fruit maturation[41]. SiMADS34 has been found to regulate inflorescence structure and panicle development in a variety of regulatory pathways[42]. Ring-type E3 ligase SGD1 and its E2 partner SiUBC32 control grain development by regulating grain weight and grain size in foxtail millet[43].

      Although several genes involved seed development have been identified, few key regulatory genes have been cloned, and the molecular mechanisms and regulatory networks in foxtail millet seed development largely remain unclear. In recent years, several studies have used RNA-seq transcriptome analysis or metabolomics analysis to identify gene expression profiles in foxtail millet seed development[4449]. Wang et al.[44] first identified the gene expression profiles of foxtail millet at five different developmental stages, including at 7, 14, 21, 28, and 35 days after anthesis (DAA), using RNA-seq analysis. These studies have identified a number of genes that exhibited dynamic or enriched expression patterns in foxtail millet seed development and highlighted many potential regulators that may be involved in seed filling and the accumulation of seed metabolites. However, detailed morphological analysis and description regarding the foxtail millet seed development process, is an important step in the study of the seed development process. In previous studies, transcriptome analysis of the seed development process was carried out according to the date of flowering, but detailed morphological analysis and description of the specific seed development process have not been provided. To improve our understanding of dynamic gene expression profiles of seed development, we conducted detailed morphological analysis and description of the development process of foxtail millet seeds, including the ovule stage of before pollination that have not been identified in previous studies. In our study, morphological analyses and definition of foxtail millet grain development were performed, and the dynamic regulatory network of grain development was depicted by RNA-seq analysis of grains at four developmental stages. Our studies have not only provided morphological and cytohistological analyses of grain development, but also developed a temporal transcriptome based on high-throughput RNA-seq atlas of grain development, and this study will provide a precious genetic resource for understanding grain developmental process for the future.

    • The cultivars 'Jingu 21' of foxtail millet were used in this study. Materials were planted and grown at Shenfeng experimental fields in Jinzhong, Shanxi, China (37°25′ N, 112°35′ E) from May to October in 2022. The surfaces of florets at anthesis were tagged using a marker pen to denote flowering dates. For morphological analysis, samples were collected daily over a period of 30 d of foxtail millet grain development. For staining assay, paraffin section analysis, and RNA-seq, samples from four stages (ovule stage, milk stage, dough stage, maturity stage) were collected. Three replicates were generated for each stage, and each replicate was made up of a mixture grain of five spikes. Samples were snap frozen in liquid nitrogen before processing and stored at −80 °C.

    • For confocal microscopy, samples were fixed with 2.0% paraformaldehyde and 2.0% glutaraldehyde fixative (Coolaber, Beijing, China) for 48 h, rinsed three times in PBS buffer, then dehydrated in alcohol concentrations (20%, 50%, 70%, 80%, 90%, 95%, 100%, 100%, 100%) for 1 h, respectively. After completing the above operations, the gradient dehydration samples were immediately mounted in a mixture of benzyl benzoate (MACKLIN, Shanghai, China) and benzyl alcohol (MACKLIN, Shanghai, China) (2v:1v) and photographed using a confocal microscopy (DMI 8, Leica, Germany), or placed into the mixture above and stored away from light in refrigerator at 4 °C. The confocal microscopy system excitation wavelength is 568 nm and emission wavelength is 550−630 nm.

    • A stereo microscope (SZX16, Olympus, Japan) was used to record the daily growth changes of Jingu 21 developing grains throughout 1−30 DAP. For staining starch or lipid, median transverse sections of foxtail millet grains were manually cut with a sharp razor and immersed into solutions of Lugol's iodine, 0.1% (w/v) Evans blue, 0.1% (w/v) toluidine blue, and 0.5% (w/v) Sudan IV (dissolved in Chloroform), respectively. After staining, the hand-sectioned grains were viewed under microscope (BX51, Olympus, Japan), and immediately photographed.

    • Paraffin section analyses were performed according to a method as described with a few modification[50]. Samples collected at four stages (ovule stage, milk stage, dough stage, maturity stage) were first fixed in 2.0% paraformaldehyde and 2.0% glutaraldehyde fixative. Fixed samples were washed, dehydrated, embedded into wax and cut into 10 μm-thick sections. Samples were redyed with Safranin O-Fast Green (Phygene, Fuzhou, China) or toluidine blue O (Coolaber, Beijing, China) and observed with microscope (BX51, Olympus, Japan).

    • Hand sections from the central region of foxtail millet grains from ovule stage to maturity stage were made using sharp razor. Evans blue, toluidine blue, Sudan IV, and Lugol's iodine staining (Coolaber, Beijing, China) were performed according to the method as described with a few modification[5153]. Sections were soaked in 0.1% Evans blue, 0.1% toluidine blue, and Lugol's iodine for 10 min, 4 h, and 5 min, respectively, and immediately washed with distilled water. For Sudan IV staining, sections were stained with 0.5% Sudan IV for 24 h, washed with 70% alcohol and immediately placed in distilled water. All samples were under the microscope for observation and photography (BX51, Olympus, Japan).

    • RNA-seq was performed by Shanghai OE Biotech Co., Ltd. (Shanghai, China). Raw reads were trimmed and mapped by Trimmomatic (version 0.36)[54] and HISAT2 (version 2.2.1.0)[55] to the reference genome of foxtail millet[56], respectively. DESeq package was used to analysis the differential expression of genes (using BaseMean value to estimate expression)[57]. Fold change > 2 and adjusted P-value < 0.05 as the criteria for differentially expressed genes (DEGs) screening. Hierarchical cluster, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis were performed by the OE Biotech cloud analysis tools (https://cloud.oebiotech.cn)[58].

    • Total RNA extracted by the FlaPure Plant Total RNA Extraction Kit (Genesand Biotech, Beijing, China). Four hundred ng total RNA was used to generate the cDNA by the Union Script First-strand cDNA Synthesis Kit (Genesand Biotech, Beijing, China). qPCR was carried out using the SYBR Green Super Mix (Mei5bio, Beijing, China) and Bio-Rad CFX96 (Bio-Rad CFX96, BIORAD, USA). The primers were listed in Supplemental Table S6. The gene expression was normalized to housekeeping gene Actin 1 (Seita.8G043100).

    • Prism 7.0 (GraphPad Software, INC., USA) was used for statistical analysis. The statistical significance were evaluated by one-way analysis of variance (ANOVA) followed by Duncan's multiple comparison range test and were considered significant at p-values < 0.05 level.

