Fig 1.
Seed and pod development of soybean.
(A) Seed phenological stages during the acquisition of seed longevity (2013 crop). (B) The relationship between seed age and phenological stages (2014 crop). Stage 9 corresponds to dry mature seeds.
Fig 2.
Physiological characterization of soybean seed maturation.
(A) Evolution of seed dry weight (back circle) and water content (blue square). Data are the means (± SE) of 3 to 5 replicates of 20 seeds. (B) Acquisition of germinability (●, black circle) and desiccation tolerance (blue circle), evaluated after fast drying to 10% moisture and longevity (red triangle) as assessed by P50 (time necessary to obtain a loss of viability of 50% during storage 35°C and 75% RH). Data are the means (± SE) of 4 replicates of 25 seeds. (C) Loss of seed germination during storage at 75% HR, 35°C. Data are the mean of 4 replicates of 25 seeds harvested at indicated phenological stages.
Table 1.
Mapping of single-end reads to the soybean genome.
Fig 3.
Correlation and principal component analyses of soybean transcriptomes during seed maturation.
(A) Pair-wise Pearson correlation coefficients were used to generate the heat map. The color scale indicates the degree of correlation (blue, low; yellow, high). (B) and (C) Principal component analysis performed using median centering of the transcriptomes of seed phenological stages. 7.2D, transcriptomes of rapidly dried seeds at stage 7.2. The letters a,b, correspond to biological replicates.
Fig 4.
Over-representation analysis of functional classes during seed maturation.
Functional classes and subclasses statistically affected are indicated according to Mapman ontology. Data were subjected to a Bin-wise Wilcoxon test and resulting p-values were adjusted according to Bonferroni. The scale bar indicates the z-score calculated from p-values (i.e. p-value of 0.05 represents a z-score of 1.96 after adjustment). Over-represented and under-represented classes are indicated respectively in blue and yellow.
Fig 5.
Changes in soluble sugar contents in axis and cotyledons during maturation.
Data are the average of triplicates (± SE) using 6 axis (A-C) and 6 cotyledon pairs (D-E) from the 2014 crop. The changes in RFO/Suc ratio are shown in panel B and E for axes and cotyledons, respectively. The increase in longevity (P50) is indicated as a grey area as a help to the eye. (A, D) Glc, glucose and Fru, fructose; (B, E) Suc, sucrose and RFO/Suc ratio; (C, F) Raf, Raffinose, Sta, Stachyose and Ver, Verbascose.
Fig 6.
Transcription factor co-expression network of soybean seed maturation.
(A) Gene co-expression network of seed maturation visualized using an organic layout in Cytoscape. Temporal analysis of nodes in the network was obtained by coloring each gene by its specific expression profile during seed development. Dark-green; high transcript levels at seed filling (stage7.1); green, light-green and yellow, transitory transcript levels maximum at stage 7.2, 7.3 and 8.1 respectively; light-orange, dark-orange and grey, transcript levels increasing during late seed development, being maximum at stage 8.2, 8.3 and stage 9. (B) Zoom on the gene module corresponding to late maturation. Nodes correlating with longevity (PCC>0.9) are colored in blue. Text labels indicate gene numbers.
Fig 7.
qPCR analysis of selected genes during seed maturation validates RNA-Seq data.
(A) Heat shock transcription factor A3 (HSFA3: Glyma.03g191100); (B) Heat shock transcription factor A6B (HSFA6B: Glyma.03g157300); (C) sHSP17a (Glyma.17g224900); (D) sHPS17b (Glyma.14g099900), (E) sHSP21 (Glyma.08g318900). Data (±SE) are the average of three biological replicates of 30 seeds.
Fig 8.
Heat map of most differentially expressed genes correlating with longevity.
Genes were retained when the log2 intensity > 4 between stage 7.2 and 9 and that correlate with longevity (PPC P50>0.85). Genes were log2 mean centered and colored from the lowest (green) to highest values (red).
Fig 9.
Venn diagrams identify transcripts correlating with longevity.
Venn diagrams comparing transcripts that are differentially expressed in immature seeds that are rapidly dried at stage 7.2 (7.2D) compared to dry, mature seeds (stage 9) and transcripts that are differentially expressed between freshly harvested seeds at stage 7.2 (7.2F) and rapidly dried seeds at stage 7.2 (7.2D). Genes were considered statistically different when the absolute ratio was at least two fold with a P(BH)<0.01. (A). Number of genes with higher transcript levels in 9 vs 7.2D. (B). Number of genes with lower transcript levels in 9 compared to 7.2 D.
Table 2.
GO enrichment analysis of the 139 differentially expressed transcripts that are significantly higher at stage 9 compared to 7.2 after drying and common between 7.2D/7.2F and 9/7.2D.
Table 3.
GO enrichment analysis of the differentially expressed transcripts that are significantly higher at stage 9 compared to 7.2 after drying and not induced upon drying between 7.2F and 7.2D.
Table 4.
GO enrichment analysis of the 1318 differentially expressed transcripts that are significantly lower at stage 9 compared to prematurely dried seeds at stage 7.2.