    • To investigate the dynamic morphology of foxtail millet grain development after pollination. We utilized a stereo microscope to record the daily growth changes of foxtail millet grains developing during 30 d after pollination. We observed that the growth period of foxtail millet grain mainly occurred at 1−15 DAP, while grain development was initially characterized by longitudinal development before 4 DAP, and then transferred to lateral growth at 5−15 DAP. However, it was noted that the transverse and longitudinal growth was slower compared with the growth before 15 DAP (Fig. 1, Supplemental Fig. S1). We found that foxtail millet grains began to turn yellow at 8 DAP (Fig. 1b), which may be responsible for the accumulation of carotenoids and flavonoids during this process. Additionally, we also observed that the grains gradually became translucent, which could be caused by dehydration and crystallization of starch and storage proteins after 24 DAP (Fig. 1b). To investigate the development patterns of grain growth, area, perimeter, length, width, and length/width ratio of foxtail millet grains were measured everyday throughout the 30 d development period (Supplemental Fig. S1). It was observed that elongation of foxtail millet grains occurred sharply from 1 DAP, and the grain length and width reached their maximum at 15 DAP (Supplemental Fig. S1c). The grain perimeter and area were found to be consistent with the grain length and width, and also reached their maximum value at 15 DAP (Supplemental Fig. S1a, b). Grain change showed that the grain grew rapidly longitudinally in 1−4 DAP, consequently, the maximum length/width ratio was seen at 4 DAP, and lateral growth began after 5 DAP and aspect ratio change was not obvious after 15 DAP (Supplemental Fig. S1d). These apparent morphological results were consistent with the statistical data obtained from 1-30 d of seed development.

      Figure 1. 

      Dynamics of foxtail millet grain development. (a) Time series of kernel development from 1 to 30 DAP; (b) Time series of grain development from 1 to 30 DAP without kernel hull. Numbers below denote the DAP. (a) Scale bars = 1 mm, (b) Scale bars = 500 μm.

      To describe the developmental morphology of the ovule before pollination. The young spikes containing all developmental stages of ovules and florets were selected to examine the morphology of the developing florets and characterize the ovule structure using confocal microscopy (Fig. 2). The developmental process can be divided into eight periods according to appearance and morphological of foxtail millet spikelets (Fig. 2b). Morphological analysis of ovule development before pollination was also defined as eight stages, starting with the initial stage of ovule primordium (Fig. 2c, d). Similarly, the development of ovules was also defined as eight stages in cell morphology by confocal microscopy (Fig. 2el). The ovule developmental processes were divided into eight stages: Ovule primordium, megassporocyte, Functional megaspore (FM), Mono-nucleate embryo sac (Mnes), Two-nucleate embryo sac (Tnes), Four-nucleate embryo sac (Fnes), Eight-nucleate embryo sac (Enes), and Mature embryo sac (Mes). The cell morphological characteristics of foxtail millet ovule before pollination are similar to those of the embryo sac in rice[59].

      Figure 2. 

      Morphological analysis of the developing ovules and spikelets in foxtail millet before pollination. (a) The young foxtail millet spike contains all developmental stages of ovules and florets. (b) Appearance morphological analysis of florets development in foxtail millet. (c) Morphology of the ovule primordia in foxtail millet. (d) Morphology of ovule development in foxtail millet before pollination. (e)−(l) The structure of ovules observed by confocal microscopy in foxtail millet before pollination. FM: Functional megaspore; Mnes: Mono nucleate embryo sac; Tnes: Two-nucleate embryo sac; Fnes: Four-nucleate embryo sac; Enes: Eight-nucleate embryo sac; Mes: Mature embryo sac. (b)−(d) Scale bars = 10 μm, (e) Scale bars = 43.7 μm; (f) Scale bars = 33.9 μm; (g) Scale bars = 45.6 μm; (h) Scale bars = 45.3 μm; (i) Scale bars = 56.7 μm; (j) Scale bars = 48.9 μm; (k) Scale bars = 34.6 μm; (l) Scale bars = 60.3 μm.

      We defined the development process of foxtail millet grains, referring to relevant research on the seed development process of rice, maize, wheat and other graminaceous cereal crop[6066]. 1−5 DAP is defined as the watery stage, characterized by grains that can be easily squeezed to release transparent liquid. 6−11 DAP is defined as the milk stage, marked by grains gradually turning yellow and releasing milky white pulp substance when squeezed. 12−15 DAP is defined as the soft dough stage, and the green color of the seed coat begins to fade, and the grains squeeze out nutrients that have the consistency of dough. 16−22 DAP is defined as the hard dough stage, and the grains become hardened and could not be easily crushed, and the embryo grows to its maximum at this stage. 23−30 DAP is defined as the mature stage, and the grains become hard and transparent due to the loss of water, and the embryo becomes clearly visible and smaller. The mature ovule before pollination is defined as the ovule stage. To ensure the accurate collection of four different development stages, we systematically summarized the morphological characteristics of foxtail millet seed development (Supplemental Table S1). Our results provide detailed cytological and morphological support for the dynamic developmental process of foxtail millet grain, and offer reference for future research on specific periods of foxtail millet grain development.

    • Grain development is a critical process of nutrient accumulation, and nutrients were mainly accumulated in starchy endosperm and the aleurone layer[67]. Starchy endosperm mainly accumulates starch, while the aleurone layer primarily accumulates storage proteins, lipids, vitamins, and minerals[30,68]. Various staining methods were used to analysis the accumulation of storage products during grain development. Evans blue staining showed that programmed cell death (PCD) was initially present in the periphery zone of the mature ovule stage (Fig. 3a1), and extended to the entire starchy endosperm at the milk stage, dough stage, and maturity stage (Fig. 3a2a4). In contrast, at the maturity stage, cells of the embryo-surrounding region (ESR) remain viable, as indicated by the absence of Evans blue staining (shown by light colour; Fig. 3a4). This suggests that active starch is still present in the starchy endosperm, and that apparently disrupting cytoplasmic membrane integrity does not affect starch biosynthesis. The most plausible explanation is that the nuclear, plastid, mitochondria, and ER membranes remain fully functional. The staining results of toluidine blue were similar to those of Evans blue (Fig. 3b1b4). We observed a gradual increase in starch accumulation in the starchy endosperm, where starch is the main component (shown by brown/black; Fig. 3d1d4, e1e4). As shown in transversal sections of husked grain (Fig. 3e1e4), lipids were predominantly stored in the aleurone layer and embryo (shown by orange-red; Fig. 3c1c4, e1e4). Through staining analysis of the developing grains, we also observed that starch grains changed from irregular state to regular crystalline starch (Fig. 3). Crystallization of starch and storage protein resulted in the formation of translucent endosperm, as has been reported[51,52].

      Figure 3. 

      Storage product accumulation in the foxtail millet hulled grain by staining analysis. (a1)−(a4) Transversely sectioned hulled grain stained with Evans blue to analyze the PCD of developing grain. (b1)−(b4) Transversely sectioned hulled grain stained with toluidine blue to observe endosperm storage product accumulation. (c1)−(c4) Transversely sectioned hulled grain dyed with Sudan Red IV to analyze the lipids accumulation. (d1)−(d4) Transversely sectioned hulled grain stained with Lugol's iodine to analyze the accumulation of starch. (e1)−(e4) Transversely sectioned hulled grain dyed with Sudan Red IV and Lugol's iodine to analyze the accumulation of lipids and starch. Scale bars = 10 μm.

      To further observe the cytomorphology characteristics of foxtail millet grain development, two staining methods were used to make paraffin sections to observe the complete morphological changes of seeds from the ovule stage before pollination to the maturity stage. Cytological morphology of mature ovule before pollination is shown (Supplemental Fig. S2a, e), and morphology of mature ovule and mature pollen grains can be observed. Milk stage is the filling stage of foxtail millet grain, it is a process of endosperm starch accumulation and differentiation (Supplemental Fig. S2b, 2f; Fig. 3a2e2). Dough stage is the process when the accumulation of foxtail millet grains storage products has been completed, at this stage foxtail millet grain has formed the complete embryo (Supplemental Fig. S2c, g; Fig. 3a3e3). At the maturity stage the endosperm starch of the foxtail millet grains entered the maturation process, and the seed begins to lose water and become translucent (Supplemental Fig. S2d, 2h).

    • To explore the molecular mechanisms of grain development, transcriptome profiling was used to detect genes potentially involved in foxtail millet grain development. According to the stage of grain development, samples used for RNA-seq were collected at four developmental phases, namely, the ovule stage before pollination (ovule stage), the milk stage after pollination (milk stage), nutrient storage stage after pollination (dough stage), and the grain maturity stage (maturity stage) (Fig. 4a). In this study, we collected 12 samples for RNA-seq analysis, with three biological replicates for each stage (Supplemental Table S1). Clean reads of the 12 samples ranged from 44.88 to 50.60 million. 94.78% to 99.02% of the clean reads were mapped to the reference genome (http://foxtail-millet.biocloud.net/home), and 83.14% to 95.09% of the reads were uniquely mapped (Supplemental Table S2). These data indicate that the quantity and quality of reads were sufficient to perform quantitative analysis of gene expression.

      Figure 4. 

      Global gene expression profiling and KEGG pathway analysis. (a) Schematic overview of the experimental approach. RNA-seq was performed for four developmental stages. (b) Distribution of genes in four samples. (c) Hierarchical cluster analysis of genes expressed during four developmental stages. (d) KEGG pathway classification of the all expression gene in four developmental stages.

      To understand the relationships between different groups, we used principal component analysis (PCA) on the full data set, which can visually demonstrate transcriptional features and developmental similarities (Supplemental Fig. S3a). To evaluate the global gene expression profiles of different samples, gene expression levels were assessed using fragments per kilobase of transcript per million mapped reads (FPKM), based on normalized read counts. Across these four stages, a total of 29,530 genes were measured (Supplemental Data Set S1). Compared to the dough and milk stages, fewer genes were detected in the remaining stages, especially in the maturity stage (Supplemental Fig. S3b). The relationship between the gene number of the four samples was plotted as a Venn diagram, with 24,528 genes identified as co-expressed genes (Fig. 4b). Based on their expression levels, we classified these genes into six groups. No expression FPKM < 1, extremely low expression FPKM < 10, low expression FPKM ranging between 10 and 30, medium expression FPKM ranging between 30 and 100, high expression FPKM ranging between 100 and 300, and very high expression FPKM > 300 (Supplemental Fig. S3c; Supplemental Table S3). Thus, the results revealed that 21,866 expressed genes with FPKM values > 1 (Supplemental Data Set S2). Hierarchical cluster analysis and KEGG pathway classification were performed on all expressed genes (Fig. 4c, 4d).

    • The expression dynamics of TFs during grain development were investigated. At least one developmental stage, 509 TFs were detected, and they belonged to 42 families and other categories of TFs (Supplemental Data Set S3; Supplemental Table S4). The genes encoding putative TFs was obtained from the Plant Transcription Factor Database (http://planttfdb.gao-lab.org/index.php)[69]. One hundred and six differentially expressed TFs were identified, including MADS-box family, WRKY family, bHLH family, MYB family, AP2/ERF family, and NAC family. Based on their expression patterns at four developmental stages, the 106 TFs were classified into four groups (Fig. 5). Among these TFs, 38 exhibited significantly higher expression levels specifically during the ovule stage, compared to the remaining three stages. These TFs were assigned as ovule stage preferential genes (Fig. 5a). In the milk stage, 60 TFs exhibited differential expression, with significantly higher expression compared to the other three stages (Fig. 5b). Only two TFs displayed considerably higher expression at the dough stage compared to other developmental stages, and these were designated as dough stage predominantly expressed genes (Fig. 5c). There were six genes whose expression were significantly higher in the maturity stage, and these genes were designated as the main TFs expressed in this stage (Fig. 5d). WRKYs that regulate seed development have been reported[35,7072]. Our results showed that eight WRKYs TFs were detected during grain development. Among these, Si3g13930-WRKY53 was found to be enriched at the ovule stage, while six WRKYs TFs (Si3g27390-WRKY24, Si2g18500-WRKY30, Si5g26980-WRKY48, Si2g18510-WRKY70, Si1g06940-WRKY71 and Si3g17470-VQ4) were enriched at the milk stage. Additionally, Si5g14790-VQ4 was enriched at the dough stage. The MADS-box is a crucial regulator of various processes of plant reproductive development and play important roles in the control of seed development[73]. SiMADS34, an E-class MADS-box protein, played a crucial role in regulating foxtail millet grain yield[42]. Among the MADS-box genes, SiMADS17 and SiMADS2 showed high expression in foxtail millet seeds[74]. Our findings demonstrated the presence of 13 MADS-box genes, with seven genes (Si6g22290-MADS7, Si2g27130-MADS8, Si3g10210-MADS13, Si7g22110-MADS17, Si5g40600-MADS21, Si3g07900-MADS58 and Si1g07790-ZMM17) being enriched at the ovule stage, and six genes (Si9g47800-MADS1, Si5g40400-MADS2, Si4g06820-MADS5, Si4g26480-MADS16, Si5g30750-MADS32, Si9g09150-MADS34) being enriched at the milk stage. An important regulatory role in plant development is played by the bHLH[75]. Opaque11 (O11) and ZmZHOUPI (ZmZOU), encoding bHLH TFs, are activated after pollination and regulate several biological processes during endosperm development in maize[26,53]. ZmbHLH121 can bind to the G-box cis-element ABA response element (ABRE) to positively regulate maize grain size and weight[76]. An-1 and OsbHLH107, also regulate grain size in rice[77,78]. Our analysis results showed that 11 bHLH genes were detected during foxtail millet grain development, including five at the ovule stage (Si9g05580-bHLH89, Si2g26780/Si9g10860-bHLH49, Si7g29990-LHW, Si1g35350-bHLH82), five at the milk stage (Si2g03880-PIF4, Si9g09960-bHLH148, Si6g19250-BIM1, Si5g08020-bHLH168, Si5g41500-bHLH128), and only one at the maturity stage (Si3g08580-bHLH68). Opaque2 (O2) and Opaque3 (O3), an extensively studied bZIP TF, its functional studies has roles in regulating grain protein accumulation, amino acid and sugar metabolism, through the transcriptional regulation of certain zein genes[79,80]. Our results also revealed the presence of a small amount of the bZIP TFs Si1g10450-HBP-1a and Si1g33540-bZIP23.

      Figure 5. 

      Hierarchical clustering of TFs expression levels during four developmental stages. TFs predominantly expressed at (a) the ovule stage, (b) the milk stage, (c) the dough stage, and (d) the maturity stage, respectively.

    • KEGG pathway classification analysis was performed on TFs, revealing a significant enrichment of TFs in hormone signaling (Supplemental Fig. S4). Our results identified 221 genes involved in phytohormones signaling, including auxin, cytokinin (CTK), gibberellin (GA), abscisic acid (ABA), and brassinosteroid (BR) signaling pathways. To investigate the temporal expression patterns of phytohormone-related genes, we investigated gene expression profile clustering. Among the phytohormones, auxin plays crucial roles in ovule and seed development[81]. Therefore, our focused primarily on genes related to auxin signal transduction, including those participating in auxin biosynthesis, polar transport, and response (Fig. 6a). Plasma membrane influx and efflux transporters mediate polar auxin transport, including three transporters[82]. These include the influx carrier AUXIN1/LIKE-AUX1 family, the efflux carrier PIN-FORMED family, and the P-glycoprotein (PGP) subfamily of the ATP-binding cassette transporter family. The auxin signaling pathway during grain development in foxtail millet included eight PIN genes, four AUX genes, and nine PGP genes (Fig. 6b). Among these, six efflux carrier genes (SiPILS2, SiPIN1a, two SiPIN1b, SiPIN1c, SiPIN3a) were preeminently expressed at the ovule and milk stages, but their expression significantly down-regulated in the remaining samples. Whereas the expression levels of the following two efflux carrier genes (SiPIN5a, SiPIN5b) increased at the dough and maturity stages. The influx carrier genes SiLAXs (SiLAX1, SiLAX2x1, SiLAX2x2, SiLAX3) was found to be up-regulated at the milk stage. SiPGP4 (Si3g06120), SiPGP11, SiPGP19, SiPGP21 (Si3g15930) showed preferential expression at the ovule stage and significantly down-regulated at other stages of grain development. However, SiPGP1, SiPGP4 (Si5g28820), SiPGP11, SiPGP14, and SiPGP21 (Si5g04220) exhibited significantly higher expression at the milk stage than other stages. It is worth noting that the tryptophan-dependent pathway is the primary pathway responsible for the production of indole-3-acetic acid (IAA)[83]. Our transcriptome data showed that 22 IAAs genes, five IAAs genes were preferentially expressed at the ovule stage, while 16 IAAs genes showed the highest expression at the milk stage (Fig. 6c). Auxin response factor (ARF) and Small Auxin-Up RNA (SAUR) families are the most important family of auxin-responsive proteins, potentially playing an important role in the regulation of foxtail millet grain development and its response to abiotic stress[84,85]. Several ARFs and SAURs genes were found to be enriched at the ovule and/or milk stages during foxtail millet grain development (Fig. 6d, f). The Gretch Hagen 3 (GH3s) family not only plays an essential role in auxin signaling, but is also involved in the plant defense response systems[86,87]. Three GH3s genes were markedly down-regulated at the milk stage, while the expression of two GH3s genes gradually increased during foxtail millet grain development (Fig. 6e). These data suggest that these auxin signaling related genes are involved in the grain development. Cytokinin, an important phytohormone, plays a key role in regulating various physiological processes such as seed production, cell proliferation, and differentiation[88]. It also plays a significant role in the expression of several important genes during grain development. There genes include three histidine kinases (Si5g431500-HK3, Si9g11580-HK4, Si9g26610-HK5), four histidine-containing phosphor transfer proteins (Si6g24170-AHP1, Si2g14080-AHP2, Si5g31910-PHP5, Si3g17280-PHP5X1), and the corresponding 10 response regulators type-A/B (ARR) family members (Si7912360-ARR1, Si1g20230-ARR2, Si1g37750-ARR3, Si5g45190-ARR4, Si3g02790-ARR6, Si8g03520-ARR9, Si9g47340-ARR21, Si1g35420-ARR23, Si1g06860-ARR24, Si5g41810-ARR26) (Supplemental Fig. S5). GA is an important endogenous hormone that regulates multiple developmental processes in plants[89,90]. Research has shown that GAs are particularly important for ovule formation in early seed development and can also regulate embryo development in late seed development[91]. The GA receptor GID1 specifically recognizes DELLA protein and prevents it from being trans activated, forming the GA-GID1-DELLA complex, which can facilitate interaction with GID2[92]. GA signaling is a complex process and involves 12 GA oxidase genes responsible for GA biosynthesis, two DELLA proteins (Si9g50510-GA1 and Si4g27600-RGL2) and one GA receptor (Si3g23940-GID1) involved in GA signal transduction. Additionally, our transcriptome data revealed the presence of numerous TFs involved in GA signaling (Supplemental Fig. S6). ABA serves as the key regulator of seed dormancy and germination[93]. ABA signaling promotes PP2Cs binding to the PYR/PYL families of ABA receptors, dephosphorylates SnRK2 kinase activity, and activates SnRK2 to phosphorylate and autophosphorates downstream effector genes[94]. Our transcriptome data revealed the genes involved in these ABA signaling pathways were highly enriched (Supplemental Fig. S7). Specifically, we detected six ABA receptor PYR/PYL family genes, including one PYR gene (Si9g39880-PYR1) and five PYL genes (Si4g22990-PYL2, Si1g02490-PYL3, Si9g42890-PYL4, Si5g37100-PYL5, Si3g08200-PYL8). Furthermore, we identified eight protein phosphatase genes (Si9g33810-PP2C27, Si9g45010-PP2C30, Si3g16340-PP2C50, Si3g12370-PP2C1, Si5g21960-PP2C6, Si2g18270-PP2C4, Si7g02990-PP2C3.1, Si5g03330-PHS1), a protein kinase SnRK2 gene (Si3g36320-SnRK2.3), and Si3g24450-ABF2 serves as a binding factor of the ABA responsive element. BR are novel and potent plant hormones that have a wide range of biological activities, which can regulate flowering time, promote anthocyanin accumulation, cell division and elongation, stimulate vegetative growth, promote flowering and fruit ripening, and improve fruit quality and yield[95,96]. In foxtail millet, studies have shown that a LRR receptor-like kinase DROOPY LEAF1 (DPY1) can regulate leaf droopiness by interacting with SiBRI1-SiBAK1, thereby achieving the purpose of regulating plant architecture[97]. RING-type E3 ligase SiSGD1 gene can interact with E2 ubiquitin-conjugating enzyme SiUBC32 and ubiquitinates the BR signaling receptor SiBRI1, affecting foxtail millet grain size[43]. These data suggested potential involvement of these BR signaling genes in foxtail millet grain development. Altogether 72 genes related to BR signaling were detected, including a series of enzymes such as receptor kinase, protein phosphatase, protein kinase, casein kinase, hydrolase, and some TFs (Supplemental Fig. S8). Among these, four receptor kinases (Si9g12450-SERK2, Si5g30180-BRI1, Si8g01300-BRI1, Si6g05160-BAK1), eight protein kinases (Si9g43630-PBL7, Si4g16710-GSK4, Si1g03360-GSK4, Si5g15090-GSK1, Si9g03700BSK5, Si3g01910-BSK3, Si9g31580-BSK2, Si9g34630-ASK8), and three hydrolases (Si9g54960-XTH24, Si4g23660-XTH22, Si4g23650-TCH4) were identified.

      Figure 6. 

      Expression patterns of genes enriched in the auxin signaling during the four developmental stages. (a) Schematic of the auxin pathway in plants. (b) Polar auxin transporter gene expression patterns, (c) IAAs, (d) ARFs, (e) GH3s, and (f) SAURs respectively. AUX1: auxin influx carrier (AUX1 LAX family); TIR1: Transport inhibitor response 1; AUX/IAA: auxin-responsive protein IAA; ARFs: Auxin response factors; GH3: Indole-3-acetic acid-amido synthetase (auxin responsive GH3 gene family); SAUR: Small auxin up-regulated RNA (SAUR family protein). PINs: Auxin efflux carrier component; LAXs: Auxin transporter-like protein; PGPs: ABC transporter B family member; IAAs: Indole-3-acetic acid.

    • A critical process in grain filling is the conversion of sucrose to starch[98]. Starch is an important component of foxtail millet grains[99]. This study identified 164 genes related to starch and sucrose metabolism were identified (Fig. 7 ; Supplemental Fig. S9). The expression of SiSUS2 and SiSUS7 gradually increased and peaked at the milk stage, whereas the expression of SiSUS1 peaked at the dough stage and then gradually dropped. SiSUS4 expression increased progressively with a peak at the maturity stage. Four SiSPSs, one SiTPP, one SiAGPase, one SiGBSS, four SiSSs, two SiSBEs, one SiAMY, and one SiBMY gene were highly expressed at the ovule stage, and their expression gradually decreased. One SiSPS, three SiTPSs, six SiTPPs, one SiAGPase, one SiAMY, and one SiBMY gene were higher expressed at the milk stage. The highest at the dough stage was observed for three SiTPSs, one SiTPP, one SiAGPase, four SiSSs, one SiSBE, three SiAMYs and one SiBMY. At the maturity stage, they were highest expressed by one SiTPP, one SiGBSS, one SiSS, one SiSBE and one SiAMY. Increasing AGPase activity may improve the sink strength of developing grains and ultimately promote grain weight at the maturity stage.

      Figure 7. 

      Analysis of expression patterns of genes in the sucrose-starch metabolism pathway during four developmental stages. AGPase: ADP glucose pyrophosphorylase; GBSS: Granule-bound starch synthase; SSs: Starch synthase; SBE: Starch-branching enzyme; BMY: Beta-amylase; AMY: Alpha-amylase isozyme; SUS: Sucrose synthase; TPS: Alpha trehalose phosphate synthase; TPP: Trehalose phosphate phosphatase; SPS: Sucrose phosphate synthase; 1-SST: Sucrose 1-fructosyltransferase.

    • Encoded by several different genes and families, zein can be divided into α-, β-, γ-, and δ- types zein, which are the most important storage proteins[100,101]. In maize, α-zein is further divided into 19-kD (Z1A, Z1B, Z1D) and 22-kD α-zeins (Z1C), β-zein contains only a class of 14-kD β-zeins, whereas the γ-zein subfamily contains 16-kD, 27-kD, 50-kD zeins, and δ-zein includes 10-kD, 18-kD zeins[102]. The expression profiles of the 16 zein genes clearly showed that zeins were strongly induced during the dough and maturity stages of endosperm development, which is characteristic of storage product accumulation (Fig. 8 & Supplemental Table S5). Our study showed a total of seven 19-kd-α-zein genes, notably Si8G18820-PMS1-19-kDα exhibited the highest expression level of 100,000 during the maturity stage. There were also seven 22-kD α -zein genes, with three of them showing expression levels exceeding 10,000 in the dough and maturity stages. Furthermore, we observed the presence of one β-zein gene (Si3g11730-β-zein) and one γ-zein gene (Si2g22180-γ-zein). Surprisingly, all the detected 19-KD-α-zein genes in the transcriptome data were found to be located on chromosome 8 of foxtail millet, forming a tandem array in a single gene cluster (Fig. 8b). Therefore, we speculate that α-zein genes in foxtail millet may be arranged in tandem in one or more clusters within the genome, which play a role in regulating the accumulation of storage proteins. Amino acids are crucial components of proteins and serve as essential nutrients for humans. Foxtail millet grains are abundant in various amino acids, including essential amino acids needed by human, such as leucine, phenylalanine, isoleucine, tryptophan, methionine, and lysine[103]. Therefore, we further analyzed the expression of genes involved in amino acid synthesis and metabolism during foxtail millet grain development. These genes can be divided into those associated with the synthesis and metabolism of essential amino acids, non-essential amino acids synthesis and metabolism, and those involved in both pathways. We found that a majority of these genes were highly expressed during the initial three stages of grain development (Supplemental Fig. S10).

      Figure 8. 

      Analysis of zein family genes expression during four developmental stages. (a) Cluster Heat map of the FPKM values of 16 zein genes at the four developmental stages. (b) Dynamic transcript levels of 19-kDa-α-zein genes at four developmental stages. (c) Dynamic transcript levels of 22-kDa-α-zein at four developmental stages. (d) Dynamic transcript levels of β- and γ-zeins at four developmental stages.

    • The selection of yellow grain in the domestication of foxtail millet presents a striking aspect, especially when compared to major cereal crops such as wheat (Triticum aestivum) and rice (Oryza sativa). Grain colour and quality were important breeding traits for foxtail millet domestication. The yellow color of most foxtail millet landraces and varieties, is typically produced by the accumulation of carotenoids and flavonoids in the pericarp, aleurone, and endosperm[104]. This study identified 25 genes, including a bunch of enzymes, that are related to the carotenoid pathway (Fig. 9a). Schematic representation of the carotenoid pathway during foxtail millet grain development (Fig. 9b). Carotenoids play a critical role in maintaining good health, its biosynthesis requires many enzymes, phytoene synthase (PSY) is the most essential rate-limiting enzyme in the carotenoid pathway[105]. Most genes involved in carotenoid biosynthesis accumulated significantly during foxtail millet grain development. This indicating that the carotenoids began to accumulate at the milk stage, which aligned with previous findings, and the foxtail millet grains started to turn yellow after 8 DAP (Fig. 1b). One of the candidate genes, SiPSY1, which affects foxtail millet grain color and quality, was identified as a potential gene for foxtail millet color variation that can be utilized in breeding[104]. In this study, we detected three PSY genes (Si4g27520-PSY1, Si3g38930-PSY2, and Si2g30580-PSY3) in the carotenoid metabolic pathway of foxtail millet. Furthermore, we further analyzed the expression levels of genes involved in flavonoid pathways from ovule stage to maturity stage. In the flavonoid biosynthesis pathway, a total of 36 genes related this pathway were identified (Supplemental Fig. S11a). Among these genes, Si2g44100-C4H, Si8g14530-CHS, two CHI genes (Si7g30460-CHIL, Si9g04210-CHI-1), Si1g33590-FLS, Si9g38590-F3'5'H, and four ANR genes (Si7g25720, Si7g25730, Si7g25740, Si7g25760) showed the highest levels at the milk stage, and then gradually decreased. Si3g03560-F3H, Si9g23880-F3'H, Si5g25220-DFR, and Si1g01130-LAR showed higher levels at the ovule and dough stages, but lower levels at the milk and maturity stage (Supplemental Fig. S11). Folate metabolism process is conserved and requires the catalysis of a variety of specific enzymes[47]. This study detected most genes related to the folate metabolism pathway in foxtail millet (Supplemental Fig. S12a). SiADCS, SiDHPS1/2, SiDHFR1/2, two SiFPGS, SiDHNA2, SiDHN-P, and SiGGH expression levels gradually decreased, SiGCHI expression levels gradually increased, SiDHFS, SiDHNA1, SiDHN-P3 expression levels decreases and then increases during foxtail millet grain development (Supplemental Fig. S12b).

      Figure 9. 

      Hierarchical clustering analysis of carotenoid metabolic pathway genes at the four developmental stages. (a) Heat map analysis of carotenoid metabolic pathway genes in the four developmental stages. (b) Schematic diagram of carotenoid metabolic pathway. PSY: phytoene synthase; PDS: phytoene desaturase; Z-ISO: ζ-carotene isomerase; ZDS: ζ-carotene desaturase; CRTISO: carotenoid isomerase; LCYE: lycopene ε-cyclase; LCYB: lycopene β-cyclase; LCYE: lycopene ε-cyclase; BCH: β-carotene hydroxylase; ZEP: zeaxanthin epoxidase; VDE: violaxanthin de-epoxidase; NCED: 9-cis-epoxycarotenoid dioxygenase; CYP97A: cytochrome P450 carotene hydroxylase.

    • To ensure the reliability of our transcriptome data, we selected 14 key metabolic genes by RT-qPCR validation. Specifically, we analyzed the expression patterns of three zein genes (Si9g29850-AZS22-16, Si3g11730-β-zein, Si2g22180-γ-zein), the rate-limiting enzyme gene (Si2g30580-PSY3) of carotenoid biosynthesis, five auxin signaling pathway-related genes (Si5g13880-IAA2, Si6g23990-SAUR36, Si1g25820-SAUR36, Si5g43630-ARF4, Si9g21430-ARF22), five transcription factors (Si9g47800-MADS1, Si6g22290-MADS7, Si3g10210-MADS13, Si5g406000-MADS21, Si6g14500-ERF058), and a house-keeping gene (Seita.8G043100). The results of our RT-qPCR analysis demonstrated that the observed variation in gene expression levels aligned with those obtained from the RNA-seq data (Supplemental Fig. S13). This concurrence strongly suggests the reliability of our analytical findings.

    • Foxtail millet grain development is crucial for grain quality and yield, although previous studies have identified several genes that regulate starch biosynthesis, grain color, carotenoid, flavonoid, and folate metabolism[40,106,107]. However, detailed morphological analyses of foxtail millet grain development process are scarce and comprehensive description of transcriptome dynamics based on morphological analyses is lacking. Therefore, conducting a, detailed morphological analysis of the ovule before pollination and the grain development stages after pollination can provide detailed morphological support for the dynamic developmental process of foxtail millet grains (Fig. 1, Fig. 2; Supplemental Fig. S1). This study provides a valuable reference for studying the specific periods of ovule and grain development in foxtail millet. In addition to the morphological analysis of grain development of Jingu 21, the accumulation characteristics of main storage substances (protein and fat) during foxtail millet grain development were also analyzed (Fig. 3; Supplemental Fig. S2). On the basis of morphological analysis and storage material analysis of foxtail millet grain development, we have established a time-dynamic transcriptome of grain development in Jingu 21. This transcriptome provides valuable genetic resources for understanding the genetic control of grain development in foxtail millet. Additionally, we have analyzed the integrated dynamic transcriptome of TFs, plant hormone signaling, starch and sugar metabolism, zein related genes, carotenoid metabolism, flavonoid metabolism, and folate metabolism in grain development. This comprehensive analysis deepens our understanding of the changes in important metabolic pathways during foxtail millet grain development, and help us provide new insights for developing better high-yield cultivars.

      Seed development is regulated by numerous TFs, many of which have been extensively reported[108,109]. In our RNA-seq analysis, we aimed to identify regulatory networks that play important role in grain development. We have discovered 106 grain-specific TFs (Fig. 5). These results are similar with the previous study on the gene expression profile of foxtail millet grain filling process[44]. MADS-box were responsible for the regulation of the development of flower organs, embryos, endosperm, and seeds[73]. SiMADS17/33/34/37/46/52 were considered key regulatory factors for grain yield in foxtail millet, and they are highly expressed in seeds. Moreover, SiMADS2/12/26/28/60 were thought to be highly expressed in pericarp and mainly expressed at the filling stage[42,74]. Many SiMADS TFs were analyzed in our transcriptome data. Combined with previous reports and our research results, we have identified highly homologous genes from the transcriptome data, namely Si1g07790-ZMM17, Si4g06820-MADS5, Si4g26480-MADS16. SiMADS34 (Si9g09150-MADS34), which has been reported to regulate foxtail millet inflorescence structure, was also found to be highly up-regulated in the ovule and milk stages of grain development, but its role in the grain filling process is remains unknown. It is noteworthy that the four TFs Si9g47800-MADS1, Si6g22290-MADS7, Si3g10210-MADS13, Si5g406000-MADS21 exhibit different expression patterns during grain development. Further investigation is required to determine their functions and roles. In maize, OPAQUE11 is a central hub in endosperm development and nutrient metabolism[110,111]. Among the numerous TFs we have discovered, 11 belong to the bHLH family. It is noteworthy that only Si3g08580-bHLH68 shows differential expression at the maturity stage, however, no studies have been reported on its regulatory mechanism at the maturity stage of foxtail millet grain development. Our dynamic transcriptome analysis of TFs enables us to identify key regulators of seed development, and gain insights into the mechanisms underlying grain development.

      Phytohormones play an extremely crucial role in seed development, so far, numerous plant hormone-related genes have been identified[112,113]. BR has been found to promote foxtail millet grain development, it has been discovered that SiBZR1 and SiUBC32-SGD1-SiBRI1 of the BR signaling pathway, can regulate grain yield in foxtail millet and other gramineous crops[43,114]. These findings provide a new strategy for BR in breeding and improving important traits in foxtail millet. In our transcriptome data, we observed the presence of several enzymes and TFs within the BR signaling pathway (Supplemental Fig. S8). Four BZR1 genes were detected in our study, among which Si2g36710-SiBZR1 gene was consistent with the genetic results reported in previous studies. The Si5g14880-SiBZR1, Si1g03480-SiBZR1, Si3g37600-SiBZR1 genes were found to be homologous genes of SiBZR1. Further analysis indicated that Si1g03480-SiBZR1 and Si3g37600-SiBZR1 were strongly induced at maturity stage, while Si5g14880-SiBZR1 was expressed throughout grain development. Based on these findings, it can reasonably speculate that these SiBZR1 genes may play distinct regulatory roles in different stages of foxtail millet grain development. What's more, our transcriptome data indicated the preferential expression of auxin influx carriers, efflux carriers, and auxin-response factors in early foxtail millet grain development, specifically at the ovule and/or milk stages (Fig. 6). Genes associated with the auxin signaling pathway may be valuable resources for foxtail millet molecular breeding.

      Starch is the main component of seeds, and its synthesis is a complex process involving many enzymes[115]. The activity of key enzymes participating in converting sucrose to starch in seeds was shown to regulate grain filling[116,117]. In our study, we identified several genes associated with starch and sucrose metabolism (Fig. 7). These genes include SiSUS, SiAGPase, SiGBSS, SiSSs, SiSBE, SiAMY, and SiBMY, which are important for grain formation. It is important to note that the amylose content is the main factor affecting the nutritional and eating quality of foxtail millet[13]. Recessive mutations in the waxy (Wx) gene, encoding granule-binding starch synthetase (GBSS), an enzyme required for amylose synthesis, control the waxy characteristic of foxtail millet grain[118,119]. However, the waxy gene not only reduces amylose content but also affects the gelatinization, aging, adhesiveness, and physicochemical properties of starch[120]. In our study, we detected two genes in granule binding starch synthetase (Si4g02980-GBSS and Si2g12020-GBSS1B). Specifically, the expression levels of the Si4g02980-GBSS gene at the milk and maturity stages were found to be 300 times and 650 times higher, respectively, compared to earlier stages. This observation suggests that the GBSS gene is essential for seed filling and starch accumulation. This study also showed that the Wx gene is an important target for domesticating of foxtail millet to improve its nutritional quality. During foxtail millet grain development, the expression of zein genes was highest at the milk stage, which is consistent with characteristics of nutrient accumulation (Fig. 8). High levels of zein expression play a key role in endosperm hardness, but it contains almost no essential amino acids such as lysine and tryptophan[121]. In maize, zein gene transcription is regulated by O2, PBF, OHP1, OHP2, ZmMADS47, and others, among O2 having the greatest effect on α-zein gene expression[102,122]. Maize and sorghum storage protein genes exist in the form of gene clusters, which are tandem array by gene family members, and this arrangement leads to the accumulation of storage proteins at extremely high levels[123,124]. Our study identified a gene cluster consisting of seven 19-KD-α-zein genes tandemly arrayed. However, it remains unclear whether other genes are also involved in this gene cluster. These results indicate that the accumulation of foxtail millet storage proteins may also be regulated by complete tandem gene clusters. Therefore, to gain a better understanding the accumulation of foxtail millet storage proteins, it is crucial to consider the zein gene cluster as a whole. Zein, a seed storage protein that is specifically expressed at the endosperm filling stage, is a representative substance for studying storage proteins at the endosperm filling stage. Despite the importance of zeins, their regulatory mechanisms are still unclear, and further investigation is needed to elucidate the molecular mechanisms of zeins in the endosperm.

      Carotenoid and flavonoid, together as nutrients of foxtail millet, greatly impact its beige color, and the more carotenoids, the more yellow the beige color and the better the quality of foxtail millet[104]. SiPSY gene plays an important role in carotenoid accumulation in foxtail millet, it was found to be up-regulated, while the SiCCD1 gene was down-regulated, resulting in continuous carotenoid accumulation in foxtail millet[40,125]. Our data analysis revealed three SiPSY genes (SiPSY1/2/3), with two genes showing the highest expression at the milk stage, and another gene being most expressed at the dough stage. This suggests that the SiPSY gene is expressed at different stages of foxtail millet grain development and may have different regulatory roles. Understanding the key genes that control grain colour and carotenoid accumulation in foxtail millet is crucial for breeding high-quality varieties.

    • Overall, we conducted a detailed morphological analysis of foxtail millet grain development, defining the grain development process, and described the characteristics of the accumulation of storage products during grain development. Subsequently, we performed transcriptome analysis to elucidate the dynamic mechanisms of TFs, plant hormone signaling, starch and sugar metabolism, carotenoid metabolism, flavonoid biosynthesis, and folate metabolism pathways during grain development in foxtail millet. The expression of key genes involved in important metabolic pathways during foxtail millet grain development were investigated using RT-qPCR. It was found that some genes exhibited high expression levels at the early stages (ovule and milk stages), whereas others showed high expression levels at the later stages (dough and maturity stages). These findings suggest different genes play different roles in grain development. The results of this study provide valuable insights into the developmental process and molecular mechanism of foxtail millet grain, as well as establish a strong theoretical foundation for its improvement.

    • The authors confirm contribution to the paper as follows: study conception and design: Wang JG, Wang D; draft manuscript: Wang D, Wang JG, Du H, and Yang G; experiments performance and manuscript revision: Wang D, Su M, Hao JH, Li ZD, Dong S, Yuan X, Li X, Gao L, and Chu X. All authors have read and agreed to the published version of the manuscript.

    • All data generated or analyzed during this study are included in this published article and its supplementary information files.

      • This work was supported by National Natural Science Foundation (32200222 for Wang JG); High-level Talents Start-up Fund of Shanxi Agricultural University (J242198006 for Wang JG); Shanxi Province Outstanding Doctoral and Post-Doctoral Scholarship Award Foundation (SXBYKY2021055 for Wang JG) and Hou Ji Laboratory Foundation (202204010910001-32 for Wang JG); Shanxi Province Key R&D Program Project (2022ZDYF119 for Du H); Shanxi Province Outstanding Doctoral and Post-Doctoral Scholarship Award Foundation (SXBYKY2021059 for Yang G).

      • The authors declare that they have no conflict of interest.

      • Supplemental Fig. S1 Dynamics of foxtail millet grain development: from 1 to 30 days after pollination (DAP). (a) Changes in the grains area of during foxtail millet grains development; (b) Changes in the perimeter of during grains development; (c) The growth of the foxtail millet grains in the length and width; (d) The change in the length/width ratio during grains development. Values for (a) , (b) , (c) and (d) are means ± s.d. (from 1 to 15 days after pollination n = 40, from 16 to 30 days after pollination n = 20).
      • Supplemental Fig. S2 Cell morphological analysis of the developing grains in foxtail millet: Paraffin section analysis. (a) and (e) Morphology of mature ovule cells before pollination; (b) and (f) Milk stage of grain development at 6-11 DAP; (c) and (g) Dough stage of grain development at 12-22 DAP; (d) and (h) Maturity stage of grain development at 23-30 DAP. DAP: days after pollination. (a-d) stained with 0.1% Toluidine Blue-O to observe. (e-f) stained with Safranin O and Fast Green to observe. Scale bars =10 μm.
      • Supplemental Fig. S3 Analysis of different samples global gene expression. (a) The principal component analysis (PCA) of the RNA-seq data of the 12 samples; (b) Total number of genes expressed at each developmental stage; (c) Percentage of gene numbers in different categories according to their expression levels in each tissue, based on FPKM values; (d) Gene ontology (GO) enrichment analysis of the all expression gene in four developmental stages.
      • Supplemental Fig. S4 Hierarchical clustering analysis of the relative expression levels of transcription factors during grain development. (a) Hierarchical clustering analysis of the relative expression levels of transcription factors during grain development; (b) Gene ontology (GO) enrichment analysis of the TFs in foxtail millet grain development; (c) KEGG pathway classification top 20 of the TFs in foxtail millet grain development.
      • Supplemental Fig. S5 Clustering expression patterns of cytokinin signaling pathway related genes in foxtail millet grains development. (a) Diagram of the cytokinin signaling pathway; (b) Expression patterns of histidine kinase in the cytokinin signaling pathway; (c) Expression patterns of histidine containing phosphotransfer protein in the cytokinin signaling pathway; (d) Expression patterns of the response regulator in the cytokinin signaling pathway; (e-f) The dynamic transcript levels of cytokinin signaling pathway in the different development stages of the foxtail millet grain. HK4/CRE1: AHK4/CYTOKININ RESPONSE1; AHP1: Histidine-containing phosphotransfer protein 1; AHP2: Histidine-containing phosphotransfer protein 2; A-ARR: two-component response regulator ORR belong to ARR family Type-A subfamily; B-ARR: two-component response regulator ORR belong to ARR-B family.
      • Supplemental Fig. S6 Dynamic transcriptome analysis of genes associated with gibberellin signaling pathway. (a) Diagram of the gibberellin signaling pathway; (b) Hierarchical clustering analysis of the relative expression levels of gibberellin signaling pathway genes during grain development; (c-d) The dynamic transcript levels of gibberellin signaling pathway in different development stages. GID1/2: GA-insensitive dwarf 1/2; DELLA: DELLA protein; TFs: Transcription factors.
      • Supplemental Fig. S7 Dynamic transcriptome analysis of genes involved in abscisic acid signaling pathway of foxtail millet grains development. (a) Diagram of the abscisic acid signaling pathway; (b) Hierarchical clustering analysis of the relative expression levels of abscisic acid signaling pathway genes during grain development; (c-f) The dynamic transcript levels of abscisic acid signaling pathway DEGs in the different development stages of the foxtail millet grain. PYRs: Abscisic acid receptor PYR family; PYLs: Abscisic acid receptor PYL family; PP2C: Protein phosphatase 2 C; SnRK2: serine/threonine-protein kinase SnRK2; ABF: ABA responsive element binding factor.
      • Supplemental Fig. S8 Dynamic transcriptome analysis of genes involved in brassinosteroid signal transduction of foxtail millet grains development. (a) Diagram of the brassinosteroid signal transduction pathway; (b) Hierarchical clustering analysis of the relative expression levels of brassinosteroid signal transduction pathway genes during grain development; (c-e) The dynamic transcript levels of brassinosteroid signal transduction pathway in the different development stages of the foxtail millet garins. BAK1: Brassinosteroid insensitive 1-associated receptor kinase 1; BRI1: Brassinosteroid insensitive 1; BKI1: BRI1 kinase inhibitor 1; BSK: BR-signaling kinase; BSU1: serine/threonine-protein phosphatase BSU1; BIN2: Brassinosteroid insensitive 2; BZR1/2: Brassinosteroid resistant 1/2; TCH4: xyloglucosyl transferase TCH4; CYCD3: Cyclin D 3.
      • Supplemental Fig. S9 Heat map analysis of genes associated with sucrose-starch conversion during grain filling from Ovule stage to maturity stage.
      • Supplemental Fig. S10 Dynamic analysis of genes involved to amino acid metabolism pathways in foxtail millet grains development. (a) Dynamic analysis of genes involved in amino acid biosynthesis during foxtail millet grains development. These genes are divided into essential amino acid biosynthesis genes, non-essential amino acid biosynthesis genes and genes synthesized by both; (b) Dynamic analysis of genes involved in amino acid metabolism and degradation during the development of foxtail millet grains. These genes can be divide into essential amino acid metabolism genes, non-essential amino acid metabolism genes and genes involved in the common metabolism of both.
      • Supplemental Fig. S11 Dynamic analysis of the flavonoid biosynthesis pathway during foxtail millet grains development. (a) Hierarchical clustering analysis of the relative expression levels of the flavonoid biosynthesis pathway genes during grain development; (b) Flavonoid biosynthesis pathway during foxtail millet grain development. The rectangles represent the expression level of genes. C4H: Trans-cinnamate 4-monooxygenase; CHS: Chalcone synthase; CHI: Chalcone isomerase; F3H: Flavanone 3-hydroxylase; F3’H: flavonoid 3'-hydroxylase; F3’5’H: flavonoid 3',5'-hydroxylase; FLS: flavonol synthase; DFR: Dihydroflavonol 4-reductase; LAR: Leucoanthocyanidin dioxygenase; ANR: Anthocyanidin reductase.
      • Supplemental Fig. S12 The biosynthetic pathway and expression patterns of folate synthesis pathway genes during foxtail millet grain development. (a) Hierarchical clustering analysis of the relative expression levels of folate metabolism pathways genes during seed development; (b) Folate synthesis pathway during foxtail millet seed development. The rectangles represent the expression level of genes. ADC: Aminodeoxychorismate; pABA: para aminobenzoic acid; GTP: guanosine-triphosphate; DHN: dihydroneopterin; ADCS: aminodeoxychorismate synthase; ADCL: aminodeoxychorismate lyase; HPPK: hydroxymethyldihydropterin pyrophospho kinase; DHPS: dihydropteroate synthase; DHFS: dihydrofolate Synthetase; DHFR: dihydrofolate reductase; FPGS: folylpolyglutamate-synthase; GGH: gamma glutamyl hydrolase; GTPCHI: GTP cyclohydrolase I; DHNA: dihydroneopterine aldolase.
      • Supplemental Fig. S13 RT-qPCR was used for quantitatively verification of the key genes during grain development. (a-e) Relative expression levels of five key TFs (SiMADS-box21, SiMADS-box13, SiMADS-box7, SiMADS-box1, SiERF058) during grain development. (f-j) Relative expression levels of fix auxin signaling pathway related genes (SiIAA2, Si6g23990-SAUR36, Si1g25820-SAUR36, SiARF4, SiARF22) during grain development. (l-n) The relative expression levels of rate-limiting enzyme gene SiPSY3 and three zein genes.
      • Supplemental Table S1 Morphological analyses of foxtail millet grain.
      • Supplemental Table S2 Summary of RNA-Seq read mapping results.
      • Supplemental Table S3 The classification of gene expression levels from the four tissues in this study.
      • Supplemental Table S4 Multiple TFs involved in grain development in foxtai millet.
      • Supplemental Table S5 Zein gene in transcriptome data.
      • Supplemental Table S6 Primers used for RT-qPCR.
      • Supplemental Data Set S1 The total number of genes detected in the four RNA-seq samples.
      • Supplemental Data Set S2 All expressed genes with FPKM value >1 in the four RNA-seq samples.
      • Supplemental Data Set S3 Many transcription factors were detected, belonging to 42 transcription factor families and other types of transcription factors.
      • Copyright: © 2023 by the author(s). Published by Maximum Academic Press on behalf of Hainan Yazhou Bay Seed Laboratory. This article is an open access article distributed under Creative Commons Attribution License (CC BY 4.0), visit https://creativecommons.org/licenses/by/4.0/.
    Figure (9)  References (125)
  • About this article
    Cite this article
    Wang D, Su M, Hao JH, Li ZD, Dong S, et al. 2023. Dynamic transcriptome landscape of foxtail millet grain development. Seed Biology 2:19 doi: 10.48130/SeedBio-2023-0019
    Wang D, Su M, Hao JH, Li ZD, Dong S, et al. 2023. Dynamic transcriptome landscape of foxtail millet grain development. Seed Biology 2:19 doi: 10.48130/SeedBio-2023-0019

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return